Mechanical and Chemical Properties: Graphene Science Handbook

Nanoscience & Technology Aliofkhazraei
Ali
Milne
Ozkan
GRAPHENE SCIENCE HANDBOOK
GRAPHENE
Mitura
Gervasoni
Mechanical and Chemical Properties
SCIENCE

“I am confident in the materials … The wide scope of information covered and the qualifications of the contributors

projects a positive image of the potential quality of the publication.”
—Albert V. Tamashausky, Asbury Carbons Inc.
“This book is a result of an impressive project to collect views from experts in every aspect of graphene science. All popular
topics in the research of this impressive material are covered. This is the best and most complete presentation that has been
published so far for the hottest material of our times. A must-have reference.”
—Ioannis Remediakis, Department of Materials Science and Technology, University of Crete, Greece
HANDBOOK
“…this set of volumes represents a complete handbook showing the state of the art of science and technology related with
graphene. This set of books is written by great specialists and competent experts. For someone who works in this field, this
set of volumes is an essential reference for the characterization and application of graphene.”
— Dr. Alex Axelevitch, Holon Institute of Technology (HIT) Mechanical and
An In-Depth Look at the Outstanding Properties of Graphene
The Graphene Science Handbook is a six-volume set that describes graphene’s special structural, electrical, and
Chemical Properties
chemical properties. The book considers how these properties can be used in different applications (including the
development of batteries, fuel cells, photovoltaic cells, and supercapacitors based on graphene) and produced on a
massive and global scale.
Volume four is dedicated to the mechanical and chemical properties of graphene and covers
• Mechanical properties using a continuum elastic model introduced to describe graphene’s elastic behavior
• Results of theoretical investigations of the mechanical properties of graphene structures
• Mechanical stabilities and properties of graphene under various strains
• Different types of graphene devices for biomolecule and gas sensing
• Printed graphene-based electrochemical sensor technology EDITED BY
• Various types of graphene-based electrochemical sensors
• The chemical vapor deposition of graphene on copper Mahmood Aliofkhazraei • Nasar Ali
• Strategies covering graphene modification
•
•
Graphene in solar cells, including transparent electrodes, active layers, and interface layer
Changes at the micrometric and nanometric scales, and more
William I. Milne • Cengiz S. Ozkan
Stanislaw Mitura • Juana L. Gervasoni
K20503
ISBN: 978-1-4665-9123-3
90000
9 781466 591233
K20503_COVER_Final.indd 1 3/17/16 5:43 PM

GRAPHENE
SCIENCE
HANDBOOK
Mechanical and
Chemical Properties
GRAPHENE
SCIENCE
HANDBOOK
Mechanical and
Chemical Properties
EDITED BY
Mahmood Aliofkhazraei • Nasar Ali
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2016 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works

Version Date: 20160330
International Standard Book Number-13: 978-1-4665-9124-0 (eBook - PDF)
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and
information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and
publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission
to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any
future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic,
mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or
retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact
the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides
licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment
has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation
without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
Contents
Preface.......................................................................................................................................................................................... ix
Editors........................................................................................................................................................................................... xi
Contributors................................................................................................................................................................................xiii
Section I Mechanical Properties
Chapter 1 Mechanical Properties of Graphene......................................................................................................................... 3

Tao Chen and Rebecca Cheung
Chapter 2 Effective Elastic Properties of a Novel Continuous Fuzzy Fiber-Reinforced Composite with
Wavy Carbon Nanotubes........................................................................................................................................ 17
M. C. Ray and S. I. Kundalwal
Chapter 3 Effects of Vacancies, Nitrogen Atoms, and sp3 Bonds on Mechanical Properties of Graphene
Using Molecular Dynamics Simulations............................................................................................................... 41
Akihiko Ito and Shingo Okamoto
Chapter 4 Mechanical Properties of Graphene Sheets........................................................................................................... 61

O. E. Glukhova
Chapter 5 Mechanical Stabilities and Properties of Graphene and Its Modification by BN Predicted from
First-Principles Calculations.................................................................................................................................. 79
Qing Peng and Suvranu De
Chapter 6 Some Mechanical Properties of Graphene and Their Role in Forming Polymer Nanocomposites...................... 93
Kausala Mylvaganam and Liangchi Zhang
Section II Chemical Fabrication/Properties

Chapter 7 Grain Boundaries in Chemical Vapor Deposition-Grown Graphene.................................................................. 107
László P. Biró and Philippe Lambin
Chapter 8 Graphene-Based Biological and Chemical Sensors............................................................................................. 127

Shirui Guo, Wei Wang, Cengiz S. Ozkan, and Mihrimah Ozkan
Chapter 9 Printed Graphene-Based Electrochemical Sensors...............................................................................................147

A. Tuantranont, A. Wisitsoraat, C. Sriprachuabwong, C. Karuwan, P. Pasakon, and D. Phokharatkul
Chapter 10 Chemical Routes to Graphene Quantum Dots: Photoluminescence Mechanism and Novel Applications......... 163
Shoujun Zhu, Xiaohuan Zhao, Yubin Song, Bo Li, Junhu Zhang, and Bai Yang
v
vi Contents
Chapter 11 Electrochemical Biosensors and Biofuel Cells Based on Graphene and Graphene Derivatives......................... 179
Jaroslav Filip, Tomas Bertok, and Jan Tkac
Chapter 12 Chemical Modification of Graphene.................................................................................................................... 207

Xiao-Rong Li, Jing-Juan Xu, and Hong-Yuan Chen
Chapter 13 Graphene Synthesis by Chemical Vapor Deposition on Copper.......................................................................... 225

Kemal Celebi, Ning Yang, Matthew T. Cole, Kenneth B. K. Teo, and Hyung Gyu Park
Chapter 14 Chemically Modified Graphene and Its Applications in Electrochemical Sensing............................................. 247
Raghu G. Kempegowda and Pandurangappa Malingappa
Chapter 15 Graphene: Electrochemical Exfoliation and Applications................................................................................... 269

Quang Duc Truong and Itaru Honma
Chapter 16 Modification of Graphene with Polymers via Addition Chemistry..................................................................... 295

Horacio J. Salavagione
Chapter 17 Molecular Theory of Graphene Chemical Modification.......................................................................................317

Elena F. Sheka
Section III Recent Progress
Chapter 18 Low-Cost and Simple Method for Graphene Synthesis....................................................................................... 343

Isaiah Owusu Gyan, Haoyu Zhu, and I. Francis Cheng
Chapter 19 Graphene-Based Solar Cells................................................................................................................................ 361

Tanvi Upreti, Vinay Gupta, and Suresh Chand
Chapter 20 Graphene Production from Chlorination of Metallocenes................................................................................... 387

P. González-García, E. Urones-Garrote, A. Gómez-Herrero, D. Ávila-Brande, and L. C. Otero-Díaz
Chapter 21 Chemical Modification of Graphene with Polymers............................................................................................ 405

Somayeh Mohamadi and Naser Sharifi-Sanjani
Chapter 22 Charge Carrier Mobility in Graphene: Strain and Screening Effects.................................................................. 435
Tariq M. G. Mohiuddin, Raheel Shah, and Ram N. Singh
Chapter 23 Graphene-Based Antibacterial Materials............................................................................................................. 447

Van Hoa Nguyen and Jae-Jin Shim
Chapter 24 Nanofluidics in Graphene-Based Material Systems............................................................................................. 465

Ling Liu and Lin Zhang
Contents vii
Chapter 25 Nanoporous Graphene Sheets for Gas Separation............................................................................................... 477

Andreas W. Hauser and Peter Schwerdtfeger
Chapter 26 Photorefractive Properties of Graphene-Based Organic Systems........................................................................ 491

N. V. Kamanina
Chapter 27 Applications of Graphene and Its Derivatives in Electrochemical Sensors and Comparison Study of
Graphene as a Modifier with Other Modifiers..................................................................................................... 499
Majid Arvand, Navid Ghodsi, and Tahereh M. Gholizadeh
Index...........................................................................................................................................................................................517
Preface
The theory behind “graphene” was first explored by the physi- can slide easily over each other. The distance between layers is
cist Philip Wallace in 1947. However, the name “graphene” 0.335 nm. Due to its unique structure and geometry, graphene
was not actually coined until 40 years later, where it was used possesses remarkable physical–chemical properties, including
to describe single sheets of graphite. Ultimately, Professor a high Young’s modulus, high fracture strength, excellent elec-
Geim’s group in Manchester (UK) was able to manufacture trical and thermal conductivity, high charge carrier mobility,
and see individual atomic layers of graphene in 2004. Since large specific surface area, and biocompatibility.
then, much more research has been carried out on the mate- These properties enable graphene to be considered as an
rial, and scientists have found that graphene has unique and ideal material for a broad range of applications, ranging from
extraordinary properties. Some say that it will literally change quantum physics, nanoelectronics, energy research, catalysis,
our lives in the twenty-first century. Not only is graphene the and engineering of nanocomposites and biomaterials. In this
thinnest possible material, but it is also about 200 times stron- context, graphene and its composites have emerged as a new
ger than steel and conducts electricity better than any other biomaterial, which provides exciting opportunities for the devel-
material at room temperature. This material has created huge opment of a broad range of applications, such as nanocarriers
interest in the electronics industry, and Konstantin Novoselov for drug delivery. The building block of graphene is completely
and Andre Geim were awarded the 2010 Nobel Prize in different from other graphite materials and three-dimensional
Physics for their groundbreaking experiments on graphene. geometric shapes of carbon, such as zero-dimensional spheri-
Graphene and its derivatives (such as graphene oxide) have cal fullerenes and one-dimensional carbon nanotubes.
the potential to be produced and used on a commercial scale, The second volume of this handbook is predominantly
and research has shown that corporate interest in the discov- about the nanostructure and atomic arrangement of graphene.
ery and exploitation of graphene has grown dramatically in The chapters in this volume focus on atomic arrangement
the leading countries in recent decades. In order to understand and defects, modified graphene, characterization of graphene
how this activity is unfolding in the graphene domain, publi- and its nanostructure, and also recent advances in graphene
cation counts have been plotted in Figure P.1. Research and nanostructures. The planar structure of graphene provides an
commercialization of graphene are both still at early stages, excellent opportunity to immobilize a large number of sub-
but policy in the United States as well as in other key coun- stances, including biomolecules and metals. Therefore, it is
tries is trying to foster the concurrent processes of research not surprising that graphene has generated great interest for
and commercialization in the nanotechnology domain. its nanosheets, which nowadays can serve as an excellent plat-
Graphene can be produced in a multitude of ways. Initially, form for antibacterial applications, cell culture, tissue engi-
Novoselov and Geim employed mechanical exfoliation by neering, and drug delivery.
using a Scotch tape technique to produce monolayers of the It is possible to produce composites reinforced with gra-
material. Liquid-phase exfoliation has also been utilized. phene on a commercial scale and low cost. In these composites,
Several bottom-up or synthesis techniques developed for gra- the existence of graphene leads to an increase in conductivity
phene include chemical vapor deposition, molecular beam and strength of various three-dimensional materials. In addi-
epitaxy, arc discharge, sublimation of silicon carbide, and epi- tion, it is possible to use cheaply manufactured graphene in
taxy on silicon carbide. these composites. For example, exfoliation of graphite is one
The first volume of this handbook concerns the fabrica- of the cheapest graphene production techniques. The behavior
tion methods of graphene. It is divided into four sections: (1) of many two-dimensional materials and their equivalent three-
fabrication methods and strategies, (2) chemical-based meth- dimensional forms are completely different. The origin of the
ods, (3) nonchemical methods, and (4) advances of fabrication aforementioned differences in the behavior of these materials
methods. is associated with the weak forces that hold a large number of
Carbon is the sixth most abundant element in nature and single layers together to create a bulk material. Graphene can
is an essential element of human life. It has different struc- be used in nanocomposites. Currently, researchers have been
tures called carbon allotropes. The most common crystalline able to produce several tough and light materials by adding
forms of carbon are graphite and diamond. Graphite is a three- small amounts of graphene to metals, polymers, and ceram-
dimensional allotrope of carbon with a layered structure in ics. The composite materials usually show better electrical
which tetravalent atoms of carbon are connected to three other conductivity characteristics compared with pure bulk materi-
carbon atoms by three covalent bonds and form a hexagonal als, and they are also more resistant against heat.
network structure. Each one of these aforementioned layers is The third volume describes graphene’s electrical and opti-
called a graphene layer or sheet. Each sheet is placed in par- cal properties and also focuses on nanocomposites and their
allel on other sheets. Hence, the fourth valence electron con- applications. The fourth volume relates to the mechanical
nects the sheets to each other via van der Waals bonding. The and chemical properties of graphene and cites recent devel-
covalent bond length is 0.142 nm. The bonds that are formed opments. The fifth volume presents other topics, such as size
by carbon atoms between layers are weak; therefore, the sheets effects in graphene, characterization, and applications based
ix
x Preface
16,000
14,000
12,000
Number of documents
10,000
8000
6000
4000
2000
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Year
FIGURE P.1 Number of documents published around graphene during recent years, extracted from Scopus search engine by searching
“graphene” in title + keywords + abstract.
on size-affected properties. In recent years, scientists have and biomedical devices. Although graphene was initially pro-
produced advanced composites using graphene, which are posed as an alternative to silicon, its initial applications have
excellent from the point of view of mechanical and thermal been in electronic inks and additives to resins and coatings.
properties. However, in some of these composites, high elec- We have identified six areas of emerging applications for gra-
trical conductivity only is desirable. For example, the Institute phene, including displays/screens, memory chips, biomedical
of Metal Research, Chinese Academy of Sciences (IMR, devices, batteries/fuel cells, coatings, inks, and materials. In
CAS) has created a polymer matrix composite reinforced the investigation of the corporate engagement in graphene, we
with graphene, which has a high electrical conductivity. In sought to understand early corporate activity patterns related
this composite, a flexible network of graphene has been added to broader research and invention trends. In traditional inno-
to a polydimethylsiloxane matrix (of the silicon family). vation models, a lag between research publication and patent-
Investigation of early corporate trajectories for graphene ing is consistent with the linear model. However, more recent
has led to three major observations. First, the discovery-to- innovation models are stressing concurrent launch, open inno-
application cycle for graphene seems to be accelerated, for vation, and strategic property management.
example, compared to fullerene. Even though the discovery The sixth volume of this handbook is about the applica-
of graphene is relatively new, large and small firms have con- tion and industrialization of graphene, starting with chapters
tributed to an upsurge in early corporate activities. Second, a about biomaterials and continues onto nanocomposites, elec-
rapid globalization has occurred by companies in the United trical/sensor devices, and also new and novel applications.
States, Europe, Japan, South Korea, and other developed The editorial team would like to thank all contributors
economies, which were involved in early graphene activities. for their excellent chapters contributed to the creation of this
Chinese companies are currently starting to enter the gra- handbook and for their hard work and patience during its
phene domain, resulting in the expansion of research capabil- preparation and production. We sincerely hope that the pub-
ity of nanotechnology. Nevertheless, science alone does not lication of this handbook will help people, especially those
guarantee commercial exploitation. To clarify the issue, the working with graphene, and benefit them from the knowledge
level of corporate patenting in the United Kingdom, which is contained in the published chapters.
a pioneer in graphene research, is slightly ahead of Canada
and Germany; however, it is dramatically lower than in the Summer 2015
United States, Japan, and South Korea. Third, the potential Mahmood Aliofkhazraei
applications of graphene are rapidly expanding. Corporate Nasar Ali
patenting trends are indicative of their enthusiasm to utilize William I. Milne
the features of graphene in various areas, including transis- Cengiz S. Ozkan
tors, electronic memory and circuits, capacitors, displays, Stanislaw Mitura
solar cells, batteries, coatings, advanced materials, sensors, Juana L. Gervasoni
Editors
Mahmood Aliofkhazraei is an assistant professor in the University, Seoul and a high-end foreign expert for the
Materials Engineering Department at Tarbiat Modares Changchun University of Science and Technology in China.
University. Dr. Aliofkhazraei’s research interests include nan- In 2015, he was elected to an Erskine Fellowship to visit the
otechnology and its use in surface and corrosion science. One University of Canterbury, New Zealand. His research inter-
of his main interests is plasma electrolysis, and he has pub- ests include large area s ilicon- and carbon-based electronics,
lished more than 40 papers and a book in this area. Overall thin film materials, and, most recently, MEMS and carbon
he has published more than 12 books and 90 journal articles. nanotubes, graphene, and other 1-D and 2-D structures for
He has delivered invited talks, including keynote addresses electronic applications, especially for field emission. He has
in several countries. Aliofkhazraei has received numerous published/presented approximately 800 papers, of which
awards, including the Khwarizmi award, IMES medal, INIC around 200 were invited/keynote/ plenary talks—his “h”
award, best-thesis award (multiple times), best-book award index is currently 57 (Web of Science).
(multiple times), and the best young nanotechnologist award
of Iran (twice). He is on the advisory editorial board of several Cengiz S. Ozkan has been a professor of mechanical engi-
materials science and nanotechnology journals. neering and materials science at the University of California,
Riverside, since 2009. He was an associate professor from
Nasar Ali is a visiting professor at Meliksah University in 2006 to 2009 and an assistant professor from 2001 to 2006.
Turkey. Earlier he held the post of chief scientific officer at Between 2000 and 2001 he was a consulting professor at
CNC Coatings Company based in Rochdale, UK. Prior to Stanford University. He earned a PhD in materials science
this Dr. Ali was a faculty member (assistant professor) at the and engineering at Stanford University in 1997. Dr. Ozkan’s
University of Aveiro in Portugal where he founded and led areas of expertise include nanomaterials for energy storage;
the Surface Engineering and Nanotechnology group. Dr. Ali synthesis/processing including graphene, III–V, and II–VI
has extensive research experience in hard carbon-coating materials; novel battery and supercapacitor architectures;
materials, including nanosized diamond coatings and CNTs nanoelectronics; biochemical sensors; and nanopatterning
deposited using CVD methods. He has over 120 interna- for beyond CMOS (complementary metal-oxide semiconduc-
tional refereed research publications, including a number of tor). He organized and chaired 20 scientific and international
book chapters. Dr. Ali serves on a number of committees for conferences. He has written more than 200 technical publica-
international conferences based on nanomaterials, thin films, tions, including journal papers, conference proceedings, and
and emerging technologies (nanotechnology), and he chairs book chapters. He holds over 50 patent disclosures, has given
the highly successful NANOSMAT congress. He served as more than 100 presentations worldwide, and is the recipient of
the fellow of the Institute of Nanotechnology for 2 years on more than 30 honors and awards. His important contributions
invitation. Dr. Ali has authored and edited several books on include growth of hierarchical three-dimensional graphene
surface coatings, thin films, and nanotechnology for lead- nanostructures; development of a unique high-throughput
ing publishers, and he was also the founder of the Journal metrology method for large-area CVD-grown graphene
of Nano Research. Dr. Ali was the recipient of the Bunshah sheets; doping and functionalization of CVD-grown and
prize for presenting his work on time-modulated CVD at the pristine graphene layers; study of digital data transmission
ICMCTF-2002 Conference in San Diego, California. in graphene and InSb materials; memory devices based on
inorganic/organic nanocomposites, novel lithium-ion batter-
William I. Milne, FREng, FIET, FIMMM, was head of ies based on nano-silicon from beach sand and silicon diox-
electrical engineering at Cambridge University from 1999 ide nanotubes; fast-charging lithium-ion batteries based on
until 2014 and has been director of the Centre for Advanced silicon-decorated three-dimensional nano-carbon architec-

Photonics and Electronics (CAPE) since 2004. He earned a tures; and high-performance supercapacitors based on three-
BSc at St. Andrews University in Scotland in 1970 and later dimensional graphene foam architectures.
earned a PhD in electronic materials at the Imperial College
London. In 2003 he was awarded a DEng (honoris causa) Stanislaw Mitura has been a professor in biomedical engi-
by the University of Waterloo, Canada, and he was elected neering at Koszalin University of Technology from 2011. He is
as Fellow of the Royal Academy of Engineering in 2006. a visiting professor at the Technical University (TU) of Liberec
He received the JJ Thomson medal from the Institution of and was awarded a doctor honoris causa from TU Liberec.
Engineering and Technology in 2008 for achievement in elec- He was professor of materials science at Lodz University of
tronics and the NANOSMAT prize in 2010. He is a distin- Technology from 2001 to 2014. He earned an MSc in phys-
guished visiting professor at Tokyo Institute of Technology, ics at the University of Lodz in 1974, a PhD in mechanical
Japan, and a distinguished visiting professor at Southeast engineering at the Lodz University of Technology (1985),
University in Nanjing, China, and at Shizuoka University, and a DSc in materials science at the Warsaw University of
Japan. He is also a distinguished visiting scholar at KyungHee Technology in 1993. Professor Mitura’s most prominent
xi
xii Editors
cognitive achievements comprise the following: from the con- scientific research involves the interactions of atomic particles
cept of nucleation of diamond powder particles to the syn- of matter, electronic excitations in solids, surfaces, and nano-
thesis of nanocrystalline diamond coatings (NDC); discovery systems, the absorption of hydrogen in metals, and study of
of diamond bioactivity; a concept of the gradient transition new materials under irradiation. Gervasoni is a researcher
from carbide forming metal to diamond film; and technology at the National Atomic Energy Commission of Argentina
development of nanocrystalline diamond coatings for medi- and the National Council of Scientific and Technological
cal purposes. Professor Mitura has published over 200 peer- Research (CONICET, Argentina). She teaches at the Instituto
reviewed articles, communications, and proceedings, over 50 Balseiro and is involved in directing graduate students and
invited talks, and contributed to 7 books and proceedings, postdoctorates. She has published over 100 articles in interna-
including Nanotechnology for Materials Science (Pergamon, tional journals, some of which have a high impact factor, and
Elsevier, 2000) and Nanodiam (PWN, 2006). He organized she has attended many international conferences. Gervasoni
and co-organized several conferences focused on materials has been a member of the Executive Committee and/or the
science and engineering, especially diamond synthesis under International Scientific Advisory Board of the International
reduced pressure. He is an elected member of the Academy of Conference on Surfaces Coatings and Nanostructured
Engineering in Poland, guest editor in few international jour- Materials (Nanosmat) since 2010, Latin American Conference
nals, including Journal of Nanoscience and Nanotechnology, on Hydrogen and Sustainable Energy Sources (Hyfusen), and
Journal of Superhards Materials and also a member of the the International Conference on Clean Energy (International
editorial boards of several journals and an elected Fellow of Conference on Clean Energy, ICCE-2010) and guest editor
various foreign scientific societies. of the International Journal of Hydrogen Energy (Elsevier).
Recently she has focused her research on the study of hydro-
Juana L. Gervasoni earned her doctorate in physics at the gen storage in carbon nanotubes. Along with her academic and
Instituto Balseiro, Bariloche, Argentina, in 1992. She has been research work, Gervasoni is heavily involved in gender issues
head of the Department of Metal Materials and Nanostructures, in the scientific community, especially in Argentina and Latin
Applied Research of Centro Atomico Bariloche (CAB), America. She is member of the Third World Organization
National Atomic Energy Commission (CNEA), since 2012. for Women in Science (TWOWS), branch of the Third World
She has been member of the Coordinating Committee of the Academy of Science (TWAS), Trieste, Italy, since 2010, as
CNEA Controlled Fusion Program since 2013. Her area of well as of Women in Nuclear (WiN) since 2013.
Contributors
Majid Arvand Matthew T. Cole Andreas W. Hauser
Department of Chemistry Department of Engineering Department of Chemical and
University of Guilan University of Cambridge Biomolecular Engineering
Rasht, Iran Cambridge, United Kingdom University of California, Berkeley
Berkeley, California
D. Ávila-Brande Suvranu De
Department of Inorganic Chemistry Department of Mechanical, Aerospace Itaru Honma
Complutense University of Madrid and Nuclear Engineering Institute of Multidisciplinary Research
Madrid, Spain Rensselaer Polytechnic Institute for Advanced Materials
Troy, New York Tohoku University
Sendai, Japan
Tomas Bertok Jaroslav Filip
Slovak Academy of Sciences Slovak Academy of Sciences Akihiko Ito
Bratislava, Slovakia Bratislava, Slovakia Composite Materials Research
Laboratories
László P. Biró Navid Ghodsi Masaki-cho, Japan
Institute of Technical Physics and Department of Chemistry
Materials Science University of Guilan N. V. Kamanina
Centre for Energy Research Hungarian Rasht, Iran Vavilov State Optical Institute
Academy of Sciences Saint-Petersburg Electrotechnical
Budapest, Hungary Tahereh M. Gholizadeh University (“LETI”)
Department of Chemistry St. Petersburg, Russia
Kemal Celebi University of Guilan
Department of Mechanical and Process Rasht, Iran C. Karuwan
Engineering National Electronics and Computer
O. E. Glukhova
Zürich, Switzerland Technology Center
Saratov State University National Sciences and Technology
Saratov, Russia
Suresh Chand Development Agency
Organic and Hybrid Solar Pathumthani, Thailand
A. Gómez-Herrero
Cell Group Centro Nacional de Microscopía
New Delhi, India Raghu G. Kempegowda
Electrónica
Department of Chemistry
Universidad Complutense
Bangalore University
Hong-Yuan Chen Madrid, Spain
Karnataka, India
State Key Laboratory of Analytical
Chemistry for Life Science P. González-García
Nanjing University Dirección de Investigación y Posgrado S. I. Kundalwal
Centro de Ingeniería y Desarrollo Department of Mechanical
Nanjing, China
Industrial Engineering
Querétaro, México Indian Institute of Technology
Tao Chen Kharagpur
Scottish Microelectronics Centre West Bengal, India
Shirui Guo
University of Edinburgh
Physical and Life Sciences
Edinburgh, United Kingdom
Directorate Philippe Lambin
Livermore, California Department of Physics of Matter and
I. Francis Cheng Radiations
Department of Chemistry Vinay Gupta University of Namur
University of Idaho Organic and Hybrid Solar Cell Group Namur, Belgium
Moscow, Idaho New Delhi, India
Bo Li
Rebecca Cheung Isaiah Owusu Gyan State Key Laboratory of Supramolecular
Scottish Microelectronics Centre Department of Chemistry Structure and Materials
University of Edinburgh University of Idaho Jilin University
Edinburgh, United Kingdom Moscow, Idaho Changchun, China
xiii
xiv Contributors
Xiao-Rong Li Mihrimah Ozkan Raheel Shah

State Key Laboratory of Analytical Material Sciences and Engineering Sultan Qaboos University
Chemistry for Life Science Program Muscat, Oman
Nanjing University and
Nanjing, China Department of Electrical Engineering Naser Sharifi-Sanjani
and School of Chemistry
Ling Liu Department of Chemistry University of Tehran
Department of Mechanical and University of California, Riverside Tehran, Iran
Aerospace Engineering Riverside, California
Utah State University
Hyung Gyu Park Elena F. Sheka
Logan, Utah
Department of Mechanical and Process Department of Theoretical Physics and
Engineering Mechanics
Pandurangappa Malingappa
Swiss Federal Institute of Technology People’s Friendship University of
Department of Chemistry
(ETH Zurich) Russia
Bangalore University
Zürich, Switzerland Moscow, Russia
Karnataka, India
P. Pasakon Jae-Jin Shim

Somayeh Mohamadi
National Electronics and Computer School of Chemical Engineering
School of Environmental Engineering
Technology Center Yeungnam University
University of Environment
National Sciences and Technology Gyeongsan, Korea
Karaj, Iran
Development Agency
Pathumthani, Thailand Ram N. Singh
Tariq M. G. Mohiuddin
Sultan Qaboos University Sultan Qaboos University
Qing Peng Muscat, Oman
Muscat, Oman
Department of Mechanical, Aerospace
and Nuclear Engineering
Kausala Mylvaganam Yubin Song
Rensselaer Polytechnic Institute
School of Mechanical and State Key Laboratory of
Troy, New York
Manufacturing Engineering Supramolecular Structure and
University of New South Wales Materials
D. Phokharatkul
New South Wales, Australia Jilin University
National Electronics and Computer
Changchun, China
Technology Center
Van Hoa Nguyen National Sciences and Technology
Department of Chemistry Development Agency C. Sriprachuabwong
Nha Trang University Pathumthani, Thailand National Electronics and Computer
Hanoi, Vietnam Technology Center
M. C. Ray National Sciences and Technology
Shingo Okamoto Department of Mechanical Engineering Development Agency
Graduate School of Science and Indian Institute of Technology Pathumthani, Thailand
Engineering West Bengal, India
Ehime University Kenneth B. K. Teo
Matsuyama, Japan Horacio J. Salavagione Aixtron Ltd.
Department of Polymer Physics, Cambridge, United Kingdom
L. C. Otero-Díaz Elastomers, and Energy
Department of Inorganic Chemistry Applications
Complutense University of Madrid Institute of Polymer Science and Jan Tkac
Madrid, Spain Technology (CSIS) Institute of Chemistry
Madrid, Spain Slovak Academy of Sciences
Cengiz S. Ozkan Bratislava, Slovakia
Material Sciences and Engineering Peter Schwerdtfeger
Program Centre for Theoretical Chemistry and Quang Duc Truong
and Physics (CTCP) Institute of Multidisciplinary Research
Department of Mechanical Engineering Massey University (Auckland for Advanced Materials
University of California, Riverside Campus) Tohoku University
Riverside, California Auckland, New Zealand Sendai, Japan
Contributors xv
A. Tuantranont Jing-Juan Xu Liangchi Zhang

National Electronics and Computer State Key Laboratory of Analytical School of Mechanical and
Technology Center Chemistry for Life Science Manufacturing Engineering
National Sciences and Technology Nanjing University University of New South Wales
Development Agency Nanjing, China New South Wales, Australia
Pathumthani, Thailand
Lin Zhang
Tanvi Upreti Bai Yang Department of Mechanical and
Organic and Hybrid Solar Cell State Key Laboratory of Aerospace Engineering
Group Supramolecular Structure Utah State University
New Delhi, India and Materials Logan, Utah
Jilin University
E. Urones-Garrote Changchun, China Xiaohuan Zhao
National Center for Microscopy State Key Laboratory of Supramolecular
Electronics Ning Yang Structure and Materials
Complutense University of Madrid Department of Mechanical and Process Jilin University
Madrid, Spain Engineering Changchun, China
Swiss Federal Institute of Technology
Wei Wang (ETH Zurich) Haoyu Zhu
Material Sciences and Engineering Zürich, Switzerland Department of Chemistry
Program University of Idaho
University of California, Riverside Moscow, Idaho
Riverside, California Junhu Zhang
State Key Laboratory of Shoujun Zhu
A. Wisitsoraat Supramolecular Structure State Key Laboratory of Supramolecular
National Electronics and Computer and Materials Structure and Materials
Technology Center Jilin University Jilin University
National Sciences and Technology Changchun, China Changchun, China
Development Agency
Pathumthani, Thailand
Section I
Mechanical Properties
1 Mechanical Properties of Graphene
Tao Chen and Rebecca Cheung
CONTENTS
Abstract.......................................................................................................................................................................................... 3
1.1 Introduction.......................................................................................................................................................................... 3
1.2 Mechanical Parameters........................................................................................................................................................ 4
1.2.1 Valence Force Model................................................................................................................................................ 4
1.2.2 Continuum Elastic Model......................................................................................................................................... 4
1.2.3 Nonlinearity and Ideal Strength............................................................................................................................... 5
1.2.4 First-Principle Calculation........................................................................................................................................ 5
1.2.5 Molecular Dynamics Simulation.............................................................................................................................. 6
1.2.6 Discussion of Calculated Results.............................................................................................................................. 7
1.3 Measurement Results............................................................................................................................................................ 7
1.4 Graphene Resonators............................................................................................................................................................ 8
1.4.1 Overview................................................................................................................................................................... 8
1.4.2 Quality Factor........................................................................................................................................................... 9
1.4.3 Dissipation Mechanism.......................................................................................................................................... 10
1.4.4 Dynamic Range and Sensitivity............................................................................................................................. 11
1.5 Graphene Derivatives......................................................................................................................................................... 12
1.6 Conclusion.......................................................................................................................................................................... 12
References.................................................................................................................................................................................... 12
ABSTRACT and graphene originates from their distinctive electrical prop-

erties [1]. Unlike bulk graphite, monolayer graphene’s valence
Graphene possesses features of excellent mechanical proper- band and conduction band touch upon each other at the Dirac
ties such as unprecedented highly versatile Young’s modu- point [3–5]. The electron energy E and momentum K follows
lus and tensile strength. These properties not only endow a linear relationship around the Dirac points; thus, in undoped
graphene with robustness to withstand mechanical abrasion graphene where the Fermi level coincides with the Dirac
but they also make graphene an ideal candidate material for point, the electrons and holes behave like massless particles.
microelectromechanical/nanoelectromechanical devices. In Such massless particles are transported at a velocity that is
this chapter, we discuss the fundamental mechanical param- comparable to that of light, leading to extremely high carrier
eters of graphene on the basis of a continuum elastic model, mobility [6–9]. When the number of layers increases, the band
Young’s modulus, quality factors, and others. Initially, the structure of graphene will become more and more like bulk
discussion will concentrate on the mechanical properties of graphite. Although the concept of graphene has been previ-
pristine graphene, and then extend to cover the properties ously known to scientists [3,10], it is only after Andre Geim
of graphene derivatives. It is aimed at providing the readers and Konstantin Novoselov first isolated monolayer graphene
with a comprehensive index to important works related to and reported its superb electrical transportation that intensive
graphene’s mechanical properties. efforts in the research of graphene or analogous materials
have been generated [11–14].
Contrasting with the transition of electrical properties, the
1.1 INTRODUCTION
mechanical properties of graphene resemble those of bulk
Academic communities involved in the study of graphite have graphite. The honeycomb 2D lattice of graphene formed by
been aware of the concept of graphene since the 1960s when sp2 carbon–carbon bonds leads to extremely high in-plane
chemist Hanns-Peter Boehm coined the term “graphene” stiffness which has been one of the most significant proper-
from “graphite” and added the suffix “ene.” In the specific ties that graphene possesses. The phenomenal in-plane stiff-
and strict senses, “graphene” only refers to an individual layer ness of the graphene/graphite lattice structure has already
of graphite’s planar honeycomb carbon lattice. However, the found widespread use in various applications. For instance,
term “graphene” has also been adopted for up to 10 layers of carbon fiber, which contains graphene flakes of a few microns
carbon sheets by convention [1,2]. The bifurcation of graphite in size, has been widely used in the aerospace industry as well
3
4 Graphene Science Handbook
as in the military for its ultrastrength and light weight. As (a) → (b) →
A B a1 b1
the large growth in volume and the processing of graphene
mature [15,16], it is rapidly emerging as a novel material for
K
future microelectromechanical/nanoelectromechanical sys-
tem (MEMS/NEMS) devices possessing great advantages → M Γ
y a2
[17–19]. First of all, its high Young’s modulus means that gra-
phene resonators could achieve higher frequencies compared →
b2
with resonators made of other materials with the same dimen- x
sions. Moreover, a higher Young’s modulus and nanoscale
dimension could boost the sensitivity of sensors based on FIGURE 1.1 Lattice structure of graphene. The two inequivalent
graphene resonators, as will be discussed later in this chapter. carbon atoms are marked with different colors. Dotted rectangle
Graphene resonators can be used as force, mass, and charge encloses a super-cell. (Reprinted with permission from F. Liu,
sensors with ultrasensitive response [20–23]. P. Ming, and J. Li, Ab initio calculation of ideal strength and phonon
There has not been an encyclopedia type of review article instability of graphene under tension, Phys. Rev. B, vol. 76, no. 6,
that encloses all aspects of the mechanical properties of gra- p. 064120, 2007. Copyright 2007 by the American Physical Society.)
phene, which has encouraged us to devote time and effort
to write this chapter so as to meet the needs of researchers system’s energy provides a convenient, and sometimes inevi-
working in this area. In this chapter, the basic mechanical table, viewpoint. To describe the potential energy in interac-
properties of graphene will be discussed. Section 1.2 will tion among carbon atoms, a valence force field model can be
present an isotropic continuum elastic model with basic adapted from the model of the bulk crystal structure. The
mechanical parameters for describing the mechanical prop- sp2 potential energy Usp2 of graphene can be expressed in the
erties of graphene, which is intended to be a comprehen- atomic scale as [26,27]
sive base for the following discussions. Theoretical work,
Np
including density functional theory (DFT) calculation, and α
∑∑ 4a (r ⋅ r − a )
1 2
molecular dynamics (MD) simulation of the mechanical U sp2 = 2 ij ij
2
0
2 0
parameters will also be discussed here. Section 1.3 reviews i =1 j
the experimental measurements of graphene’s Young’s mod- Np 2 Np

β  a2 
ulus and ideal strength by atomic force microscope (AFM)
nanoindentation to provide a comparison with theoretical
+ ∑∑ j<k
a0 
r ⋅ rjk + 0  +
2  ij
2 ∑∑γD ⋅ D
i =1 j < k
i i (1.1)
calculations. In Section 1.4, we will look at graphene reso- i =1
nators. The quality factors of graphene resonators will be
summarized and compared with resonators made from other Here, i indexes individual carbon atoms whose nearest
materials. The energy dissipation mechanism in graphene atoms are indexed by j and k. Np denotes the number of primi-
bridges will also be touched upon. Finally, the mechanical tive cells in the graphene sheet under consideration. rij rep-
properties of some graphene derivatives will be discussed in resents the bond vectors pointing from one atom toward its
Section 1.5. nearest atoms. Di is the dangling bond vector perpendicular
to the local tangential plane. α, β, and γ are constants. For gra-
phene, their values are 155.9, 25.5, and 7.4 J/m2, respectively
1.2 MECHANICAL PARAMETERS [28]. The first term on the right-hand side of the equation is
the deformation energy caused by the uniaxial length change
1.2.1 Valence Force Model of the bonds. The second term denotes energy associated with
In this section, we introduce a simple continuum elastic model the in-plane angle change between adjacent bonds. The last
for describing the mechanical behavior of graphene. Some of term represents the out-of-plane deformation energy of gra-
the notations and definitions have been adapted from their phene. As the coefficient γ is rather small compared with α
original literature to make the discussion consistent in this and β, the bending deformation energy can be neglected in
chapter. Figure 1.1a illustrates the hexagonal lattice structure some cases.
of graphene, which can also be viewed as a flattened carbon
nanotube (CNT) cut along its longitudinal plane [24,25]. The
1.2.2 Continuum Elastic Model
first Brillouin zone is shown in Figure 1.1b. The two coor-
dinates x and y are mutually orthogonal, corresponding to The application of a valence force model in graphene mechan-
the armchair and zigzag chirality of CNTs, respectively. The ical components is limited by the countless number of atoms.
primitive cell of such a lattice contains two carbon atoms as Therefore, in practice, a continuum isotropic shell has been
marked by different colors, which are inequivalent due to the used to model the elastic behavior of graphene [28–30]. In a
directions of their adjacent bonds. The lattice constant a0 is linear deformation range, the strain energy U for a shell can
approximately equal to 2.46 Å in equilibrium [10]. For study- be expanded in a Taylor series of local in-plane strain εx, εy,
ing the static or dynamic behavior of such a many-bodied εxy and curvature k x, k y, k xy [31]; for simplicity, higher moduli
ensemble containing an extremely large number of atoms, the are omitted
Mechanical Properties of Graphene 5
1  E2 D 1
U=
2 ∫∫  (1 − υ ) (ε
2 x + ε y )2 − 2(1 − υ)(ε x ε y − ε 2xy )  Θ ac = − (1 + ν)(C111 − C222 )
4
}
3
+ D2 D (k x + k y )2 − 2(1 − ν)(k x k y − k xy
2
)  dS (1.2) + (1 − ν)(1 + ν)2 (C222 − C112 )
8
1
where E2D, υ, and D2D are the in-plane stiffness, Poisson ratio, + (1 − ν)3 (2C111 − C222 + 3C112 ) (1.4)
8
and flexural rigidity for the shell. These three parameters are
most important for describing the mechanical properties of
1
graphene and to be investigated in detail from both theoreti- Θ zz = (1 + ν)(C111 − C222 )
cal and experimental perspectives. Owing to graphene’s 2D 4
nature, the in-plane stiffness E2D has the dimensionality of 3
surface tension (N/m). The flexural rigidity D2D originates + (1 − ν)(1 + ν)2 (C222 − C112 )
8
from the third term in Equation 1.1, so its value is also very
small and can, therefore, also be neglected. The terms in the 1
+ (1 − ν)3 (2C111 − C222 + 3C112 ) (1.5)
integration, as a whole, represent the deformation energy den- 8
sity. There are two conditions for the validity of the equa-
tion. First, the length scale of the deformation should be much and the average third-order elastic modulus is
larger than the lattice constant [28], and second, the three
parameters must be isotropic in graphene’s 2D plane. As will
3
be seen, the assumed 2D isotropy of the model can be justified Θ av = (1 − ν)(1 + ν)2 (C222 − C112 )
by graphene’s hexagonal symmetry and the close resemblance 8
of calculated Young’s modulus in armchair and zigzag direc- 1
tions in a small deformation range. E2D, υ, and G 2D can be + (1 − ν)3 (2C111 − C222 + 3C112 ) (1.6)
8
treated as constants only when the deformation remains in
linear regime. For applications such as solving graphene reso-
The value of Θ is normally negative, leading to decreasing
nators, the linear deformation condition is normally fulfilled.
of stiffness at large tensile strains. The maximum strength σI
The bulk Young’s modulus E is related to the 2D stiffness E2D
achievable in a material with perfect lattice structure is called
by E = E2D /d0. d0 is the thickness of the graphene, normally
ideal strength [33,34] beyond which the material will be torn.
taking the value of 3.34 Å, the interlayer spacing in graph-
The ideal strength characterizes the physical properties of
ite. Another classic mechanical parameter of graphene is the
bonds in lattice structures. Nonetheless, materials normally
shear modulus G 2D, which is not explicitly included in the
break when far below the ideal strength due to stress concen-
above equation. For isotropic materials, G 2D = E2D /2(1 + ν).
tration around defects; thus it has been difficult to measure
the ideal strength of a material due to its imperfect lattice
1.2.3 Nonlinearity and Ideal Strength structure. However, a graphene sheet exfoliated from highly
ordered pyrolytic graphite (HOPG), or kish graphite, can be
In Equation 1.2, only the second-order elastic modulus E2D
defect free up to a millimeter, thus enabling direct character-
has been taken into account. This model does not allow the
ization of ideal strength which will be discussed later. The
softening of stiffness under high tension. If the tensile strain
deformation energy can even be expressed as even high-order
in graphene continues to increase, the stress will reach a max-
elastic modulus [30], which, however, will not be discussed in
imum, or peak, where the graphene begins to fracture. This
this chapter as most of these higher-order parameters are not
stiffness softening is common to almost every material. To
directly measurable, and are of less practical value in applica-
account for the nonlinearity of stiffness under great strain,
tion. Readers interested in the anisotropic continuum elastic
another parameter should also be considered; so again, if the
models with higher-order moduli are referred to References
isotropy is presumed, the stress should be expressed as
29 and 30.
σ = E ε + Θε 2 (1.3)
1.2.4 First-Principle Calculation
where σ is the symmetric second Piola–Kirchhoff stress, ε is The mechanical properties of a graphene sheet can be inves-
the uniaxial Lagrangian strain, and Θ is the third-order elastic tigated by first-principle methods. A first-principle calcu-
modulus. According to Cadelano [29], the third-order elas- lation is also called an ab initio calculation which is based
tic modulus is related to the third-order elastic constants C111, on quantum mechanics and does not require empirical data
C222, and C112, which are defined in crystal elasticity [32]. The but only locations of constitutive elements in the periodic
third-order elastic modulus along armchair Θac, zigzag direc- table. Instead of solving Schrödinger’s equation, the practi-
tion Θzz, and all other directions are different. The third-order cal implementation of a first-principle calculation is made
moduli along armchair and zigzag directions are expressed as possible by applying the DFT [35]. The DFT states that the
wave functions and observations of a physical system are

120 0.18
functional of the density of electrons. Compared with solv-
ing many-bodied Schrödinger’s equations based on the wave
100 0.15
Equivalent stress (GPa)

functions which has 3N variables, DFT considers the density
of electrons with just three variables of coordinates. Here,
Poisson’s ratio
80 0.12
N is the number of atoms. Therefore, DFT requires much less
computing resources, and has been proven to be a successful 60 0.09
and economic tool in investigating various physical proper-
ties of atomic and molecular systems. It is capable of calculat- 40 0.06
ing the relationship between the total ground-state electron
20 x (zigzag) direction 0.03
energy and the locations of atoms, which is the fundamental
y (armchair) direction
requirement for optimizing the geometry of a molecule or lat-
0
tice. The most commonly used algorithm for optimizing the 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
equilibrium geometry is the Broyden–Fletcher–Goldfarb– Strain ε
Shanno (BFGS) algorithm [36], which iterates calculating the
energy of a system and adjusting the lattice configuration. The FIGURE 1.2 DFT-calculated stresses and Poisson ratios along
configuration with the minimum total energy is found to be both zigzag (circle) and armchair direction (triangle), plotted against
the optimized lattice configuration. strain. (Reprinted with permission from F. Liu, P. Ming, and J. Li,
Ab initio calculation of ideal strength and phonon instability of gra-
The DFT calculation requires the lattice to be periodical
phene under tension, Phys. Rev. B, vol. 76, no. 6, p. 064120, 2007.
in all three directions. To rule out the mutual interaction of Copyright 2007 by the American Physical Society.)
graphene layers, the interlayer spacing, or the period along the
unshown z direction perpendicular to the x–y plane in Figure stress in the vertical axis refers to the bulk stress, which has
1.1a, should be much larger than graphite’s interlayer spacing. been derived from 2D super-cell stress using E = E2D /d0. It can
The dotted rectangle in Figure 1.1a encloses a super-cell of be observed from the curve that, in a small strain regime, the
graphene. The super-cell of such a rectangular shape allows equivalent bulk Young’s modulus is approximately a constant
independent investigation of stress–strain relationship along of 1050 GPa in both the armchair and zigzag directions. The
both armchair and zigzag directions, which are the two most Poisson ratio of armchair graphene at 5% strain is about 0.18,
representative directions. To obtain the deformation energy, a little larger than zigzag graphene’s 0.15 at the same strain.
a geometry optimization can be performed on the unstrained The Poisson ratio scales adversely as the strain. If the two
graphene super-cell to find the lattice structure of minimum Poisson ratio curves are reversely extrapolated to an infinitesi-
energy. Then a lattice constant of the super-cell, in the x direc- mal strain, they will intersect, which means that, under a small
tion, for example, can be increased to achieve a target strain strain condition, the Poisson ratios along both the armchair
and fixed so that the nonboundary atoms are allowed to relax and zigzag directions are almost equal. The maximum stress
to a configuration with which the system energy is minimized points, as colored in red for uniaxial tension in the x direction,
again. The potential energy discrepancy between the strained or zigzag direction, is 110 GPa when εx = 0.194, while for gra-
structure and the original graphene lattice is the total energy phene which is tensioned in the y direction, or armchair direc-
strain U. Such a calculation process mimics the procedure of tion, the maximum stress is 121 GPa at εx = 0.266. The stress
stretching a shell to acquire the strain–stress curve. The only in the x direction reaches its maximum in a smaller strain,
difference is that the stress is not a direct readout, but deducted indicating that the absolute value of a third-order modulus
from the strain energy–strain curve. As mentioned above, in the x direction is larger than that in the y direction, lead-
since the x and y directions are independent, the stiffness and ing to faster softening in the x direction. The deduced third-
Poisson ratio in each direction can be calculated separately. order modulus Θzz along the zigzag direction is −946.9 N/m,
For instance, if a homogeneous strain is applied to the super- while that along the armchair direction is −759.7 N/m. These
cell in the x direction, the strain energy density W becomes calculated values are in consistency with other DFT calcula-
tions [30].
U 1
W= = E2 Dx ε 2x (1.7)
A 2 1.2.5 Molecular Dynamics Simulation
Here, A represents the area of the graphene super-cell As opposed to the first-principle methods, a molecular
under consideration. The second-order differentiation of the dynamic simulation utilizes an empirical potential function
above equation gives the E2Dx. The lattice constant of the of atoms to evaluate the forces of interaction among atoms
super-cell along the y direction should see a contraction strain under tension. Also, by using Newton’s law, the location and
εy when a tensile strain along the x direction is applied. Thus, trajectory of atoms can be obtained at each simulation step
the Poisson ratio is obtained as ν = |ε y /ε x |. [37]. The molecular dynamic simulation is a transient simula-
Figure 1.2 shows the stress and Poisson ratios against the tion from which information can be extracted. To acquire the
straining of graphene in both the armchair and zigzag direc- stress–stain relationship of graphene, a strain can be applied
tions calculated with the ab initio methods [25]. The equivalent to the edges of graphene sheets either in the armchair or
140
Armchair graphene 0.45
120
Zigzag graphene
Armchair
100 0.4
Tensile stress (GPa)
Armchair direction Zigzag direction Zigzag
Shear modulus (TPa)

80
0.35
60 60 1.398
Shear stress (GPa)
Bond-length (Å)
0.3 40
40
Armchair 1.396
20
20 Zigzag
0.25
0 1.394
0 0.1 0.2 0.3 0 1000 2000
0 Shear strain Temperature (K)
0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 0.3 0.2
0 500 1000 1500 2000
Strain
Temperature (K)
FIGURE 1.3 Stress–strain relationship of graphene from MD

simulation. (Reprinted from Carbon N. Y., vol. 48, no. 3, Q. X. Pei, FIGURE 1.4 The shear modulus of graphene as a function of
Y. W. Zhang, and V. B. Shenoy, A molecular dynamics study of the temperature. (Reproduced from K. Min et al. (2011), by permission
mechanical properties of hydrogen functionalized graphene, pp. of AIP Publishing.)
898–904, Copyright 2010, with permission from Elsevier.)
Table 1.1 for comparison. It can be seen from the table that
the in-plane stiffness of graphene is very well concentrated at
zigzag direction, and the stress can be retrieved directly. The around 350 N/m. The stiffness along the zigzag and armchair
capability of MD simulation depends on the potential func- directions does not differ from each other very much, which
tion it uses, for example, the adaptive intermolecular reactive justifies the assumed isotropy of the continuum elastic model
bond-order potential allows for covalent bond breaking and discussed previously. The ideal strengths are also pretty close
creation with associated changes in hybridization of atomic to each other. The commonly observed phenomenon is that
orbitals within a classical potential [38,39]. the ideal strength along the armchair direction is larger than
Figure 1.3 shows the stress–strain curve obtained from along the zigzag direction, and indicates that the third-order
MD simulation [40]. Young’s modulus along the armchair modulus is higher along the armchair direction. Since single-
direction is 0.89 TPa, and the stress reaches its maximum walled CNTs can be viewed as wrapped-up graphene, their
at about 105 GPa. Young’s modulus along the zigzag direc- Young’s modulus and Poisson ratio are very comparable with
tion is about 0.83 TPa, and the stress reaches its maximum those of graphene’s [42–44]. For the comparison the mechani-
at 137 GPa. Young’s moduli along different directions at low cal properties of graphene nanoribbons and CNTs, the reader
strain are just slightly different from each other. The reduc- can also refer to Reference 45.
tion in the stiffness is also quicker in the zigzag direction than
in the armchair direction, in accordance with DFT calcula-
1.3 MEASUREMENT RESULTS
tions. The stress drops sharply after reaching the maximum
strain, indicating the breaking of carbon–carbon bonds and Nanoindentation has been a well-established technique
fracture of the graphene sheet. The values obtained from the for characterizing stiffness of materials and other elastic
MD simulation are considerably close to ab initio calculations parameters of either bulk material or suspended thin films
as discussed above. [46,47]. For characterizing suspended thin films, AFM
The MD simulation does not require the super-cell to tips are popularly used as indenters due to their accurate
be periodical. Therefore, it is much more flexible to impose positioning and precise application of force. The radius of
boundary conditions on the super-cell. In an MD simula- the nanosized tip of the AFM ensures that the load can be
tion, shear strain can be applied to either edge of the gra- modeled as point-load, which simplifies interpretation of
phene sheet to study the shear modulus. Figure 1.4 shows the the measured data. By applying a preset force on the tip
shear modulus as a function of temperature [41]. The insets while contacting the film surface, the AFM system retrieves
show the shear stress–shear strain curve and the bond length the deflection of the cantilever where the tip is located as
against temperature, respectively. At room temperature, the well as the vertical displacement of the tip holder, so as to
shear moduli G is 450 GPa (zigzag) and 350 GPa (armchair), acquire a force–displacement relationship for calculation of
respectively. mechanical properties. To perform the nanoindentation, the
graphene sheet can be suspended in two forms, either as a
doubly clamped bridge, or a round membrane clamped at its
1.2.6 Discussion of Calculated Results
circumference.
To summarize the calculated mechanical parameters of gra- Figure 1.5 shows suspended graphene flakes straddling
phene, some of the values from different sources are listed in across prepatterned trenches in a 280 nm SiO2/Si substrate
TABLE 1.1
Graphene’s Mechanical Properties from DFT Calculation and MD Simulations
Source
Method Material E2D (N/m) υ σI (N/m) D (N/m) G2D (N/m)
K. N. Kudin Gr 345 0.149 – – –
DFT [31]
D. M. Clatterbuck Gr 348 0.169 39.5 – –
DFT [33]
F. Liu Gr 350.7 0.186 36.74(zz) −946.9(zz) –
DFT [25] 40.4(ar) −759.7(ar)
Q. X. Pei Gr 287.24 – 45.8(zz) – –
MD [40] 35.7(ac)
K. Min – – – – – 150.3(zz)
MD [41] 116.9(ar)
G. Kalosakas Gr 320.6 0.22 33 ± 1(zz) 700(zz, DFT) 22(zz)
DFT & MD [78] 42 ± 3(ar) −670(zz, MD) 84(ar)
−670(ar, DFT)
−560(ar, MD)
C. Lee [47] Gr 340 – 42 −690 –
Nanoindentation
Note: ar, armchair; zz, zigzag; DFT, density functional theory; Gr, graphene; MD, molecular dynamics.
[46]. Graphene sheets are mechanically exfoliated by it allows more operational margin. The extracted 2D stiffness
scratching HOPG against the substrate; thus graphene sheets from the measured data is approximately 342 N/m, remark-
will be shoved out or peeled off the bulk graphite by the fric- ably consistent with first-principle calculations [47]. Breaking
tion force, and will settle on the substrate, some crossing the tests have also been performed on these samples, and it has
trenches. The width and thickness of the suspended part of been observed that the breaking stress only depends on the
the graphene flakes are entirely random, while the length is diameter of the tip of the AFM, but not the radius of the gra-
determined by the length of the trench underneath. Figure 1.6 phene membrane. For a graphene disk clamped at its circum-
plots the deduced spring constant K against w(t/L)3 of eight ference, the maximum stress σm induced by a spherical tip
different devices with similar lengths [46]. E is derived from pressing at the center is a function of the applied point-load
the slope of the fitted line to be 0.5 TPa, and the built-in ten- F [47]:
sion is estimated to be around 300 nN. The derived Young’s
modulus of a few layers of graphene is lower than that of bulk 1/ 2
 FC 
graphite, which is about 1 TPa. Young’s modulus measured σm =  (1.8)
 4πR 
this way tends to be inherently inaccurate. The suspended
graphene tends to be corrugated so that the evolvement of its
shape after the AFM tip touches upon it is much more com- This is in conformity with the observed adverse depen-
plicated, and the hypotheses are not valid any more [46,48]. dency of the tip’s radius and breaking force. The break-
Moreover, the shape of the graphene is also not well defined, ing point occurs when the σ–ε curve flattens, where dσ/
as can be seen from Figure 1.5, contributing to the error in dε = E + 2Θε = 0 according to Equation 1.3. So the maxi-
the results. mum strain will be −E/2Θ, and the maximum stress equals to
Similar to the previous preparation of suspended graphene −E2/2Θ − E/2 = 55 N/m. The solved value of the third-order
bridges, suspended graphene membranes can be fabricated moduli Θ2D is about −690 N/m, this value is remarkably con-
by exfoliating graphene onto prepatterned holes [47]. Finite sistent with ab initio calculations.
element analysis demonstrates that the force–displacement
curve is insensitive to the radius of the tip of the AFM when
the radius of the tip radius is much smaller than the radius 1.4 GRAPHENE RESONATORS
of the graphene disk. The simulation also suggests that, as 1.4.1 Overview
long as the tip is positioned within one-tenth of the radius
of the graphene disk radius from the membrane’s center, The implementation of graphene resonators does not only
the force–displacement behavior does not vary much. Such lead to various applications but also provides another route to
insensitivity in measuring a graphene disk proves to be a study the mechanical properties of graphene. The fundamen-
notable advantage over measuring a graphene bridge because tal frequency of harmonic resonators f 0 follows the general
(a) 6
t < 5 nm
1 μm t ≥ 5 nm 7 nm
5 Fit (R = 0.959)
Fit for 7/8 data (R = 0.996)
4
3 7 nm
k (N/m)
2 nm 8 nm
2
5 nm 5 nm
8 nm
1
2 nm
–1
(b) 0.0 0.1 0.2 0.3 0.4 0.5 0.6
w(t/L)3 (picometers)
FIGURE 1.6 Equivalent spring constants of graphene bridges plot-

ted as a function of w(t/L)3, (Reproduced from I. W. Frank et al.,
Mechanical properties of suspended graphene sheets, J. Vac. Sci.
Technol. B Microelectron. Nanom. Struct., vol. 25, no. 6, p. 2558,
2007, by permission of AVS Publications.)
of 1.03. The tension T 0 comes from various aspects, includ-

ing the fabrication process, the van der Waals force between
graphene and substrates, and electrostatic force.
J. S. Bunch et al., fabricated graphene resonators by plac-
2 μm ing graphene over a prepatterned trench in an SiO2 layer,
with a similar technique elaborated in Section 1.2 for the
nanoindentation experiment [20]. The graphene sheet has
been actuated and detected optically. A total of 33 reso-
nators have been characterized, and their fundamental
FIGURE 1.5 SEM images of graphene bridges exfoliated over frequencies have been plotted against d 0/L 2 [20]. With the
trenches. (Reproduced from I. W. Frank et al., Mechanical properties built-in tension assumed to be zero, theoretical frequency
of suspended graphene sheets, J. Vac. Sci. Technol. B Microelectron. against d 0/L 2 calculated with Young’s modulus of 0.5 and
Nanom. Struct., vol. 25, no. 6, p. 2558, 2007, by permission of AVS
1 TPa encloses most of the data points; thus, it is plausible to
Publications.)
conclude that Young’s modulus of graphene is close to that
of bulk graphite.
rule of f0 ∝ E /ρ, where f 0 is the fundamental frequency,
and ρ is the mass density. With ρ known, detecting the fun- 1.4.2 Quality Factor
damental resonance frequency allows deduction of Young’s
modulus. Graphene resonators can be actuated and sensed The quality factor Q is one of the most important specifi-
in different ways. Commonly applied techniques include cations of resonators, which can be extracted from the full-
thermal actuation by laser and optical detection, electrostatic width-half-maximum (FWHM) Δf of each Lorentzian peak,
actuation and electrical detection, or electrostatic actuation and expressed as Q = f0/Δf. Quality factor describes how fast
and optical detection. For a doubly clamped graphene bridge the total mechanical energy of the resonance mode gets dis-
under tension T 0 , its fundamental frequency can be expressed sipated away. The quality factors of graphene resonators vary
as [49] enormously in different reports, and are strongly related to
their ambient pressure and temperature. Table 1.2 lists a few
 2

1/ 2 quality factors of different graphene resonators, including
  E  d0  A 2 0.57T 0 
1/ 2
characterization environment and the quality factors of other
f0 =   A    +  (1.9)
   ρ  L  ρL2 wd0 
2
resonators for comparison. At room temperature, the quality
  factors of graphene resonators compare with those of CNT
resonators, on the scale of a few hundreds, which is more than
where ρ can use the value of graphite (2200 kg/m3), and A is one order of magnitude lower than resonators made from dia-
a constant. For a doubly clamped bridge, A takes the value mond or silicon nitride.
TABLE 1.2
List of Quality Factors of Graphene Resonator and Resonator Made of Other
Material
Author Q Material Structure Pressure Temperature
C. Chen [21] 10,000 Graphene Beam Unknown 5K
J. S. Bunch [20] 20–850 Graphene Beam <10−6 torr RTa
1800 50K
A. M. van der Zande [51] 9000 Graphene – <5 × 10−5 torr 10K
V. Sazonova [79] 40–200 CNT Beam <10 torr RT
J. Wang [80] 6225 Diamond Beam <5 × 10−2 torr RT
T. S. Biswas [81] >1.6 × 105 Si3N4 Nanostring ~2 × 10−6 torr RT
Note: CNT, carbon nanotube.

a Room temperature.
As observed from other MEMS resonators [50], the tempera- against the diameter [53]. The quality factor is proportional
ture largely affects the quality factor of a graphene resonator. to the diameter of the graphene membrane. With a larger
When the temperature drops toward absolute zero, the quality diameter, the quality factor could improve by one order of
factor increases. The quality factor of a graphene resonator magnitude.
is found to be proportional to 1/T λ (see Figure 1.7) [51]. λ is
around 0.35 ± 0.05 when the temperature is between 9 and
1.4.3 Dissipation Mechanism
40 K, and approximately 2.3 ± 0.1 when it is between 40 K
and room temperature. A similar phenomenon has also been MD simulation predicts that the energy dissipation of CNT
reported in Reference 21, and is independent of the size and resonators follows the 1/T 0.36 power law [54] from absolute
shape of the graphene sheet, which indicates that such a seg- zero to room temperature, thus matching the experimental
mental dependence of quality factor on temperature is not an observations at low temperature perfectly well. The simula-
isolated case. The scaling as 1/T 0.36 has been observed in a tion reveals that the external vibration energy is continuously
CNT resonator at low temperature as well [52]. converted to internal energy, which causes an increase in tem-
The dimensions of the devices are another influential fac- perature. The coincidence of a 0.36 power in both calculated
tor in the aspect of quality. According to Reference 51, the and experimental results at low temperature may suggest that
quality factor of suspended doubly clamped graphene resona- thermalelastic dissipation dominates at low temperature for
tors seems to be independent of the thickness of graphene. both CNT and graphene resonators. However, similar MD
However, the lateral size does affect the quality factor. Figure simulation performed on a graphene resonator shows the
1.8 shows the quality factors of a graphene circular membrane quality factor scales as 1/T [55]. The reason behind this dis-
crepancy remains unknown.
A full understanding of the energy dissipation in gra-
75 phene resonators is still lacking, since energy dissipation in
10–2 resonators is very complicated and is associated with various
f 1 (MHz)
55 energy loss mechanisms. The common mechanisms include

clamp loss, Ohmic loss, surface loss and thermalelastic loss,
and so on [55–64]. Seoanez and Guinea attempted to cat-
35
egorize the possible mechanisms of dissipation in graphene
Q–1
0 150 300
10–3
T (K) and nanotube resonators [63]. They have analyzed six pos-
sible dissipation approaches and the scaling of the energy
loss mechanisms. However, the analysis of the energy loss
only provides a qualitative understanding since the losses are
10–4 related to material, dimension, and characterizing method
10 100
T (K) and environment. The theoretical estimation of quality fac-
tor turns out to be a tough task.
FIGURE 1.7 The inverse of the quality factor of graphene reso- Of all the loss mechanisms, surface loss may play an impor-
nator is segmentally proportional to temperature. (Reprinted with tant role in graphene resonators. Surface loss is caused by sur-
permission from A. M. Van Der Zande et al., Large-scale arrays face defects, including the defects of the lattice structure or
of single-layer graphene resonators, Nano Lett., pp. 4869–73, 2010. absorbates [61]. The defects introduce stress or charge gradi-
Copyright 2010 American Chemical Society.) ents, thus increasing the entropy of the resonator, so the energy
(a) 30 (b) (c)

25
20
Frequency (MHz)
15 1000
1000
10 1.0
Q
Amp (a.u.)
0.5
5
0.0
3.65 3.66
Frequency (MHz)
100 100
5 10 15 20 25 5 10 15 20 25 5 10 15 20 25
Diameter (μm) Diameter (μm) Diameter (μm)
FIGURE 1.8 Quality factors of graphene membrane resonators, plotted against the diameter. The quality factors increase with larger
diameter. (Reprinted with permission from R. A. Barton et al., High, size-dependent quality factor in an array of graphene mechanical
resonators, Nano Lett., vol. 11, no. 3, pp. 1232–6, 2011. Copyright 2011 American Chemical Society.)
decreases. Reports have shown that the quality factor drops where x2(t) is the displacement, Meff is the effective mass, Γ
as the dimension scales down, while the surface-to-volume represents the damping coefficient, ζ is the cubit spring con-
ratio scales up [61]. Therefore, as the surface-to-volume ratio stant, and η is the nonlinear damping coefficient. M eff ω 20
of graphene is the highest of all materials due to its nature, can be abbreviated by one coefficient, which is the effective
the surface loss greatly limits the quality factor of graphene spring constant κeff. The onset of nonlinearity xc is, by conven-
resonators. tion, defined at the 1 dB compression point [69]. An analytical
MD simulation of CVD-grown graphene shows that the expression for xc may be solved using the above equation by
quality factor is adversely proportional to the density of the omitting certain terms for a rectangular resonator-like sus-
defects [55]. The defects induce spurious vibration modes that pended graphene bridge [70]:
drain the energy of the fundamental vibration. The simulation
predicts that a tensile strain could possibly greatly enhance
2 1  d2 T0 L2 
the quality factor by flattening out the buckles and suppress- xc = + (1.11)
ing the spurious mode. Previous studies in graphene resona-
4
3 Q  3 π 2 Etd 

tors also support this hypothesis [64].
which determines the upper bound of a graphene resonator’s
1.4.4 Dynamic Range and Sensitivity linear response range [69]. The lower bound of graphene
resonator’s response range is set by the thermal noise floor.
Dynamic range refers to the window in which the magnitude Thermal noise refers to the thermal oscillation of a resona-
of vibration magnitude of a resonator responds linearly to tor without external excitation. The power density of thermal
the stimuli, which has been commonly used to characterize noise is expected to have the form
the performance of amplifiers and sensors [65]. A higher
dynamic range is preferable for most applications such as
mass sensors or radio frequency (RF) devices. The resonance 4ω 0 kBT 1
S x (ω ) = (1.12)
amplitude increases linearly with larger actuation force up to M eff Q (ω 20 − ω 2 )2 + (ωω 0 /Q)2

a certain point and starts to level off. Nonlinearity in reso-
nators originates from different internal and external factors and root mean square displacement at the center of graphene
[66,67]. As pointed out in the previous section, the stiffness of bridge is expressed as
the material softens under high stress, thus introducing non-
linearity. Moreover, the actuation force on graphene has been ∞
assumed to be proportional to actuation forces in most cases kBT
[21,51], but in fact, the graphene will bend under actuation,
xth =
∫S (ω)dω =
x
κ eff
(1.13)
adding nonlinear terms to the driving force. All the nonlin- 0

ear terms will lead to nonlinear driving and resilient force
of the vibrating beam. Other sources of nonlinearity include following Reference 70. The DR is defined as
nonlinear damping, geometry, and so on. In general, the non-
linear motion of a resonator under a sinusoidal force can be 0.745 xc
described by the Duffing equation [68]: DR = 20 log
2S x ∆f
M eff x ′′(t ) + Γx ′(t ) + M eff ω 20 x(t ) + ζx 2 (t )   d

5/2
E 3/2 
= 20 log 3.9 dt    (1.14)
+ ηx 2 (t ) x ′(t ) = F cos(ωt ) (1.10)   L Q kBT ∆f ρ 
2

It can be clearly seen that the dynamic range depends

strongly on the d/L ratio. The dynamic range for the resona- TABLE 1.3
tor in Reference 20 is around 60 dB, lower than most NEMS List of the Mechanical Properties of Functionalized
resonators made from bulk material. This is due to the atomic Graphene
thickness of graphene. Source Method Material E2D (N/m) υ σI (N/m)
The extremely high surface area of graphene makes gra- R. R. Nair [76] CF 100 – 15
phene resonators ideal for mass sensing. Assuming that Nanoindentation
the absorbed substance does not couple strongly with the J. W. Suk [77] GO 145.3 – –
mechanical properties of graphene, the frequency shift caused Nanoindentation
by effective mass change of the graphene resonator can be E. Cadelano [72] CH 248 – –
expressed as DFT
M. Topsakal [73] CH 243 0.07 –
∂ω 0 ω0 DFT
δω = δm ≈ δm (1.15) E. Cadelano [74] CH 219 0.21 –
∂M eff 2 M eff
DFT
K. N. Kudin [31] C2F 328 0.039
So the mass sensitivity is DFT
Note: DFT, density functional theory.

δm ∂M eff 2 M eff
= ≈ (1.16)
δω ∂ω 0 ω0
silicon and steel. In graphene oxide, Young’s modulus is also
The force sensitivity of the resonator limited by the ther- largely reduced [77]. The degradation of Young’s modulus in
mal fluctuation is expressed as the aforementioned three graphene derivatives is caused by
transforming sp2 carbon–carbon bonds to sp3 bonds by an
4κ eff kBT attached radical [71–74]. Some of the representative values of
dF = (1.17) the mechanical properties have been listed in Table 1.3.
Qω 0

For the resonator discussed in the above section, the force 1.6 CONCLUSION
sensitivity is 0.9 fN/ Hz , which translates to a charge sensi- A general continuous isotropic model has been introduced
tivity of 8 × 10 −4 e/ Hz [51]. As discussed above, the qual- to describe graphene’s elastic behavior. Within the model,
ity factor scales adversely with temperature, so the sensitivity the most important parameters are in-plane stiffness E2D and
will see an increase with a decrease in temperature. The sen- Poisson ratio ν. The 2D in-plane stiffness is related to bulk
sitivity of the charge sensor based on a graphene resonator can Young’s modulus E by E = E2D /d0. These parameters can be
rival RF single-electron transistor electrometers [51]. calculated by both the first-principle method and MD simula-
tion. We reviewed the calculated results along different direc-
1.5 GRAPHENE DERIVATIVES tions and the experimental characterization of graphene’s
mechanical properties by AFM nanoindentation. The experi-
Functionalization of graphene endows the family of 2D mental results agree very well with theoretical results. Owing
materials with various mechanical features, thus broadening to the perfect structure of graphene, its ideal strength can be
the scope of application of graphene. The functionalization ascertained by the indentation. Such an experiment proves the
of graphene encloses a wide range of treatment processes, accuracy of the theoretical prediction about the ideal strength.
including hydrogenation, oxidation, fluorination, and so on. With large Young’s modulus, the graphene resonator can
Hydrogenation is of particular interest to hydrogen storage, achieve very high frequencies, however, its quality factor is
where the hydrogen atoms adsorb to carbon atoms in the gra- greatly limited by various other factors. The quality factor of
phene lattice [71]. Calculation based on the continuum elastic a graphene resonator also resembles CNT resonators, which
theory and first principles reveals that the hydrogenated gra- may suggest that the dominating dissipation mechanism
phene suffers a degradation of Young’s modulus [71–74]. It is surface loss. The surface loss greatly reduces the quality
has been found that Young’s modulus decreases by as much factors due to their extremely high surface-to-volume ratio.
as 37% for 100% hydrogen coverage (CH) [74]. Nevertheless, the sensitivity of sensors based on graphene
The structure of fluorinated graphene, alias fluorog- resonators is still comparable to those made of other material.
raphene, resembles that of hydrogenated graphene sheet.
Fluorographene is a semiconductor material whose band-gap
scales with the proportion of fluorine atoms, thus it is almost REFERENCES
transparent [75,76]. The Young’s modulus of fluorographene 1. B. Partoens and F. Peeters, From graphene to graphite:
is found to degrade even more to near 100 N/m, nevertheless, Electronic structure around the K point, Phys. Rev. B, vol. 74,
such stiffness still rivals that of other common materials like no. 7, p. 075404, 2006.
2. Y. H. Ho, J. Y. Wu, Y. H. Chiu, J. Wang, and M. F. Lin, 21. C. Chen, S. Rosenblatt, K. I. Bolotin, W. Kalb, P. Kim, I.
Electronic and optical properties of monolayer and bilayer Kymissis, H. L. Stormer, T. F. Heinz, and J. Hone, Performance
graphene, Philos. Trans. Series A, Math. Phys. Eng. Sci., vol. of monolayer graphene nanomechanical resonators with electri-
368, no. 1932, pp. 5445–58, 2010. cal readout, Nat. Nanotechnol., vol. 4, no. 12, pp. 861–7, 2009.
3. P. Wallace, The band theory of graphite, Phys. Rev., vol. 71, 22. A. Sakhaee-Pour, M. T. Ahmadian, and A. Vafai, Applications
no. 9, pp. 622–34, 1947. of single-layered graphene sheets as mass sensors and atom-
4. K. Geim and K. S. Novoselov, The rise of graphene, Nat. istic dust detectors, Solid State Commun., vol. 145, no. 4,
Mater., vol. 6, no. 3, pp. 183–91, 2007. pp. 168–72, 2008.
5. H. Castro Neto, N. M. R. Peres, K. S. Novoselov, and A. K. 23. F. Schedin, A. K. Geim, S. V Morozov, E. W. Hill, P. Blake,
Geim, The electronic properties of graphene, Rev. Mod. Phys., M. I. Katsnelson, and K. S. Novoselov, Detection of individ-
vol. 81, no. 1, pp. 109–62, 2009. ual gas molecules adsorbed on graphene, Nat. Mater., vol. 6,
6. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. no. 9, pp. 652–5, 2007.
Zhang, S. V. Dubonos, I. V. Grigorieva, and A. A. Firsov, 24. L. Xie, L. Jiao, and H. Dai, Selective etching of graphene
Electric field effect in atomically thin carbon films, Science, edges by hydrogen plasma, J. Am. Chem. Soc., vol. 132, no.
vol. 306, no. 5696, pp. 666–9, 2004. 42, pp. 14751–3, 2010.
7. K. I. Bolotin, K. J. Sikes, Z. Jiang, M. Klima, G. Fudenberg, 25. F. Liu, P. Ming, and J. Li, Ab initio calculation of ideal strength
J. Hone, P. Kim, and H. L. Stormer, Ultrahigh electron and phonon instability of graphene under tension, Phys. Rev.
mobility in suspended graphene, Solid State Commun.,
B, vol. 76, no. 6, p. 064120, 2007.
vol. 146, no. 9–10, pp. 351–5, 2008. 26. C. Lobo and J. L. Martins, Valence force field model for gra-
8. X. Du, I. Skachko, A. Barker, and E. Y. Andrei, Approaching phene and fullerenes, Zeitschrift fur Phys. D Atoms, Mol.
ballistic transport in suspended graphene, Nat. Nanotechnol., Clust., vol. 39, no. 2, pp. 159–64, 1997.
vol. 3, no. 8, pp. 491–5, 2008. 27. P. Keating, Effect of invariance requirements on the elas-
9. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, M. I. tic strain energy of crystals with application to the diamond
Katsnelson, I. V. Grigorieva, S. V. Dubonos, and A. A. Firsov, structure, Phys. Rev., vol. 145, no. 2, pp. 637–45, 1966.
Two-dimensional gas of massless Dirac fermions in graphene, 28. J. Atalaya, A. Isacsson, and J. M. Kinaret, Continuum elastic
Nature, vol. 438, no. 7065, pp. 197–200, 2005. modeling of graphene resonators, Nano Lett., vol. 8, no. 12,
10. J. McClure, Band structure of graphite and de Haas-van pp. 4196–200, 2008.
Alphen effect, Phys. Rev., vol. 108, no. 3, pp. 612–18, 1957. 29. E. Cadelano, P. Palla, S. Giordano, and L. Colombo, Nonlinear
11. A. K. Geim, Graphene: Status and prospects, Science, vol. elasticity of monolayer graphene, Phys. Rev. Lett., vol. 102,
324, no. 5934, pp. 1530–4, 2009. no. 23, p. 235502, 2009.
12. H. Wang, L. Yu, Y.-H. Lee, Y. Shi, A. Hsu, M. L. Chin, L.-J. 30. X. Wei, B. Fragneaud, C. Marianetti, and J. Kysar, Nonlinear
Li, M. Dubey, J. Kong, and T. Palacios, Integrated circuits elastic behavior of graphene: Ab initio calculations to con-
based on bilayer MoS2 transistors, Nano Lett., vol. 12, no. 9, tinuum description, Phys. Rev. B, vol. 80, no. 20, p. 205407,
pp. 4674–80, 2012. 2009.
13. C. R. Dean, A. F. Young, I. Meric, C. Lee, L. Wang, S. 31. K. Kudin, G. Scuseria, and B. Yakobson, C2F, BN, and C
Sorgenfrei, K. Watanabe, T. Taniguchi, P. Kim, K. L. Shepard, nanoshell elasticity from ab initio computations, Phys. Rev. B,
and J. Hone, Boron nitride substrates for high-quality gra- vol. 64, no. 23, p. 235406, 2001.
phene electronics, Nat. Nanotechnol., vol. 5, no. 10, pp. 722– 32. S. Bhagavantam and J. Bhimasenachar, Elastic constants of
6, 2010. crystals, Proc. Indian Acad. Sci. Sect. A, vol. 20, no. 5, pp.
14. Z. A. Piazza, H.-S. Hu, W.-L. Li, Y.-F. Zhao, J. Li, and L.-S. 298–303, 1945.
Wang, Planar hexagonal B(36) as a potential basis for extended 33. D. Clatterbuck, C. Krenn, M. Cohen, and J. Morris, Phonon
single-atom layer boron sheets, Nat. Commun., vol. 5, p. 3113, instabilities and the ideal strength of aluminum, Phys. Rev.
2014. Lett., vol. 91, no. 13, p. 135501, 2003.
15. T. Hesjedal, Continuous roll-to-roll growth of graphene films 34. A. Kelly and N. H. Macmillan, Strong Solids, Clarendon
by chemical vapor deposition, Appl. Phys. Lett., vol. 98, no. Press, 1986, p. 423.
13, p. 133106, 2011. 35. J. Kohanoff, N. I. Gidopoulos, and S. Wilson, Density func-
16. S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, tional theory: Basics, new trends and applications, vol. 2, pp.
J. Balakrishnan, T. Lei, H. R. Kim, Y. Il Song, Y.-J. Kim, 532–68, 2003.
K. S. Kim, B. Ozyilmaz, J.-H. Ahn, B. H. Hong, and S. Iijima, 36. J. D. Head and M. C. Zerner, A Broyden–Fletcher–Goldfarb–
Roll-to-roll production of 30-inch graphene films for trans- Shanno optimization procedure for molecular geometries,
parent electrodes, Nat. Nanotechnol., vol. 5, no. 8, pp. 574–8, Chem. Phys. Lett., vol. 122, no. 3, pp. 264–70, 1985.
2010. 37. W. F. van Gunsteren and H. J. C. Berendsen, Computer simu-
17. M. Dragoman, G. Konstantinidis, D. Dragoman, D. Neculoiu, lation of molecular dynamics: Methodology, applications, and
a. Cismaru, F. Coccetti, R. Plana, H. Harnagel, a. Kostopoulos, perspectives in chemistry, Angew. Chemie Int. Ed. English,
and R. Buiculescu, RF NEMS based on carbon nanotubes and vol. 29, no. 9, pp. 992–1023, 1990.
graphene, 2008 Int. Semicond. Conf., pp. 103–6, 2008. 38. S. J. Stuart, A. B. Tutein, and J. A. Harrison, A reactive
18. C. Chen and J. Hone, Graphene nanoelectromechanical sys- potential for hydrocarbons with intermolecular interactions,
tems, Proc. IEEE, vol. 101, pp. 1766–79, 2013. J. Chem. Phys., vol. 112, no. 14, p. 6472, 2000.
19. T. Miao, S. Yeom, P. Wang, B. Standley, and M. Bockrath, 39. G. Kalosakas, N. N. Lathiotakis, C. Galiotis, and K. Papagelis,
Graphene nanoelectromechanical systems as stochastic- In-plane force fields and elastic properties of graphene, p. 7,
frequency oscillators, Nano Lett., 2014. 2012.
20. J. S. Bunch, A. M. van der Zande, S. S. Verbridge, I. W. Frank, 40. Q. X. Pei, Y. W. Zhang, and V. B. Shenoy, A molecular
D. M. Tanenbaum, J. M. Parpia, H. G. Craighead, and P. L. dynamics study of the mechanical properties of hydrogen
McEuen, Electromechanical resonators from graphene sheets, functionalized graphene, Carbon N. Y., vol. 48, no. 3, pp.
Science, vol. 315, no. 5811, pp. 490–3, 2007. 898–904, 2010.
41. K. Min and N. R. Aluru, Mechanical properties of graphene 59. Y.-H. Park and K. C. Park, High-fidelity modeling of MEMS
under shear deformation, Appl. Phys. Lett., vol. 98, no. 1, resonators—Part I: Anchor loss mechanisms through sub-
p. 013113, 2011. strate, J. Microelectromech. Syst., vol. 13, no. 2, pp. 238–47,
42. R. S. Ruoff and D. C. Lorents, Mechanical and thermal prop- 2004.
erties of carbon nanotubes, Carbon N. Y., vol. 33, no. 7, pp. 60. M. A. Abdelmoneum, M. U. Demirci, and C. T.-C. Nguyen,
925–30, 1995. Stemless wine-glass-mode disk micromechanical resonators,
43. J. N. Coleman, U. Khan, W. J. Blau, and Y. K. Gun’ko, Small in The Sixteenth Annual International Conference on Micro
but strong: A review of the mechanical properties of carbon Electro Mechanical Systems, 2003. MEMS-03 Kyoto. IEEE,
nanotube–polymer composites, Carbon N. Y., vol. 44, no. 9, 2003, pp. 698–701.
pp. 1624–52, 2006. 61. T. Ono and M. Esashi, Energy dissipation in submicrometer
44. J.-P. Salvetat, J.-M. Bonard, N. H. Thomson, A. J. Kulik, L. thick single-crystal silicon cantilevers, J. Microelectromech.
Forro, W. Benoit, and L. Zuppiroli, Mechanical properties of Syst., vol. 11, no. 6, pp. 775–83, 2002.
carbon nanotubes, Appl. Phys. A: Mater. Sci. Process., vol. 69, 62. K. Y. Yasumura, T. D. Stowe, E. M. Chow, T. Pfafman, T. W.
pp. 255–60, 1999. Kenny, B. C. Stipe, and D. Rugar, Quality factors in micron-
45. R. Faccio, P. A. Denis, H. Pardo, C. Goyenola, and A. W. and submicron-thick cantilevers, J. Microelectromech. Syst.,
Mombrú, Mechanical properties of graphene nanoribbons, vol. 9, no. 1, pp. 117–25, 2000.
J. Phys. Condens. Matter, vol. 21, no. 28, p. 285304, 2009. 63. C. Seoanez, F. Guinea, and A. Neto, Dissipation in graphene
46. I. W. Frank, D. M. Tanenbaum, a. M. van der Zande, and and nanotube resonators, Phys. Rev. B, 2007.
P. L. McEuen, Mechanical properties of suspended graphene 64. S. Y. Kim and H. S. Park, The importance of edge effects on
sheets, J. Vac. Sci. Technol. B Microelectron. Nanom. Struct., the intrinsic loss mechanisms of graphene nanoresonators,
vol. 25, no. 6, p. 2558, 2007. Nano Lett., vol. 9, no. 3, pp. 969–74, 2009.
47. C. Lee, X. Wei, J. W. Kysar, and J. Hone, Measurement of 65. B. Ho Eom, P. K. Day, H. G. LeDuc, and J. Zmuidzinas, A
the elastic properties and intrinsic strength of monolayer gra- wideband, low-noise superconducting amplifier with high
phene, Science, vol. 321, no. 5887, pp. 385–8, 2008. dynamic range, Nat. Phys., vol. 8, no. 8, pp. 623–7, 2012.
48. J. C. Meyer, A. K. Geim, M. I. Katsnelson, K. S. Novoselov, 66. R. Legtenberg, H. A. C. Tilmans, J. H. J. Fluitman, and M.
T. J. Booth, and S. Roth, The structure of suspended graphene Elwenspoek, Nonlinearity and hysteresis of resonant strain
sheets, Nature, vol. 446, no. 7131, pp. 60–3, 2007. gauges, J. Microelectromech. Syst., vol. 7, no. 1, pp. 122–7,
49. W. Weaver, Jr., S. P. Timoshenko, and D. H. Young, Vibration 1998.
Problems in Engineering, John Wiley & Sons, 1974, pp. 425–7. 67. M. V. Andres, K. W. H. Foulds, and M. J. Tudor, Nonlinear
50. B. Kim, M. A. Hopcroft, R. N. Candler, C. M. Jha, M. vibrations and hysteresis of micromachined silicon resonators
Agarwal, R. Melamud, S. A. Chandorkar, G. Yama, and T. W. designed as frequency-out sensors, Electron. Lett., vol. 23, no.
Kenny, Temperature dependence of quality factor in MEMS 18, p. 952, 1987.
resonators, J. Microelectromech. Syst., vol. 17, no. 3, pp. 755– 68. R. Lifshitz and M. Cross, Nonlinear dynamics of nano-
66, 2008. mechanical and micromechanical resonators, in Review of
51. A. M. Van Der Zande, R. A. Barton, J. S. Alden, C. S. Ruiz- Nonlinear Dynamics and Complexity, Weinheim: Wiley-VCH
Vargas, W. S. Whitney, P. H. Q. Pham, J. Park, J. M. Parpia, Verlag GmbH & Co. KGaA, 2008.
H. G. Craighead, and P. L. McEuen, Large-scale arrays of sin- 69. K. L. Ekinci and M. L. Roukes, Nanoelectromechanical sys-
gle-layer graphene resonators, Nano Lett., pp. 4869–73, 2010. tems, Rev. Sci. Instrum., vol. 76, no. 6, p. 061101, 2005.
52. A. K. Hu, G. A. Steele, B. Witkamp, M. Poot, L. P. 70. H. W. C. Postma, I. Kozinsky, a. Husain, and M. L. Roukes,
Kouwenhoven, and H. S. J. Van Der Zant, Carbon Nanotubes Dynamic range of nanotube- and nanowire-based electrome-
as Ultrahigh Quality Factor Mechanical Resonators, 2009. chanical systems, Appl. Phys. Lett., vol. 86, no. 22, p. 223105,
53. R. A. Barton, B. Ilic, A. M. van der Zande, W. S. Whitney, 2005.
P. L. McEuen, J. M. Parpia, and H. G. Craighead, High, size- 71. E. Muñoz, A. K. Singh, M. A. Ribas, E. S. Penev, and B. I.
dependent quality factor in an array of graphene mechanical Yakobson, The ultimate diamond slab: GraphAne versus
resonators, Nano Lett., vol. 11, no. 3, pp. 1232–6, 2011. graphEne, Diam. Relat. Mater., vol. 19, no. 5–6, pp. 368–73,
54. H. Jiang, M.-F. Yu, B. Liu, and Y. Huang, Intrinsic energy loss 2010.
mechanisms in a cantilevered carbon nanotube beam oscilla- 72. E. Cadelano, P. L. Palla, S. Giordano, and L. Colombo, Elastic
tor, Phys. Rev. Lett., vol. 93, no. 18, p. 185501, 2004. properties of hydrogenated graphene, Phys. Rev. B, vol. 82,
55. Z. Qi and H. S. Park, Intrinsic energy dissipation in CVD- pp. 1–8, 2010.
grown graphene nanoresonators, Nanoscale, vol. 4, no. 11, pp. 73. M. Topsakal, S. Cahangirov, and S. Ciraci, The response of
3460–5, 2012. mechanical and electronic properties of graphane to the elas-
56. D. S. Bindel, E. Quevy, T. Koyama, S. Govindjee, J. W. tic strain, Appl. Phys. Lett., vol. 96, 2010.
Demmel, and R. T. Howe, Anchor loss simulation in resona- 74. E. Cadelano and L. Colombo, Effect of hydrogen coverage on
tors, in 18th IEEE International Conference on Micro Electro the Young’s modulus of graphene, Phys. Rev. B, vol. 85, no.
Mechanical Systems, 2005. MEMS 2005, 2005, pp. 133–6. 24, p. 245434, 2012.
57. F. Ayazi, Support loss in micromechanical disk resonators, 75. J. T. Robinson, J. S. Burgess, C. E. Junkermeier, S. C.
in 18th IEEE International Conference on Micro Electro Badescu, T. L. Reinecke, F. K. Perkins, M. K. Zalalutdniov
Mechanical Systems, 2005. MEMS 2005, 2005, pp. 137–41. et al., Properties of fluorinated graphene films, Nano Lett.,
58. Z. Hao, A. Erbil, and F. Ayazi, An analytical model for sup- vol. 10, no. 8, pp. 3001–5, 2010.
port loss in micromachined beam resonators with in-plane 76. R. R. Nair, W. Ren, R. Jalil, I. Riaz, V. G. Kravets, L. Britnell,
flexural vibrations, Sensors Actuators A Phys., vol. 109, no. P. Blake et al., Fluorographene: A two-dimensional counter-
1–2, pp. 156–64, 2003. part of Teflon, Small, vol. 6, no. 24, pp. 2877–84, 2010.
77. J. W. Suk, R. D. Piner, J. An, and R. S. Ruoff, Mechanical 80. J. E. Butler, D. S. Y. Hsu, and T.-C. Nguyen, CVD polycrystal-
properties of monolayer graphene oxide, ACS Nano, vol. 4, pp. line diamond high-Q micromechanical resonators, Tech. Dig.
6557–64, 2010. MEMS 2002 IEEE Int. Conf. Fifteenth IEEE Int. Conf. Micro
78. G. Kalosakas, N. N. Lathiotakis, C. Galiotis, and K. Papagelis, Electro Mech. Syst. (Cat. No.02CH37266), vol. 3, no. 2, pp.
In-plane force fields and elastic properties of graphene, 657–60, 2002.
J. Appl. Phys., vol. 113, no. 13, p. 134307, 2013. 81. T. S. Biswas, A. Suhel, B. D. Hauer, A. Palomino, K. S. D.
79. V. Sazonova, Y. Yaish, H. Ustünel, D. Roundy, T. A. Arias, Beach, and J. P. Davis, High-Q gold and silicon nitride bilayer
and P. L. McEuen, A tunable carbon nanotube electromechan- nanostrings, Appl. Phys. Lett., vol. 101, no. 9, p. 093105,
ical oscillator, Nature, vol. 431, no. 7006, pp. 284–7, 2004. 2012.
2 Effective Elastic Properties of
a Novel Continuous Fuzzy
Fiber-Reinforced Composite with
Wavy Carbon Nanotubes
M. C. Ray and S. I. Kundalwal
CONTENTS
Abstract........................................................................................................................................................................................ 17
2.1 Introduction........................................................................................................................................................................ 17
2.2 Architecture of the FFRC Containing Wavy CNTs........................................................................................................... 20
2.3 Models of Wavy CNTs....................................................................................................................................................... 22
2.4 Effective Elastic Properties of the Continuous FFRC Containing Wavy CNTs................................................................ 22
2.4.1 MOM Approach...................................................................................................................................................... 22
2.4.1.1 Effective Elastic Properties of the PMNC............................................................................................... 22
2.4.1.2 Effective Elastic Properties of the CFF................................................................................................... 26
2.4.1.3 Effective Elastic Properties of the FFRC................................................................................................ 27
2.4.2 MT Method............................................................................................................................................................. 28
2.5 Results and Discussion....................................................................................................................................................... 30
2.6 Conclusions......................................................................................................................................................................... 36
References.................................................................................................................................................................................... 37
ABSTRACT progress has shown that graphene-based materials can have

a profound impact on electronic and optoelectronic devices,
This chapter deals with the investigation of the effect of chemical sensors, nanocomposites, and energy storage (Singh
carbon nanotube (CNT) waviness on the effective elastic et al., 2011). However, until 2004, single-layer graphene was
properties of a novel continuous fuzzy fiber-reinforced com- believed to be thermodynamically unstable under ambi-
posite (FFRC). Effective elastic properties of this composite ent conditions (Mermin, 1968; Landau and Lifshitz, 1980).
estimated by the mechanics of materials (MOM) approach Novoselov et al. (2004) successfully identified single layers of
have been compared with those predicted by the Mori– graphene in a simple tabletop experiment. This revolutionary
Tanaka (MT) method. The interphase between a CNT and the discovery has added a new dimension of research in the fields
polymer matrix which models the nonbonded van der Waals of physics, chemistry, biotechnology, and materials science.
interaction between them is also considered for estimating the The “thinnest” known material, graphene exhibits excellent
effective elastic properties of the FFRC. The effect of CNT multifunctional properties (Novoselov et al., 2004; Huang
waviness on the effective elastic properties of the FFRC is et al., 2006; Balandin et al., 2008; Bao et al., 2009; Blake
investigated when wavy CNTs are coplanar with either of the et al., 2008; Ghosh et al., 2008; Li et al., 2008; Wang et al.,
two mutually orthogonal planes. Development of wavy CNTs 2008; Jiang et al., 2009a,b; Lee et al., 2009; Timo et al., 2009;
and the micromechanics models are presented in detail. Tsoukleri et al., 2009; Wang et al., 2009; Dreyer et al., 2010;
Seol et al., 2010; Milowska et al., 2012). Table 2.1 summarizes
the mechanical, thermal, and electrical properties of single
2.1 INTRODUCTION and multilayer graphene. Recently, graphene has been used as
Graphene, a monolayer of sp2-hybridized carbon atoms an alternative carbon-based reinforcement in the preparation
arranged in a two-dimensional lattice, has emerged as an of polymer nanocomposites demonstrating improved multi-
exotic material in recent years owing to its exceptional functional properties (Stankovich et al., 2006; Ramanathan
charge transport, thermal, electrical, optical, and mechani- et al., 2008; Ansari and Giannelis, 2009; Bose et al., 2010;
cal properties. Graphene and its derivatives are being stud- Fan et al., 2010; Lv et al., 2010; Zhang et al., 2010; Zhao et al.,
ied in nearly every field of science and engineering. Recent 2010; Kuila et al., 2011; Potts et al., 2011; Song et al., 2014;
17
TABLE 2.1
Mechanical, Thermal, and Electrical Properties of Graphene and Graphene Oxide-Based Materials
References Method Material Properties
Huang et al. (2006) Molecular mechanics (using Single-layer graphene, T = 0.0574 nm E = 4.23 TPa and ν = 0.397 (uniaxial tension)
second-generation Brenner Single-layer graphene, T = 0.0678 nm E = 3.58 TPa and ν = 0.397 (uniaxial stretching)
potential) Single-layer graphene, T = 0.0811 nm E = 2.99 TPa and ν = 0.397 (equibiaxial
stretching)
Balandin et al. (2008) Confocal micro-Raman spectroscopy Single-layer graphene (4.84 ± 0.44) × 103 to (5.30 ± 0.48) × 103 W/mK
at RT
Ghosh et al. (2008) Confocal micro-Raman spectroscopy Suspended graphene flake ~3080–5150 W/mK at RT
Lee et al. (2008) AFM Monolayer graphene E = 1 ± 0.1 TPa, σint = 130 ± 10 GPa at
εint = 0.25
Bao et al. (2009) Experimental Single-layer graphene CTE = −7 × 10−6 K−1 at 300 K
Jiang et al. (2009a,b) Nonequilibrium Green’s function Single-layer graphene CTE = −6 × 10−6 K−1 at 300 K
approach
Lee et al. (2009) AFM Monolayer graphene E = 1.02 TPa and σ = 130 GPa
Bilayer graphene E = 1.04 TPa and σ = 126 GPa
Trilayer graphene E = 0.98 TPa and σ = 101 GPa
Tsoukleri et al. (2009) Raman Graphene Strain ~1.3% in tension;
strain ~0.7% in compression
Timo et al. (2009) Electrical four-point measurement Reduced graphene oxide flake 0.14–2.87 W/mK
6.2 × 102 to 6.2 × 103 Ω−1 m−1
Seol et al. (2010) Thermal measurement Single-layer (suspended) 3000–5000 W/mK at RT (suspended)
Single-layer (on SiO2 support) 6000 W/mK at RT (on a silicon dioxide support)
Milowska et al. DFT Monolayer graphene E = 1.05 TPa, ν = 0.11, K = 0.47 TPa, and
(2012) G = 0.46 TPa
Note: AFM, atomic force microscopy; DFT, density functional theory.
Yang et al., 2014). Graphene is expected to play an important (MWCNT). Within a couple of years, Iijima and Ichihashi
role in the fabrication of sensor, fuel cell, nanoelectronic, and (1993) discovered the synthesis of single-walled carbon nano-
bioelectronic devices in the near future (Geim and Novoselov, tube (SWCNT). The other manufacturing processes, such as
2007; Kauffman and Star, 2010). Graphene is also capable of laser ablation, chemical vapor deposition (CVD) and plasma-
replacing metal conductors in electronic and electrical devices enhanced CVD are being employed to synthesize CNTs on a
due to its excellent electrical conductivity and mechanical large scale.
flexibility (Kim et al., 2009; Li et al., 2009; Bae et al., 2010). Researchers probably thought that CNTs may be useful as
Ongoing research shows that graphene can replace brittle and nanoscale fibers for developing novel nanocomposites, and
chemically unstable indium in oxide, in flexible displays, and this conjecture motivated them to accurately predict the prop-
touch screens. Therefore, graphene, as an electronic circuit erties of CNTs. Hence, since the discovery of CNTs, research-
material, is considered to be potentially superior to other car- ers have been carrying out extensive studies to estimate their
bon-based nanofillers (Huang et al., 2007; Wu et al., 2008; elastic properties. Both experimental measurements and theo-
Park et al., 2010). retical studies confirmed that SWCNTs have axial Young’s
The research on the synthesis of molecular carbon struc- modulus in the terapascal range (Treacy et al., 1996; Lu, 1997;
ture by an arc-discharge method for evaporation of carbon led Krishnan et al., 1998; Lourie and Wagner, 1998; Popov et al.,
to the discovery of an extremely thin needle-like graphitic car- 2000; Jin and Yuan, 2003; Shen and Li, 2004, 2005; Liu et al.,
bon molecule (Iijima, 1991). While examining under an elec- 2005; Tsai et al., 2010). Table 2.2 summarizes the elastic prop-
tron microscope at the level of atomic resolution, Iijima (1991) erties of different types of CNTs in which the symbols have
observed that such needle-like carbon materials are seamless their usual meaning. The quest for utilizing such exceptional
coaxial tubes made of carbon atom sheets. The thickness of mechanical properties of CNTs and their high aspect ratio led
the carbon atom sheet was less than a nanometer while the to the opening of an emerging area of research on the devel-
separation between the walls of the tubes was found to be opment of CNT-reinforced composites (Odegard et al., 2003;
0.34 nm. The outer diameter of such needle-like material is in Gao and Li, 2005; Seidel and Lagoudas, 2006; Jiang et al.,
the range of a few nanometers. Iijima (1991) named such nee- 2009a,b; Meguid et al., 2010; Tsai et al., 2010; Wernik and
dle-like carbon materials as multiwalled carbon nanotubes Meguid, 2010; Wernik et al., 2012).
Effective Elastic Properties of a Novel Continuous Fuzzy Fiber-Reinforced Composite 19
TABLE 2.2
Comparison of Engineering Constants for Different Types of CNTs
References Method (Symmetry) CNT Type E11 (Gpa) E22 (Gpa) G12 (Gpa) G23 (GPa) ν12
Jin and Yuan Energy approach SWCNT (6, 6) 1376 1324 551 – 0.267
(2003) (about 1-axis) SWCNT (10, 10) 1355 1338 550 – 0.262
SWCNT (15, 15) 1346 1350 543 – 0.252
SWCNT (20, 20) 1342 1370 545 – 0.253
Force approach SWCNT (6, 6) 1245 1217 483 – 0.267
(about 1-axis) SWCNT (10, 10) 1241 1235 493 – 0.262
SWCNT (15, 15) 1244 1238 495 – 0.252
SWCNT (20, 20) 1243 1243 495 – 0.253
Shen and Li Molecular mechanics SWCNT (5, 5) 2080 421.13 791 132 0.172
(2004) (about 1-axis) SWCNT (10, 10) 1060 63.88 442 17 0.162
SWCNT (15, 15) 707 19.45 301 5 0.161
SWCNT (20, 20) 531 7.88 227 2 0.160
SWCNT (50, 50) 213 0.4 92 0.1 0.159
Shen and Li Molecular mechanics MWCNT (10, 10) 1580 40.80 493 10.57 0.156
(2005) (about 1-axis) MWCNT (15, 15) 1410 37.84 398 9.073 0.150
MWCNT (20, 20) 1320 31.22 353 8.151 0.147
MWCNT (50, 50) 1170 22.84 277 5.975 0.139
Song et al. Strain energy approach SWCNT (5, 5) 867.22 341.92 – 136.44 0.163
(2009) (about 1-axis) SWCNT (10, 10) 878.37 338.76 – 136.22 0.171
SWCNT (10, 0) 916.21 311.59 – 124.94 0.142
SWCNT (15, 0) 920.31 316.20 – 124.33 0.143
SWCNT (17, 0) 919.51 320.91 – 126.35 0.125
Tsai et al. (2010) Molecular dynamics SWCNT (10, 0) 1382.5 645 1120 268.75 0.272
simulation (about 1-axis) SWCNT (14, 0) 981.5 504 779.2 210 0.27
SWCNT (18, 0) 759.9 425 596.3 177.08 0.27
It has been experimentally observed that CNTs are actu- the influence of CNT waviness on the elastic moduli of the
ally curved cylindrical tubes with a relatively high aspect CNT-reinforced composite.
ratio (Shaffer and Windle, 1999; Qian et al., 2000; Vigolo Additionally, with CNT waviness, the manufacturing of
et al., 2000; Berhan et al., 2004; Tsai et al., 2011; Tyson et al., unidirectional continuous CNT-reinforced composites in
2011). Images analyzed by Qian et al. (2000) and Berhan large scale has to encounter some challenging difficulties,
et al. (2004) have been illustrated in Figure 2.1a and b, which such as the agglomeration of CNTs, the misalignment, and
clearly demonstrate that the embedded CNTs remains highly the difficulty in manufacturing very long CNTs. Further
curved when dispersed in the polymer matrix. It is hypothe- research on improving the out-of-plane properties of com-
sized that their affinity to become curved is due to their high posites and the better application of short CNTs led to the
aspect ratio and the associated low bending stiffness. Many growth of CNTs on the surface of advanced fibers. For exam-
research investigations revealed that CNT curvature reduces ple, Bower et al. (2000) have grown aligned CNTs on the
the effective elastic properties of the CNT-reinforced com- substrate surface using high-frequency microwave plasma-
posite (Fisher et al., 2002; Berhan et al., 2004; Shi et al., enhanced CVD. Veedu et al. (2006) demonstrated that the
2004; Anumandla and Gibson, 2006). The effect of CNT remarkable improvements in the interlaminar fracture tough-
curvature on the polymer matrix nanocomposite (PMNC) ness, hardness, delamination resistance, in-plane mechani-
stiffness has been investigated by Pantano and Cappello cal properties, damping, and thermoelastic behavior of the
(2008). They concluded that in the presence of weak bond- laminated composite can be obtained by growing MWCNTs
ing, the enhancement of the nanocomposite stiffness can be of about 60 µm long on the surface of fibers. Qiu et al. (2007)
achieved through the bending energy of CNTs rather than presented a method to fabricate multifunctional multiscale
through the axial stiffness of CNTs. Tsai et al. (2011) studied composites through an effective infiltration-based vacuum-
the effects of CNT waviness and its distribution on the effec- assisted resin transfer-molding process. Ray (2010) devel-
tive nanocomposite stiffness. In their study, elastic moduli oped a shear lag model of a novel hybrid smart composite
were overestimated when CNT aspect ratio or waviness fol- to analyze the load transferred to the coated piezoelectric
lowed a symmetric distribution. Farsadi et al. (2012) devel- fibers from the CNT-reinforced matrix in the absence and
oped a three-dimensional finite element model to investigate the presence of the electric field. Such a fiber augmented with
(a) 1.4 (b)
Nanotube wt%
1.2
1
0.8
0.6
2A
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Length scale (μm)
T/2
300 nm 100 nm Mag = 50.00 K X
FIGURE 2.1 Images of CNT-reinforced polymers showing that the embedded CNTs exhibit significant curvature with the polymer. (a)
Transmission electron microscopy image of MWCNT (1 wt%) in polystyrene. (Reproduced from Qian, D. et al. 2000. Applied Physics
Letters 76(20):2868–70. With permission.) (b) Scanning electron microscopy image of CNT showing straight and sinusoidal segments.
(Reproduced from Berhan, L., Yi, Y. B., and Sastry, A. M. 2004. Journal of Applied Physics 95(9):5027–5034. With permission.)
radially grown CNTs on its circumferential surface is being circumferential surface of carbon fiber. Such a resulting
called “fuzzy fiber” (Garcia et al., 2008; Mathur et al., 2008; fuzzy fiber is shown in Figure 2.3. Here, wavy CNTs are
Chatzigeorgiou et al., 2011), and the resulting composite is modeled as sinusoidal solid CNTs (Fisher et al., 2002;
called fuzzy fiber-reinforced composite (FFRC). Recently, a Berhan et al., 2004; Anumandla and Gibson, 2006;
novel continuous FFRC reinforced with zigzag SWCNTs and Pantano and Cappello, 2008; Tsai et al., 2011), while at
carbon fibers was proposed by the authors (Kundalwal and any location along the length of CNT, it is considered as
Ray, 2011). transversely isotropic (Jin and Yuan, 2003; Shen and Li,
The distinct constructional feature of the unidirectional 2004; Shi et al., 2004; Liu et al., 2005; Li and Guo, 2008;
continuous FFRC studied by the authors is that the uni- Song et al., 2009; Tsai et al., 2010) with CNT axis being
formly spaced straight CNTs are radially grown on the cir- the axis of symmetry. Also, the CNTs are grown on the
cumferential surface of carbon fibers. However, in practice, circumferential surface of carbon fiber in such a way that
such radially grown CNTs may be wavy. Since CNT wavi- their axes of transverse isotropy at their roots are normal
ness influences the effective elastic properties of the CNT- to the circumferential surface of carbon fiber. The radially
reinforced composite, CNT waviness may also affect the
effective elastic properties of the FFRC. Such an effect cannot
be predicted from the above-mentioned existing studies as the 2, z
constructional feature of the FFRC is distinct and not simi-
lar to CNT-reinforced composites studied elsewhere. Thus,
a critical issue that is yet to be addressed is to examine the
effect of waviness of CNTs on the effective elastic properties
of the FFRC. In this chapter (Kundalwal and Ray, 2013), an 3, y
endeavor has been made to investigate the effect of waviness
of radially grown CNTs on the effective elastic properties of
the continuous FFRC.
2.2 ARCHITECTURE OF THE FFRC Carbon fiber Polymer Wavy CNTs CFF
CONTAINING WAVY CNTs 1, x
Figure 2.2 shows a schematic of a lamina of the FFRC FIGURE 2.2 Schematic diagram of a lamina made of the FFRC
containing wavy CNTs being studied here. The novel containing wavy CNTs. (Reproduced from Kundalwal, S. I. and
constructional feature of the FFRC is that wavy CNTs Ray, M. C. 2013. ASME Journal of Applied Mechanics 80:021010.
of equal length are radially and uniformly grown on the With permission.)
Polymer CNTs Carbon fiber
PMNC
CFF
Wavy CNTs
FFRC
Carbon fiber
FIGURE 2.5 Modeling of the FFRC and its phases. (Reproduced
from Kundalwal, S. I. and Ray, M. C. 2013. ASME Journal of
Applied Mechanics 80:021010. With permission.)
FIGURE 2.3 Fuzzy fiber with wavy CNTs radially grown on its
circumferential surface.
grown wavy CNTs eventually reinforce the polymer matrix coplanar with the 2–3 (2′–3′) plane or the 1–3 (1′–3′) plane
surrounding the carbon fiber along the direction transverse as shown in Figure 2.4a and b, respectively. Note that the
to the length of carbon fiber. Thus, the combination of the plane passing through the axis of carbon fiber is the 1–3 or
fuzzy fiber with wavy CNTs and the polymer matrix can be 1′–3′ plane, while the plane transverse to the carbon fiber
viewed as a composite fuzzy fiber (CFF) in which a carbon is the 2–3 or 2′–3′ plane. For the various steps involved
fiber is embedded in wavy CNT-reinforced PMNC. It may in the micromechanical modeling of the continuous FFRC
be noted that the variations of the constructional feature containing wavy CNTs, the schematic shown in Figure 2.5
of the CFF can be such that the planes of wavy CNTs are will be followed.
(a) 2 2
Carbon fiber
2′
Wavy CNTs Carbon fiber
3′
ϕ
2R 3 2a 1, 1′
Polymer Polymer
Wavy CNTs 2L
Transverse cross-section Longitudinal cross-sections
3 3
Carbon fiber
(b) 3′
Wavy CNTs Carbon fiber
2′
ϕ
2R 2 2a 1, 1′
Wavy CNTs Polymer Polymer

2L
Transverse cross-section Longitudinal cross-section
FIGURE 2.4 Transverse and longitudinal cross sections of the CFF in which wavy CNTs are coplanar with either the 2–3 (2′–3′) or
the 1–3 (1′–3′) plane. (Reproduced from Kundalwal, S. I. and Ray, M. C. 2013. ASME Journal of Applied Mechanics 80:021010. With
permission.)
2.3 MODELS OF WAVY CNTs of thickness dy. Averaging the effective elastic properties of
these slices over the length (Ln) of the RVE (i.e., the thickness
As mentioned in Section 2.2, a CNT wave is assumed to of the unwound PMNC lamina), the homogenized effective
have a sinusoidal shape. Considering a carbon fiber and an elastic properties of the unwound PMNC can be estimated.
unwound PMNC lamina while this lamina is composed of Each slice can be treated as an off-axis unidirectional PMNC
sinusoidally wavy CNTs which are coplanar with either the lamina, and its elastic coefficients can be determined by trans-
2–3 (2′–3′) plane or the 1–3 (1′–3′) plane as shown in Figure forming the elastic coefficients of the corresponding specially
2.6, the CFF can be viewed to be formed by wrapping the orthotropic lamina.
carbon fiber with the unwound PMNC lamina as shown in Now, these wavy CNTs are characterized by
Figure 2.7. As shown in Figure 2.8, the representative vol-
ume element (RVE) is divided into infinitesimally thin slices
nπ
z = A sin (ωy) or x = A sin (ω y); ω = (2.1)
Ln
(a) Carbon fiber Carbon fiber
2 Polymer
2 corresponding to the plane of CNT waviness being copla-
2′ 3′ nar with the 2–3 plane and the 1–3 plane, respectively. In
Polymer
Equation 2.1, A and Ln are the amplitude of CNT wave and
ϕ the linear distance between the CNT ends, respectively, and n
3 represents the number of CNT waves. The running length of
3
CNT (Lnr) can be expressed in the following form:
Wavy CNTs
1
1 Ln
Wavy CNTs
Unwound PMNC lamina

L nr =
∫ 1 + A 2ω 2 cos2 (ωy) dy (2.2)
0
(b) Carbon fiber Carbon fiber in which the angle θ shown in Figure 2.8 is given by
2 Wavy CNTs Polymer
2
2′ dz dx
3′ tan θ = cos (ωy) or tan θ = cos (ω y)
Polymer
dy = Aω dy = Aω
ϕ
3 corresponding to the plane of CNT waviness being copla-
3
nar with the 2–3 plane and the 1–3 plane, respectively. Note
that for a particular value of ω, the value of θ varies with the
1 1 amplitude of CNT wave.
Unwound PMNC lamina
2.4 EFFECTIVE ELASTIC PROPERTIES
FIGURE 2.6 Unwound lamina of the PMNC containing wavy OF THE CONTINUOUS FFRC
CNTs. CONTAINING WAVY CNTs
This section deals with the procedures for employing the two
2 Interphase micromechanics methods; namely, the mechanics of materi-
Interphase Polymer 2′ Wavy CNTs als (MOM) approach and the Mori–Tanaka (MT) method for
2 estimating the effective elastic properties of the FFRC con-
taining wavy CNTs.
3′
ϕ
2.4.1 MOM Approach
3 3
This section presents the derivation of analytical microme-
chanics models based on the MOM approach for estimating
Carbon fiber
the effective elastic properties of the PMNC material, CFF,
Wavy CNTs
Interphase and FFRC.
Unwound PMNC lamina Carbon filter
2.4.1.1 Effective Elastic Properties of the PMNC
FIGURE 2.7 Transverse cross sections of the CFF with unwound In order to estimate the effective elastic properties of the PMNC
and wound PMNC. (Reproduced from Kundalwal, S. I. and Ray, material surrounding the carbon fiber, the consideration of the
M. C. 2013. ASME Journal of Applied Mechanics 80:021010. With nonbonded van der Waals interaction between an atom of a
permission.) CNT and an atom of the polymer matrix is an important issue.
2
(a) Interphase
2, z
θ
3
CNT Interphase
dy
A
3, y
λ Polymer
Ln
Carbon fiber
1
(b) 1, x Interphase
θ
3
CNT Interphase
dy
A
3, y
λ Polymer
Ln
Carbon fiber
FIGURE 2.8 RVEs of the unwound PMNC material containing a wavy CNT radially grown on the carbon fiber. (a) CNT is coplaner with
yz plane, (b) CNT is coplaner with xy plane. (Reproduced from Kundalwal, S. I. and Ray, M. C. 2013. ASME Journal of Applied Mechanics
80:021010. With permission.)
The atomic structure of CNTs consisting of sp2-hybridized wavy CNTs where CNT axis makes an angle θ with the 3
carbon atoms hinders the formation of the strong covalent (3′)–axis can be approximated by transforming the effective
bonds between the carbon atoms of a CNT and the surround- elastic properties of the unwound lamina of PMNC contain-
ing polymer matrix. Naturally, the bonding between a CNT ing straight CNTs. Figure 2.9 illustrates the cross sections
and the surrounding polymer takes place through van der of the RVE of the unwound lamina with straight CNTs. The
Waals noncovalent interactions which are obviously weaker three-phase micromechanics model based on the MOM
than covalent bonding between carbon–carbon bonds (Li and approach derived by Ray (2010) has been modified to predict
Chou, 2003). Following the previous research (Odegard et al.,
2003; Seidel and Lagoudas, 2006; Tsai et al., 2010; Li et al.,
2011a,b), an interphase may be considered between a CNT
2 Polymer matrix
and the polymer matrix which characterizes the nonbonded 2
CNT fiber
van der Waals interaction, and researchers treated this inter-
Interphase
phase as an equivalent solid continuum. The effective elastic
properties of such interphase resembling a solid continuum
3 1
can be determined by molecular dynamics simulation and
are readily available in the open literature (Tsai et al., 2010).
CNT fiber Polymer matrix
Thus, because of the consideration of this interphase between Interphase
a CNT and the polymer matrix, the micromechanical model
of the unwound PMNC by the MOM approach turns out to be FIGURE 2.9 Cross sections of the RVE of the unwound PMNC
a three-phase MOM model. material. (Reproduced from Kundalwal, S. I. and Ray, M. C. 2011.
It may be noted that the effective elastic properties at any International Journal of Mechanics and Materials in Design
point in the unwound PMNC lamina containing sinusoidally 7(2):149–166. With permission.)
the effective elastic constant matrix [Cnc] of the unwound σ1n  σ1i  σ1p  σ1nc 
PMNC material with straight CNTs. On the basis of the prin-  n   i   p   nc 
cipal material coordinate (1–2–3) axes shown in Figure 2.2, σ 2  σ 2  σ 2  σ 2 
the constitutive relations for the constituent phases of the 3n  3i  3p  3nc 
PMNC are given by  n  =  i  =  p  =  nc  (2.4)
σ 23  σ 23  σ 23  σ 23 
σ13 n  σ1i 3  σ13p  σ13 nc 
{σ r} = [Cr ]{ r}; r = n, i, and p (2.3)  n   i   p   nc 

σ12  σ12  σ12  σ12 
where the state of stress vector, the state of strain vector and and
the elastic coefficient matrix of the rth phase are
1n  1i  1p  1nc 
 n   i   p   nc 
{σ } = [σ1 σ 2 σ 3 σ 23 σ13 σ12 ] , { } = [1 2 3 23 13 12 ]
r r r r r r r T r r r r r r r T
 2   2   2   2 
and σ 3n  σ 3i  σ p  σ nc 
v n  n  + v i  i  + v p  p3  =  nc3  (2.5)
23  23  23  23 
 C11
r r
C12 r
C13 0 0 0  13n  13
i  13p  nc 
13
 r r r   n   i   p   nc 
C12 C22 C23 0 0 0 
12  12  12  12 
 Cr r
C23 r
C33 0 0 0 
[Cr ] =  13 .
 0 0 0 r
C44 0 0  In Equation 2.5, vn, vi, and vp represent the volume frac-
 0 tions of CNT fiber, interphase, and polymer material, respec-
0 0 0 r
C55 0 
 r
 tively, present in the RVE of the PMNC, and are determined in
 0 0 0 0 0 C66  Section 2.5. Also, the superscript nc represents the unwound

PMNC material containing straight CNTs. Substituting
In Equation 2.3, the superscripts n, i, and p denote, respec- Equation 2.3 into Equations 2.4 and 2.5, the stress and strain
tively, CNT, interphase, and polymer matrix. For the con- vectors in the unwound PMNC material containing straight
stituent phase denoted by r, σ1r , σ 2r , and σ 3r are the normal CNTs can be expressed in terms of the corresponding stress
stresses along the principal material coordinate axes 1, 2, and and strain vectors of the constituent phases as follows:
3, respectively; 1r , 2r , and 3r are the corresponding normal
strains; σ12
r
, σ13
r
, and σ 23
r r
are the shear stresses; 12 r
, 13 r
, and 23 {σ nc} = [C1 ]{ n} + [C2 ]{ i} + [C3 ]{ p} (2.6)
r
are the shear strains; and Cij (i, j = 1, 2, 3,…, 6) are the elastic
coefficients. { nc} = [V1 ]{ n} + [ V2 ]{ i} + [V3 ]{ p} (2.7)

Several researchers (Gao and Li, 2005; Li and Chou,
2003, 2009; Odegard et al., 2003; Thostenson and Chou,
[C4 ]{ n} − [C5 ]{1} = 0 (2.8)
2003; Seidel and Lagoudas, 2006; Jiang et al., 2009a,b; Li
et al., 2011b) assumed a perfectly bonding condition between
a CNT and the polymer matrix. Esteva and Spanos (2009) [C5 ]{ i} − [C6 ]{ p} = 0 (2.9)
emphatically reported that the imperfect bonding does not
affect the effective longitudinal Young’s modulus of the The various matrices appearing in Equations 2.6 through
CNT-reinforced polymer matrix composite and marginally 2.9 are given by
affects the transverse elastic properties of the composite
for high-volume fraction (>0.8) of CNTs. In their extensive  C11
n
C12 n
C13 n
0 0 0 
 n n n 
research, Tsai et al. (2010) investigated the effect of the inter-  C12 C22 C23 0 0 0 
phase between a CNT and the polymer matrix formed due v Cn v n C23n
v n C33n
0 0 0 
to the nonbonded van der Waals interaction for estimating [C1 ] =  n 13 ,
 0 0 0 Cn44 0 0 
the effective elastic properties of CNT-reinforced polymer  0
matrix composite. They found that the interphase margin- 0 0 0 n
C55 0 
 n

ally enhances the effective Young’s modulus of the com-  0 0 0 0 0 C66 
posite transverse to CNT fiber over that of the composite
without the consideration of the interphase. Hence, in the  0 0 0 0 0 0
MOM approach being presented here, it is assumed that  0 0 0 0 0 0 

CNT is perfectly bonded to the neighboring phases. The iso- C13i i
C23 i
C33 0 0 0
field conditions and the rules of mixture for satisfying the [C2 ] = v i  ,
perfect bonding conditions between a CNT and the neigh-  0 0 0 0 0 0
 0 0 0 0 0 0
boring phases can be expressed as (Smith and Auld, 1991;  
Benveniste and Dvorak, 1992; Ray, 2010).  0 0 0 0 0 0 

 0 0 0 0 0 0 Using Equations 2.7 through 2.9, the local strain vectors

 0 { n} and { p} can be expressed in terms of the composite
 0 0 0 0 0  strain { nc} and subsequently, using them in Equation 2.6, the
C13p p
C23 p
C33 0 0 0 following constitutive relation between the states of stresses
[ C3 ] = v p  ,
 0 0 0 0 0 0 and strains at any point in the unwound PMNC material con-
 0 0 0 0 0 0 taining straight CNTs is obtained:
 
 0 0 0 0 0 0 
{σ nc} = [Cnc ]{ nc} (2.11)

 C11
n n
C12 n
C13 0 0 0 
 n n n  where the effective elastic coefficient matrix [Cnc] of the lam-
C12 C22 C23 0 0 0 
 0 ina of the unwound PMNC containing straight CNTs is given
0 1 0 0 0 
[C4 ] =  , by
 0 0 0 Cn44 0 0 
 0 n
0 

0 0 0 C55
 [Cnc ] = [C1 ][V5 ]−1 + [C7 ][ V6 ]−1 (2.12)
n

 0 0 0 0 0 C66 
in which
 C11
i
C1i 2 i
C13 0 0 0 
 i i i 
C12 C22 C23 0 0 0  [C7 ] = [C3 ] + [C2 ][C5 ]−1[C6 ],
 0 0 1 0 0 0 
[C5 ] =  , [V4 ] = [V3 ] + [V2 ][C5 ]−1[C6 ],
 0 0 0 Ci44 0 0 
 0 0 0 0 i
C55 0  [V5 ] = [V1 ] + [V4 ][C6 ]−1[C4 ], and
 i
 (2.13)
 0 0 0 0 0 C66 
[V6 ] = [V4 ] + [ V1 ][C4 ]−1[C6 ]
 C11
p p
C12 p
C13 0 0 0 
 p p p  Once [Cnc] is computed by Equation 2.12, the effective
C12 C22 C23 0 0 0 
 0 elastic coefficients (CNC
ij ) at any point in the unwound PMNC
0 1 0 0 0 
[C6 ] =  , lamina where CNT is inclined at an angle θ with the 3 (3′)–
 0 0 0 Cp44 0 0  axis can be derived in a straightforward manner by employing
 0 0 0 0 p
C55 0  the appropriate transformation law. Thus, if the plane of wavy
 
 0 0 0 0 0 p
C66  CNTs is coplanar with the 2–3 (2′–3′) plane, the effective
elastic coefficients (CNC
ij ) at any point in the unwound PMNC
v n 0 0 0 0 0 lamina are given by
0 vn 0 0 0 0 
 NC
C11 = C11
nc NC
, C12 = C12 k + C13
nc 2 nc 2 NC
1 , C13 = C12 1 + C13
nc 2 nc 2
k ,
0 0 1 0 0 0
[V1 ] =  ,
2 = C22 k + C331 + 2(C23 + 2C 44 ) k 1 ,
C2NC nc 4 nc 4 nc nc 2 2
0 0 0 vn 0 0
0 0 0 0 vn 0
  NC
C23 = (C22
nc
+ C33
nc
− 4Cnc
44 ) k 1 + C23 ( k + 1 ),
2 2 nc 4 4
 0 0 0 0 0 v n 
NC
C33 = C22 1 + C33
nc 4
k + 2(C23
nc 4 nc
+ 2 Cnc 2 2
44 ) k 1 ,
vi 0 0 0 0 0
0 vi 0 0 0 0 
NC
C44 = (C22
nc
+ C33
nc
− 2C23
nc
− 2C44
nc
)k 212 + C44
nc
( k 4 + 14 ),

0 0 0 0 0 0
C55 = C55 k + C661 and C66 = C551 + C66 k
NC nc 2 nc 2 NC nc 2 nc 2
[V2 ] =   and
0 0 0 vi 0 0
(2.14)
0 0 0 0 v1 0
 
 0 0 0 0 0 v1  in which
v p 0 0 0 0 0 − 21
0   nπA  nπy  
2
 vp 0 0 0 0  k = cos θ = 1 +  cos  
0   L n  L n  
0 0 0 0 0  and
[V3 ] =  
0 0 0 vp 0 0 − 21
 nπy    nπA  nπy   
2
0 0 0 0 vp 0 nπA
1 = sin θ = cos  1 +  cos  L   
  Ln  L n    L n  
 0 0 0 0 0 v p   n
(2.10)
Similarly, if the plane of wavy CNTs is coplanar with From Equation 2.17, it is obvious that the effective elastic
the 1–3 (1′–3′) plane, then the effective elastic coefficients properties at any point of the PMNC surrounding the carbon
I(cI NC
ij ) at any point in the unwound PMNC lamina where fiber with respect to the principal material coordinate (1–2–3)
CNT is inclined at an angle θ with the 3 (3′)–axis are axes of the FFRC vary over an annular cross section of the
given by PMNC phase of the RVE of the CFF. The volume average of
these effective elastic properties [CPMNC ] over the volume of
NC
C11 = C11 k + C33
nc 4
1 + 2(C13
nc 4 nc
+ 2C55
nc
)k 212 , the PMNC can be treated as the homogenized effective elastic
properties of the PMNC with sinusoidally wavy CNTs sur-
NC
C12 = C12 k + C2nc312 ,
nc 2
rounding the carbon fiber. However, without loss of general-
ity, it may be considered that the volume averages of these
NC
C13 = (C11
nc
+ C33
nc
− 4C55
nc
)k 212 + C13
nc
( k 4 + 14 ),
effective elastic properties [CPMNC ] over the volume of the
NC
C22 = C2nc2 , C23
NC
= C12 1 + C23
nc 2 nc 2
k , PMNC can be treated as the homogenized effective elastic
properties [CPMNC] of the PMNC material surrounding the
NC
C33 = C11 1 + C33
nc 4
k + 2(C13
nc 4 nc
+ 2C55
nc
)k 212 , carbon fiber with respect to the 1–2–3 coordinate axes of the
FFRC and are given by
44 = C 44 k + C661 ,
CNC nc 2 nc 2
2π R
NC
C55 = (C11
nc
+ C33
nc
− 2C13
nc
− 2C55
nc
)k 212 + C55
nc
( k 4 + 14 ), 1
NC
= C 1 +C k
nc 2 nc 2
[C PMNC
]=
π( R − a 2 )
2 ∫ ∫ [C PMNC
] r dr dθ (2.18)
and C (2.15)
0 a
66 44 66
It is now obvious that the effective elastic properties of Thus, the effective constitutive relations for the PMNC
the unwound PMNC lamina with wavy CNTs vary along the material with respect to the principal material coordinate
length of CNT as the value of θ varies over the length of CNT. (1–2–3) axes of the FFRC can be expressed as
The average effective elastic coefficient matrix [CNC ] of the
lamina of such unwound PMNC material containing wavy
{σ PMNC} = [CPMNC ]{ PMNC} (2.19)
CNTs can be obtained by averaging the transformed elastic
coefficients (CNC
ij ) over the linear distance between CNT ends
as follows (Hsiao and Daniel, 1996): 2.4.1.2 Effective Elastic Properties of the CFF
The effective elastic properties of the CFF can be predicted
Ln
1 by estimating the effective elastic properties of a lamina
[CNC ] =
Ln ∫ [C NC
] dy (2.16) of continuous unidirectional fiber reinforced composite in
which the carbon fiber is the reinforcement and the matrix
0
phase is the PMNC material. The cross sections of the RVE
It may also be noted that when the carbon fiber is viewed of such lamina have been illustrated in Figure 2.10. Here,
to be wrapped by such unwound PMNC lamina, the matrix the length of carbon fiber aligns with the 1–direction. The
[CNC ] provides the effective elastic properties at a point MOM approach derived in Section 2.4.1.1 is augmented to
located in the PMNC where CNT axis (3′–axis) is oriented estimate the effective elastic properties of the CFF. Similar
at an angle ф with the 3–axis in the 2–3 plane. Hence, at any to Equations 2.4 and 2.5, the iso-field conditions and the
point in the PMNC surrounding the carbon fiber, the effective rules of mixture for the RVE shown in Figure 2.10 can be
elastic coefficient matrix [CPMNC ] of the PMNC with respect written as
to the 1–2–3 coordinate system turns out to be location depen-
dant and can be determined by the following transformations:
[CPMNC ] = [ T]− T [CNC ][T]−1 (2.17) 2

PMNC 2
Carbon fiber
where
1 0 0 0 0 0
0 m2 n2 mn 0 0 
1 3

0 n2 m2 −mn 0 0
[T] =   Carbon fiber PMNC
0 −2mn 2mn m2 − n2 0 0
0 0 0 0 m −n 
  FIGURE 2.10 Cross sections of the RVE of the CFF. (Reproduced
0 0 0 0 n m  from Kundalwal, S. I. and Ray, M. C. 2011. International Journal
of Mechanics and Materials in Design 7(2):149–166. With
with m = cosф and n = sinф. permission.)
1f  1PMNC  1CFF  v PMNC C11

PMNC PMNC
v PMNC C12 PMNC
v PMNC C13
 f   PMNC   CFF  
σ 2  σ 2  σ 2 
PMNC PMNC PMNC
 C12 C22 C23
σ 3f  σ 3PMNC  σ 3CFF   C13 PMNC PMNC
C23 PMNC
C33
 f  =  PMNC  =  CFF  (2.20) [C9 ] = 
σ 23  σ 23  σ 23   0 0 0
σ13f  σ13PMNC  σ13CFF   0 0 0
 f   PMNC   CFF  
σ12  σ12  σ12   0 0 0
and 0 0 0 
0 0 0 

σ1f  σ1PMNC  σ1CFF  0 0 0 
 f   PMNC   CFF  ,
 2   2   2  CP44MNC 0 0 
 
f  PMNC  CFF  0 PMNC
C55 0 
vf  3f  + v PMNC  3PMNC  =  3CFF  (2.21) PMNC

23  23  23  0 0 C66 
13f  13 PMNC  13
CFF 
 f   PMNC   CFF   1 0 0 0 0 0 
12  12  12  CPMNC C PMNC
C PMNC
0 0 0 
 12 22 23 
C13
PMNC
C PMNC
23 C PMNC
33 0 0 0 
In Equation 2.21, vf and vPMNC are the volume fractions of [C11 ] =  ,
 0 0 0 CPMNC
44 0 0 
the carbon fiber and the PMNC material, respectively, with  0 PMNC 
0 0 0 C55 0
respect to the volume of the RVE of the CFF, and have been  
PMNC
determined in Section 2.5. Using Equations 2.19 through 2.21,  0 0 0 0 0 C66 
and following the procedure for deriving Equation 2.12, the
0 0 0 0 0 
constitutive relations of the CFF can be obtained as follows: 0 v
 f 0 0 0 0 
{σ CFF} = [CCFF ]{CFF} (2.22) 0 0 vf 0 0 0
[V7 ] =   and
0 0 0 vf 0 0
in which the effective elastic coefficient matrix [CCFF] of the 0 0 0 0 vf 0
 
CFF is given by 0 0 0 0 0 vf 
[CCFF ] = [C8 ][V9 ]−1 + [C9 ][ V10 ]−1 (2.23) 1 0 0 0 0 0 

0 v 0 0 0 0 
 PMNC 
The various matrices appearing in Equation 2.23 are 0 0 v PMNC 0 0 0 
[V8 ] =   (2.24)
0 0 0 v PMNC 0 0 
[V9 ] = [V7 ] + [V8 ][C11 ]−1[C10 ], 0 0 0 0 v PMNC 0 
 
[ V10 ] = [ V8 ] + [V7 ][C10 ]−1[C11 ], 0 0 0 0 0 v PMNC 

C11f f
C12 f
C13 0 0 0
  2.4.1.3 Effective Elastic Properties of the FFRC
 0 0 0 0 0 0
It may be reiterated that the RVE of the FFRC lamina can
 0 0 0 0 0 0
[C8 ] = vf  , be viewed as being comprised of a CFF and the polymer
 0 0 0 0 0 0 material. The cross sections of such an RVE are shown in
 0 0 0 0 0 0 Figure 2.11. In order to satisfy the perfectly bonding situation
  between the CFF and the polymer, the iso-field conditions and
 0 0 0 0 0 0 
the rules of mixture appropriate for this RVE are given by
 1 0 0 0 0 0 
 Cf 1CFF  1P  1 
 12 Cf22 Cf23 0 0 0   CFF   P   
 C13
f
Cf23 f
C33 0 0 0  σ 2  σ 2  σ 2 
[C10 ] =  , σ 3CFF  σ 3P  σ 3 
 0 0 0 Cf44 0 0   CFF  =  P  =   (2.25)
 0 σ 23  σ 23  σ 23 
0 0 0 Cf55 0  σ13 CFF  σ13P  σ13 
 
 0 0 0 0 0 f
C66   CFF   P   

σ12  σ12  σ12 
2 Polymer matrix 2 1 0 0 0 0 0
CFF 0
 vP 0 0 0 0 
0 0 vP 0 0 0
[V12 ] =  
1 3 0 0 0 vP 0 0
0 0 0 0 vP 0
 
CFF Polymer matrix 0 0 0 0 0 vP 
C11
CFF CFF
C12 CFF
C13 0 0 0
FIGURE 2.11 Cross sections of the RVE of the FFRC.  
(Reproduced from Kundalwal, S. I. and Ray, M. C. 2011.  0 0 0 0 0 0
International Journal of Mechanics and Materials in Design  0 0 0 0 0 0
7(2):149–166. With permission.) [C12 ] = vCFF  ,
 0 0 0 0 0 0
 0 0 0 0 0 0
and  
 0 0 0 0 0 0 
σ1CFF  σ1P  σ1   vpC11

P P
vpC12 P
vpC12 0 0 0 
 CFF   P    P P P 
 2   2   2   C12 C11 C12 0 0 0 
3CFF   P  3   CP CP
CP
0 0 0 
vCFF  CFF  + vp  3P  =   (2.26) [C13 ] =  12 12 11
P
,
23  23  23   0 0 0 C44 0 0 
13 CFF  13
P  13   0 0 0 0 P
C44 0 
 CFF   P    P

12  12  12   0 0 0 0 0 C44 

 1 0 0 0 0 0 
In Equation 2.26, vCFF and vP are the volume fractions of CCFF
 12 CCFF CCFF 0 0 0 
the CFF and the polymer matrix, respectively, with respect 22 23
to the volume of the FFRC, and have been determined in C13

CFF
CCFF
23
CFF
C33 0 0 0 
[C14 ] =  
Section 2.5. It may be noted that unlike the constituent  0 0 0 CCFF
44 0 0 
phases, the symbols denoting the states of stresses and strains  0 0 0 0 CCFF 0 
55
in the FFRC are written without using the superscript. Using  CFF

Equations 2.22, 2.25, and 2.26, and following the procedure  0 0 0 0 0 C66 
for deriving Equation 2.12, the constitutive relations for the
 1 0 0 0 0 0 
FFRC are derived as follows: CP
 12 C P
11 CP
12 0 0 0 
{σ} = [C]{} (2.27) C12P
C P
CP
0 0 0 
and 
12 11

 0 0 0 CP44 0 0 
where the effective elastic coefficient matrix [C] of the FFRC  0 0 0 0 CP44 0 
is given by  
 0 0 0 0 0 CP44 

−1 −1
[C] = [C12 ][V13 ] + [C13 ][ V14 ] (2.28) (2.29)

with 2.4.2 MT Method

The micromechanics model (i.e., the MOM approach) of
[V13 ] = [V11 ] + [V12 ][C15 ]−1[C14 ], the FFRC, derived in the preceding section, is based on the
assumptions delineated by Equations 2.4 and 2.5, respec-
[V14 ] = [V12 ] + [V11 ][C14 ]−1[C15 ], tively, which imply the perfect bonding condition between a
0 0 0 0 0 0  CNT fiber and the neighboring phases. However, it is required
0 to justify the validity of these assumptions for modeling the
 vCFF 0 0 0 0  perfect bonding conditions of the MOM approach. For this
0 0 vCFF 0 0 0  purpose, another micromechanics model based on the MT
[V11 ] =  ,
0 0 0 vCFF 0 0  method which does not require to satisfy the relations similar
0 0 0 0 vCFF 0  to Equations 2.4 and 2.5 will be presented here. If the inter-
  phase between a CNT and the polymer matrix is considered,
0 0 0 0 0 vCFFF 
then the micromechanical model of the unwound PMNC with
straight CNTs will be a three-phase MT model. Utilizing the 3 − 4 νi 5 − 4 νP

elastic properties of CNT, interphase, and polymer matrix
n
S1212 = , Si
1111 = Si
2222 = , S13333 = 0,
8(1 − νi ) 8(1 − νP )
properties, the three-phase MT method can be derived for the
unwound PMNC material with straight CNTs. The explicit
formulation of such three-phase MT method can be derived 4 νp − 1 νp
as (Dunn and Ledbetter, 1995)
i
S1122 = S2211
i
= , Si
1133 = Si
2233 = ,
8(1 − νp ) 2(1 − νp )
[Cnc ] = v p [Cp ][I] + v n [Cn ][A n ] + v i [Ci ][A i ] S13311 = S13322 = 0,
−1
× v p [I] + v n [A n ] + v i [A1 ] (2.30)
3 − 4v p
i
S1313 = S2323
i
=1 i
and S1212 =
4 8(1 − v p )
The concentration tensors [An] and [Ai] appearing in
Equation 2.30 are given by
where νi and νp denote the Poisson’s ratios of the interphase,
−1
[A n ] = [I] + [Sn ]{([C p ])−1 ([Cn ] − [Cp ])} (2.31) and the polymer matrix, respectively. Once [Cnc] is computed
from Equation 2.30, Equations 2.14 through 2.18 are used
to estimate the average effective elastic coefficient matrix
−1
[A i ] = [I] + [Si ]{([C p ])−1 ([Ci ] − [Cp ])} (2.32) [CPMNC] of the PMNC material surrounding the carbon fiber
containing wavy CNTs.
Also, in the above matrices, [Sn] and [Si] indicate the The effective elastic properties of the CFF can be predicted
Eshelby tensors for the domains denoted by n and i, respec- by estimating the effective elastic properties of a composite in
tively, and [I] is an identity matrix. As assumed in Section which the carbon fiber is the reinforcement and the matrix
2.4.1.1, a CNT may be treated as a solid circular cylinder. phase is the PMNC material. Thus, according to the two-
Thus, the specific form of the Eshelby tensor for the cylindri- phase model by the MT method (Mori and Tanaka, 1973),
cal inclusion given by Qui and Weng (1990) is utilized to com- the effective elastic coefficient matrix for the CFF is given by
pute the matrices [S↓ n] and [Si]. The elements of the Eshelby
tensors for the cylindrical CNT reinforcement in the isotropic [CCFF ] = [CPMNC ] + vf ([Cf ] − [CPMNC ])[A 2 ] (2.34)
interphase and the cylindrical interphase in the isotropic poly-
mer matrix are explicitly written as follows (Qui and Weng, in which the matrices of the strain concentration factors are
1990): as follows:
 S1111
n n
S1122 n
S1133 0 0 0 −1
 n  2 ] v PMNC[I] + vf [A
[A 2 ] = [A 2 ]
S2211
n
S2222 n
S2233 0 0 0   and
Sn n
S3322 n
S3333 0 0 0
[Sn ] =  3311  and
 0 0 0 n
S2323 0 0 2 ] = [I] + [Sf ]([CPMNC ])−1 ([Cf ] − [CPMNC ])  −1 (2.35)
[A
 
 0 0 0 0 n
S1313 0
 n

 0 0 0 0 0 S1212  where the Eshelby tensor [Sf ] is computed based on the prop-
erties of the PMNC matrix and the shape of the carbon fiber.
 S1111
i i
S1122 i
S1133 0 0 0 It is worth noting that the PMNC matrix is transversely isotro-
 i i i  pic and, consequently, the Eshelby tensor (Li and Dunn, 1998)
S2211 S2222 S2233 0 0 0 
Si i
S3322 i
S3333 0 0 0  corresponding to transversely isotropic material is utilized for
[Si ] =  3311 i
 computing [Sf ] while the inclusion is a circular cylindrical
 0 0 0 S2323 0 0  fiber. The elements of the Eshelby tensor for the cylindrical
 0 0 0 0 i
S1313 0 
  carbon fiber embedded in the transversely isotropic PMNC
 0 0 0 0 0 i
S1212  material are explicitly given by (Li and Dunn, 1998)

(2.33)
 S1111
f f
S1122 f
S1133 0 0 0 
in which  f 
S2211 Sf2222 Sf2233 0 0 0 
Sf f
S3322 f
S3333 0 0 0 
5 − 4 νi 4 νi − 1 [Sf ] =  3311 
n
S1111 = S2222
n
= n
, S3333 = 0, S1122
n
= S2211
n
= , f
8(1 − ν )
i
8(1 − νi )  0 0 0 S2323 0 0 
 0 f 
0 0 0 S1313 0
 f

νi 1  0 0 0 0 0 S1212 
n
S1133 = S2233
n
= n
, S3311 = S3322
n
= 0, S1313
n
= S2323
n
=
2(1 − ν )
p
4 (2.36)
in which
 S1111 S1122 S1133 0 0 0 
S S2222 S2233 0 0 0 
PMNC
5C11 + C12
PMNC  2211 
f
S1111 = Sf2233 = , S3311 S3322 S3333 0 0 0 
PMNC
8C11 [SCFF ] =  
 0 0 0 S2323 0 0 
3C PMNC
12 −C
PMNC
11  0 0 0 0 S1313 0 
f
S1122 = Sf2211 = PMNC
,  
8C11  0 0 0 0 0 S1212 

(2.38)
PMNC
C13 1
f
S1133 = S2233
f
= PMNC
f
, S3311 = S3322
f
= 0, S1313
f
= Sf2323 =
2C11 4 in which
5 − 4 νp νp
3CPMNC − CPMNC S1111 = 0, S2222 = S3333 = , S2211 = S3311 = ,
S f
3333 = 0 and S f
1212 = 11 PMNC12 . 8(1 − ν )
p
2(1 − νp )
8C11
4v p − 1 1
Finally, considering the CFF as the cylindrical inclusion S2233 = S3322 = , S1122 = S1133 = 0, S1313 = S1212 =
embedded in the isotropic polymer matrix, the effective elas- 8(1 − v p ) 4
tic properties [C] of the FFRC can be determined by utiliz- and
ing the two-phase MT method (Mori and Tanaka, 1973) as
follows: 3 − 4 νp
S2323 =
8(1 − νp )
[C] = [C ] + vCFF ([C

p CFF
] − [C ])[A 3 ]
p
(2.37)
2.5 RESULTS AND DISCUSSION
in which the matrices of the strain concentration factors are In this section, numerical results for the effective elastic prop-
given by erties of the FFRC containing wavy CNTs are evaluated using
the two different micromechanical models derived in Sections
3 ][vp [I] + vCFF [A
[A 3 ] = [A 3 ]]−1 and 2.4.1 and 2.4.2. Zigzag SWCNTs, carbon fiber, and polyimide
material are used for evaluating the numerical results. Their
3 ] = [I] + [SCFF ](Cp )−1 ([CCFF ] − [Cp ])  −1
[A material properties are summarized in Table 2.3. The effec-
  tive elastic properties and the thickness of the hollow circular
cylindrical continuum representing the interphase between a
in which the Eshelby tensor [SCFF] is computed based on the CNT and the polyimide matrix are also listed in Table 2.3.
properties of the polymer matrix and the shape of the CFF. Volume fraction of CNTs (VCNT) in the FFRC depends on
The elements of the Eshelby tensor for the cylindrical CFF CNT diameter, running length of CNT, carbon fiber diam-
reinforcement in the isotropic polymer matrix are explicitly eter, and surface-to-surface distance between two adjacent
given by (Qui and Weng, 1990) radially aligned CNTs at their roots. If there are no other
TABLE 2.3
Material Properties of the Constituent Phases of the FFRC
Material C11 (GPa) C12 (GPa) C13 (GPa) C23 (GPa) C33 (GPa) C44 (GPa) C66 (GPa) (nm)
(10, 0) CNTa 709.90 172.40 240.00 240.00 1513.10 1120.00 268.70 dn = 0.780
Interphasea 29.60 15.20 15.20 15.20 29.60 7.20 7.20 Gap = 0.333
(14, 0) CNTa 557.50 137.50 187.70 187.70 1082.80 779.20 210.00 dn = 1.100
Interphasea 27.63 14.23 14.23 14.23 27.63 6.70 6.70 Gap = 0.324
(18, 0) CNTa 472.90 118.70 159.70 159.70 846.10 596.30 177.10 dn = 1.420
Interphasea 27.55 14.19 14.19 14.19 27.55 6.68 6.68 Gap = 0.316
Carbon fiberb 236.40 10.60 10.60 10.70 24.80 7.00 25.00 d = 10000.000
Polyimidec 9.00 6.00 6.00 6.00 9.00 1.50 1.50 −
a Tsai et al. (2010).

b Honjo (2007).
c Odegard et al. (2005).
phases or materials in between CNTs, it is reported (Jiang where D = 2R. The volume (Vf ) of the carbon fiber is
et al., 2009a,b) that the minimum surface-to-surface distance
between two adjacent CNTs is the equilibrium van der Waals π 2
distance, which is about 0.34 nm. Since in the PMNC mate- Vf = dL (2.40)
8
rial, polymer molecules fill the gap between CNTs and the
formation of the interphase is also considered, the surface-to- where d = 2a. Thus, the carbon fiber volume fraction in the
surface distance between the two adjacent CNTs at their roots FFRC (νf ) can be expressed as
is considered as 1.7 nm. Recall that the FFRC lamina can be
viewed as being comprised of CFFs and the polymer matrix
(see Figure 2.2). For fibers with circular cross section, it is a Vf π d2
vf = FFRC
= 2
(2.41)
well-known fact that the hexagonal packing array is the opti- V 2 3 D
mal packing of fibers and the corresponding maximum fiber
volume fraction is 0.9069. Hence, in the FFRC, the hexagonal Using Equation 2.41, the carbon fiber volume fraction (vf )
packing array of CFFs is considered as shown in Figure 2.12 and the PMNC volume fraction (vPMNC) in the CFF can be
for evaluating the numerical results. It is also assumed that the derived as
radially grown CNTs are uniformly spaced on the circumfer-
ential surface of carbon fiber. Note that although the volume π 2
fraction of the CFF is 0.9069, the volume fraction of carbon dL
2 3
vf = 8 = vf and v PMNC = 1 − vf (2.42)
fiber with respect to the volume of FFRC is much less than π 2 π
0.9069. DL
8
The determination of VCNT in the FFRC containing wavy
CNTs is an important issue. It is obvious that the construc- The maximum number (NCNT)max of radially grown aligned
tional feature of the FFRC imposes a constraint on the maxi- wavy CNTs on the circumferential surface of carbon fiber is
mum value of VCNT as the number of wavy CNTs grown given by
on the circumferential surface of a particular carbon fiber
diameter is limited. Since the radially grown wavy CNTs are
πdL
uniformly spaced on the circumferential surface of carbon (N CNT )max = (2.43)
fiber, the maximum number of CNTs grown on the circum- 2(d n + 1.7)2
ferential surface of carbon fiber of particular diameter can
be determined. Referring to Figure 2.12, the cross section The volume (VCNT) of wavy CNTs is
of the RVE of the FFRC can be considered as an equilateral
triangle. The volume fractions of different constituent phases π 2
VCNT = d n L nr (N CNT )max (2.44)
of the PMNC, CFF, and FFRC can be determined from this 4
equilateral triangle. First, the volume of the RVE of the
FFRC (VFFRC) is determined from this equilateral triangle and the volume (Vi) of the CNT/polymer interphase is
as follows:
π 2 2
3 2 Vi = [d i − d n ](R − a)(N CNT )max (2.45)
V FFRC = DL (2.39) 4
4
where di is the outer diameter of the RVE of the hollow cylin-
CFFs
drical CNT/polymer interphase.
Wavy CNTs
Thus, the maximum CNT volume fraction (VCNT)max with
respect to the volume of the FFRC containing wavy CNTs can
be determined as follows:
VCNT πd 2n L nr
(VCNT )max = FFRC
= vf (2.46)
V d(d n +1.7)2
Since the lamina of the FFRC has been made of the hex-
agonal packing array of CFFs and the polymer matrix, the
volume fractions of the CFF (vCFF) and the polymer matrix
Carbon fiber Polymer (vp ) with respect to the volume of the FFRC are given by
FIGURE 2.12 Hexagonal packing array comprised of CFFs. π

(Reproduced from Kundalwal, S. I. and Ray, M. C. 2013. ASME vCFF = ≅ 0.9069 and v p = 1 − vCFF (2.47)
2 3
Journal of Applied Mechanics 80:021010. With permission.)
The maximum CNT volume fraction with respect to the 0.12

volume of the PMNC (vn)max can be determined in terms of CNT (10, 0)
CNT (14, 0)
(VCNT)max as follows: 0.1
CNT (18, 0)
CNT volume fraction (VCNT)

VCNT 2 3 0.08
(v n )max = PMNC
= (VCNT )max (2.48)

V (π − vf 2 3 )
0.06
Finally, the volume fractions of the CNT/polymer matrix
interphase (vi) and the polymer matrix (vp) with respect to the 0.04
volume of the PMNC can be determined in terms of (Vn)max
as follows:
0.02
Vi (d 2 − d 2n )
vi = = 2i and v p = 1 − (v n )max − (v i )
V PMNC (d n )(v n )max
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
(2.49) A/Ln
The carbon fiber volume fraction can vary according to

FIGURE 2.13 Variation of the maximum CNT volume fraction in
the manufacturing route, typically from 0.3 to 0.7. In this
the FFRC with the waviness factor (ω = 12π/Ln).
study, the value of the diameter of carbon fiber is assumed
as 2a = 10 µm, and its volume fraction (vf ) is considered as
0.6. The diameter of the CFF and the length of straight CNT value of A/L n. It may also be noted from Figure 2.13 that for
become 12.2943 µm and 1.1472 µm, respectively, when the a particular value of waviness factor (A/L n ), the value of VCNT
value of vf is 0.6. The degree of CNT waviness is charac- is increased with the increase in the value of integer designat-
terized by waviness factor, A/L n. It should be noted that for ing a zigzag CNT. This is due to the fact that as the integer
straight CNT, the value of waviness factor is zero. Unless oth- for designating a zigzag CNT increases, the diameter of CNT
erwise mentioned, amplitude variations of CNT waviness in increases.
the 1–3 and 2–3 planes have been considered for a fixed value It may be noted that the micromechanics models derived
of CNT wave frequency (ω = 12π /L n ). Also, unless other- in Sections 2.4.1 and 2.4.2 are applicable for estimating the
wise mentioned, the maximum amplitude for CNT is consid- effective elastic properties of any continuous CNT-reinforced
ered as A = 100dn based on the diameter of a zigzag (10, 0) composite. Hence, in order to verify the validity of these
CNT which yields the upper limit for the value of waviness micromechanics models, CNT and the matrix material of the
factor as A/L n = 0.068. It is evident from Equations 2.2 and nanocomposite studied by Liu and Chen (2003) are consid-
2.46 that the increase in the amplitude of CNT increases the ered for the constituents of the unwound PMNC material with
running length of CNT which eventually increases the maxi- straight CNTs. The engineering constants of this unwound
mum CNT volume fraction (VCNT)max in the FFRC. Figure PMNC material computed by these micromechanics mod-
2.13 illustrates the variation of the maximum value of CNT els are compared with those of the same predicted by Liu
volume fraction in the FFRC with waviness factor for the three and Chen (2003) using the finite element model. Table 2.4
types of zigzag CNTs. It may be observed from this figure that illustrates this comparison and it may be observed that the
the maximum value of VCNT increases with the increase in the three sets of results are in excellent agreement validating the
TABLE 2.4
Comparison of the Engineering Constants of the Unwound PMNC Material with Straight CNTs
E1/Ep E2/Ep ν12, ν13
En/Epa FEMb MOM MT FEMb MOM MT FEMb MOM MT
5 1.1948 1.1948 1.1948 1.1737 1.0528 1.0666 0.3 0.3 0.3
10 1.4384 1.4384 1.4384 1.3336 1.0722 1.0912 0.3 0.3 0.3
Note: En and Ep are the Young’s moduli of CNT and the polymer matrix, respectively; E1 and E2 are the axial Young’s modulus and
the transverse Young’s modulus of the unwound PMNC, respectively; νn and νp are the Poisson’s ratios of CNT and the
polymer matrix, respectively; ν12 and ν23 are the axial Poisson’s ratio and the transverse Poisson’s ratio of the unwound
PMNC, respectively.
a En = 1000 GPa, νn = 0.3, νp = 0.3, and CNT volume fraction, ν = 0.04871.
n
b Liu and Chen (2003).
micromechanics models derived in this chapter. This agree- 15

ment also ensures the validity of the assumptions adopted for
the MOM approach.
First, the effective elastic properties of the PMNC are
computed by employing the MOM approach and the MT 10
method. It should be noted that the models by the MOM
C12 (GPa)
approach and the MT method estimate the effective elastic
properties of the PMNC with the presence of the CNT/poly-
mer interphase. Subsequently, the estimated effective elastic
properties of the PMNC are used to compute the effective 5
elastic properties of the CFF in which the carbon fiber is MOM approach (1–3 plane)
the reinforcement and the matrix phase is the PMNC mate- MT method (1–3 plane)
rial. However, for the sake of brevity, the effective elastic MOM approach (2–3 plane)
MT method (2–3 plane)
properties of the PMNC and the CFF are not presented here. 0
Figure 2.14 illustrates the variation of the effective elastic 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
A/Ln
coefficient C11 of the FFRC with waviness factor. It may be
observed that the effective values of C11 predicted by the MT
method excellently agree with those predicted by the MOM FIGURE 2.15 Variation of the effective elastic coefficient C12 of
approach while wavy CNTs are coplanar with both the 1–3 the FFRC with the waviness factor (ω = 12π/Ln).
(1′–3′) and the 2–3 (2′–3′) planes. It may also be observed
from Figure 2.14 that the effective values of C11 of the FFRC coplanar with the 1–3 plane. The constructional feature of
are not affected by the variations of the amplitude of wavy the FFRC reveals that the FFRC would be a transversely
CNTs in the 2–3 plane. This is attributed to the fact that isotropic material. This is corroborated by the prediction of
when wavy CNTs are coplanar with the 2–3 plane, the value C13 which is identical to that of C12 as shown in Figure 2.16.
NC
of C11 of the PMNC is not affected by CNT waviness as Figure 2.17 demonstrates that the increase in the value of
given by Equation 2.14. When wavy CNTs are coplanar with waviness factor decreases the value of C22 when CNT wavi-
the 1–3 plane, Equation 2.15 reveals that the transformed ness is coplanar with the 1–3 plane, whereas the value of
NC
value of C11 of the PMNC varies with the amplitude of CNT C22 enhances for the higher values of waviness factor when
waves. Hence, the effective value of the elastic coefficient CNT waviness is coplanar with the 2–3 plane. Also, it may
C11 of the FFRC varies with the amplitude of CNT waves be noted that the MT method slightly overestimates the
being coplanar with the 1–3 plane as shown in Figure 2.14. value of C22 as compared to that estimated by the MOM
Both the MOM approach and the MT method yield identi- approach. Since the FFRC is a transversely isotropic mate-
cal estimates for the value of C12 as shown in Figure 2.15. rial, identical results are also obtained for the effective elastic
It may be observed that CNT waviness causes significant coefficient C33 but are not presented here. Similar trend of
increase in the effective value of C12 when wavy CNTs are results has been obtained for the effective elastic coefficient
15
180
160
140
120 10
C13 (GPa)
C11 (GPa)
100
80
60 5
MOM approach (1–3 plane) MOM approach (1–3 plane)
40
MT method (1–3 plane) MT method (1–3 plane)
20 MOM approach (2–3 plane) MOM approach (2–3 plane)
MT method (2–3 plane) MT method (2–3 plane)
0 0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
A/Ln A/Ln
FIGURE 2.14 Variation of the effective elastic coefficient C11 of FIGURE 2.16 Variation of the effective elastic coefficient C13 of
the FFRC with the waviness factor (ω = 12π/Ln). the FFRC with the waviness factor (ω = 12π/Ln).
35 C23 as shown in Figure 2.18. Figure 2.19 demonstrates that

the increase in the value of waviness factor increases the
30 value of C55 when CNT waviness is coplanar with the 1–3
plane, whereas the value of C55 is slightly improved for the
25 higher values of waviness factor when CNT waviness is
coplanar with the 2–3 plane. Also, it may be noted that
C22 (GPa)
20
the MT method overestimates the values of C55 as com-
15
pared to those estimated by the MOM approach. Although
not shown here, identical results are also obtained for the
10 effective elastic coefficient C66.
MOM approach (1–3 plane) It is important to note from Figures 2.14 through 2.19 that
5 if wavy CNTs are coplanar with the 1–3 (1′–3′) plane, then
MOM approach (2–3 plane)
MT method (2–3 plane) the effective values of C11, C12, C13, and C55 of the FFRC are
0 significantly improved over their values with straight CNTs
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
A/Ln (ω = 0) for the higher values of waviness factor. The pre-
dicted elastic coefficients of the FFRC containing sinusoi-
dally wavy CNTs in the present study are found to be coherent
FIGURE 2.17 Variation of the effective elastic coefficient C22 of
with the previously reported results (Pantano and Cappello,
the FFRC with the waviness factor (ω = 12π/Ln).
2008; Tsai et al., 2011; Farsadi et al., 2012) for the CNT-
15
reinforced composite containing sinusoidally wavy CNTs
with symmetric distributions. When a wavy CNT is copla-
nar with the 1–3 (1′–3′) plane, the amplitudes of CNT waves
become parallel to the 1–axis. This results into the aligning
of the projections of parts of CNT lengths with the 1–axis
10
leading to the stiffening of the polymer matrix along the 1–
C23 (GPa)
axis. The enhancement of the effective elastic coefficients of

the FFRC presented in Figures 2.14, 2.15, 2.16, and 2.19 for
the values of C11, C12, C13, and C55, respectively, is attributed
5 to such increase in the axial stiffness along the 1–direction
MOM approach (1–3 plane) when CNT waves are coplanar with the 1–3 or the (1′–3′)
MT method (1–3 plane) plane. On the other hand, if wavy CNTs are coplanar with the
MOM approach (2–3 plane) 2–3 (2′–3′) plane then the transverse elastic coefficients of
0 the FFRC are improved over their values with straight CNTs
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 (ω = 0), and the reverse is true when wavy CNTs are coplanar
A/Ln
with the 1–3 (1′–3′) plane.
Comparison of the model by the MOM approach with
FIGURE 2.18 Variation of the effective elastic coefficient C23 of the MT method reveals that the MOM approach yields con-
the FFRC with the waviness factor (ω = 12π/Ln). servative estimates for the effective elastic coefficients of
the FFRC. Hence, predictions by the MOM approach have
25 been considered in the subsequent results for investigat-
MOM approach (1–3 plane)
MT method (1–3 plane) ing the effect of CNT diameter, presence of CNT/polymer
20 MOM approach (2–3 plane) matrix interphase and higher values of CNT wave frequen-
MT method (2–3 plane) cies on the effective elastic properties of the FFRC con-
sidering wavy CNTs to be coplanar with the 1–3 (1′–3′)
15 plane. Figures 2.20 and 2.21 illustrate the comparisons of
C55 (GPa)
the effective elastic coefficients C11 and C22 of the FFRC for
10
different values of CNT diameter, respectively. Figure 2.20
reveals that CNT diameter does not influence the effective
elastic coefficient C11 of the FFRC, and the same is true
5 for the elastic coefficients C12 and C55 but are not presented
here. Figure 2.21 depicts that the increase in CNT diameter
marginally increases the value of C22 for the higher values
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 of waviness factor. Although not presented here, the same
A/Ln is true for the effective elastic coefficients C23 and C44. This
may be attributed to the fact that as the diameter of CNT
FIGURE 2.19 Variation of the effective elastic coefficient C55 of increases the value of CNT volume fraction in the FFRC
the FFRC with the waviness factor (ω = 12π/Ln). increases as depicted in Figure 2.13 which results in the
180 180
160 160
140 140
120 120
100 100
C11 (GPa)
C11 (GPa)
80 80
60 60
Three-phase MOM approach, CNT (10, 0)
40 40
CNT (10, 0) Two-phase MOM approach, CNT (10, 0)
20 CNT (14, 0) 20 Three-phase MOM approach, CNT (18, 0)
0 0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
A/Ln A/Ln
the FFRC with the waviness factor while the wavy CNTs are copla- the FFRC with the waviness factor while the wavy CNTs are copla-
nar with the 1–3 (1′–3′) plane (ω = 12π/Ln). nar with the 1–3 (1′–3′) plane (ω = 12π/Ln).
25 25
20 20
15 15
C22 (GPa)
C22 (GPa)
10 10

5 CNT (10, 0) 5 Two-phase MOM approach, CNT (10, 0)
CNT (14, 0) Three-phase MOM approach, CNT (18, 0)
0 0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
A/Ln A/Ln
nar with the 1–3 (1′–3′) plane (ω = 12π/Ln). nar with the 1–3 (1′–3′) plane (ω = 12π/Ln).
increase in the values of the effective elastic coefficients of the two-phase MOM approach for the higher values of wavi-
the PMNC. ness factor. Although not presented here, similar predictions
Next, the effective elastic properties of the FFRC con- have also been obtained for the effective elastic coefficients
taining wavy CNTs are computed by employing the MOM C23 and C44.
approach with the presence and absence of the CNT/polymer So far, in this chapter, the effect of wavy CNTs on the
interphase as illustrated in Figures 2.22 through 2.24. To effective elastic properties of the FFRC have been studied by
investigate the effect of the interphase on the effective elastic considering CNT wave frequency (ω) as 12π/L n for a par-
properties of the FFRC, zigzag (10, 0) and (18, 0) CNTs are ticular value of waviness factor (A/L n ). To this end, however,
considered to be coplanar with the 1–3 (1′–3′) plane. From the variation of CNT wave frequency for a particular value of
Figures 2.22 and 2.24, it may be observed that the models A/L n would be an important study. For this, the discrete val-
by the two- and three-phase MOM approach predict almost ues of wave frequency of zigzag (18, 0) CNT are considered
identical estimates for the effective values of C11 and C55, as 5π/L n 10π/L n 15π/L n and 20π/L n. Figures 2.25 through
respectively. From Figure 2.23 it may be observed that the 2.27 illustrate the comparisons of the effective elastic coef-
three-phase MOM approach provides a slightly enhanced ficients of the FFRC for different values of CNT wave fre-
estimate for the value of C22 as compared to that predicted by quencies. Figure 2.25 illustrates that the increase in the value
10 25
8 20
6 15
C22 (GPa)
C55 (GPa)
4 10
ω = 5π/Ln
Three-phase MOM approach, CNT (10, 0) ω = 10π/Ln
2 Two-phase MOM approach, CNT (10, 0) 5
ω = 15π/Ln
Two-phase MOM approach, CNT (18, 0) ω = 20π/Ln
0 0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
A/Ln A/Ln
nar with the 1–3 (1′–3′) plane (ω = 12π/Ln). nar with the 1–3 (1′–3′) plane ((18, 0) CNT).
12
225
200 10
175
8
150
C 55 (GPa)
C11 (GPa)
125 6
100
4 ω = 5π/Ln
75 ω = 5π/Ln
ω = 10π/Ln
50 ω = 10π/Ln
2 ω = 15π/Ln
ω = 15π/Ln
25 ω = 20π/Ln
ω = 20π/Ln
0
0 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
A/Ln A/Ln

the FFRC with the waviness factor while the wavy CNTs are copla-
the FFRC with the waviness factor while the wavy CNTs are copla-
nar with the 1–3 (1′–3′) plane ((18, 0) CNT).
nar with the 1–3 (1′–3′) plane ((18, 0) CNT).
of CNT wave frequency significantly enhances the value of of the FFRC is significantly enhanced initially and then stabi-
C11. This is attributed to the fact that as CNT wave frequency lized for the higher values of waviness factor.
increases the value of CNT volume fraction and the aligned
projections of parts of CNT lengths with the 1–axis increases.
2.6 CONCLUSIONS
The more the value of CNT wave frequency, the more will be
such projections, and hence the axial stiffness increases with In this chapter, the effect of CNT waviness on the effective
the increase in CNT wave frequency. Similar predictions for elastic properties of a novel continuous FFRC has been stud-
the effective elastic coefficient C12 have also been obtained ied considering the sinusoidally wavy CNTs to be coplanar
but are not presented here. Figure 2.26 reveals that, in general, with either of the two mutually orthogonal planes. Analytical
the value of C22 of the FFRC decreases with increase in the micromechanics models based on the MOM approach and the
values of CNT wave frequency and waviness factor. Although MT method are employed to determine the effective elastic
not shown here, the same is true for the effective elastic coef- properties of this novel composite. The effective elastic prop-
ficients C23 and C44. Figure 2.27 depicts that the value of C55 erties of the FFRC are estimated in the presence and absence
of an interphase between a CNT and the polymer matrix. Blake, P., Brimicombe, P. D., Nair, R. R., Booth, T. J., Jiang, D.,
The following important conclusions are drawn from the Schedin, F. et al. 2008. Graphene-based liquid crystal device.
study carried out in this chapter. Nano Letters 8:1704.
Bose, S., Kuila, T., Uddin, M. E., Kim, N. H., Lau, A. K. T., and
Lee, J. H. 2010. In-situ synthesis and characterization of elec-
1. The effective values of the axial elastic coefficients trically conductive polypyrrole/graphene nanocomposites.
of the FFRC containing wavy CNTs being coplanar Polymer 51:5921–5928.
with the 1–3 plane are significantly improved over Bower, C., Zhu, W., Jin, S., and Zhou, O. 2000. Plasma-induced
their values with straight CNTs (ω = 0) for the higher alignment of carbon nanotubes. Applied Physics Letters
values of waviness factor and CNT wave frequency. 77(7):830–832.
Chatzigeorgiou, G., Seidel, G. D., and Lagoudas, D. C. 2011.
The effect of wavy CNTs being coplanar with the
Effective mechanical properties of “fuzzy fiber composites.
2–3 plane is found to be marginal on the effective Composites Part B 43:2577–2593.
axial elastic coefficients of the FFRC. Dreyer, R. D., Park, S., Bielawski, C. W., and Ruoff, R. S. 2010.
2. CNT waviness in the 2–3 plane causes improvement The chemistry of graphene oxide. Chemical Society Reviews
in the effective transverse elastic coefficients of the 39:228–240.
FFRC compared to that of the composite with straight Dunn, M. L. and Ledbetter, H. 1995. Elastic moduli of composites
CNTs (ω = 0). The effective transverse elastic coeffi- reinforced by multiphase particles. ASME Journal of Applied
Mechanics 62(4):1023–1028.
cients are significantly improved for the higher values
Esteva, M. and Spanos, P. D. 2009. Effective elastic properties of
of waviness factor and CNT wave frequency when nanotube reinforced composites with slightly weakened inter-
wavy CNTs are coplanar with the 2–3 plane. faces. Journal of Mechanics of Materials and Structures
3. The diameter of zigzag CNT negligibly affects the 4(5):887–900.
effective elastic coefficients of the FFRC containing Fan, H., Wang, L., Zhao, K., Li, N., Shi, Z., Ge, Z. et al. 2010.
wavy CNTs. Fabrication, mechanical properties, and biocompatibility of
4. The nonbonded van der Waals interaction between a graphene-reinforced chitosan composites. Biomacromolecules
11:2345.
CNT and the polymer matrix negligibly affects the
Farsadi, M., Ochsner, A., and Rahmandoust, M. 2012. Numerical
effective elastic coefficients of the FFRC containing investigation of composite materials reinforced with
wavy CNTs. Thus, for conservative and intuitive esti- waved carbon nanotubes. Journal of Composite Materials
mates, one may neglect the nonbonded van der Waals 47:1425–1434.
interaction between a CNT and the polymer matrix, Fisher, F. T., Bradshaw, R. D., and Brinson, L. C. 2002. Effects of
and consider the perfect bonding between them. nanotube waviness on the modulus of nanotube-reinforced
polymers. Applied Physics Letters 80(24):4647–4649.
Gao, X. L. and Li, K. 2005. A shear-lag model for carbon nano-
The present study reveals that wavy CNTs can be properly tube-reinforced polymer composites. International Journal of
used to construct novel advanced nanocomposites with supe- Solids and Structures 42:1649–1667.
rior elastic properties. Garcia, E. J., Wardle, B. L., Hart, A. J., and Yamamoto, N. 2008.
Fabrication and multifunctional properties of a hybrid lami-
nate with aligned carbon nanotubes grown in situ. Composites
REFERENCES
Science and Technology 68(9):2034–2041.
Ansari, S. and Giannelis, E. P. 2009. Functionalized graphene sheet- Geim, A. K. and Novoselov, A. K. 2007. The rise of graphene.
poly(vinylidene fluoride) conductive nanocomposites. Journal Nature Materials 6:183–191.
of Polymer Science Part B: Polymer Physics 47(9):888–897. Ghosh, S., Calizo, I., Teweldebrhan, D., Pokatilov, E. P., Nika, D. L.,
Anumandla, V. and Gibson, R. F. 2006. A comprehensive closed Balandin, A. A. et al. 2008. Extremely high thermal conduc-
form micromechanics model for estimating the elastic modu- tivity of graphene: Prospects for thermal management appli-
lus of nanotube-reinforced composites. Composites Part A cations in nanoelectronic circuits. Applied Physics Letters
37(12):2178–2185. 92:151911.
Bae, S., Kim, H., Lee, Y., Xu, X., Park, J. S., Zheng, Y. et al. 2010. Honjo, K. 2007. Thermal stresses and effective properties calculated
Roll-to-roll production of 30-inch graphene films for transpar- for fiber composites using actual cylindrically-anisotropic
ent electrodes. Nature Nanotechnology 5:574–578. properties of interfacial carbon coating. Carbon 45(4):865–872.
Balandin, A. A., Ghosh, S., Bao, W., Calizo, I., Teweldebrhan, D., Hsiao, H. M. and Daniel, I. M. 1996. Elastic properties of compos-
Miao, F. et al. 2008. Superior thermal conductivity of single- ites with fiber waviness. Composites Part A 27(10):931–941.
layer graphene. Nano Letters 8:902. Huang, J., Wang, X., deMello, A. J., deMello, J. C., and Bradley, D.
Bao, W., Miao, F., Chen, Z., Zhang, H., Jang, W., Dames, C., and Lau, D. C. 2007. Efficient flexible polymer light emitting diodes
C. N. 2009. Controlled ripple texturing of suspended graphene with conducting polymer anodes. Journal of Materials
and ultrathin graphite membranes. Nature Nanotechnology Chemistry 17:3551–3554.
4:562–566. Huang, Y., Wu, J., and Hwang, K. C. 2006. Thickness of gra-
Benveniste, Y. and Dvorak, G. J. 1992. Uniform fields and uni- phene and single-wall carbon nanotubes. Physical Review
versal relations in piezoelectric composites. Journal of the B 74:245413.
Mechanics and Physics of Solids 40(6):1295–1312. Iijima, S. 1991. Helical microtubules of graphitic carbon. Nature
Berhan, L., Yi, Y. B., and Sastry, A. M. 2004. Effect of nanorope 354:56–58.
waviness on the effective moduli of nanotube sheets. Journal Iijima, S. and Ichihashi, T. 1993. Single-shell carbon nanotubes of
of Applied Physics 95(9):5027–5034. 1-nm diameter. Nature 363:603–605.
Jiang, B., Liu, C., Zhang, C., Liang, R., and Wang, B. 2009a. Lourie, O. and Wagner, H. D. 1998. Evaluation of Young’s modulus
Maximum nanotube volume fraction and its effect on over- of carbon nanotubes by micro-Raman spectroscopy. Journal
all elastic properties of nanotube-reinforced composites. of Materials Research 13(9):2418–2422.
Composites Part B 40(3):212–217. Liu, J. Z., Zheng, Q. S., Wang, L. F., and Jiang, Q. 2005. Mechanical
Jiang, J. W., Wang, J. S., and Baowen, L. 2009b. Thermal expan- properties of single-walled carbon nanotube bundles as bulk
sion in single-walled carbon nanotubes and graphene: materials. Journal of the Mechanics and Physics of Solids
Nonequilibrium Green’s function approach. Physical Review 53(1):123–142.
B 80:205429. Liu, Y. J. and Chen, X. L. 2003. Evaluations of the effective mate-
Jin, Y. and Yuan, F. G. 2003. Simulation of elastic properties of rial properties of carbon nanotube-based composites using
single-walled carbon nanotubes. Composites Science and a nanoscale representative volume element. Mechanics of
Technology 63:1507–1515. Materials 35(1):69–81.
Kauffman, D. R. and Star, A. 2010. Graphene versus carbon nano- Lu, J. P. 1997. Elastic properties of carbon nanotubes and nanoropes.
tubes for chemical sensor and fuel cell applications. Analyst Physical Review Letters 79(7):1297–1300.
135(11):2745–3012. Lv, C., Xue, Q., Xia, D., Ma, M., Xie, J., and Chen, H. 2010. Effect
Kim, K. S., Zhao, Y., Jang, H., Lee, Y. S., Kim, J. M., Kim, K. S. of chemisorptions on the interfacial bonding characteristics
et al. 2009. Large-scale pattern growth of graphene films for of graphene-polymer composites. The Journal of Physical
stretchable transparent electrodes. Nature 457:706–710. Chemistry C 114:6588–6594.
Krishnan, A., Dujardin, E., Ebbesen, E. W., Yianilos, P. N., and Mathur, R. B., Chatterjee, S., and Singh, B. P. 2008. Growth of car-
Treacy, M. M. J. 1998. Young’s modulus of single-walled bon nanotubes on carbon fiber substrates to produce hybrid/
nanotubes. Physical Review B 58(20):14013–14019. phenolic composites with improved mechanical properties.
Kuila, T., Bose, S., Hong, C. E., Uddin, M. E., Khanra, P., Kim, Composites Science and Technology 68:1608–1615.
N. H. et al. 2011. Preparation of functionalized graphene/lin- Meguid, S. A., Wernik, J. M., and Cheng, Z. Q. 2010. Atomistic-
ear low density polyethylene composites by a solution mixing based continuum representation of the effective properties of
method. Carbon 49:1033–1037. nano-reinforced epoxies. International Journal of Solids and
Kundalwal, S. I. and Ray, M. C. 2011. Micromechanical analysis Structures 47:1723–1736.
of fuzzy fiber reinforced composites. International Journal of Mermin, N. D. 1968. Crystalline order in two dimensions. Physical
Mechanics and Materials in Design 7(2):149–166. Review 176:250–254.
Kundalwal, S. I. and Ray, M. C. 2013. Effect of carbon nanotube Milowska, K., Birowska, M., and Majewski, J. A. 2012. Mechanical
waviness on the elastic properties of the fuzzy fiber reinforced and electrical properties of carbon nanotubes and graphene
composites. ASME Journal of Applied Mechanics 80:021010. layers functionalized with amines. Diamond and Related
Landau, L. D. and Lifshitz, E. M. 1980. Statistical Physics Part 1. Materials 23:167–17.
Pergamon Press, Oxford. Mori, T. and Tanaka, K. 1973. Average stress in matrix and aver-
Lee, C., Wei, X., Kysar, J. W., and Hone, J. 2008. Measurement of age elastic energy of materials with misfitting inclusions. Acta
the elastic properties and intrinsic strength of monolayer gra- Metallurgica 21(5):571–574.
phene. Science 321:385. Novoselov, K. S., Geim, A. K., Morozov, S. V., Jiang, D., Zhang,
Lee, C., Wei, X. D., Li, Q. Y., Carpick, R., Kysar, J. W., and Hone, Y., Dubonos, S. V., Grigorieva, I. V., and Firsov, A. A. 2004.
J. 2009. Elastic and frictional properties of graphene. Physica Electric field effect in atomically thin carbon films. Science
Status Solidi B—Basic Solid State Physics 246:2562–2567. 306:666–669.
Li, C. and Chou, T. W. 2003. Multiscale modeling of carbon nano- Odegard, G. M., Clancy, T. C., and Gates, T. S. 2005. Modeling of
tube reinforced polymer composites. Journal of Nanoscience the mechanical properties of nanoparticle/polymer compos-
and Nanotechnology 3(5):423–430. ites. Polymer 46(2):553–562.
Li, C. and Chou, T. W. 2009. Failure of carbon nanotube/polymer Odegard, G. M., Gates, T. S., Wise, K. E., Park, C., and Siochi,
composites and the effect of nanotube waviness. Composites E. J. 2003. Constitutive modeling of nanotube-reinforced
Part A 40(10):1580–1586. polymer composites. Composites Science and Technology
Li, H. and Guo, W. 2008. Transversely isotropic elastic properties of 63(11):1671–1687.
single-walled carbon nanotubes by a rectangular beam model Pantano, A. and Cappello, F. 2008. Numerical model for compos-
for the C–C bonds. Journal of Applied Physics 103:103501. ite material with polymer matrix reinforced by carbon nano-
Li, J. Y. and Dunn, M. L. 1998. Anisotropic coupled-field inclu- tubes. Meccanica 43(2):263–270.
sion and inhomogeneity problems. Philosophical Magazine A Park, H., Rowehl, J. A., Kim, K. K., Bulovic, V., and Kong, J.
77(5):1341–1350. 2010. Doped graphene electrodes for organic solar cells.
Li, X., Cai, W., An, J., Kim, S., Nah, J., Yang, D. et al. 2009. Large- Nanotechnology 21:505204.
area synthesis of high-quality and uniform graphene films on Popov, V. N., Van Doren, V. E., and Balkanski, M. 2000. Elastic
copper foils. Science 324:1312–1314. properties of single-walled carbon nanotubes. Physical
Li, X., Wang, X., Zhang, L., Lee, S., and Dai, H. 2008. Chemically Review B 61(4):3078–3084.
derived, ultrasmooth graphene nanoribon semiconductor. Potts, J. R., Dreyer, D. R., Bielawski, C. W., and Ruoff, R. S.
Science 319:1229–1232. 2011. Graphene-based polymer nanocomposites. Polymer
Li, Y., Waas, A. M., and Arruda, E. M. 2011a. A closed-form, hier- 52:5–25.
archical, multi-interphase model for composites—Derivation, Qian, D., Dickey, E. C., Andrews, R., and Rantell, T. 2000. Load
verification and application to nanocomposites. Journal of the transfer and deformation mechanisms in carbon nano-
Mechanics and Physics of Solids 59(1):43–63. tube-polystyrene composites. Applied Physics Letters
Li, Y., Waas, A. M., and Arruda, E. M. 2011b. The effects of the 76(20):2868–2870.
interphase and strain gradients on the elasticity of layer by Qiu, J., Zhang, C., Wang, B., and Liang, R. 2007. Carbon nano-
layer (LBL) polymer/clay nanocomposites. International tube integrated multifunctional multiscale composites.
Journal of Solids and Structures 48(6):1044–1053. Nanotechnology 18(27):275708.
Qui, Y. P. and Weng, G. J. 1990. On the application of Mori- Treacy, M. M. J., Ebbesen, T. W., and Gibson, J. M. 1996.
Tanaka’s theory involving transversely isotropic spheroidal Exceptionally high Young’s modulus observed for individual
inclusions. International Journal of Engineering Science carbon nanotubes. Nature 381:678–680.
28(11):1121–1137. Tsai, C. H., Zhang, C., Jack, D. A., Liang, R., and Wang, B. 2011.
Ramanathan, T., Abdala, A. A., Stankovich, S., Dikin, D. A., Alonso, The effect of inclusion waviness and waviness distribution on
M. H., Piner, R. D. et al. 2008. Functionalized graphene elastic properties of fiber-reinforced composites. Composites
sheets for polymer nanocomposites. Nature Nanotechnology Part B 42(1):62–70.
3:327–331. Tsai, J. L., Tzeng, S. H., and Chiu, Y. T. 2010. Characterizing
Ray, M. C. 2010. The concept for a novel hybrid smart composite elastic properties of carbon nanotube/polyimide nanocom-
reinforced with radially aligned zigzag carbon nanotubes posites using multi-scale simulation. Composites Part B
on piezoelectric fibers. Smart Materials and Structures 41(1):106–115.
19:035008. Tsoukleri, G., Parthenios, J., Papagelis, K., Jalil, R., Ferrari, A. C.,
Seidel, G. D. and Lagoudas, D. C. 2006. Micromechanical analysis Geim, A. K. et al. 2009. Subjecting a graphene monolayer to
of the effective elastic properties of carbon nanotube rein- tension and compression. Small 5:2397.
forced composites. Mechanics of Materials 38:884–907. Tyson, B. M., Abu Al-Rub, R. K., Yazdanbakhsh, A., and Grasley,
Seol, J. H., Jo, I., Moore, A. L., Lindsay, L., Aitken, Z. H., Pettes, M. Z. 2011. Carbon nanotubes and carbon nanofibers for enhanc-
T. et al. 2010. Two-dimensional phonon transport in supported ing the mechanical properties of nanocomposite cementitious
graphene. Science 328:213. materials. ASCE Journal of Materials in Civil Engineering
Shaffer, M. S. P. and Windle, A. H. 1999. Fabrication and charac- 23:1028–1035.
terization of carbon nanotube/poly(vinyl alcohol) composites. Veedu, V. P., Cao, A., Li, X., Ma, K., Soldano, C., Kar, S., Ajayan, P.
Advanced Materials 11(11):937–941. M., and Ghasemi-Nejhad, M. N. 2006. Multifunctional com-
Shen, L. and Li, J. 2004. Transversely isotropic elastic properties of posites using reinforced laminae with carbon-nanotube for-
single-walled carbon nanotubes. Physical Review B 69:045414. ests. Nature Materials 5:457–462.
Shen, L. and Li, J. 2005. Transversely isotropic elastic properties of Vigolo, B., Penicaud, A., Coulon, C., Sauder, C., Pailler, R.,
multi-walled carbon nanotubes. Physical Review B 71:035412. Journet, C., Bernier, P., and Poulin, P. 2000. Macroscopic
Shi, D. L., Feng, X. Q., Huang, Y. Y., Hwang, K. C., and Gao, H. fibers and ribbons of oriented carbon nanotubes. Science
2004. The effect of nanotube waviness and agglomeration on 290(5495):1331–1334.
the elastic property of carbon nanotube-reinforced compos- Wang, G., Shen, X., Wang, B., Yao, J., and Park, J. 2009. Synthesis
ites. ASME Journal of Engineering Materials and Technology and characterization of hydrophilic and organophilic gra-
126(3):250–257. phene nanosheets. Carbon 47:1359–1364.
Singh, V., Daeha, J., Zhai, L., Das, S., Khodarkar, S. I., and Seal, Wang, G., Yang, J., Park, J., Gou, X., Wang, B., Liu, H., and
S. 2011. Graphene based materials: Past, present and future. Yao, J. 2008. Facile synthesis and characterization of gra-
Progress in Material Science 56:1178–1271. phene nanosheets. The Journal of Physical Chemistry C
Smith, W. A. and Auld, B. A. 1991. Modeling 1-3 composite piezo- 112(22):8192–8195.
electrics: Thickness mode oscillations. IEEE Transactions on Wernik, J. M., Cornwell-Mott, B. J., and Meguid, S. A. 2012.
Ultrasonics, Ferroelectrics and Frequency Control 38(1):40–47. Determination of the interfacial properties of carbon nanotube
Song, W. L., Cao, M. S., Lu, M. M., Bi, S., Wang, C. Y., Liu, J., Yuan, reinforced polymer composites using atomistic-based contin-
J., and Fan, L. Z. 2014. Flexible graphene/polymer compos- uum model. International Journal of Solids and Structures
ite films in sandwich structures for effective electromagnetic 49:1852–1863.
interference shielding. Carbon 66:67–76. Wernik, J. M. and Meguid, S. A. 2010. Recent developments in mul-
Song, X., Ge, Q., and Yen, S. C. 2009. A first-principles study on tifunctional nanocomposites using carbon nanotubes. ASME
the elastic properties of single-walled carbon nanotubes. Applied Mechanics Review 63:050801.
Proceedings of the Institution of Mechanical Engineers, Part Wu, J., Becerril, H. A., Bao, Z., Liu, Z., Chen, Y., and Peumans,
N: Journal of Nanoengineering and Nanosystems 223:163. P. 2008. Organic solar cells with solution-processed graphene
Stankovich, S., Dikin, D. A., Dommett, G. H. B., Kohlhaas, K. M., transparent electrodes. Applied Physics Letters 92:263302.
Zimney, E. J., Stach, E. A., Piner, R. D., Nguyen, S. T., and Yang, X., Mei, T., Yang, J., Zhang, C., Lv, M., and Wang, X. 2014.
Ruoff, R. S. 2006. Graphene based polymer composites. Synthesis and characterization of alkylamine-functionalized
Nature 442:282–286. graphene for polyolefin-based nanocomposites. Applied
Thostenson, E. T. and Chou, T. W. 2003. On the elastic properties of Surface Science 305:725–731.
carbon nanotube-based composites: Modelling and character- Zhang, K., Zhang, L. L., Zhao, X. S., and Wu, J. 2010. Graphene/
ization. Journal of Physics D: Applied Physics 36(5):573–582. polyaniline nano fiber composites as supercapacitor elec-
Timo, S., Burg, B. R., Schirmer, N. C., and Poulikakos, D. 2009. trodes. Chemistry of Materials 22:1392–1401.
An electrical method for the measurement of the thermal and Zhao, X., Zhang, Q., and Chen, D. 2010. Enhanced mechanical
electrical conductivity of reduced graphene oxide nanostruc- properties of graphene-based poly(vinyl alcohol) composites.
tures. Nanotechnology 20:405704. Macromolecules 43:2357–2363.
3 Effects of Vacancies, Nitrogen
Atoms, and sp3 Bonds on Mechanical
Properties of Graphene Using
Molecular Dynamics Simulations
Akihiko Ito and Shingo Okamoto
CONTENTS
Abstract........................................................................................................................................................................................ 41
3.1 Introduction........................................................................................................................................................................ 41
3.2 Computational Method....................................................................................................................................................... 43
3.2.1 Potential.................................................................................................................................................................. 43
3.2.2 Analysis Model....................................................................................................................................................... 44
3.2.2.1 Pristine Graphene.................................................................................................................................... 44
3.2.2.2 Vacancy-Containing Graphene................................................................................................................ 44
3.2.2.3 Nitrogen-Containing Graphene............................................................................................................... 44
3.2.2.4 Graphene Containing Both Vacancies and Nitrogen Atoms................................................................... 45
3.2.2.5 Graphene Containing Both Interlayer sp3 Bonds and Nitrogen Atoms................................................... 45
3.2.3 MD Simulation....................................................................................................................................................... 46
3.3 Result and Discussion......................................................................................................................................................... 47
3.3.1 Validation of Calculation Method.......................................................................................................................... 47
3.3.2 Vacancy................................................................................................................................................................... 48
3.3.3 Nitrogen.................................................................................................................................................................. 49
3.3.4 Comparison of Effects between Vacancy and Nitrogen......................................................................................... 51
3.3.5 Comparison of Effects between sp3 Bond and Nitrogen........................................................................................ 54
3.3.6 Comparison of Effects between Vacancy and sp3 Bond......................................................................................... 56
3.4 Conclusions......................................................................................................................................................................... 57
References.................................................................................................................................................................................... 58
ABSTRACT basic structures in carbon materials, in order to develop high-

performance carbon materials. Molecular dynamics (MD) is
Mechanical properties of graphene containing atomic size the most suitable method for investigation of nanostructures.
defects, such as vacancy, nitrogen atom, and sp3 bond are Recently, studies aiming to clarify the relationship between
discussed in this chapter. Molecuar dynamics (MD) simu- atomic-scale defects and mechanical properties using molec-
lations on tensile and shear loadings of defective graphene ular simulation have increased in number. Literature on the
were performed to estimate the mechanical properties, such mechanical properties of carbon materials including carbon
as strength and modulus. We showed the usefulness of MD nanotubes (CNT) and diamonds are summarized in Table 3.1.
simulation for understanding the relationship between nano- There is a large number of research on the mechanical proper-
structures and mechanical properties. ties of CNTs that were carried out before 2007. Consequently,
the reports on graphene have increased in number. The tensile
properties of graphene and CNTs containing multiple Stone–
3.1 INTRODUCTION Wales (SW) defects have been investigated by Xiao et al. using
Defects often affect the mechanical and electronic properties MD simulations [5]. These studies have clarified the relation-
of carbon materials quite significantly. This, therefore, has led ship between the number of defects and the mechanical prop-
to more studies having been recently conducted on defects (i.e., erties. The influence of GBs on the tensile strength of graphene
vacancies [1,2], sp3-type defect [2], dislocations [3], and grain has been investigated by Shenoy et al. [6]. Pao et al. studied the
boundaries (GB) [4]) in graphene and graphite. It is crucial to effect of GBs on the shear properties of graphene using MD
clarify the mechanical properties of graphene and graphite, the simulations [7]. The MD simulations on the tensile loadings of
41
TABLE 3.1
Literature on the Mechanical Properties of Carbon Materials Using Molecular Simulations
Author Year Method Materials Defect Load
Xia et al. [12] 2002 MD CNT SW Tension
Ogata and Shibutani [13] 2003 DFT CNT − Tension, compression
Sammalkorpi et al. [14] 2004 MD CNT Vacancy Tension
Mielke et al. [10] 2004 DFT, QM, MM CNT Vacancy Tension
Zhang et al. [9] 2005 QM, MM CNT Vacancy Tension
Papanikos et al. [15] 2006 FEM CNT Vacancy Tension
Komanduri et al. [16] 2006 MD CNT Vacancy Tension
Schatz et al. [17] 2007 QM/MM CNT Vacancy Tension
Han et al. [18] 2007 MD CNT Vacancy Compression
Tserpes and Papanikos [19] 2007 FEM CNT SW Tension
Shen and Chen [20] 2007 MD Diamond Nitrogen Tension
Li et al. [21] 2007 DFT Graphene − Tension
HaiYang and XinWei [22] 2008 MD CNT Vacancy, sp3 Tension, compression
Xiao et al. [5] 2009 FEM Graphene, CNT SW Tension
Jiang et al. [23] 2009 MD Graphene Isotopic carbon Tension
HaiYang and XinWei [24] 2009 MD CNT sp3 Shear
Shen and Chen [25] 2009 MD Diamond Nitrogen Tension
Aluru et al. [26] 2009 MD Graphene − Tension
Zhao and Aluru [27] 2010 MD Graphene Vacancy Tension
Wong [28] 2010 MD CNT Vacancy Tension
Kim et al. [29] 2010 MD, MM Graphene sp3 Shear
Shenoy et al. [30] 2010 MD Graphene sp3 Tension
Shenoy et al. [6] 2010 MD, DFT Graphene GB Tension
Neek-Amal and Peeters [31] 2010 MD Graphene Vacancy Tension
Tsai and Tu [32] 2010 MD Graphene − Tension, shear
Shokrieh and Rafiee [33] 2010 CM Graphene, CNT − Tension
Anifantis et al. [34] 2010 FEM Graphene − Tension, shear
Ansari et al. [35] 2011 MD Graphene Vacancy Tension
Zhang et al. [36] 2011 MD Graphene sp3 Tension, shear
Min and Aluru [37] 2011 MD Graphene − Shear
Peón-Escalante et al. [38] 2012 FEM Graphene Vacancy Tension, shear
Ahzi et al. [39] 2012 MD Graphene Vacancy, nitrogen Tension
Yan et al. [40] 2012 MD Graphene Vacancy, SW Tension
Wong and Vijayaraghavan [41] 2012 MD Graphene Vacancy Tension
Ansari et al. [8] 2012 MD Graphene Vacancy, SW Tension
Cao and Qu [42] 2012 MD Graphene GB Tension
Cao and Yuan [43] 2012 MD Graphene GB Tension
Okamoto and Ito [44] 2012 MD Graphene Nitrogen Tension
Ito and Okamaoto [11] 2012 MD Graphene, Graphite Vacancy Tension
Cao and Qu [45] 2013 MD Graphene GB Tension
Chang et al. [46] 2013 MD Graphene GB Tension
Wu and Wei [47] 2013 MD Graphene GB Tension
Pao et al. [7] 2013 MD Graphene GB Shear
Okamoto and Ito [48] 2013 MD Graphene Nitrogen Tension, shear
Ito and Okamoto [49] 2013 MD Graphene Vacancy Tension, shear
Le and Batra [50] 2014 MD Graphene Crack Tension
Rajabpour, A. et al. [51] 2014 MD Graphene sp3 Shear
Im et al. [52] 2014 MD Graphene GB Tension
Liu et al. [53] 2014 MD Graphene SW Tension
graphene with SW defects or vacancies have been performed calculations [10]. The mechanical properties of graphene
by Ansari et al. [8]. The influence of the single and double and graphite-containing vacancies have been investigated
vacancies on the tensile strength has been investigated through by the authors [11] who have clarified the effects of the size
molecular mechanics (MM) calculations by Zhang et al. [9]. and distribution of vacancies on the tensile and shear proper-
They compared their results obtained using MM calculations ties of graphene and graphite. In addition, the carbon materi-
with Mielke’s results obtained using quantum mechanics als generally contain impurities such as oxygen, nitrogen, or
Effects of Vacancies, Nitrogen Atoms, and sp3 Bonds on Mechanical Properties of Graphene 43
hydrogen atoms, and these impurities may affect the mechani- where the parameter bij is the bond-order term that depends
cal and electronic properties of the materials. Recently, Shen on the local environment.
et al. [20,25] investigated the effects of nitrogen (N) doping
on the mechanical properties of ultrananocrystalline diamond
( )
−1/ 2 ni
(UNCD) films, using MD simulations. They demonstrated bij = χij 1 + β ni ζijni , (3.3)

that the strength of the N-doped UNCD films decreases with
increase in number and density of the dopant N atoms. The MD
simulations of N-containing graphene were also performed by ζij = ∑ f (r )g(θ
C ik ijk ), (3.4)
Ahzi et al. [39] and the authors [44,48], who have clarified the k ≠i, j

relationships between the tensile properties and N content in
graphene. In addition, the authors investigated the effect of the
ci2 ci2
distributional forms of N atoms in graphene. g(θijk ) = 1 + − 2 , (3.5)
However, the mechanical properties of graphene that condi di + (hi − cos θijk )2
2

tain both vacancies and N atoms have yet to be fully clarified.
In the present study, the authors investigated the influences where θijk is the angle between bonds ij and ik.
of the vacancies and N atoms on the tensile strength and The parameters Aij, Bij, λij, and μij depend on the atom type,
Young’s modulus of graphene when these two types of defects namely, carbon or nitrogen. For atoms i and j (of different
are present together in graphene. In addition, the authors dis- types), these parameters are
cussed which defect of vacancy or N atom greatly affects the
mechanical properties of graphene. Aij = ( Ai × A j )1/ 2 , Bij = ( Bi × B j )1/ 2 , (3.6)
Perfect graphene consists of sp2 bonds which, mutually,
make up an angle of 120° and formulate a planar structure. On
the other hand, real carbon materials also contain sp3 bonds (λ i + λ j ) (µ + µ j )
λ ij = , µ ij = i , (3.7)
derived from raw materials, or are generated during the man- 2 2
ufacturing processes. The mechanical properties of bilayer
graphene sheets coupled by sp3 bonding were investigated by where the parameters with a single index represent the inter-
Zhang et al. [36]. It was verified from the MD simulations action between atoms of the same type.
by Zhang et al. that interlayer sp3 bonds decrease both the The parameter χij in Equation 3.3 is determined similar to
tensile strength and Young’s modulus. However, the mechani- the previous work [44,48].
cal properties of graphene containing both sp3 bonds and N The 2nd REBO and Tersoff potentials contain the same
atoms have not been investigated so far. In the present study, cutoff function fC (r), as given in Equation 3.8. It is known that
the authors clarified the influence on the tensile strength when for these potential functions, the interatomic force increases
both sp3 bonds and N atoms are present together in graphene. dramatically when r exceeds Rmin owing to the discontinu-
ity of the second derivatives of the cutoff function, and this
3.2 COMPUTATIONAL METHOD dramatic increase in the interatomic force greatly affects
the tensile strength. In the author’s works, the cutoff length
3.2.1 Potential Rmin of the 2nd REBO potential is set to 2.0 Å in the studies
In the present study, three types of interatomic potentials are using graphene models with vacancies to avoid the dramatic
used: the second-generation reactive empirical bond order increase in interatomic forces, similar to the previous work
(2nd REBO) [54], Tersoff [55,56], and Lennard–Jones poten- by Brenner et al. [57]. And then, the cutoff length Rmin of both
tials. The 2nd REBO potential for covalent C─C bonds is the 2nd REBO and Tersoff potentials is set to 2.1 Å in the
given in Equation 3.1: studies using graphene models with N atoms.
EREBO = ∑ ∑ [V (r ) − B V (r )], (3.1)

R ij
*
ij A ij 1,
 {1 + cos[ π(r − R ) /R − R ]}
r < Rmin
i j >i  min max min
fC (r ) =  , Rmin < r < Rmax
 2
where the terms VR (rij) and VA(rij) denote the pair-additive
0, r > Rmax
interactions that reflect interatomic repulsions and attractions,
respectively. The Bij* denotes the bond-order term. (3.8)
The Tersoff potential for covalent C─N bonds is given in
Equation 3.2: The Lennard–Jones potential for interlayer interaction in
the bilayer graphene models is given in Equation 3.9:
∑ [ f (r ) A exp(−λ r ) − b f (r )B exp(−µ r )],
1
V= C ij ij ij ij ij C ij ij ij ij
2
i≠ j
 r  12  r  6 
(3.2) V LJ
= 4ε  0  −  0   . (3.9)
r   rij  
 ij 
The ε is set to 0.00284 eV, and the equilibrium distance r0 models of pristine graphene used under two types of shear
is set to 3.2786 Å, so that the interplanar spacing of graph- loadings have the same number of atoms and dimensions as
ite at 300 K is 3.35 Å, which is a known experimental value the analysis models used under tensile loadings, as shown in
[58]. The use of the 2nd REBO and Lennard–Jones potentials Figure 3.1. The details of the analysis models are explained
is switched according to the interatomic distance and bond in the previous papers [49] by the authors.
order, similar to the method used in the adaptive intermolecu-
lar REBO potential [59]. The use of the Tersoff and Lennard– 3.2.2.2 Vacancy-Containing Graphene
Jones potentials is also switched in the same way. The analysis models of the AGR with a cluster-type vacancy
are shown in Figure 3.3. Periodic boundary conditions and
boundary zones are defined in a way similar to the method for
3.2.2 Analysis Model
the pristine AGR model. The analysis models with randomly
The analysis models used in the present studies are shown distributed vacancies are defined in the previous papers [11]
below. by the authors.
3.2.2.1 Pristine Graphene 3.2.2.3 Nitrogen-Containing Graphene

Analysis models, namely, zigzag graphene ribbon (ZGR) The analysis models with randomly distributed N atoms are
and armchair graphene ribbon (AGR) models of pristine gra- defined in the previous papers [44,48] by the authors. The
phene used under zigzag and armchair tensions are shown in analysis models of the ZGR containing two N atoms that
the previous research [11], as shown in Figure 3.1. The analy- are located at an interval d are shown in Figure 3.4. Three
sis models of ZGR with a double length in the zigzag tensile cases with the differing intervals, namely, 1.76 Å, 2.38 Å, and
direction, as shown in Figure 3.2 are also used under tensile 7.31 Å, are investigated. Periodic boundary conditions and
loadings in order to investigate which defect of vacancy or N boundary zones are defined in a way similar to the method for
atom affects greatly the mechanical properties. The analysis the pristine ZGR model.
(a) (b)
l l l l
30 Å
30 Å
Y
52 Å 51 Å
O X
: Tensile loading
: Shear loading
: Carbon atom
FIGURE 3.1 Configurations of graphene used under tensile and shear loadings. (a) ZGR and (b) AGR models.
l l
30 Å
Y
104 Å
O X
FIGURE 3.2 Configurations of ZGR with a double length in the tensile direction.
(a) (b) (c)
O X
FIGURE 3.3 Analysis models of AGR containing a cluster-type vacancy. (a) Single, (b) double, and (c) sextuple vacancies.
O X 1.76 Å 2.38 Å 7.31 Å

: Nitrogen atom
FIGURE 3.4 Analysis models of ZGR containing two N atoms that are located at an interval d.
3.2.2.4 Graphene Containing Both Vacancies significantly affects the mechanical properties of graphene.
and Nitrogen Atoms Therefore, two types of the analysis models are used in order
The example of the analysis model of the ZGR containing to compare the effects of vacancy and N atoms on the fracture
both vacancies and N atoms which are randomly distributed in behavior of graphene. One is the ZGR model containing a pair
graphene, is shown in Figure 3.5. The positions of the vacan- of two adjoining N atoms and a cluster-type vacancy. In that
cies and N atoms are set so that the clustered vacancies and model, three types of vacancies namely, single, double, and
adjoining N atoms are not generated. Five cases with different sextuple vacancies are investigated. In the case of the single
vacancy density and N contents, namely, 1% vacancies—4% vacancy, the configuration is shown in Figure 3.6. The other
N atoms, 2% vacancies—4% N atoms, 4% vacancies—1% N model contains a single vacancy and a pair of N atoms which
atoms, 4% vacancies—2% N atoms, and 4% vacancies—4% are located parallel to the tensile axis at a different interval,
N atoms, are investigated. namely, 1.76 Å, 2.38 Å, and 7.31 Å, as shown in Figure 3.7.
In the previous studies [44,48], it was found that the No periodic boundary conditions are imposed. The bound-
decrease in strength becomes large when two N atoms are ary zones are defined in a way similar to the method for the
present and adjoining each other in graphene. It is worth pristine ZGR model.
clarifying which defect of two adjoining N atoms or vacancy 3.2.2.5 Graphene Containing Both Interlayer
sp3 Bonds and Nitrogen Atoms
The analysis models of the bilayer ZGR containing both
interlayer sp3 bonds and N atoms are used in order to inves-
tigate which defect of interlayer sp3 bond or N atom greatly
affects the mechanical properties of graphene. Two types of
bilayer ZGR which are constructed by two layers of ZGR are
used. One is the bilayer ZGR containing two N atoms per
Y layer, which is located parallel to the tensile axis at a different
interval d, namely, 1.76 Å, 2.38 Å, and 7.31 Å, and an inter-
layer sp3 bond as shown in Figure 3.8. The other model con-
O X : Nitrogen atom tains randomly distributed interlayer sp3 bonds and N atoms.
The positions of the interlayer sp3 bonds and N atoms are
FIGURE 3.5 Analysis model of ZGR containing both vacancies set so that the clustered interlayer sp3 bonds and the adjoin-
and N atoms, which are randomly distributed. ing N atoms are not generated. Four cases with different
: Nitrogen atom
O X
FIGURE 3.6 Analysis model of ZGR containing a pair of two adjoining N atoms and a cluster-type vacancy.
O X : Nitrogen atom d
FIGURE 3.7 Analysis model of ZGR containing a single vacancy and a pair of two N atoms which are located parallel to the tensile axis
at a different interval d.
(a)
Z X
(b)
Z sp3 bond
: Nitrogen atom
Y X
FIGURE 3.8 Analysis model of bilayer ZGR containing two N atoms per layer, which are located parallel to the tensile axis at a different
interval d, and an interlayer sp3 bond. Viewed in the (a) Z direction and (b) Y direction.
N contents, namely, 0, 1, 4, and 10%, at the interlayer sp3 3.2.3 MD Simulation

bond density of 5%, are investigated. The interlayer sp3 bonds
are generated in a way similar to the method by Zhang et al. The velocity Verlet method is used for the time integral of
No periodic boundary conditions are imposed. The boundary the equations of motion of atoms. The velocities of all atoms
zones are defined in a way similar to the method used for the are adjusted simultaneously using the velocity scaling method
pristine ZGR model. [60] so that the temperature of the object can be maintained at
TABLE 3.2
Computational Conditions of Molecular Dynamics Simulations
Without N Atoms With N Atoms
Initial position of atoms Given so that the analysis model can become the crystal structure of
graphene at a preset temperature
Relaxation method in Directions of relaxation Active zone Boundary zone Active zone Boundary zone
unloaded states All directions Clamped All directions Left-hand side:
Y direction
Right-hand side:
X and Y directions
Relaxation time 7000 MD steps 10,000–35,000 MD steps
Method of loadings Directions of enforced Tensile: X direction, shearing: Y direction
displacements
Incremental strain Tensile: ΔεXX = 0.004, shearing: ΔγXY = 0.0023
Relaxation method in Directions of relaxation Active zone Boundary zone Active zone Boundary zone
loaded states All directions Clamped All directions Tensile:
Y direction
Shearing:
Clamped
Relaxation time 7000 MD steps 7000 MD steps
Method to determine Stress Sampled and averaged for last 2000 MD steps for each strain
mechanical properties Modulus Obtained from the slopes of the straight lines in the range where the
relationship between the stress and strain is linear
Strength Given by the largest peak of the nominal stress–nominal strain curves
the preset temperature TSET. The mass of a carbon atom, mC, at 300 K. First, the atoms of the analysis model are relaxed
and the mass of a nitrogen atom, mN, are 1.9927 × 10−26 kg and until the stresses are stabilized for 10,000–35,000 MD simu-
2.3253 × 10−26 kg, respectively. The time step is 0.2 fs. lation steps. The atoms in the active zone are relaxed in all
The atomistic and global stresses are obtained on the basis the directions. The atom shown by a gray circle (see Figures
of the Virial theorem shown in the previous papers [11] by the 3.1a and 3.2) is relaxed in only the X direction. The atoms
authors. in the left-hand side boundary zone are relaxed in only the
The details of the computational conditions are summa- Y direction. The atoms in the right-hand side boundary zone
rized in Table 3.2. In the case of not containing N atoms, the except the atom shown by a gray circle are relaxed in only
methods of tensile and shear loadings are as follows. The the X and Y directions. After the atoms are relaxed, constant
initial positions of the atoms are given so that the analysis displacements are applied to the atoms in the boundary zones
model represents the crystal structure of graphene at a preset to simulate uniaxial tensile loading in the X direction or
temperature. First, the atoms in the active zone of the analy- shear loading in the Y direction. The atoms in the boundary
sis model are relaxed in unloaded states for 7000 MD steps. zones are restrained in the X and Z directions, and in all the
The atoms in the boundary zone are fixed. After constant directions in the case of tensile and shear loadings, respec-
displacements are applied to the atoms in both the boundary tively. The atoms in the active zone of the analysis model are
zones to simulate uniaxial tensile loading in the X direction or relaxed for all the directions in the case of both tensile and
shear loading in the Y direction, the atoms in the active zone shear loadings for 7000 MD simulation steps. The atoms in
are relaxed for 7000 MD steps. The tensile strain increment, the boundary zones of the analysis model are relaxed for only
ΔεXX, is 0.004. The shearing strain increment, ΔγXY, is 0.0023. the Y direction in the case of tensile loading. The tensile strain
The output stresses are sampled for the last 2000 MD steps increment ΔεXX is 0.004. The shearing strain increment ΔγXY is
for each strain and are averaged. Young’s and shearing moduli 0.0023. The Young’s and shearing moduli are obtained by the
are obtained from the slopes of the straight lines in the range same way as the above mentioned method in the case of not
where the relationship between the stress and strain is lin- containing N atoms.
ear, and tensile and shearing strengths are given by the largest
peak of the nominal stress–nominal strain curves.
In the cases containing N atoms, the methods of tensile and 3.3 RESULT AND DISCUSSION
shear loadings are as follows, because larger MD simulation
3.3.1 Validation of Calculation Method
steps are required to stabilize the initial structures compared
with the case of not containing N atoms. The initial positions The authors performed the MD simulations on tensile and
of the atoms are given such that the analysis model becomes shear loadings of pristine graphene at 300 K to verify the pro-
identical to the crystal structure of N-containing graphene priety of their method of calculation. The results of tensile
TABLE 3.3 TABLE 3.4

Comparison of the Computed Tensile Properties of Comparison of the Computed Shear Properties of
Pristine Graphene and the One Reported by the Pristine Graphene and the One Reported in the
Literature Literature
Young’s Tensile Shear Shear Strength
Author Method Modulus (TPa) Strength (GPa) Author Method Modulus (TPa) (GPa)
Present work MD 0.836 83 Present work MD 0.412 51
Lee et al. [61] Experimental 1 130 Blakslee [65] Experimental 0.434 −
Frank et al. [62] Experimental 0.5 − Min and Aluru [37] MD − 60
Aluru et al. [26] MD 1.01 99 Pao et al. [7] MD 0.421 60
Shenoy et al. [30] MD 0.86 121 LiJun and MD 0.08–0.21 6–45
Jiang et al. [23] MD 1.1 − TienChong [66]
Shen et al. [63] MD 0.933 − Shen et al. [63] MD 0.403 −
Neek-Amal and MD 0.501 ± 0.032 − Kim et al. [29] MD 0.284 16.6
Peeters [31] Tsai and Tu [32] MD 0.358 −
Tsai and Tu [32] MD 0.912 − Faccio et al. [67] MD 0.408 −
Li et al. [21] DFT 1.05 116 Anifantis et al. [34] FEM 0.280 −
Shokrieh and CM 1.04 − Sakhaee-Pour [64] FEM 0.225 −
Rafiee [33]
Xiao et al. [5] FEM 1.13 110
Anifantis et al. [34] FEM 1.367 −
Sakhaee-Pour [64] FEM 1.025 −
The relation between the vacancy size and relative
strength, that is, the strength of the crack-containing mate-
rial relative to the strength of pristine graphene under tensile
and shear loadings, is shown in Figure 3.10. The ratio of the
loadings are shown in Table 3.3. The results in the literature strength of vacancy-containing graphene to that of pristine
are also shown for comparison in Table 3.3. graphene under the tensile loading decreases as the vacancy
The computed tensile strength is 83 GPa, which nearly size increases following the rule of Griffith [11]. Moreover,
agrees with the 121 GPa calculated by Shenoy et al. through the relation between the relative strength under the shear load-
MD simulations [30] and the experimentally obtained value of ing and vacancy size is almost identical to that during the ten-
130 GPa [61]. The computed Young’s modulus is 0.836 TPa, sile loading. Snapshots of AGRs during the shear loadings are
which is within the range of results obtained by the DFT shown in Figure 3.11. In the vacancy-containing graphene, the
method [21] (1.05 TPa), by the MD method [30] (0.86 TPa), concentration of stress occurs around the vacancy just before
and by experiment (0.5 TPa [62] and 1 TPa [61]). It is esti- the fracture. The direction of fracture progression is at an
mated that the lower value obtained in this work is due to angle of about 150° to the X-axis in all cases. In the case of
the effect of size on the elastic properties of graphene [23]. the tensile loadings [11], a fracture in the zigzag direction was
On the other hand, the results of shear loadings are shown
in Table 3.4. The computed shear strength is 51 GPa, which
nearly agrees with the 60 GPa calculated by Min and Aluru TABLE 3.5
[37] using MD simulations. The computed shear modu- Moduli of the Vacancy-Containing Graphene
lus is 0.412 TPa, which also nearly agrees with the value
Materials Young’s Modulus (TPa) Shear Modulus (TPa)
of 0.421 TPa calculated by Pao et al. through MD simula-
tions and the experimentally obtained value of 0.434 TPa by AGR
Blakslee et al. [61]. Pristine 0.879 0.408
Single vacancy 0.868 (−1.2%) 0.399 (+0.5%)
Double vacancy 0.870 (−1.0%) 0.392 (–1.3%)
3.3.2 Vacancy
Sextuple vacancy 0.848 (−3.6%) 0.385 (–3.9%)
The moduli and strengths of vacancy-containing graphene
ZGR
obtained at 300 K are listed in Tables 3.5 and 3.6, respec-
Pristine 0.794 0.442
tively. The nominal stress–nominal strain curves under the
Single vacancy 0.782 (−1.5%) 0.418 (–2.1%)
shear loading are given in Figure 3.9.
Double vacancy 0.765 (−3.6%) 0.420 (–1.6%)
The results for pristine graphene are also shown for refer-
Sextuple vacancy 0.767 (−3.4%) 0.402 (–5.8%)
ence. Both the tensile and shear strength of AGR and ZGR
containing a vacancy decrease greatly. On the other hand, Note: The values in parentheses represent the differences between the
both the Young’s and shear modulus hardly change with the pristine and vacancy-containing materials.
vacancy size.
TABLE 3.6
Strength of the Vacancy-Containing Graphene
Tensile Strength (GPa) Shear Strength (GPa)
MD MM [9] QM [10] MD
Materials (This Work) (Zhang et al.) (Mielke et al.) (This Work)
AGR
Pristine 76 87.9 124 51
Single vacancy 61 (–19%) 64.8 (–26%) 101 (–18%) 36 (–30%)
Double vacancy 62 (–18%) 64.4 (–26%) 107 (–13%) 36 (–30%)
Sextuple vacancy 51 (–32%) 33 (–34%)
ZGR
Pristine 91 105.5 135 53
Single vacancy 75 (–17%) 70.4 (–33%) 100 (–26%) 42 (–17%)
Double vacancy 64 (–29%) 71.3 (–32%) 105 (–22%) 39 (–23%)
Sextuple vacancy 65 (–28%) 34 (–33%)
Note: The values in parentheses represent the differences between the pristine and vacancy-con-
taining materials.
(a) (b)
60 60
Pristine Pristine
50 Single vacancy 50 Single vacancy
Nominal stress, τXY (GPa)
Double vacancy Double vacancy

40 Sextuple vacancy 40 Sextuple vacancy
30 30
20 20
10 10
0 0
0 0.1 0.2 0.3 0 0.1 0.2 0.3
Nominal strain, γXY Nominal strain, γXY
FIGURE 3.9 Stress–strain curves of graphene containing a cluster-type vacancy under the shear loading. (a) AGR and (b) ZGR.
1.2 propagated more easily than the armchair one. The results of
Tension the present work agree with those results.
1.0 The tensile properties of graphene containing randomly
Shear
distributed vacancies are summarized in Table 3.7 [11]. The
Relative strength
0.8
Young’s modulus and tensile strength significantly decrease
0.6
as the density of vacancies increases. The decline in tensile
strength is 59% at the density of 4%.
0.4
0.2 3.3.3 Nitrogen
The computed mechanical properties of graphene containing
0.0
0 2 4 6 8 randomly distributed N atoms are listed in Tables 3.8 and 3.9
Number of atom defects [44,48]. The same trends can be seen in the cases of tensile
and shear loadings. The modulus hardly changes as N content
FIGURE 3.10 Relative strengths and vacancy size, namely, the increases. On the other hand, the strength decreases espe-
number of atom defects under tensile and shear loadings. cially when containing adjoining N atoms.
(a-1) (a-2) (a-3) (a-4)
Initial structure of Just before fracture Fracture started Fracture progressed

the pristine AGR occurred
(b-1) (b-2) (b-3) (b-4)
Initial structure of Just before fracture Fracture started Fracture progressed

the AGR with a occurred
single vacancy
(c-1) (c-2) (c-3) (c-4)
Initial structure of Just before fracture Fracture started Fracture progressed O X

the AGR with a occurred
double vacancy
(d-1) (d-2) (d-3) (d-4) i
τ xy (GPa)
70
60
50
40
Initial structure of Just before fracture Fracture started Fracture progressed 30
the AGR with a occurred 20
sextuple vacancy 10
0
FIGURE 3.11 Stages of fracture progression in the AGRs containing a cluster-type vacancy under shear loading. (a-1)–(a-4): pristine,
(b-1)–(b-4): single vacancy, (c-1)–(c-4): double vacancy, (d-1)–(d-4): sextuple vacancy.
It is found that both the tensile and shear strengths

TABLE 3.7 decrease with decreasing distance between two N atoms.
Tensile Properties of Vacancy-Containing Graphene The decrease in the tensile strength relative to that of pris-
Density of
tine ZGR is 25% when two N atoms adjoin each other, that
Vacancies (%) Young’s Modulus (TPa) Tensile Strength (GPa) is, when d is 1.76 Å. Then, the decrease in the shear strength
0 0.794 91
relative to that of pristine ZGR is 29% at the same d. On the
1 0.752 (−5.4%) 56 (−38%)
other hand, both the Young’s modulus and shear modulus
2 0.657 (−17%) 48 (−48%)
hardly change.
4 0.617 (−22%) 37 (−59%)
The relations of the tensile and shear strengths with the
distance between two N atoms are shown in Figure 3.13. The
Note: The values in parentheses represent the differences between the larger the distance between two N atoms, the higher are both
pristine and vacancy-containing materials. the tensile and shear strengths. These relations are almost
constant when d is over 2.38 Å.
Snapshots taken during the shear loadings are shown in
Figure 3.14. In the case of d = 1.76 Å ((a-1)–(a-4)), fracture
The effects of the distance between N atoms on the occurs at the point where two N atoms adjoin each other, sim-
mechanical properties were investigated. The stress–strain ilar to the case of tensile loading [39]. In the case of d = 2.38
curves of the ZGR containing two N atoms that are located and 7.31 Å ((b-1)–(b-4) and (c-1)–(c-4), respectively), fracture
at an interval d under the tensile and shear loading are shown starts at the cleavage of the C─C bond adjoining the N atom
in Figure 3.12. and not at a C─N bond, similar to the case of tensile loading.
The calculated tensile and shear properties are listed in In all the cases, fractures propagate in the direction of 120°
Table 3.10. to the X-axis.
TABLE 3.8
Moduli of the N-Containing Graphene
Young’s Modulus (TPa) Shear Modulus (TPa)
Nitrogen Content Without Adjoining With Adjoining Without Adjoining With Adjoining
(%) N Atoms N Atoms N Atoms N Atoms
0 0.786 0.786 0.442 0.442
1 0.773 (−1.6%) 0.770 (−2.0%) 0.465 (+5.2%) 0.452 (+2.3%)
2 0.801 (+1.9%) 0.776 (−1.3%) 0.479 (+8.4%) 0.462 (+4.5%)
4 0.822 (+4.6%) 0.823 (+4.7%) 0.466 (+5.4%) 0.466 (+5.4%)
Note: The values in parentheses represent the differences between the pristine and N-containing materials.
TABLE 3.9
Strength of the N-Containing Graphene
Tensile Strength (GPa) Shear Strength (GPa)
Nitrogen Without Adjoining With Adjoining N Without Adjoining With Adjoining
Content (%) N Atoms Atoms N Atoms N Atoms
0 94 94 53 53
1 89 (−5.3%) 64 (−32%) 52 (−1.8%) 40 (−25%)
2 87 (−7.4%) 69 (−27%) 51 (−3.8%) 41 (−23%)
4 82 (−13%) 69 (−27%) 50 (−5.7%) 42 (−21%)
Note: The values in parentheses represent the differences between the pristine and N-containing materials.
(a) 100 (b) 60

Pristine Pristine
90
d = 1.76 Å 50 d = 1.76 Å
Nominal stress, σX (GPa)
80
d = 2.38 Å d = 2.38 Å
70 d = 7.31 Å d = 7.31 Å
40
60
50 30
40
30 20
20
10
10
0 0
0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2 0.25
Nominal strain, εX Nominal strain, γXY
FIGURE 3.12 Stress–strain curves of graphene containing two N atoms located at an interval d. (a) Tensile and (b) shear loadings.
3.3.4 Comparison of Effects between

TABLE 3.10 Vacancy and Nitrogen
Mechanical Properties of Graphene Containing
The mechanical properties of graphene containing both
Two N Atoms Located at a Distance d vacancies and N atoms are summarized in Tables 3.11 and
Young’s Tensile Shearing Shearing 3.12.
Modulus Strength Modulus Strength The tensile and shear strengths decrease greatly as the den-
d (Å) (GPa) (TPa) (TPa) (GPa) sity of vacancies increases at the constant N content of 4%. On
1.76 0.795 71 0.452 37 the other hand, the tensile and shear strengths hardly change
2.38 0.796 84 0.450 50 as the N content increases at the constant vacancy density of
7.31 0.809 90 0.570 51 4% as shown in Table 3.12.
100
TABLE 3.11
80
Effect of Vacancy on the Strength of N-Containing
Strength (GPa)
60 Graphene
40 Density of Nitrogen Tensile Shear Strength
Tensile loading Vacancies (%) Content (%) Strength (GPa) (GPa)
20
Shear loading 0 4 83 (−12%) 42 (−21%)
0 1 4 64 (−32%) 38 (−28%)
0 2 4 6 8
2 4 62 (−34%) 32 (−40%)
Distance d between nitrogen atoms ()
4 4 45 (−52%) 27 (−49%)
FIGURE 3.13 Relations of tensile and shear strengths with the The values in parentheses represent the differences between the pristine and
distance between two N atoms. defect-containing materials.
The relation between the tensile strength and vacancy

density is shown in Figure 3.15a. The relation between the
tensile strength and N content is also shown in Figure 3.15b.
For the randomly distributed vacancies and N atoms, the aver-
age values of the two results calculated using the models with TABLE 3.12
different vacancy and N atom arrangements, is plotted. The Effect of N Atom on the Strength of
error bar in the graph represents the range between these two Vacancy-Containing Graphene
values. The tensile strength decreases by 52% at the vacancy
Density of Nitrogen Tensile Shear Strength
density of 4% and N content of 4%. Vacancies (%) Content (%) Strength (GPa) (GPa)
Snapshots taken during the tensile and shear loadings are
4 0 45 (−52%) 27 (−49%)
shown in Figures 3.16 and 3.17, respectively. In all the cases, a
4 1 44 (−5.3%) 27 (−49%)
fracture occurs due to cleavage of the C─C bond, which is on
4 2 45 (−5.2%) 27 (−49%)
the periphery of the vacancy.
4 4 45 (−52%) 27 (−49%)
The mechanical properties of graphene containing both a
cluster-type vacancy and a pair of adjoining N atoms were The values in parentheses represent the differences between the pristine
investigated. The relation between the relative strength, that and defect-containing materials.
is, the ratio of the tensile strength of graphene containing
(a-1) (a-2) (a-3) (a-4)
Initial structure Before fracture Beginning of fracture Progress of fracture
(b-1) (b-2) (b-3) (b-4)
Initial structure Before fracture Beginning of fracture Progress of fracture

(c-1) (c-2) (c-3) (c-4)
O X Initial structure Before fracture Beginning of fracture Progress of fracture
: Nitrogen atom
FIGURE 3.14 Stages of fracture progress under shear loading in graphene containing two N atoms at different intervals d: 1.76 Å ((a-1)–
(a-4)), 2.38 Å ((b-1)–(b-4)), and 7.31 Å ((c-1)–(c-4)).
(a) 90 (b) 60
80 Tension
50
70 Shear
Strength (GPa)
Strength (GPa)
60 40
50
30
40
30 20
20 Tension
10
10 Shear
0 0
0 1 2 3 4 5 0 1 2 3 4 5
Vacancy density (%) Nitrogen content (%)
FIGURE 3.15 Relation between strength and density of defects. Strength with (a) vacancy density and (b) N-content.
(a-1) (a-2) (a-3) (a-4)
Initial structure Cleavage of C–C Fracture developed Tearing of the sheet

(b-1) (b-2) (b-3) (b-4)

(c-1) (c-2) (c-3) (c-4)

(d-1) (d-2) (d-3) (d-4)

O X
: Carbon atom : Nitrogen atom
FIGURE 3.16 Stages of fracture progressions of ZGRs containing vacancies and N atoms randomly under tensile loading. (a-1)–(a-4): 1%
vacancies and 4% N atoms, (b-1)–(b-4): 4% vacancies and 1% N atoms, (c-1)–(c-4): 4% vacancies and 4% N atoms, (d-1)–(d-4): 4% vacan-
cies and 0% N atoms.
defects to that of pristine graphene, and vacancy size, is shown shown in Figure 3.19. In the case of the single vacancy ((a-
together with the results [11] for the ZGR containing only the 1)–(a-4)), a fracture occurs on the periphery of the adjoining
vacancy in Figure 3.18. N atoms. On the other hand, in the other cases, a fracture
There is no difference in the behavior of relative strength occurs around the vacancy, instead of the periphery of the
when the vacancy size increases for the two cases containing adjoining N atoms.
both a vacancy and N atoms, and only a vacancy. Thus, it was found that the effect of adjoining N atoms is
Snapshots of ZGR containing both a vacancy and a pair larger than that of the single vacancy, and smaller than that of
of adjoining N atoms taken during the tensile loading are the double vacancy.
(a-1) (a-2) (a-3) (a-4)

(b-1) (b-2) (b-3) (b-4)

(c-1) (c-2) (c-3) (c-4)

(d-1) (d-2) (d-3) (d-4)

O X : Carbon atom : Nitrogen atom
FIGURE 3.17 Stages of fracture progressions of ZGRs containing vacancies and N atoms randomly under shear loading. (a-1)–(a-4): 1%
vacancies and 4% N atoms, (b-1)–(b-4): 4% vacancies and 1% N atoms, (c-1)–(c-4): 4% vacancies and 4% N atoms, (d-1)–(d-4): 4% vacan-
cies and 0% N atoms.
Next, the effects of distance between two N atoms 3.20. The larger the distance of separation between two N
when containing the single vacancy and two N atoms were atoms, the higher is the tensile strength as shown in Figure 3.21,
investigated. similar to the case of containing only two N atoms [44].
The stress–strain curves of the ZGRs containing a single Snapshots taken during the tensile loading are shown in
vacancy and a pair of two N atoms which are located parallel Figure 3.22. In the case of the single vacancy and the pair
to the tensile axis at a different interval d, are shown in Figure of adjoining N atoms ((a-1)–(a-4)), a fracture occurs on the
periphery of the adjoining N atoms. On the other hand, in the
other cases, that is, when the distance of separation between
1.2
Vacancy and N-atoms two N atoms becomes large, a fracture occurs around the
1.0 Only vacancy vacancy.
Thus, graphene is more sensitive to vacancies than N atoms,
Relative strength
0.8
except when two N atoms adjoin each other in graphene.
0.6
0.4 3.3.5 Comparison of Effects between

sp3 Bond and Nitrogen
0.2
The stress–strain curves of the bilayer ZGRs containing an
0.0
0 2 4 6 8 sp3 bond in the interlayer and two N atoms per layer are shown
Number of defects in Figure 3.23. The relations of the tensile strengths with the
distance between two N atoms are shown in Figure 3.24. The
FIGURE 3.18 Relative strength and number of atom defects, that greater the distance between two N atoms, the higher is the
is, vacancy size in graphene containing a cluster-type vacancy and a tensile strength, similar to the results of graphene with vacan-
pair of adjoining N atoms, or only a cluster-type vacancy. cies and N atoms.
(a-1) (b-1) (c-1)
Initial structure Initial structure Initial structure

(a-2) (b-2) (c-2)
Beginning of a fracture Beginning of a fracture Beginning of a fracture

(a-3) (b-3) (c-3)
Progression of the fracture Progression of the fracture Progression of the fracture

(a-4) (b-4) (c-4)
Y
Tearing of the sheet Tearing of the sheet Tearing of the sheet
FIGURE 3.19 Stages of fracture progressions of ZGR containing a cluster-type vacancy and a pair of adjoining N atoms. (a-1)–(a-4): single
vacancy and N atoms, (b-1)–(b-4): double vacancy and N atoms, (c-1)–(c-4): sextuple vacancy and N atoms.
Snapshots taken during the tensile loading are shown in defects to the tensile strength of perfect graphene and N con-
Figure 3.25. In the case where the d = 1.76 Å ((a-1)–(a-3), tents, is shown by filled diamonds in Figure 3.26. The result
(a′- 1)–(a′-3)), that is, the pair of adjoining N atoms, a frac- [44] in the case of containing only N atoms is also shown for
ture occurs on the periphery of the adjoining N atoms. On the reference by open diamonds. The tensile strength decreases
other hand, when the d becomes larger (d = 7.31 Å ((c-1)–(c-3), by about 20% when containing sp3 bonds at the density of 5%,
(c′-1)–(c′-3))), a fracture occurs around the sp3 bond. but it hardly changes as the N content increases. It is found
Finally, the mechanical properties of bilayer ZGRs con- that the decrease in strength when both sp3 bonds and N atoms
taining both randomly distributed N atoms and sp3 bonds are present together becomes greater than that in the case of
were investigated. The relation between the relative strength, including only N atoms.
that is, the ratio of the tensile strength of bilayer ZGR with Snapshots of the bilayer ZGR with 10% N atoms and 5%
sp3 bonds taken during the tensile loading are shown in Figure
3.27. A fracture occurs due to cleavage of the C─C bond,
120 which is in proximity to both a C─N bond and interlayer sp3
Pristine
100 d = 1.76 Å
Nominal stress, σx (GPa)
d = 2.38 Å
80 d = 7.31 Å 100
Tensile strength (GPa)
80
60
60
40
40
20
20
0 0
0 0.05 0.1 0.15 0.2 0 2 4 6 8
Nominal strain, εx Distance between nitrogen atoms ()
FIGURE 3.20 Stress–strain curves of ZGRs containing a single FIGURE 3.21 Tensile strength and distance between N atoms in
vacancy and a pair of two N atoms which are located parallel to the ZGR containing a single vacancy and a pair of adjoining N atoms
tensile axis at a different interval d. which are located parallel to the tensile axis at the different interval d.
(a-1) (b-1) (c-1)
Initial structure Initial structure Initial structure

(a-2) (b-2) (c-2)
Beginning of a fracture Beginning of a fracture Beginning of a fracture

(a-3) (b-3) (c-3)
Progression of the fracture Progression of the fracture Progression of the fracture

(a-4) (b-4) (c-4)
Y
Tearing of the sheet Tearing of the sheet Tearing of the sheet
FIGURE 3.22 Stages of fracture progressions of ZGR containing a single vacancy and a pair of adjoining N atoms which are located
parallel to the tensile axis at a different interval d : 1.76 Å ((a-1)–(a-4)), 2.38 Å ((b-1)–(b-4)), and 7.31 Å ((c-1)–(c-4)).
120
Pristine 100
Tensile strength (GPa)
100 d = 1.76 Å
80
Nominal stress, σx (GPa)
d = 2.38 Å
80 d = 7.31 Å 60
40
60
20
40
0
0 2 4 6 8
20 Distance between nitrogen atoms ()
0 FIGURE 3.24 Tensile strength and distance between N atoms

0 0.05 0.1 0.15 0.2
in bilayer ZGR containing an interlayer sp3 bond and two N atoms
Nominal strain, εx
per layer, which are located parallel to the tensile axis at a different
interval d.
FIGURE 3.23 Stress–strain curves of bilayer ZGRs containing an
interlayer sp3 bond and two N atoms per layer which are located
3.3.6 Comparison of Effects between
parallel to the tensile axis at a different interval d.
Vacancy and sp3 Bond
bond as shown in Figure 3.28. In the case of containing only In the case of containing randomly distributed vacancies,
N atoms [44], it was found that a fracture occurs due to cleav- the reduction in the tensile strength stands at about 59% at
age of the C─C bond which is in proximity to a C─N bond. the vacancy density of 4% as shown in Table 3.7. On the
On the other hand, in the case of containing both N atoms other hand, the reduction stands at 17% at the sp3 bond den-
and sp3 bonds, it was found that the C─C bond which is in sity of 5.81% by Zhang et al. [36]. Recently, Zandiatashbar
proximity to both a C─N bond and interlayer sp3 bond is more et al. investigated the effect of vacancy and sp3 bond on the
easily cleaved. The greater decrease in strength when both sp3 mechanical properties of graphene using AFM nanoindenta-
bonds and N atoms are present together seems to be owing tion [2]. They reported that vacancy affects the mechanical
to those behaviors. Thus, graphene is slightly more sensitive properties of graphene more significantly than the sp3 bond.
to sp3 bonds than N atoms, except when two nitrogen atoms Thus, graphene must be more sensitive to vacancy than the
adjoin each other in graphene. sp3 bond.
(a-1) (a-2) (a-3)
Z X Initial structure Beginning of a fracture Tearing of the layer

(viewed in the Z direction) (viewed in the Z direction) (viewed in the Z direction)
(a′-1) (a′-2) (a′-3)
Z sp3 bond
Y× X Initial structure Beginning of a fracture Tearing of the layer

(viewed in the Y direction) (viewed in the Y direction) (viewed in the Y direction)
(b-1) (b-2) (b-3)

(b′-1)
Z sp3 bond (b′-2) (b′-3)
Y× Beginning of a fracture Tearing of the layer

X Initial structure
(viewed in the Y direction) (viewed in the Y direction)
(viewed in the Y direction)
(c-1) (c-2) (c-3)

(c′-1) (c′-2) (c′-3)
Z sp3 bond
Initial structure Beginning of a fracture Tearing of the layer

× X
Y (viewed in the Y direction) (viewed in the Y direction) (viewed in the Y direction)
FIGURE 3.25 Stages of fracture progressions of bilayer ZGRs containing an interlayer sp3 bond and two N atoms per layer which are
located parallel to the tensile axis at a differing interval d: 1.76 Å ((a-1)–(a-3)), 2.38 Å ((b-1)–(b-3)), and 7.31 Å ((c-1)–(c-3)).
1.2 or sp3 bonds, to investigate the influence of these defects on

the mechanical properties.
1.0 It was found that for the cluster-type vacancy, the relation-
Relative strength
0.8 ship between the vacancy size and the shear strength agrees
with that during the tensile loading at any size of vacancy.
0.6
Fractures develop in the zigzag direction, similar to the results
0.4 of tensile loading.
: N-atoms and 5 % sp3 bonds
0.2
It was also found that the shear strength decreases greatly
: Only N-atoms
when two N atoms adjoin each other, similar to the results of
0.0 tensile loading. In addition, fracture starts at the cleavage of
0 2 4 6 8 10 12
Nitrogen content (%) the C─C bond adjoining the N atom and not at a C─N bond,
similar to the case of tensile loading.
FIGURE 3.26 Relative strength as a function of nitrogen content The authors compared the effects of vacancies and N
in bilayer ZGR with N atoms and 5% sp3 bonds, or in ZGR with only atoms using the analysis models containing both vacancies
N atoms. and N atoms. It was clarified that graphene is more sensitive
to vacancies than N atoms, except when two nitrogen atoms
adjoin each other in graphene.
3.4 CONCLUSIONS
The authors also compared the effects of interlayer sp3
The authors performed the MD simulations of tensile or shear bonds and N atoms using the analysis models containing
loadings on graphene containing vacancies, nitrogen atoms, both sp3 bonds and N atoms. It was clarified that graphene
(a) (b) (c)

Z
Y× X
FIGURE 3.27 Snapshots of the bilayer ZGR with 10% N atoms and 5% sp3 bonds during tensile loading. (a) Initial structure, (b) just before
a cleavage occurs, and (c) fractures occurred.
(a) (b)
sp3 bond
Carbon
×
Nitrogen Carbon Cleavage
FIGURE 3.28 Enlargements of the boxed section shown in Figure 3.27b. (a) Just before a cleavage occurs and (b) the cleavage occurred.
is slightly more sensitive to sp3 bonds than N atoms. It was 10. Mielke, S. L., Troya, D., Zhang, S., Li, J.-L., Xiao, S., Car, R.,
also found that a fracture occurs due to cleavage of the C─C Ruoff, R., Shatz, G. C., and Belytschko, T. 2004. The role of
bond which is in proximity to both a C─N bond and inter- vacancy defects and holes in the fracture of carbon nanotubes.
Chem. Phys. Lett. 390:413–420.
layer sp3 bond.
11. Ito, A. and Okamoto, S. 2012. Molecular dynamics analysis
In summary, it was found that the effect on the strength on effects of vacancies upon mechanical properties of gra-
of graphene increases in the order of N atom, sp3 bond, and phene and graphite. Eng. Lett. 20:271–278.
vacancy. 12. Xia, Y., Zhao, M., Ma, Y., Ying, M., Liu, X., Liu, P., and Mei, L.
2002. Tensile strength of single-walled carbon nanotubes with
defects under hydrostatic pressure. Phys. Rev. B 65:155415.
REFERENCES
13. Ogata, S. and Shibutani, Y. 2003. Ideal tensile strength and
1. Thrower, P. A. 1969. The study of defects in graphite by trans- band gap of single-walled carbon nanotubes. Phys. Rev. B
mission electron microscopy. Chem. Phys. Carbon 5:217. 68:165409.
2. Zandiatashbar, A., Lee, G.-H., An, S. J., Lee, S., Mathew, N., 14. Sammalkorpi, M., Krasheninnikov, A., Kuronen, A.,
Terrones, M., Hayashi, T., Picu, C. R., Hone, J., and Koratkar, Nordlund, K., and Kaski, K. 2004. Mechanical properties of
N. 2014. Effect of defects on the intrinsic strength and stiff- carbon nanotubes with vacancies and related defects. Phys.
ness of graphene. Nat. Commun. 5, article 3186. Rev. B 70:245416.
3. Hashimoto, A., Suenaga, K., Gloter, A., Urita, K., and Iijima, 15. Tserpes, K. I., Papanikos, P., and Tsirkas, S. A. 2006. A pro-
S. 2004. Direct evidence for atomic defects in graphene layers. gressive fracture model for carbon nanotubes. Composites:
Nature 430:870–873. Part B 37:662–669.
4. Osing, J. and Shvets, I. V. 1998. Bulk defects in graphite 16. Agrawal, P. M., Sudalayandi, B. S., Raff, L. M., and
observed with a scanning tunnelling microscope. Surf. Sci. Komanduri, R. 2006. A comparison of different methods of
417:145–150. Young’s modulus determination for single-wall carbon nano-
5. Xiao, J. R., Staniszewski, J., and Gillespie Jr., J. W. 2010. tubes (SWCNT) using molecular dynamics (MD) simulations.
Tensile behaviors of graphene sheets and carbon nano- Comp. Mater. Sci. 38:271–281.
tubes with multiple Stone-Wales defects. Mater. Sci. Eng. A 17. Khare, R., Mielke, S. L., Paci, J. T., Zhang, S., Ballarini, R.,
527:715–723. Schatz, G. C., and Belytschko, T. 2007. Coupled quantum
6. Grantab, R., Shenoy, V. B., and Ruoff, R. S. 2010. Anomalous mechanical/molecular mechanical modeling of the fracture of
strength characteristics of tilt grain boundaries in graphene. defective carbon nanotubes and graphene sheets. Phys. Rev. B
Science 330:946–948. 75:75412.
7. Liu, T.-H., Pao, C.-W., and Chang, C.-C. 2013. Thermal response 18. Xin, H., Han, Q., and Yao, X.-H. 2007. Buckling and axially
of grain boundaries in graphene sheets under shear strain compressive properties of perfect and defective single-walled
from atomistic simulations. Comp. Mater. Sci. 70:163–170. carbon nanotubes. Carbon 45:2486–2495.
8. Ansari, R., Ajori, S., and Motevalli, B. 2012. Mechanical prop- 19. Tserpes, K. I. and Papanikos, P. 2007. The effect of Stone–
erties of defective single-layered graphene sheets via molecu- Wales defect on the tensile behavior and fracture of single-
lar dynamics simulation. Superlatt. Microstruct. 51:274–289. walled carbon nanotubes. Compos. Struct. 79:581–589.
9. Zhang, S., Mielke, S. L., Khare, R., Troya, D., Ruoff, R. S., 20. Shen, L. and Chen, Z. 2007. An investigation of grain size
Schatz, G. C., and Belytschko, T. 2005. Mechanics of defects and nitrogen-doping effects on the mechanical properties
in carbon nanotubes: Atomistic and multiscale simulations. of ultrananocrystalline diamond films. Int. J. Sol. Struct.
Phys. Rev. B 71:115403. 44:3379–3392.
21. Liu, F., Ming, P., and Li, J. 2007. Ab initio calculation of ideal 41. Wong, C. H. and Vijayaraghavan, V. 2012. Nanomechanics
strength and phonon instability of graphene under tension. of free form and water submerged singlelayer graphene sheet
Phys. Rev. B 76:064120. under axial tension by using molecular dynamics simulation.
22. Hai-Yang, S. and Xin-Wei, Z. 2008. Molecular dynamics Mater. Sci. Eng. A 556:420–428.
study of effects of sp3 interwall bridging and initial vacancy- 42. Cao, A. and Qu, J. 2012. Study on the mechanical behavior
related defects on mechanical properties of double-walled of tilt bicrystal graphene by molecular dynamics simulations:
carbon nanotube. Phys. B 403:3798–3802. Bulk verse nanoribbons. J. Appl. Phys. 112:43519.
23. Jiang, J.-W., Wang, J.-S., and Li, B. 2009. Young’s modu- 43. Cao, A. and Yuan, Y. 2012. Atomistic study on the strength of
lus of graphene: A molecular dynamics study. Phys. Rev. B symmetric tilt grain boundaries in graphene. Appl. Phys. Lett.
80:113405. 100:211912.
24. Hai-Yang, S. and Xin-Wei, Z. 2009. Molecular dynamics study 44. Okamoto, S. and Ito, A. 2012. Effects of nitrogen atoms on
of effects of sp3 interwall bridging upon torsional behavior of mechanical properties of graphene by molecular dynamics
double-walled carbon nanotube. Phys. Lett. A 373:682–685. simulations. Eng. Lett. 20:169–175.
25. Shen, L. and Chen, Z. 2009. A study of mechanical proper- 45. Cao, A. and Qu, J. 2013. Atomistic simulation study of brittle
ties of pure and nitrogen-doped ultrananocrystalline diamond failure in nanocrystalline graphene under uniaxial tension.
films under various loading conditions. Int. J. Sol. Struct. Appl. Phys. Lett. 102:71902.
46:811–823. 46. Yi, L., Yin, Z., Zhang, Y., and Chang, T. 2013. A theoreti-
26. Zhao, H., Min, K., and Aluru, N. R. 2009. Size and chirality cal evaluation of the temperature and strain-rate dependent
dependent elastic properties of graphene nanoribbons under fracture strength of tilt grain boundaries in graphene. Carbon
uniaxial tension. Nano Lett. 9:3012–3015. 51:373–380.
27. Zhao, H. and Aluru, N. R. 2010. Temperature and strain-rate 47. Wu, J. and Wei, Y. 2013. Grain misorientation and grain-
dependent fracture strength of graphene. J. Appl. Phys. 108: boundary rotation dependent mechanical properties in poly-
64321. crystalline graphene. J. Mech. Phys. Solids 61:1421–1432.
28. Wong, C. H. 2010. Elastic properties of imperfect single- 48. Okamoto, S. and Ito, A. 2013. Molecular dynamics analysis
walled carbon nanotubes under axial tension. Comp. Mater. on tensile and shearing properties of nitrogen-containing gra-
Sci 49:143–147. phene. J. Comm. Comp. 10:19–27.
29. Zheng, Q., Geng, Y., Wang, S., Li, Z., and Kim, J.-K. 2010. 49. Ito, A. and Okamoto, S. 2013. Tensile and shearing properties
Effects of functional groups on the mechanical and wrinkling of vacancy-containing graphene using molecular dynamics
properties of graphene sheets. Carbon 48:4315–4322. simulations. J. Comm. Comp. 10:9–18.
30. Pei, Q. X., Zhang, Y. W., and Shenoy, V. B. 2010. A molecu- 50. Le, M.-Q. and Batra, R. C. 2014. Crack propagation in pre-
lar dynamics study of the mechanical properties of hydrogen strained single layer graphene sheets. Comp. Mater. Sci.
functionalized graphene. Carbon 48:898–904. 84:238–243.
31. Neek-Amal, M. and Peeters, F. M. 2010. Linear reduction of 51. Kheirkhah, A. H., Iranizad, E. S., Raeisi, M., and Rajabpour,
stiffness and vibration frequencies in defected circular mono- A. 2014. Mechanical properties of hydrogen functionalized
layer graphene. Phys. Rev. B 81:235437. graphene under shear deformation: A molecular dynamics
32. Tsai, J.-L. and Tu, J.-F. 2010. Characterizing mechanical study. Solid State Commun. 177:98–102.
properties of graphite using molecular dynamics simulation. 52. Han, J., Ryu, S., Sohn, D., and Im, S. 2014. Mechanical
Mater. Design 31:194–199. strength characteristics of asymmetric tilt grain boundaries in
33. Shokrieh, M. M. and Rafiee, R. 2010. Prediction of Young’s graphene. Carbon 68:250–257.
modulus of graphene sheets and carbon nanotubes using 53. He, L., Guo, S., Lei, J., Sha, Z., and Liu, Z. 2014. The effect
nanoscale continuum mechanics approach. Mater. Design 31: of Stone–Thrower–Wales defects on mechanical properties
790–795. of graphene sheets—A molecular dynamics study. Carbon
34. Georgantzinos, S.K., Giannopoulos, G.I., and Anifantis, N.K. 75:124–132.
2010. Numerical investigation of elastic mechanical proper- 54. Brenner, D. W., Shenderova, O. A., Harrison, J. A., Stuart,
ties of graphene structures. Mater. Design 31:4646–4654. S. J., Ni, B., and Sinnott, S. B. 2002. A second-generation
35. Ansari, R., Motevalli, B., Montazeri, A., and Ajori, S. 2011. reactive empirical bond order (REBO) potential energy
Fracture analysis of monolayer graphene sheets with double expression for hydrocarbons. J. Phys.: Condens. Matter
vacancy defects via MD simulation. Solid State Commun. 14:783–802.
151:1141–1146. 55. Tersoff, J. 1989. Modeling solid-state chemistry: Interatomic
36. Zhang, Y.Y., Wang, C.M., Cheng, Y., and Xiang, Y. 2011. potentials for multicomponent systems. Phys. Rev. B
Mechanical properties of bilayer graphene sheets coupled by 39:5566–5568.
sp3 bonding. Carbon 49:4511–4517. 56. Tersoff, J. 1998. Structural properties of amorphous silicon
37. Min, K. and Aluru, N.R. 2011. Mechanical properties of gra- nitride. Phys. Rev. B 58: 8323–8328.
phene under shear deformation. Appl. Phys. Lett. 98:013113. 57. Shenderova, O. A., Brenner, D. W., Omeltchenko, A., Su,
38. Tapia, A., Peón-Escalante, R., Villanueva, C., and Avilés, X., Yang, L. H., and Young, M. 2000. Atomistic modeling
F. 2012. Influence of vacancies on the elastic properties of a of the fracture of polycrystalline diamond. Phys. Rev. B 61:
graphene sheet. Comp. Mater. Sci. 55:255–262. 3877–3888.
39. Mortazavi, B., Ahzi, S., Toniazzo, V., and Rémond, Y. 2012. 58. Ruland, W. 1965. X-ray studies on the structure of graphitic
Nitrogen doping and vacancy effects on the mechanical prop- carbons. Acta Cryst. 18:992–996.
erties of graphene: A molecular dynamics study. Phys. Lett. A 59. Stuart, S. J., Tutein, A. B., and Harrison, J. A. 2000. A reactive
376:1146–1153. potential for hydrocarbons with intermolecular interactions.
40. Wang, M. C., Yan, C., Ma, L., Hu, N., and Chen, M. W. 2012. J. Chem. Phys. 112:6472–6486.
Effect of defects on fracture strength of graphene sheets. 60. Woodcock, L. V. 1971. Isothermal molecular dynamics calcu-
Comp. Mater. Sci. 54:236–239. lations for liquid salts. Chem. Phys. Lett. 10:257–261.
61. Lee, C., Wei, X., Kysar, J. W., and Hone, J. 2008. Measurement 65. Blakslee, O. L., Proctor, D. G., Seldin, E. J., Spence, G. B.,
of the elastic properties and intrinsic strength of monolayer and Weng, T. 1970. Elastic constants of compression-annealed
graphene. Science 321:385–388. pyrolytic graphite. J. Appl. Phys. 41:3373–3382.
62. Frank, I. W., Tanenbaum, D. M., van der Zande, A. M., and 66. LiJun, Y. and TienChong, C. 2012. Loading direction depen-
McEuen, P. L. 2007. Mechanical properties of suspended gra- dent mechanical behavior of graphene under shear strain. Sci.
phene sheets. J. Vac. Sci. Technol. B 25:2558–2561. China—Phys. Mech. Astron. 55:1083–1087.
63. Shen, L, Shen, H.-S., and Zhang, C.-L. 2010. Temperature- 67. Faccio, R., Denis, P. A., Pardo, H., Goyenola, C., and Mombrú,
dependent elastic properties of single layer graphene sheets. Á. W. 2009. Mechanical properties of graphene nanoribbons.
Mater. Design 31:4445–4449. J. Phys.: Condens. Matter 21:285304.
64. Sakhaee-Pour, A. 2009. Elastic properties of single-layered
graphene sheet. Solid State Commun. 149:91–95.
4 Mechanical Properties
of Graphene Sheets
O. E. Glukhova
CONTENTS
Abstract........................................................................................................................................................................................ 61
4.1 Introduction........................................................................................................................................................................ 61
4.2 Computational Methods..................................................................................................................................................... 63
4.2.1 Method for Calculating the Local Stress Field for the Atomic Grid...................................................................... 63
4.2.2 Atom–Atom Potential Method............................................................................................................................... 65
4.2.3 Tight-Binding Method............................................................................................................................................ 65
4.3 Compression of the Single-Layer Graphene Nanoribbons................................................................................................. 66
4.3.1 Nanoribbons of Armchair Type.............................................................................................................................. 66
4.3.2 Nanoribbons of Zigzag Type.................................................................................................................................. 69
4.4 Compression of the Bilayer Graphene Nanoribbons.......................................................................................................... 71
4.5 Investigation of the Bending for Single-Layer Graphene Nanoparticles............................................................................ 72
4.6 Investigation of the Bending for Bilayer Graphene Nanoparticles..................................................................................... 75
4.7 Conclusion.......................................................................................................................................................................... 76
References.................................................................................................................................................................................... 77
ABSTRACT [1,2], spin filters [3], supercapacitors [4], and memory ele-
ments [5] are already being built on graphene bases. For suc-
The results of the theoretical investigations of the mechanical cessful realization of such devices, it is necessary to make
properties of graphene structures are presented in this work. higher demands on better characteristics of graphene. First
We have investigated the mechanical properties of one-layer of all, graphene must withstand a required exploitation load.
and bilayer graphene sheets by using the molecular dynam- Hence, graphene should have high mechanical strength.
ics method and quantum-chemical tight-binding method. To Strength characteristics of graphene and its derivations
evaluate the mechanical properties of graphene, we applied a are the subject of many theoretical and experimental studies.
compression load and deflection load to graphene sheet. As a In particular, authors in a Reference 6 preset results of the
result of the investigations, it was found that graphene became atomistic simulation for mechanical testing of polycrystalline
wave-like with the increase in the compression value. The num- graphene samples. In this work, it is shown that the presence
ber of half-waves generated on the graphene surface, depends of grain boundaries reduces the strength of graphene by about
on the size of the graphene sheet. It was established that the 50% (down to ~46 GPa). Also, authors have established that
configuration of the graphene structure with a single half-wave the crack formation occurs at points at which the boundaries
is stable for all nanoribbons, independently of the intermediate meet, and that the cracks propagate through the bulk of the
half-wave amount, and is reachable at the individually found neighboring grains typically along the armchair and zigzag
values of compression. The stress field of deformed graphene directions. The failure strain for polycrystalline graphene
was calculated by using the developed method. This method is with grain sizes ≫15 nm was found to be ~0.09, with a cor-
based on the quantum model of the finite graphene nanoribbon responding Young’s modulus of ~600 GPa.
and the empirical calculation method of the single atom energy. Scientists have devoted a great part of their attention to the
As a result of comparison of the strength characteristics for gra- study that the influence of the atomic framework defects has
phene structures it was revealed that bilayer graphene is able to on the strength characteristics of graphene. In one such work
withstand a greater load than its one-layer counterpart owing to [7], the authors investigated the influence of 5-8-5 defects
the hardening effect of the second layer. on the tensile strength of graphene sheets containing such
defects. Results of computer simulations have shown multi-
stage characters of fracture processes occurring in graphene
4.1 INTRODUCTION
sheets containing “pentagon–octagon–pentagon” defects. It
Presently, graphene is one of the most prospective materials is found that the final stage of structure fracture is realized
used in nano- and bioelectronics. Devices such as transistors by formation of carbon monatomic chains that join separate
61
parts of graphene sheets under mechanical load. As a result used method to measure the mechanical properties of gra-
of calculations of strain–stress dependencies using molecular phene. According to experimental data, graphene can be sub-
dynamics (MD) simulations, it is revealed that presence of jected to very high deformation—about 25% along the load
5-8-5 defects in graphene sheets leads to a dramatic degra- axis [9]. Graphene sheet can withstand loads of up to 42 N/m,
dation (approximately tens of percent) of tensile strength as which corresponds to 130 GPa [10].
compared to tensile strength of their pristine counterparts. It was found that bilayer graphene has properties of high
L. He et al. [8] investigated the effect of Stone–Thrower– strength. By means of MD simulation using the Brenner
Wales (STW) (5-7-7-5) defects on mechanical properties of potential, M. Neek-Amal and F. M. Peeters found that Young’s
graphene sheets with different defect orientation angles and modulus of bilayer graphene is equal to 0.8 TPa [11]. Also, in
different tilting angles of defects array using MD simula- this study, the temperature dependence of the elastic modu-
tions. During the investigation it was revealed that the ini- lus of bilayer graphene was revealed. At low temperature
tial failure state and the strength of defective graphene sheets (about 20 K), Young’s modulus of bilayer graphene is 14%
with one centered STW defect including STW-1 defect and smaller than Young’s modulus at room temperature. Results
STW-2 defect are sensitive to both defect orientation and ten- of investigations into the bending rigidity of bilayer graphene
sile direction. For STW-1 defective graphene, the strength of nanoribbons with interlayer shear effect are presented in
the armchair direction is much lower than that of the zigzag Reference 12. Two interlayer shear moduli, 0.25 and 4.6 GPa,
direction. In case of STW-2 defective graphene, the strength were obtained by sliding a small flake of graphene on a large
of the armchair direction is much higher than the strength of graphene substrate using AIREBO potential. By means of the
the zigzag direction. Also, it was found that the mechanical sandwich beam model, the authors predict the natural fre-
properties of graphene depend on the tilting angle of STW quencies of bilayer graphene nanoribbons for various values
defects, namely, the breaking strength of graphene decreases of the bending rigidity and the interlayer shear modulus.
with the increasing tilting angle. Special attention is paid to the influence of the interlayer
One effective way to estimate the strength characteristics sp3 bonding on the strength properties of bilayer graphene.
of material is external load testing in the form of various Y. Y. Zhang, C. M. Wang et al., carried out an MD study of
types of deformations: tension, compression, and bending. bilayer graphene sheets coupled with sp3 bonding [13]. The
There are some experimental methods to study the deforma- interaction between atoms was described by the method
tion of graphene. Probing the shift of phonon frequencies is based on the potential REBO. As a result of the calculations,
an effective way of assessing the degree of stress transfer of it was found that Young’s modulus of bilayer graphene with
any material under an applied stress or strain along a given sp3 bonding is equal to 0.907 TPa, and tensile strength and
axis. Raman spectroscopy has proven very successful in breaking strength of such a structure are equal to 86.55 and
monitoring phonons for a whole range of graphitic materi- 0.135 GPa, respectively.
als, including graphene under uniaxial stress or hydrostatic Construction of nanodevices based on graphene, and using
pressure. Reference 9 demonstrates results of a study of graphene as a part of composite materials requires high-
exfoliated monolayer graphene flakes that were embedded in strength properties and resistance to deformation for this
a polymer matrix and loaded under axial compression. By material. However, determination of characteristics, such as
monitoring the shifts of the 2D Raman phonons of rectangu- strength and stiffness are complicated due to the indetermi-
lar flakes of various sizes under load, the authors found that nacy of concentration of the local stresses in the fixed points
the critical strain values for first failure to be independent of of the deformed structure. Along with it, as is well known,
flake size at a mean value of –0.60% which corresponded to nonuniform distribution of stress leads to defects and also to
a yield stress of up to −6 GPa. Using Euler mechanics and the destruction of the atomic grid of the structure [14]. The clas-
Winkler approach, scientists have shown that the presence of sical methods for determination of material strength and stiff-
the polymer constraint results in graphene buckling at a fixed ness cannot be applied at the nanoscopic level. These methods
value of strain with an estimated wrinkle wavelength of the are based on the continuum theory of stress distribution.
order of 1–2 nm. According to the classical formulation, stress is a measure
Along with the study of linear elastic properties of gra- of the internal forces occurring in the deformed body under
phene structures (Young’s modulus and Poisson’s ratio), the external impact. This influence can be mechanical, thermal,
nonlinear elastic behavior of graphene under deformation is electrical, magnetic, and gravitational. In its simplified mean-
also actively investigated. In particular, the nonlinear elas- ing, stress is a force which is acting on the infinitesimal area.
tic behavior of graphene nanoribbons under quasistatic uni- This formulation is considered to be correct if the force acting
axial tension was studied using the atomistic simulation and on the infinitesimal area is uniform. Within the continuum,
continuum model [8]. Q. Lu, W. Gao et al., showed that the it is assumed that the homogeneous stress of the structure
stretching of the graphene sheet by 20% or more induces the exists within the limit of an appropriately chosen infinitesi-
appearance of an energy gap in the electronic spectrum. mal volume surrounding this point. The choice of the appro-
One of the most widely used experimental methods (such priate volume depends on the degree of the inhomogeneity of
as Young’s modulus, Poisson’s ratio, residual stresses, etc.) to the material in every considered case. The inhomogeneity of
determine the elastic constants of single-layer and bilayer gra- stress can be caused by the inhomogeneity of the material and
phene is the deflection test. Nanoindentation is also a widely the inhomogeneity of the loading conditions.
Mechanical Properties of Graphene Sheets 63
Various methods of stress determination exist within the Thus, the virial stress, shown in Equation 4.1, can be used
MD method. These are the following types of stresses: the if the stress state is homogeneous in the entire volume; the
virial stress, the BDT stress (or the atomic stress), the Lutsko Lutsko stress, shown in Equation 4.3, is calculated for the par-
stress, and the mechanical stress introduced by Cheung and tial average volume, and the homogeneity restriction is appli-
Yip. The virial stress [15] based on the Clausius virial theo- cable to that volume. If it is necessary to calculate the stress
rem for gas pressure—is used most often around a single atom, then the calculation method shown in
Equation 4.2 is used; however, its application does not always
 i  explain the observed physical phenomena. For example, in
1 1 α α α rαβ rαβj dV
∑ ij = ∑  m vi v j + ∑ . Reference 14, the maximum stress was observed on the adja-
Ωtot 2 dr  (4.1) cent carbon atoms with sp2 hybridization at the destruction of
α = 1, n  β = 1, n r αβ 
 αβ  the graphene nanoribbons as a result of the defect induced by

adsorption of six hydrogen atoms on the hexagon ring of gra-
The summation for all the atoms occupies the total volume phene. Thus, the breaking of C–C bonds, simulated by means
Ωtot. m and v denote the mass and velocity of atoms, r is the of the MD method, was carried out in the area of the reduced
distance between the atoms, dV/dr is the force acting on atom stress for the atomic grid, although it is logically assumed that
α by atom β, i and j are the indexes of the stress tensors in the the break must occur in the area of the maximum stress. This
Cartesian coordinate system. This determination is correct effect was not explained by the authors [14].
only in case the uniform stress mode exists in the overall vol- Therefore, all the above-listed methods are justified within
ume of the considered area. The other method of the atomic the determined approximation, and are applied to the concrete
structure stress calculation was proposed in Reference 16. In objects of investigation. One can say that in the present time,
this work, the authors describe a new bulk stress measure for there is no effective method for the experimental and theoreti-
a small volume Ωα around atom α. The expression for the cal determination of the local stress distribution in the atomic
calculation of the atomic stress for atom α is expressed as grid of the graphene nanostructure. However, the calculation
follows: of the local stress field for graphene is actual and gives a per-
spective for the study of its mechanical properties.
1 1 α α α 
σ αij =  m vi v j + ∑ rαβ fαβ 
j i
(4.2)
Ωα  2 β = 1, n
 4.2 COMPUTATIONAL METHODS

4.2.1 Method for Calculating the Local
The formula for the calculation of the stress for the atomic Stress Field for the Atomic Grid
structure presented above is justified only for the homoge- In all methods used for calculating the atomic grid stress
neous system. [14–19] listed above, the value of stress was determined as
Other methods of calculation of the atomic grid stress a pressure (the bulk density of energy) that an atom within a
were proposed by Lutsko, Cheung, and Yip for the study structure experiences from the environment. Since each atom
of the inhomogeneous systems. According to Cheung and permanently experiences an influence from adjacent atoms,
Yip’s definition [17], the mechanical stress is calculated as then even in the absence of the external load and external
the sum of the time rate of the change of the momentum flux influences the stress calculated by formulas 4.1 through 4.3
and forces divided by the considered area. The concept of are not always equal to zero. Notably, the stress indicates that
the local stress by Lutsko was introduced in a work [18] and no degree of deformation of the atomic structure is induced
extended by Cormier [19]. This concept is based on the local by the external factors, but the internal interaction between
stress tensor of statistical mechanics. The Lutsko stress can atoms inside the considered structure, which is primarily
be expressed as caused by the topology of the structure. Even though the
structure has no defects, and the external forces do not affect
1 1  it, the stress would not equal to zero. However, as good known
α α α
σ ijL = ∑  m vi v j + ∑ rαβ fαβlαβ 
j i
(4.3) the mechanical stress is a measure forces, incipient in the
Ω Avg 2
α = 1, n  β = 1, n

deformed body under influence of the different factors. This
stress reflects the body’s reaction to these factors.
Here, in comparison with Equation 4.2 is introduced the In this work, an original method for calculation of stress
parameter lαβ. This parameter denotes the fraction of the near the atom was proposed. The following idea is based on
length of α–β bond lying inside the average volume ΩAvg. the aforementioned method. The stress on all the atoms of a
Thereby, Equation 4.3 allows taking into account the inho- nanostructure, being in a ground state, is supposedly equal
mogeneity of the structure. The Lutsko stress has been used to zero independently on where the atom is located (on the
to evaluate the local elastic properties of grain boundaries in boundary or in the middle of the structure). Such a situation
metals. As was shown by Cormier [19], this method of calcu- corresponds to the unloaded free state of the nanostructure.
lating the stress gives a better match for the problem solution And with it, each atom possesses self-energy which is deter-
within the continuum. mined by its nature and its environment. It is assumed that the
stress is nonzero if the external factors have an effect on the chemically unbounded atoms; Vi = (4 / 3)πr0 is the occupancy
3
atomic grid: external deformed force, adsorbed atoms, elimi- volume of the atom; r 0 is the van der Waals radius of the car-
nated atoms and others. Since as a result of the external influ- bon atoms which is equal to 1.7 Å.
ence the distribution of the energy for atoms changes, then the The stress of the atomic grid near the atom with number is
value of the difference between the energy of an atom for the calculated as
deformed grid and unloaded grid would reflect the degree of
deformation in the current point of the structure, notably the
atomic grid stress near the current atom. We would consider σ i = wi − wi0 (4.5)
this value as a stress.
It is proposed to carry out the calculation of the local stress 0
field according to the following algorithm: where wi is the bulk energy density of ith atom of the gra-
phene sheet which is in equilibrium; wi is the bulk energy
1. Optimization of the initial structure by means of the density of the atom of the graphene sheet subjected to the
quantum-chemical method external influence (deformation, occurrence of the defects,
0
2. Calculation of distribution for the bulk density of etc.). The value of wi in the center of the graphene sheet is
energy by atoms using the empirical method equal to − 58.60 GPa. At the edges of the graphene sheet
3. Search of the atomic configuration for nanostruc- bulk energy density is higher since the atoms of the edges
ture subjected by external influence as a result of have only two links with other carbon atoms. It is equal to
the energy minimization by coordinates using the −41.54 GPa on the armchair edge, and on the zigzag edge is
quantum-chemical method equal to − 40.64 GPa. It is suggested that stress is equal to
4. Calculation of the distribution for the bulk density of zero for the atoms in the center and at the edges.
energy by atoms of the structure subjected to exter- To test the developed method of calculation for the local
nal influence by means of the empirical method stress field, the graphene nanoribbon with defect induced by
5. The calculation of the local stress for the atomic grid hydrogenation of one hexagon was investigated [21]. Hydrogen
by the difference between values of the bulk density atoms attached over and the bottom plane of graphene through
of energy for atoms of the structure subjected to the one: three—over plane and three—bottom plane of graphene.
external influence and initial structure In a work [14] by means of the molecular modeling method,
it was shown that in the area of the same local defect the C–C
In our investigation, the optimization of the structure after bond breaks. They breaks in hexagon, where carbon atoms
and before deformation was carried out by means of the tight- in sp3 hybridization state. In the current investigation, the
binding method. The optimization was performed by the min- distribution of the local stress in graphene nanoribbon with
imizing of the structure’s total energy for coordinates of all defect was calculated. The fragment of graphene with defect
atoms. The calculation of distribution of the bulk energy denis shown in Figure 4.1. As shown in the figure, the attached
sity on atoms was performed by the empirical method. This hydrogen atoms appreciably deform the atomic grid: a part of
method is based on the Brenner’s potential with the intro- the atoms near the defect rises over the XY plane, and a part
duced torsion interaction energy, and van der Waals interac- of the atoms go below the XY plane. Atoms denoted by circles
tion energy, in polynomial of energy function [20]. The bulk of a larger diameter are located over the plane of graphene,
energy density of the atom is calculated by the formula and a smaller diameter, at the bottom plane of graphene.
However, perturbation of the atomic grid is observed only
directly near the hydrogenated hexagon. Gray and light gray
 (∑ j ( ≠ i ) (VR (rij ) − BijVA (rij )) 
 
 + ∑ j ≠ i (∑ k ≠ i, j (∑ l ≠ i, j , k Vtors (ω ijkl ))) + ∑ j ( ≠ i ) VVdW (rij ))
wi = 1.82
Vi 1.79
0.09
0.06
(4.4)
where VR (rij) and VA(rij) are the pair of potentials of repulsion

and attraction between chemically bonded atoms which is
determined by the atom’s type and the distance between them.
rij is the distance between atoms i and j; i and j are the numbers
X
of interaction atoms; Bij is the multiparticle term correcting
interaction energy of the atoms pair i – j considering specific-
ity of interaction of σ- and π-electron clouds; Vtors(ωijkl) is the Y
torsional potential, which is the function of the linear dihedral
angle constructed on the basis of atoms with an edge on bond FIGURE 4.1 Distribution of stress in graphene fragment.
i – j (k, l are the atoms forming chemical bonds with i, j). (Adapted from O. Glukhova, M. Slepchenkov. Nanoscale 2012.
V VdW (rij) is the van der Waals interaction potential between the (11). pp. 3335–3344.)
circles correspond to the atoms in the graphene plane. The The presented empirical model allows us to calculate the
distribution of the local stress is denoted by color. Maximum atomic structure of the carbon nanostructures with an error
stress is observed in hydrogenated hexagon and quickly less than 3%.
decreases up to zero, away from the defect. According to our
map of the stress distribution, the break of the C–C bonds
4.2.3 Tight-Binding Method
can be observed at the atoms with sp3 hybridization of the
electronic clouds. This fact completely corresponds with the Within this method, the total energy of the system is calcu-
numerical experiment of the work [14]. Taking that in work lated by the formula
[14], the maximum stress is observed from the defect where
atoms are in the sp2 hybridization state. Etot = Ebond + Erep , (4.8)

One can conclude that the proposed method for calculation
of the local stress distribution for the atomic grid yields a true where Ebond is the bond structure energy calculated as the sum
view of deformation for the structure and correctly allows to of energies of the single-particle-occupied states. Erep is the
predict the place of its possible destruction. phenomenon energy. Phenomenological energy is the energy
that takes into account the nonorthogonal, electrostatic and
4.2.2 Atom–Atom Potential Method exchange-correlation interaction, and allows to estimate the
interaction energy between the atoms in the structure.
The empirical model of graphene is based on the valence force The bond structure energy is determined by the formula
field theory taking into account the van der Waals interaction
of the unbound atoms. The total energy of the finite graphene Ebond = 2∑ ε n (4.9)
sheet is presented as a polynomial whose components have n
their weight coefficients. The weight coefficients were chosen
as the result of processing the experimental data: This expression is the sum of energies of the molecular
orbitals obtained by diagonalizing the Hamiltonian. n is the
K K  number of the occupied orbitals, and εn the energy of the
Etot = ∑ K r (r − r0 )2 + ∑ K θ (θ − θ0 )2 + ∑  a − b  (4.6) single-particle orbitals.
 r12 r 6 
The phenomenon energy can be expressed as a sum of
two-body potentials as
Here, the first term takes into account the change of the
binding length in the nanostructure in relation to the binding Erep = ∑ Vrep (|ri − rj |)
length in the graphite (r0 = 1.42 Å). The second term consid- i < j
(4.10)

ers the change of angles between bindings in relation to the
angle between bindings in the graphite (θ = 120°); and the where i, j are the number of the interaction atoms; ri, rj are the
third term takes into account the van der Waals interaction Cartesian coordinates.
(Lennard–Jones potential), Kr, Kθ, Ka, Kb -weight coefficients. The overlap matrix elements are calculated by the formu-
This way of defining the total energy of carbon tubular nano- las which take into account four types of interaction: ssσ, spσ,
structures was selected as one of the most optimal. When the ppσ, and ppπ, and the pair-repulsive potential
number of terms increases for one, it becomes more difficult
to find the weight coefficients using the experimental data. A
p1   p p 
smaller number of the weight coefficients can accurately be p    r  4 p  4 
selected for the experimental data. Vijα (r ) = Vij0α  3  exp  p1  −   +  3    (4.11)
 r     p2   p2   
Weight coefficients were selected as a result of the com-   
parison with the experimental data:
p6   p p 
K r = 3.25 ⋅ 102
J
, p     r  4  p3  4  
Vrep (r ) = p5  3  exp  p6  −   +     (4.12)
rad 2  r     p2   p2   
  
J
K θ = 4.4 ⋅ 10 −19 ,
rad 2 where i and j are the orbital moments of wave function, α
presents the bond type (σ or π). The atomic terms εs, εp are
J the atomic orbital energies of carbon which are found on the
K a = 4.0 ⋅ 10 −139 , main diagonal of the Hamiltonian. The values of the εs, εp
m12
atomic terms, the parameters pn (n = 1,2,3,4,5,6) and the equi-
0 0 0
librium overlap integrals Vij0 (Vssσ , Vspσ , Vppσ
0
, Vppπ ) are given in
J Table 4.1. The coefficients p4, p1, p6 are the power exponents,
K b = 1.5 ⋅ 10 −80 . (4.7)
m6 the coefficients p5 and p3 are determined by the overall shape
length of the surface becomes obvious. The further increase

TABLE 4.1 of the axial compression again led to the topology reconstruc-
Values of the Parameters for Tight- tion of the atomic grid for the graphene nanoribbon: from two
Binding Method half-waves to one arc. The current transition was performed
εs, eV −10.932 at a compression of up to 95% of the initial length. Any fur-
εp, eV −5.991 ther compression did not change the situation. The ribbon
0
Vssσ , eV −4.344
represented the half-wave and its curvature increase with
0
compression. That is, two half-waves can be characterized as
V , eV 3.969 the intermediate state of the graphene nanoribbon before for-
spσ
V
0
, eV 5.457 mation of the stable mode which is one half-wave [21].
ppσ
0 To establish regularity of the behavior for the graphene
V , eV −1.938
ppπ structures subjected to the axial compression, a set number
p1 2.796 of numerical experiments for nanoribbons of different lengths
p2, nm 0.232 were performed. The armchair edge of the structure was elon-
p3, nm 0.154 gated, and the width remained constant. The structure was
p4 22 sequentially lengthened by means of an increase in the number
p5, eV 10.92
of hexagons per one row. The phase transition for nanoribbons
p6 4.455
of different lengths occurred at the compression of the struc-
ture of 0.3%–0.4%. Along with it, the axial compression up to
and steepness of the function, and the parameter p2 is a cut- 98% also led to the formation of a few half-waves, although
off distance for the hopping matrix elements and repulsive they were not always identical in amplitude and length.
interactions. The picture of a few identical half-waves appeared only at
The geometric and energy parameters for the carbon nano- the specified initial length. Values of the ribbon lengths were
tubes and fullerenes obtained by the tight-binding method are found. The nanoribbons with three, four, and five identical
in a good agreement with experimental data [22,23] and ab half-waves, symmetrically located along the compression
initio data [24]. The distance between carbon atoms and ion- axis, were obtained. Geometrical characteristics of armchair
ization potential are equal to 0.14 nm and 7.61 eV, respectively. graphene nanoribbons, subjected to the axial compression,
are presented in Table 4.2.
From the table data, it can be seen that the long nanorib-
4.3
COMPRESSION OF THE SINGLE- bons with three, four, and five half-waves are characterized
LAYER GRAPHENE NANORIBBONS by an identical amplitude of half-waves and an approxi-
mately identical half-wave length. Only a short nanoribbon
4.3.1 Nanoribbons of Armchair Type
has a much smaller length and amplitude of half-wave. It was
The first model for the graphene nanoribbon subjected to established that the length of half-wave ~64 Å, amplitude
axial compression is an armchair nanoribbon consisting of ~5.5 Å and 15 hexagons forming a half-wave are characteris-
646 atoms and having a length of 71 Å and a width of 22.4 Å. tic parameters for the graphene narrow armchair nanoribbon.
The topology of the atomic grid for the graphene structures The length of these narrow armchair nanoribbons exceeds the
remained planar at the initial stage of the compression. width by eight times and more. The nanoribbons subjected
However, the compression of the structures on 0.4% induced to the axial compression of up to 98% of its initial length,
appearance of the phase transition: the planar configuration from Table 4.2 and maps of the local stress distribution by
became wave-like. The further compression of the nanorib- atoms of the structure, are presented in Figures 4.2 through
bon led to the appearance of two half-waves. At 98% of the 4.5. X and Y coordinates of atoms were selected as two geo-
initial length, the picture of two half-waves distribution on the metrical coordinates, and the third axis demonstrates stress
TABLE 4.2
Geometrical Characteristics of Curved Armchair Graphene Nanoribbons Compressed Up
to 98% of Initial Length
Number of Length of Length of Amplitude of Number of Hexagons Width of
Half-Waves Nanoribbon, Å Half-Wave, Å Half-Wave, Å in Half-Wave Nanoribbon, Å
2 71.0 35.5 2.2 9 22.4
3 181.7 60.5 5.3 14
4 258.4 64.6 5.65 15
5 335.12 66.2 5.4 15
Source: O. Glukhova, M. Slepchenkov. Nanoscale 2012. (11). pp. 3335–3344.

(a)
GPa
1.8
(b)
1.6
20 1.4
0. 0. 1.2
X coordinate
0 .6
6 6
0.
0 .2
0 .2
0. 2
0.6
0 .6 1
10
0 .6 0.6
6
0.2
0. 1 0.8
0.6
0 10 20 30 40 50 60
0.4
Y coordinate
0.2
FIGURE 4.2 (a) Armchair single-layer graphene nanoribbons in a case of two half-waves. (b) Map of distribution of the local stress for
armchair nanoribbon in a case of two half-waves. (Adapted from O. Glukhova, M. Slepchenkov. Nanoscale 2012. (11). pp. 3335–3344.)
(a)
(b) GPa
1.8
1.6
1.4
1.2
20
X coordinate
0.2
0.
0.2
0.2
6
0 .2
1
0.6
10 6
0. 0. 6
0.6 0. 6
0 .2
0.2
0 0.8
20 40 60 80 100 120 140 160
Y coordinate 0.6
0.4
0.2
FIGURE 4.3 (a) Armchair single-layer graphene nanoribbons in a case of three half-waves. (b) Map of distribution of the local stress for
armchair nanoribbon in a case of three half-waves. (Adapted from O. Glukhova, M. Slepchenkov. Nanoscale 2012. (11). pp. 3335–3344.)
(a)
(b) GPa
1.8
1.6
1.4
1.2
X coordinate
20
0 .5 0.
0.5 0.5
1
0 .5
10 0.5 0.5 5
5
0.
0
50 100 150 200 250 0.8
Y coordinate
0.6
0.4
0.2
FIGURE 4.4 (a) Armchair single-layer graphene nanoribbons in a case of four half-waves. (b) Map of distribution of the local stress for
armchair nanoribbon in a case of four half-waves. (Adapted from O. Glukhova, M. Slepchenkov. Nanoscale 2012. (11). pp. 3335–3344.)
(a)
(b) GPa
1.7
1.5
1.3
X coordinate
1.1
20
10 0 .3 0.3 0.3 0.3 0.3 0.3
0.9
0
50 100 150 200 250 300 0.7
Y coordinate
0.5
0.3
0.1
FIGURE 4.5 (a) Armchair single-layer graphene nanoribbons in a case of five half-waves. (b) Map of distribution of the local stress for
armchair nanoribbon in a case of five half-waves. (Adapted from O. Glukhova, M. Slepchenkov. Nanoscale 2012. (11). pp. 3335–3344.)
on atoms. Thus, the curved ribbon is projected at the X–Y Then the simulation of the axial compression repeated
plane which allows locating of the most stressed sections. The for nanoribbons of the larger length. The zigzag edge of the
largest stress is observed near the armchair edge. It reaches nanoribbon is elongated. The phase transition for nanorib-
a value close to the critical value ~1.7–1.8 GPa. This critical bons of different length occurred at the compression of the
value was established as a result of the numerical experiments structure on 0.4%, as also for armchair ribbons. The axial
in [21]. The calculations of the energy for a nanoribbon with compression up to 98% led to the formation of a few unsym-
some remote boundary atoms that possesses maximum stress metrical half-waves. We found nanoribbons which formed
were performed. It was found that the removal of these atoms few identical half-waves at the loss of stability. Geometrical
is not energetically efficient. Therefore, the process of the characteristics of such graphene nanoribbons are presented in
removal of the atoms is endothermic and possible under the Table 4.3.
condition of imposition of the additional energy, for example, It was established that at the further increase of the compres-
using the heating process. The visible individual small islands sion, the map of the half-waves distributions began to change,
of the increased stress are caused only by mathematical fea- and the number of half-waves formed by the structure decrease
tures of the modeling during the compression process. The step-by-step. The configuration of the nanoribbon surface with
reconstruction of the atomic grid occurred with increase of one half-wave forming at the different values of compression
the compression, but the grid does not have time to transform is the stable mode for the zigzag graphene nanoribbon of dif-
in order to obtain a smaller value of the total energy. This fact ferent lengths. The symmetry of the half-waves distribution
is caused by the dynamical character of the compression pro- for the compressed nanoribbons gets violated at the further
cess. Stress fluctuations exist on each step of the compression increase of the atomic grid length faster than for the armchair
in the atomic grid. However, all of them are less than 1.8 GPa nanoribbons. The reason for such behavior of the zigzag gra-
and do not lead to appearance of defects. phene nanoribbons can be caused by the topology of its struc-
During the further growth of the atomic skeleton length of ture. The side edges of hexagons for the zigzag nanoribbons
the nanoribbon, the symmetry in the half-wave distribution on are located perpendicular to the compression axis, thereby
the surface of the structure has not been found. It was estab- blocking the symmetrical deformation of the structure.
lished that the configuration of the structure with a single From the table data it can be seen that at the increase in
half-wave is stable for all nanoribbons, independently of the length of the graphene sheet, the amplitude of half-waves and
half-wave amount, and is reachable at the individually found length of half-waves at first rapidly increase at the transition
values of compression. from two half-waves to three, and then changes insignificantly
for three and four half-waves.
Thus, on the basis of the performed simulation of the
4.3.2 Nanoribbons of Zigzag Type
axial compression process for armchair and zigzag graphene
Investigation of the zigzag graphene nanoribbons started with nanoribbons, one can conclude that the number of half-waves
a nanoribbon consisting of 550 atoms and a length of 65 Å forming at the surface of the compressed nanoribbons, is
and width of 19.88 Å. The selected nanoribbon is comparable determined by topology of the atomic grid and length of the
with the small armchair nanoribbon consisting of 646 atoms. structure [21]. The picture of the compressed nanoribbon with
The simulation of the axial compression was performed identical symmetrical half-waves of amount 5, 6 and more
according to the algorithm described above for the armchair was not obtained during the compression process with 20 m/s
nanoribbons. The phase transition from the planar structure velocity.
to the wave-like structure was observed at the compression of The pictures of the local stress distribution by coordi-
0.4%. We obtained an explicit form of the half-wave distribu- nates of all atoms were built by means of the aforementioned
tion at the compression of the nanoribbon up to 98% of the ini- method for each graphene nanoribbon subjected to the axial
tial length. At further compression, the relief of the graphene compression. The results of the calculations obtained for zig-
nanoribbon continued its transformation, and the structure zag nanoribbons of different lengths are presented in Figures
obtained a single half-wave form at a compression of 3%. 4.6 through 4.8.
TABLE 4.3
Geometrical Characteristics of Curved Zigzag Graphene Nanoribbons Compressed Up to
98% of the Initial Length
Number of Length of Length of Amplitude of Number of Hexagons Width of
2 65 32.5 2.8 12 19.88
3 165.18 55.06 5.4 20
4 198.7 49.6 5.6 20
Source: O. Glukhova, M. Slepchenkov. Nanoscale 2012. (11). pp. 3335–3344.

(a)
(b) GPa
1.7
1.6
1.5
1.4
20 1.3
1.2
1.1
Y coordinate
0. 5
1
0.
0.5
0 .5
5
10 5 0.9
0. 0.5 0.8
1
0.5
0 .5
1
0 .5
0 .5
0.7
0.6
0.5
10 20 30 40 50 60 0.4
X coordinate 0.3
0.2
0.1
0
FIGURE 4.6 (a) Zigzag single-layer graphene nanoribbons in a case of two half-waves. (b) Map of distribution of the local stress for zigzag
nanoribbon in a case of two half-waves. (Adapted from O. Glukhova, M. Slepchenkov. Nanoscale 2012. (11). pp. 3335–3344.)
(a)
(b) GPa
1.6
1.5
1.4
1.3
1.2
1.1
1
Y coordinate
0.9
10 0. 2 0.2
0 .6
0.8
0.
0 .6
2
0.6
0.6
0.7
0
0.6
20 40 60 80 100 120 140 160
0.5
X coordinate
0.4
0.3
0.2
0.1
0
FIGURE 4.7 (a) Zigzag single-layer graphene nanoribbons in a case of three half-waves. (b) Map of distribution of the local stress for
zigzag nanoribbon in a case of three half-waves. (Adapted from O. Glukhova, M. Slepchenkov. Nanoscale 2012. (11). pp. 3335–3344.)
(a)
(b) GPa
1.6
1.5
1.4
1.3
1.2
1.1
1
0.9
Y coordinate
20 0.8
10 0. 5 0.5 0.5 0. 2 0.7
0.5
0 0.6
20 40 60 80 100 120 140 160 180 0.5
X coordinate 0.4
0.3
0.2
0.1
0
FIGURE 4.8 (a) Zigzag single-layer graphene nanoribbons in a case of four half-waves; (b) Map of distribution of the local stress for zigzag
nanoribbon in a case of four half-waves. (Adapted from O. Glukhova, M. Slepchenkov. Nanoscale 2012. (11). pp. 3335–3344.)
From the figures, one can see that the stress distribution half-waves along the length of each nanoribbon are clearly
pattern is analogous to the stress distributions of armchair seen. The further increase in the axial compression again led
nanoribbons. Maximum values of the stress are concentrated to the reconstruction of the atomic grid from two half-waves
on the atoms, which are located in the area of bending of the to one. This further compression did not change the picture.
structure. It was established that the stress does not exceed the The half-wave just increased its curvature. Therefore, one
critical value of 1.7 GPa. can consider the configuration of the atomic structure with
half-wave as a stable mode for the curved bilayer graphene
4.4
COMPRESSION OF THE BILAYER nanoribbons.
To establish the behavior for the bilayer graphene struc-
GRAPHENE NANORIBBONS
tures subjected to the axial compression, a set number of
The simulation of the uniform axial compression with veloc- numerical experiments were performed for bilayer nanorib-
ity of 20 m/s was performed for bilayer graphene nanoribbons bons of different lengths. The length of each nanoribbon has
by means of the molecular modeling method. The objects been increased, whereas its width remained constant. The
of study were zigzag graphene nanoribbons. The boundary reconstruction of the atomic grid for nanoribbons of differ-
atoms, located perpendicularly to the compression axis, were ent lengths has occurred at 0.3%–0.4% compression of the
rigidly fixed. structure. Along with it, the axial compression by up to 98%
The first bilayer graphene model subjected to axial com- of its initial length led to the formation of several half-waves,
pression is presented by two zigzag nanoribbons consisting although they were not always identical by amplitudes and
of 550 atoms and having a length 65 Å and width 19.8 Å. The lengths. The picture of a few identical half-waves appeared
distance between nanoribbons was equal to 3.4 Å. It should only for the specific initial lengths, which were found. The
be noted that behavior of both graphene nanoribbons at com- nanoribbons with three and four identical half-waves, sym-
pression was identical. The topology of the atomic grid for metrically located along the compression axis, were obtained.
each of the graphene structures remained planar at the ini- Geometrical characteristics of the zigzag bilayer graphene
tial stage of the compression. However, the planar configu- nanoribbons, subjected to the axial compression, are pre-
ration of the graphene nanoribbon became wave-like at the sented in Table 4.4.
0.4% compression. Further compression led to the appear- From the table data, it can be seen that the long nanorib-
ance of two half-waves. At 2% of the compression, the two bons with three and four half-waves are characterized by
TABLE 4.4
Geometrical Characteristics of Curved Zigzag Bilayer Graphene Nanoribbons Compressed Up
to 98% of the Initial Length
Number of Length of Each Length of Amplitude of Number of Hexagons Width of
2 65 32.3 3.1 13 19.88
3 165.18 55.4 5.48 20
4 198.7 49.8 5.55 20
an approximately identical amplitude of half-waves, and an is observed near the zigzag edge. Stress fluctuations of the
almost identical number of hexagons forming the structures atomic framework were found at each step of the compression.
of half-waves. The nanoribbons subjected to the axial com-
pression by up to 98% of its initial length from Table 4.4 and 4.5
INVESTIGATION OF THE BENDING FOR
maps of the local stress distribution per atom are presented
SINGLE-LAYER GRAPHENE NANOPARTICLES
in Figures 4.9 through 4.11. The local stress on the atoms of
the curved bilayer graphene was calculated by means of the We have considered an armchair graphene fragment, con-
original developed method. The largest stress ~1.6–1.7 GPa sisting of 620 atoms with geometrical characteristics: length
(a)
(b) GPa
20
1.7
0.5
Y coordinate
0. 5 1.6
0.5
0.5 1.5
10
1.4
0. 5
1 0.5
0.5
1
0.5 0.5 1.3

1.2
1.1
10 20 30 40 50 60 1
Y coordinate 0.9
0.8
0.7
0.6
0.5 0.5
Y coordinate
0.5
0.5
0.5
1
10 0.4
0.5
0.3
0.5
0.5
0 .5
0.5 0.2
1
0.1
0
0
10 20 30 40 50 60
X coordinate
FIGURE 4.9 (a) Zigzag bilayer graphene nanoribbons in a case of two half-waves. (b) Map of distribution of the local stress for bilayer
zigzag nanoribbons in a case of two half-waves.
(a)
(b) GPa
1.7
1.6
1.5
Y coordinate
1.4
10 1.3
1.2
0 1.1
20 40 60 80 100 120 140 160
1
X coordinate 0.9
0.8
0.7
Y coordinate
0.6
10 0.3 0.3
0.3 0.3 0.5 0.5
0 0.4
20 40 60 80 100 120 140 160 0.3
X coordinate 0.2
0.1
0
FIGURE 4.10 (a) Zigzag bilayer graphene nanoribbons in a case of three half-waves. (b) Map of distribution of the local stress for bilayer
zigzag nanoribbons in a case of three half-waves.
(a)
(b) GPa
1.7
1.6
1.5
Y coordinate
20 1.4
10 1.3
1.2
20 40 60 80 100 120 140 160 180 1.1
X coordinate 1
0.9
0.8
0.7
Y coordinate
20
0.6
10 0.5 0.2 0.6 0.5
0.4
20 40 60 80 100 120 140 160 180 0.3
X coordinate 0.2
0.1
0
FIGURE 4.11 (a) Zigzag bilayer graphene nanoribbons in a case of four half-waves. (b) Map of distribution of the local stress for bilayer
zigzag nanoribbons in a case of four half-waves.
FIGURE 4.12 The deflection of the single-layer graphene sheet by the platinum pyramid.
a = 3.9 Å, width b = 41.18 Å. The deflection of the graphene 500

sheet was carried out by nanoindentation using the MD Experimental data
method of atom–atom potentials. The sheet was rigidly fixed Approximating function
at the edges. The needle of an atomic-force microscope was 400
simulated by means of the platinum pyramid, having a face-
centered cubic lattice. The distance between neighboring
Applied force, nN
atoms of the pyramid was equal to 1.42 Å. The area of the 300
upper layer of the pyramid was equal to 243.98 Å2, the area
of the bottom layer 18.14 Å2, and surface area of the pyramid
346.78 Å2 (see Figure 4.12). 200
After each step consisting of 200 time iterations, the pyra-
mid was brought closer to the graphene sheet at a distance of
δ = 0.2 Å. Single iteration in our simulations corresponds to
100
the time of Δt = 1 fs. Thus, in our study, the velocity of deflec-
tion for graphene fragments was equal to
0
δ 0 4 8 12 16
υ= = 100 m/ sec . (4.13)
200∆t Value of deflection, A
FIGURE 4.13 The dependence of the applied force on the deflec-

At each step, the values of deflection in the center of the
tion in the center of the graphene plate. The dashed line corresponds
plate and the total energy W were fixed. The force F, required to data obtained in the experiment, and the solid line corresponds to
for the deflection ζ, was calculated by the formula the approximating function.
W
F= (4.14) can be found as the ratio of limit force to the surface of the
ζ pyramid:
The dependence of the applied force on the deflection in Flim

σ critical = = 126GPa (4.16)
the center of the plate is approximated by a cubic parabola Ssurface
(see Figure 4.13):
This result agrees well with the experimentally obtained

F = 0.18ζ3 + 1.57ζ. (4.15)
data [0], and the results obtained using the density functional
theory [25].
This result is in good agreement with the experimental In one step from the breaking (13,000 iterations, t = 13 ps),
data [10]. for the considered graphene fragment, we have calculated
After 13,200 iterations, corresponding to time t = 13.2 ps, the map of the local stress experienced by the atoms of the
destruction of the bonds was observed in the graphene sheet deformed structure (see Figure 4.16). The calculation of the
(see Figures 4.14 and 4.15). stress was carried out by means of the original developed
Thus, the value of the maximum applied force that method, based on the quantum model of finite-dimensional
the studied graphene fragment can withstand, is equal to graphene nanoribbons and the empirical method for calculat-
Flimit = 437.84 nN. The critical stress for the graphene sheet ing the energy of one atom. Dark areas on this map correspond
FIGURE 4.14 The image of the platinum pyramid and graphene sheet through 13.2 ps after the start of the experiment.
40
GPa
35
40
30
28
34
28
25
28
Y coordinate
20 22
28
1
15
10
28
28
4
10
0
5
0
0 5 10 15 20 25 30 35
X coordinate
FIGURE 4.15 Image of graphene sheet through 13.2 ps after the FIGURE 4.16 Map of the local stress, calculated thought 13 ps
start of the experiment. after the start of the experiment.
to the highest local stresses, therefore, in these areas at higher pyramid. This fragment consisted of 1240 carbon atoms
stresses, it is possible to have an occurrence of defects in the and has geometrical characteristics, which are identical to
form of the bond breaking, or elimination of the atoms in the those considered above for the single-layer fragment (length
structure. Indeed, continuing the nanoindentation of the gra- a = 36.9 Å, width b = 41.18 Å). The distance between the
phene fragment (14,200 iterations, t = 14.2 fs), it is easy to see plates of bilayer graphene was equal to 3.4 Å (see Figure 4.18).
that the breaking of bonds occurred in these areas (see Figure The diagram of deflection for the bilayer graphene sheet
4.17). Thus, the local stresses Σ41.82 ± 0.82 GPa are critical completely coincided with the deflection for the single-layer
for the considered structure. graphene sheet. However, the first destructions of the C–C
bonds in the upper graphene sheet are observed earlier—after
4.6
INVESTIGATION OF THE BENDING FOR 12,200 iterations, and the force required for breaking of the
bonds in the structure is higher at 679.81 nN.
BILAYER GRAPHENE NANOPARTICLES
The dependence of the applied force on the bending deflec-
In the next stage of our study, the bilayer graphene frag- tion in the center of the upper plate can also be approximated
ment was subjected to nanoindentation of the same platinum by means of the cubic parabola (see Figure 4.19)
40
GPa
35
38
30 18
18 18
34
30
25 26
Y coordinate
22
20 18
18
14
15 10
6
18
10 2
0
5
0
0 5 10 15 20 25 30 35
X coordinate
FIGURE 4.20 Map of the local stress, calculated for the upper plate
of bilayer graphene through 12 ps after the start of the experiment.
FIGURE 4.17 Image of the graphene sheet through 14.2 ps after
the start of the experiment.
TABLE 4.5
Comparative Characteristics of the Graphene
Monolayer and the Upper Sheet of Bilayer Graphene
Time, Critical
Needed to Limit Stress for Critical
Break the Force Structure Local Stress
Object Bonds t, ps F, nN σ, GPa ∑, GPa
FIGURE 4.18 The deflection of the bilayer graphene sheet by the Graphene sheet 13.2 437.83 126 41.82 ± 0.82
platinum pyramid. Upper sheet of 12.2 679.81 196 39 ± 0.82
bilayer graphene
800
Experiment data
Approximating function F = 0.35ζ3 + 1.7ζ. (4.17)

600
This result agrees with the results of the simulation for
nanoindentation using the Brenner’s method [12].
Applied force, nN
A map of the local stress for the upper graphene sheet in

400 a single time step (200 fs) before the breaking of the bonds
was made (see Figure 4.20). The values of the local stress
Σ39 ± 0.82 GPa are critical for the current structure.
Comparison characteristics of the critical and local
200 stresses, limit force, time and magnitude of the deflection,
required for breaking of the bonds for a single-layer graphene
sheet and the upper sheet of bilayer graphene, is presented in
Table 4.5.
0
0 4 8 12 16
Value of deflection, Å 4.7 CONCLUSION
FIGURE 4.19 The dependence of the applied force on the bending A new method for calculation of the local stress field was
deflection in the center of the upper plate bilayer graphene plate. developed. The validity of this method was certified by ade-
The dashed line corresponds to data obtained in the experiment, the quacy of the correspondence for the maximum of the local
solid line corresponds to the approximating function. stress field to the curvature of the graphene nanoribbon.
Investigation of the uniform compression process of the 3. A. Saffarzadeh, R. Farghadan. A spin-filter device based on
graphene nanoribbons with a velocity of 20 m/s showed that armchair graphene nanoribbons. Appl. Phys. Lett. 2011. Vol.
their behavior differs from the behavior of the compressed 98. pp. 023106-1–023106-3.
4. Y. Wang, Z. Shi, Y. Huang, Y. Ma, C. Wang, M. Chen, Y.
nanoribbon. All nanoribbons lost stability at the compression
Chen. Supercapacitor devices based on graphene materials.
of 0.3%–0.4%, and were transformed into a few half-waves at J. Phys. Chem. C 2009. Vol. 113. pp. 13103–13107.
the compression of 2%. Along with it, the picture of the for- 5. S. K. Hong, J. E. Kim, S. O. Kim, S. Y. Choi, B. J. Cho. Flexible
mation of a few half-waves was intermediate, and at a further resistive switching memory device based on graphene oxide.
compression of 8%–10%, the nanoribbon transformed into IEEE Electron Dev. Lett. 2010. Vol. 31. (9). pp. 1005–1007.
one half-wave. At a further compression, the nanoribbon did 6. B. Mortazavi, G. Cuniberti. Atomistic modeling of mechani-
not undergo any changes. Only the chord length had changed, cal properties of polycrystalline graphene. Nanotechnology
2014. Vol. 25. p. 215704.
and the amplitude increased. The number of half-waves on
7. A. S. Kochnev, I. A. Ovid’ko, B. N. Semenov. Tensile strength
the intermediate stage is determined only by the nanoribbon of graphene containing 5-8-5 defects. Rev. Adv. Mater. Sci.
length. The length of half-waves and their amplitude is deter- 2014. Vol. 37. pp. 105–110.
mined by the topology of the atomic grid. The realization of 8. L. He, S. Guo, J. Lei, Z. Sha, Z. Liu. The effect of Stone–
the compressive strain in graphene is possible in practice. In Thrower–Wales defects on mechanical properties of graphene
particular, it was shown [26] that by employing a cantilever sheets—A molecular dynamics study. Carbon 2014. Vol. 75.
beam, one can subject single graphenes to various degrees of pp. 124–132.
9. C. Androulidakis, E. N. Koukaras, O. Frank, G. Tsoukleri, D.
axial compression. In Reference 26, the authors have mea-
Sfyris, J. Parthenios, N. Pugno, K. Papagelis, K. S. Novoselov,
sured, in detail, the compression buckling strain in single gra- C. Galiotis. Failure processes in embedded monolayer gra-
phene flakes of different geometries. phene under axial compression. Sci. Rep. 2014. Vol. 4. pp. 1–8.
Also, it was established that the elimination of the atoms, 10. M. Huang, T. A. Pascal, H. Kim, W. A. Goddard, J. R. Greer.
the rotation of the bond, or breaking of the C–C bonds was Electronic-mechanical coupling in graphene from in situ
not observed at the compression by few percent. nanoindentation experiments and multiscale atomistic simu-
During the theoretical investigation, it was found that lations. Nano Lett. 2011. Vol. 11. (3). pp. 1241–1246.
11. M. Neek-Amal, F. M. Peeters. Nanoindentation of a circular
bilayer graphene nanoribbons lost stability at 0.2%–0.4%,
sheet of bilayer graphene. Phys. Rev. B 2010. Vol. 81. (23).
and were transformed into a few half-waves at 2% compres- p. 235421.
sion. The stable state of the curved bilayer graphene is the 12. Y. Y. Zhang, C. M. Wang, Y. Cheng, Y. Xiang. Mechanical
configuration of the atomic structure of each nanoribbon properties of bilayer graphene sheets coupled by sp3 bonding.
with one half-wave which formed at 7%–10% compres- Carbon 2011. Vol. 49. (13). pp. 4511–4517.
sion. The number of half-waves on the intermediate stage is 13. R. Nazemnezhad, H. Shokrollahi, S. Hosseini-Hashemi.
defined only by the length of the nanoribbon. The length of Sandwich beam model for free vibration analysis of bilayer
graphene nanoribbons with interlayer shear effect. J. Appl.
half-waves and their amplitude is defined by the topology of
Phys. 2014. Vol. 115. p. 174303.
the atomic framework. 14. Q. X. Pei, Y. W. Zhang, V. B. Shenoy. A molecular dynamics
As a result of nanoindentation modeling by means of the study of the mechanical properties of hydrogen functionalized
MD method, the values of the limiting loading forces, the crit- graphene. Carbon 2010. Vol. 48. (3). pp. 898–904.
ical stress of the structure, the critical local stress for a single 15. S. Shen, S. N. Atluri. Atomic-level stress calculation and con-
layer, and the bilayer graphene were obtained. Comparison tinuum-molecular system equivalence. Computer Model. Eng.
characteristics of these two structures show that bilayer gra- Sci. 2004. Vol. 6. (1). pp. 91–104.
16. S. Namilae, N. Chandra, C. Shet. Mechanical behavior of
phene is able to withstand a greater load than its single-layer
functionalized nanotubes. Chem. Phys. Lett. 2004. Vol. 387.
counterpart due to the hardening effect of the second layer. pp. 247–252.
The critical stresses for single-layer and bilayer graphene 17. K. S. Cheung. S. Yip. Atomic—Level stress in an inhomoge-
are equal to 126 GPa and 196 GPa, respectively. The value neous system. J. Appl. Phys. 1991. Vol. 70. (10). pp. 5688–5690.
of the limiting strength for single-layer graphene is equal to 18. S. Namilae, N. Chandra, C. Shet. Local elastic properties of
437.83 nN, for the bilayer 679.81 nN. carbon nanotubes in the presence of Stone–Wales defects.
It was found that the chemical carbon–carbon bonds are Phys. Rev. B. 2004. Vol. 69. (9). p. 094101.
19. J. Cormier, J. M. Rickman, T. J. Delph. Stress calculation in
destroyed in the areas of the graphene network, corresponding
atomistic simulations of perfect and imperfect solids. J. Appl.
to the corner lattice sites of the platinum tip. The local stress Phys. 2001. 89. pp. 99–104.
on the atoms at the same time is equal to 41.82 ± 0.82 GPa for 20. O. E. Glukhova, A. S. Kolesnikova. Empirical modeling of lon-
single-layer graphene and 39 ± 0.82 GPa for bilayer graphene. gitudinal tension and compression of graphene nanoparticles
and nanoribbons. Phys. Solid State 2011. Vol. 53. pp. 1957–1962.
REFERENCES 21. O. Glukhova, M. Slepchenkov. Influence of the curvature
of deformed graphene nanoribbons on their electronic and
1. C. Stampfer, E. Schurtenberger, F. Molitor, J. Guttinger, T. adsorptive properties: Theoretical investigation based on the
Ihn, K. Ensslin. Carbon nanotubes: Synthesis, properties and analysis of the local stress field for an atomic grid. Nanoscale
applications. Nano Lett. 2008. Vol. 8. (8). pp. 2378–2383. 2012. (11). pp. 3335–3344.
2. Y. M. Lin, K. A. Jenkins, A. Valdes-Garcia, J. P. Small, D. B. 22. T. Yatsuhashi, N. Nakashima. Direct ionization desorption of
Farmer, P. Avouris. Operation of graphene transistors at giga- fullerene by intense femtosecond laser fields. Rev. Laser Eng.
hertz frequencies. Nano Lett. 2009. Vol. 9. (1). pp. 422–426. Suppl. 2008. Vol. 36. pp. 1000–1003.
23. L. N. Sidorov. Buckminsterfullerene, higher fullerenes, and 25. F. Liu, P. Ming, J. Li. Ab initio calculation of ideal strength
their endohedral and fluorine derivatives. Phys. Solid State and phonon instability of graphene under tension. Phys. Rev.
2002. Vol. 44. (3). pp. 413–418. B. 2007. Vol. 76. p. 064120.
24. L. Stobinski, J. Peszke, H. M. Lin. Computational studies of 26. O. Frank, G. Tsoukleri, J. Parthenios, K. Papagelis, I. Riaz, R.
SWCNTs capped by hemispheres of C60 fullerene, based on Jalil, K. S. Novoselov, and C. Galiotis. Compression behav-
semiempirical methods. Rev. Adv. Mater. Sci. 2003. Vol. 5. ior of single-layer graphenes. ACS Nano 2010. Vol. 4. (6).
pp. 363–370. pp. 3131–3138.
5 Mechanical Stabilities and
Properties of Graphene and Its
Modification by BN Predicted from
First-Principles Calculations
Qing Peng and Suvranu De
CONTENTS
Abstract........................................................................................................................................................................................ 79
5.1 Introduction........................................................................................................................................................................ 79
5.2 DFT Calculations................................................................................................................................................................ 80
5.3 Structural Properties........................................................................................................................................................... 81
5.4 Potential Surfaces............................................................................................................................................................... 81
5.5 Stress–Strain Response...................................................................................................................................................... 82
5.6 Mechanical Failures........................................................................................................................................................... 83
5.7 Elastic Constants................................................................................................................................................................. 87
5.8 Conclusions......................................................................................................................................................................... 89
Acknowledgments........................................................................................................................................................................ 89
References.................................................................................................................................................................................... 89
ABSTRACT 5.1 INTRODUCTION
We investigate the mechanical stabilities and properties of With sp2 bonds in a monatomic layer, graphene presents mar-
graphene under various strains using first-principles plane- velous electrical and mechanical properties as well as excit-
wave calculations based on density functional theory. We ing applications [1–6]. However, graphene is not perfect: a
report the mechanics dependence of the graphene-like boron zero bandgap limits its applications in field-effect transis-
nitride (g-BN) concentration, including the high-order elastic tors with high on-off ratios, as well as logic and high-speed
constants and mechanical failure. Acting anisotropically, a switching devices, since the current can never be turned off
large nonlinear elastic deformation up to the ultimate strength completely [7–10]. A significant amount of research has been
of the material, followed by a strain softening is observed. The conducted in order to modify the graphene bandgap for elec-
ultimate strains in heterogeneous configurations are smaller tronics applications. Several methods have been proposed
than those in pure graphene and g-BN, which is related to the for this purpose; however, most of them can only improve
heterogeneity of the g-BNC monolayer. The in-plane stiffness the semiconducting properties of graphene up to a limited
as well as third-order elastic constants of graphene can be lin- point [11,12]. The graphene analog of boron nitride (g-BN)
early tuned with g-BN concentration. The fourth- and fifth- has a large bandgap [13], indicating an efficient way to engi-
order elastic constants have a more complex response to BN neer the bandgap of graphene through chemical modification
modification. The longitudinal mode elastic constants are sen- using g-BN [14–27]. Recently, a promising method has been
sitive to the BN modification in contrast with the shear mode reported in which atomistic monolayers have been generated
elastic constants. The third-, fourth-, and fifth-order elastic consisting of g-BN phases in graphene (g-BNC, Figure 5.1)
constants are required for accurate modeling of the mechani- using a thermal catalytic chemical vapor deposition method
cal properties of g-BNC under strains greater than 0.02, 0.06, [27–29]. An atomically thin integrated circuitry made of
and 0.12, respectively. This study may provide guidance in g-BNC was freshly reported [30]. These hybrid monolayers
tuning the mechanical properties through chemical modifica- have been shown to have isotropic physical properties which
tion of graphene by BN to optimize the function of graphene- can be tailored by controlling the kinetic factors affecting the
based nanodevices, as well as their safe ranges of strain for g-BN domain size within the graphene layer. This is different
the demanded engineering. from B-doped or N-doped graphene, where the integrity of
79
last two domains are nonreversible. Defects are nucleated

and accumulate with the increase of the strain until rupture.
The elastic properties of g-BNC within a small strain domain
(±0.015) were reported recently [44]. However, particularly in
graphene, the nonlinear mechanical properties are prominent
since the nanoindentation tests strongly suggest that the gra-
phene films were free from defects and remained elastic until
the intrinsic strength was reached [45,46]. When the atomic
displacement is no longer small with respect to the interatomic
spacing, the anharmonicity of the crystal becomes noticeable
and needs to be treated by the higher-order elasticity theory
when the continuum approximation is adopted [47]. High-
order elastic constants are important quantities [48–50] and
can be determined by measuring the changes of sound veloci-
ties under the application of hydrostatic and uniaxial stresses
Zigzag
[51,52]. The high-order elastic constants can be utilized to

study the nonlinear elasticity [53], thermal expansion (through
the Gruneisen parameter) [54], temperature dependence of
Armchair
elastic constants [54,55], harmonic generation [52], phonon–
phonon interactions [56], photon–phonon interactions [57],
FIGURE 5.1 Atomic super cell (24 atoms) of five g-BNC configura-
lattice defects [58], phase transitions [59], echo phenomena
tions in undeformed reference configurations, corresponding to g-BN [60], and so on. The study of the nonlinear elastic properties is
concentration of 0% (graphene), 25%, 50%, 75%, and 100% (g-BN). necessary to understand the strength and reliability of struc-
tures and devices made of graphene and g-BNC.
The mechanical properties of graphene have been stud-
the g-BN structure is missing. Opposed to h-BN, which is a ied extensively [45,46,61], including theoretical predictions
bulk hexagonal structure, g-BNC and g-BN are only monoa- of the high-order elastic constants describing the nonlinear
tomically thick. Compared with the electronic [13,31–33], elastic properties [46,61]. These predictions are based on first-
optical [34], and optoelectronic properties [35], the mechan- principles calculations in a unit cell containing two or four
ics dependence of the modification of graphene by g-BN is carbon atoms. However, it was reported that the finite wave
less studied. vector soft modes, which are the key factors in limiting the
Mechanical properties are critical in designing parts or strength of monolayer materials, can only be captured in unit
structures with these advanced 2D nanomaterials regarding cells with hexagonal rings [62]. Considering the overwhelm-
their practical applications, including high-frequency field- ing importance, the nonlinear elastic properties of graphene,
effect transistors [4], graphene-based spintronics [8], ferro- g-BN, and g-BNC deserve a rigorous investigation.
magnetic [36], antiferromagnetics [37], mass sensors [38], and The goal of this chapter is to study the dependence of
nanoelectronics [39]. Aside from electronic properties such mechanical behaviors on the g-BN concentration in g-BNC,
as bandgaps [13], tunable mechanical properties can provide in order to modulate the mechanics of graphene by BN. Our
an optimal function in designing and manufacturing the gra- study also provides an accurate continuum description of the
phene-based nano- or microelectronics. Furthermore, strain elastic properties of g-BNC, including nonlinear high-order
engineering is a common and important approach to tailor elastic properties from density functional theory (DFT) calcu-
the functional and structural properties of nanomaterials [40]. lations. The total energies of the system, forces on each atom,
In addition, these 2D materials are vulnerable to strain with and stresses on the simulation boxes are directly obtained
or without intent because of the monatomic thickness [41,42]. from DFT calculations. The response of g-BNC under the
For example, there are strains caused by the mismatch of lat- nonlinear deformation and fracture is studied, including
tices or surface corrugation due to the presence of a substrate ultimate strength and ultimate strain. The high-order elastic
[39,43]. Therefore, knowledge of the mechanical properties constants are obtained by fitting the stress–strain curves to
of modified graphene by g-BN (namely, g-BNC) is highly analytical stress–strain relationships that belong to the con-
desired. Since graphene and g-BN are the lower and upper tinuum formulation [63]. The remainder of the chapter is
bounds of g-BNC (0% and 100% of g-BN concentration), we organized as follows. Section 5.2 presents the computational
consider graphene and g-BN as a special case of g-BNC in details of DFT calculations. The results and analysis are in
this study. Section 5.3, followed by conclusions in Section 5.4.
Depending on the loading, the mechanical properties are
divided into four strain domains: linear elastic, nonlinear
5.2 DFT CALCULATIONS
elastic, plastic, and fracture. Materials in the first two strain
domains are reversible, that is, they can return to equilibrium In g-BNC materials, the g-BN phase is presented by the hex-
status after the release of the loads. On the other hand, the agonal structure of BN locally, which is a six-atom hexagonal
Mechanical Stabilities and Properties of Graphene and Its Modification by BN 81
ring unit and can be denoted as (B3N3). The g-BN domain in The 14 independent elastic constants of g-BNC are deter-
g-BNC hybrid structures is modeled by adjusting the g-BN mined by a least-squares fit to the stress–strain results from
concentrations in g-BNC while maintaining the integrity of DFT calculations in two steps, detailed in our previous work
B3N3 within the system. The g-BN domain size in a system [63], which had been well used to explore the mechanical
can be equivalently expressed by the g-BN concentration in properties of 2D materials [26,71–82]. A brief introduction is
the model as (B3N3)x(C6)1−x where (C6) denotes the six-atom that, in the first step, we use a least-squares fit of five stress–
hexagonal ring unit of graphene. The six-atom hexagonal ring strain responses. Five relationships between stress and strain
unit cell is chosen to capture the “soft mode,” which is a par- are necessary because there are five independent fifth-order
ticular normal mode exhibiting an anomalous reduction in its elastic constants (FFOEC). We obtain the stress–strain rela-
characteristic frequency and leading to mechanical instabil- tionships by simulating the following deformation states: uni-
ity. This soft mode is a key factor in limiting the strength of axial strain in the zigzag direction (zigzag); uniaxial strain
monolayer materials, and can only be captured in unit cells in the armchair direction (armchair); and equibiaxial strain
with hexagonal rings [62]. This g-BNC domain model was (biaxial). From the first step, the components of the second-
successfully used to study the electronic band structures and order elastic constants (SOEC), the third-order elastic con-
linear elastic properties of g-BNC heterostructures in our pre- stants (TOEC), and the fourth-order elastic constants (FOEC)
vious work [13,44]. are overdetermined (i.e., the number of linearly independent
The stress–strain relationship of g-BN under the desired variables is greater than the number of constraints), and the
deformation configurations is characterized via first-princi- fifth-order elastic constants are well determined (the num-
ples calculations based on DFT. DFT calculations were carried ber of linearly independent variables are equal to the number
out within the Vienna Ab-initio Simulation Package [64,65] of constraints). Under such circumstances, the second step
which is based on the Kohn–Sham DFT (KS-DFT) [66] with is needed: the least-square solution to these over- and well-
the generalized gradient approximations as parameterized by determined linear equations.
Perdew, Burke, and Ernzerhof for exchange–correlation func-
tions [67]. The electrons explicitly included in the calculations
5.3 STRUCTURAL PROPERTIES
are the (2s2p1) electrons of boron, the (2s22p2) electrons of car-
bon, and (2s22p3) electrons of nitrogen. The core electrons The optimized atomistic structures of the five configurations
(1s2) of boron, carbon, and nitrogen are replaced by the pro- are shown in Figure 5.1, in ascending order of g-BN concen-
jector-augmented wave approach [68,69]. A plane-wave cutoff tration as 0% (graphene), 25%, 50%, 75%, and 100% (g-BN).
of 520 eV is used in all the calculations. The calculations are Owing to the intrinsic difference between pure g-BN and
performed at zero temperature. graphene, the lattice constants of the g-BNC mixtures are
The criterion to stop the relaxation of the electronic obtained by averaging the lattice vectors of the super cells for
degrees of freedom is set by requiring the total energy change a direct comparison with pure g-BN and graphene. We found
to be smaller than 0.000001 eV. The optimized atomic geom- that the lattice constant increases with g-BN concentration x.
etry was achieved through minimizing Hellmann–Feynman Our results of the lattice constants are summarized in Table
forces acting on each atom until the maximum forces on the 5.1, which are in good agreement with experiments conducted
ions were smaller than 0.001 eV/Å. on g-BN (2.51 Å) [83] and graphene (2.46 Å) [84].
The atomic structures of all the deformed and undeformed When the strains are applied, all the atoms are allowed
configurations are obtained by fully relaxing the 24-atom unit full freedom of motion. A quasi-Newton algorithm is used to
cells where all atoms are placed in one plane. Periodic bound- relax all atoms into equilibrium positions within the deformed
ary conditions are applied for the two in-plane directions. unit cell that yields the minimum total energy for the imposed
The irreducible Brillouin zone was sampled with a gamma- strain state of the super cell.
centered 19 × 19 × 1 k-mesh. Such a large k-mesh was used Both compression and tension are considered here in order
to reduce the numerical errors caused by the strain of the to sample a larger elastic deformation region. We study the
systems. The initial charge densities were taken as a super- behavior of the system under the Lagrangian strain ranged
position of atomic charge densities. There was a 14 Å thick from −0.1 to 0.3 with an increment of 0.02 in each step in the
vacuum region to reduce the interlayer interaction to model three deformation modes for each configuration.
the single-layer system. To eliminate the artificial effect of
the out-of-plane thickness of the simulation box on the stress,
5.4 POTENTIAL SURFACES
we used the second Piola–Kirchhoff (P–K) stress [63,70] to
express the 2D forces per length with units of N/m. Both compression and tension are considered with Lagrangian
For a general deformation state, the number of independent strains ranging from −0.1 to 0.3 with an increment of 0.02
components of the second-, third-, fourth-, and fifth-order in each step for all three deformation modes. The system’s
elastic tensors are 21, 56, 126, and 252, respectively. However, energy will increase when strains are applied. We define
there are only 14 independent elastic constants that need to strain energy per atom Es = (Etot − E0)/n, where Etot is the
be explicitly considered due to the symmetries of the atomic total energy of the strained system, E0 is the total energy of
lattice point group D6h, which consists of a sixfold rotational the strain-free system, and n is the number of atoms in the
axis and six mirror planes [46]. unit cell. This intensive quantity is used for the comparison
TABLE 5.1
Lattice Constants a, In-Plane Stiffness Ys, Poisson Ratio ν, and High-
Order Elastic Constants of g-BNC from DFT Calculations, Compared to
Previous Calculations of Graphene
0.0 0.25 0.5 0.75 1.00 Graphenea
a 2.468 2.484 2.496 2.506 2.512 2.446
Ys 344.6 321.9 300.0 285.7 278.3 348
v 0.179 0.1839 0.1937 0.1998 0.2254 0.169
C11 356 333.2 311.8 279.6 293.2 358.1
C12 63.7 61.3 60.4 59.4 66.1 60.4
C111 −3121 −2888 −2666 −2560 −2514 −2817
C112 −472 −551 −407 −383 −425 −337
C222 −2987 −2722 −2466 −2381 −2284 −2693
C1111 19980 18220 20195 18030 16547 13416
C1112 2706 3643 4267 2628 2609 759
C1122 2843 4065 2012 6755 2215 2582.8
C2222 16568 16085 19746 13234 12288 10358.9
C11111 −80198 −77944 −130712 −98102 −65265 −31383.8
C11112 −13378 −18430 −30317 −13725 −8454 −88.4
C11122 −12852 29572 −52986 −62434 −28556 −12960.5
K12222 −28504 −34777 −34287 −47317 −36955 −13046.6
K22222 −79311 −146507 −328759 −121135 −100469 −33446.7
a Graphene in a 2-atom unit cell [45].
between different systems. Figure 5.2 shows the Es as a func- the Es is a quadratic function of applied strain, can be taken
tion of strain in uniaxial armchair, uniaxial zigzag, and equib- between −0.02 < η < 0.02. The stresses, derivatives of the
iaxial deformation. strain energies, linearly increase with the increase of applied
Es is seen to be anisotropic with strain direction, consistent strains in the harmonic region. The anharmonic region is the
with the nonisotropic structure of the monolayer g-BNC. Es is range of strain where the linear stress–strain relationship is
nonsymmetrical for compression (η < 0) and tension (η > 0) invalid, and higher-order terms are not negligible. With even
for all three modes. This nonsymmetry indicates the anhar- larger loading of strains, the systems will undergo irreversible
monicity of the g-BNC structures. The harmonic region, where structural changes, and the systems are in a plastic region,
where they might fail. The maximum strain before the failure
1.2 1.2 is the critical strain. The ultimate strains are determined as
0.8 0.8 the corresponding strain of the ultimate stress, which is the
0.9 0.9
maxima of the stress–strain curve, as discussed in the follow-
Es (eV)
Es (eV)
0.6 0.6 0.6 0.6

ing section.
0.3 0.3
0.0 0.0 5.5 STRESS–STRAIN RESPONSE
–0.1 0.0 0.1 0.2 0.3 –0.1 0.0 0.1 0.2 0.3 The stress–strain (second P–K stress) (Lagrangian strain)
η armchair η zigzag relationship for uniaxial strain in the armchair and zigzag, as
0.00
well as biaxial, are shown in Figure 5.3. These stress–strain
0.9 1.6 0.25 curves reflect the facts of the nonisotropic g-BNC atomic
structures, in addition to the anharmonic responses in com-
Es (eV)
0.6 0.50
1.2
0.75 pression and tension. Overall, the stress increases with strain
0.3 1.00
monotonically in all compression cases. The linear relation-
0.0
ship between stress and strain is valid in strain from −0.02 to
–0.1 0.0 0.1 0.2 0.3 0.02, which is the linear elastic region.
η equibiaxial When the tensile strain is beyond 0.02, a nonlinear rela-
tionship of stress–strain occurs. There is a maximum stress
FIGURE 5.2 Energy–strain responses for uniaxial strain in arm- when tensile strain becomes larger than 0.14 for all five
chair and zigzag directions, and equibiaxial strains as a function of configurations. We denote this maximum stress as ultimate
g-BN concentrations. strength and the corresponding strain as ultimate strain. The
10 32
20 0.30 єam,1
5
єam,2 24
Ideal strain
0 0.25 σ am,1
Σ1(N/m)
Σ2(N/m)
0 30
5.5 16 σ am,2
–5 –2 0.20
a
a
5.0 8
–10 –40 25 0.15
4.5 0.35 0
–15 –60 32
–0.1 0.0 0.1 0.2 0.3 –0.1 0.0 0.1 0.2 0.3 єzm,1
Ideal stress (N/m)

η armchair η armchair 0.30 24
єzm,2
Ideal strain
σ zm,1
40 8 0.25 16 σ zm,2
20 4 0.20 8
Σ1(N/m)
Σ2(N/m)
0 0 8 0.15
30 0
0.35
–20 32
єbm,1
z
–4
–40 25 6 0.30
Ideal strain
24
–8
–60 0.25 16
–0.1 0.0 0.1 0.2 0.3 –0.1 0.0 0.1 0.2 0.3
η zigzag η zigzag 0.20 8 σ bm,1
0.15 0
30 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
0.00 DFT fit h-BN concentration h-BN concentration
0
Σ1(N/m)
0.25 DFT fit

30 0.50 DFT fit
–30 0.75 DFT fit FIGURE 5.4 The ideal stress (ultimate stress) as well as the cor-
b
1.00 DFT fit responding ideal (ultimate) strain at three deformation modes
–60
20 (armchair, zigzag, and equibiaxial) and two components (1,2) as a
–0.1 0.0 0.1 0.2 0.3 function of g-BN concentrations CBN.
η equibiaxial
found that the ultimate strain of a certain configuration is a

FIGURE 5.3 Stress–strain responses for uniaxial strain in arm-
chair and zigzag directions, and equibiaxial strains as a function function of both deformation mode and the directions.
of g-BN concentrations CBN. Σ1(Σ2) denotes the x(y) component of
stress. “fit” stands for the fitting of DFT calculations (“DFT”) to 5.6 MECHANICAL FAILURES
continuum elastic theory.
When a heavy strain is applied, for example, 0.3 in this
study, the system fails. Figure 5.5 shows the atomistic con-
stress–strain curves of these five configurations are quite sim- figurations of CBN = 0.25 in the Lagrangian strain of 0.16
ilar, due to their similar 2D hexagonal structures. As a con- (shown in panels a, b, and c) and 0.30 (in panels d–f). The
sequence, the stress–strain curves are overlapped. The insets failure of the g-BNC of CBN = 0.25 at uniaxial deformation
are zoomed-in plots showing different mechanical behaviors along the armchair direction starts with the debonding of
around the ultimate strains. C–C bonds which are parallel to the direction of deforma-
The stresses are the derivatives of the strain energies with tion. There are new bonds between N–N in this failure. This
respect to the strains. The ultimate strength (the maxima in could be explained by the first-neighbor valence-bond model,
the stress–strain curve) is the maximum stress that a mate- introduced by Mazzoni et al. [86]. The first-neighbor bond
rial can withstand while being stretched, and the correspond- energies between B, N, and C atoms from first-principle cal-
ing strain is the ultimate strain. Under ideal conditions, the culations are listed as follows (in units of eV) in descend-
critical strain is larger than the ultimate strain. The systems ing order of the bond strength: eNN = −178.49, eCN = −141.67,
of g-BNC under strains beyond the ultimate strains are in a eBN = −116.73, eCC = −103.24, eCB = −77.09, eBB = −50.40.
metastable state, which can be easily destroyed by long wave- There are four kinds of bonds in strain-free configurations
length perturbations, vacancy defects, as well as high tem- of CBN = 0.25: C–N, B–N, C–C, and C–B. We define a bond
perature effects [85]. The ultimate stress and strain are also as broken when the bond length is 1.1 times of the original
called ideal stress and strain. The ultimate strain is deter- bond length. It is interesting that although the weakest bonds
mined by the intrinsic bonding strength that acts as a lower are C–B bonds, the C–C bonds along armchair, as marked
limit of the critical strain. Thus, it has practical meaning in in panel (a), break first. Owing to the intrinsic nonisotropic
consideration for its applications. structure of g-BNC, applied strain on the system box will be
The ultimate stress and strain of all five configurations distributed heterogeneously on these local bonds. It seems
at three deformation modes are summarized and plotted in that the C–C bonds have higher strain concentrations and
Figure 5.4 as a function of the g-BN concentrations CBN. break prior to a C–B bond. At a strain of 0.3, as shown in
Please note that we use CBN instead of x to represent the con- panel (d), new B–B bonds and voids are generated after the
centration of the g-BN phase in the rest of the chapter. We failure.
(a) (b) (c)
c
c
c c
ηa = 0.16 ηz = 0.16 ηb = 0.16
(d) (e) (f )
ηa = 0.30 ηz = 0.30 ηb = 0.30
FIGURE 5.5 Atomic configurations of CBN = 0.25 in the Lagrangian strain of 0.16 (a)–(c) and 0.30 (d)–(f).
For the uniaxial deformation along the armchair direction, For the pure graphene (CBN = 0.0), there is only one kind
the ultimate stress of Σ1a = 5.08 and Σ a2 = 23.56 (N/m) are at of bond: C–C. The C–C bonds along the armchair break first
ηa = 0.2 and 0.18, respectively. For the uniaxial deformation during uniaxial armchair deformation. The C–C bonds along
along the zigzag, the maximum stresses are Σ1z = 26.26 (N/m) zigzag break first during uniaxial zigzag deformation. All
at ηz = 0.26 and Σ z2 = 7.82 (N/m) at ηz = 0.26. In this case also, bonds break simultaneously under equibiaxial deformation,
the C–C bonds, as marked in panel (b), break first, followed as shown in Figure 5.8.
by the breaking of B–N bonds along zigzag. At a heavy strain In the pure, g-BN (CBN = 1.0), monolayer system, only the
of 0.3, as shown in panel (e), all bonds on g-BN along zigzag B–N bonds are present. Similar to pure graphene, the B–N
are broken. For the equibiaxial deformation, the maximum bonds along the armchair break first during uniaxial armchair
stresses are Σ1b = Σ 2b = 27.81(N/m) at ηb = 0.24. C–C bonds deformation; the B–N bonds along zigzag break first during
that are parallel and next to B–N bonds break first, as shown uniaxial zigzag deformation; all bonds break simultaneously
in panel (c). At a heavy strain of 0.3, all C–B and C–N bonds under equibiaxial deformation, as shown in Figure 5.9.
are broken, as shown in panel (f). Figure 5.3 also shows that in all three deformation cases,
For the g-BNC configuration at CBN = 0.5, the B–N bonds both pure g-BN and graphene monolayers can withstand larger
along the armchair direction break first when the system strains up to the point of failure, whilst the g-BNC structures
deforms along the armchair direction, while the C–C bonds break at lower strains. The failure behavior of g-BNC is highly
along zigzag break first during uniaxial zigzag and equibi- anisotropic and is a function of the deformation direction and
axial deformation, as shown in Figure 5.6. At a heavy strain state, as well as g-BN concentration.
of 0.3, the system breaks into pieces in all cases, as shown in The strength of g-BNC heterostructures will decrease first,
Figure 5.6d through f. down to the minimum when CBN = 0.5, and increase to that of
For the g-BNC configuration at CBN = 0.75, the C–N bonds pure g-BN, as shown in Figure 5.4. The nonisotropic nature of
along armchair break first when the system deforms along ultimate strain is more sensitive to the g-BN concentration in
the armchair. The C–B bonds along zigzag break first during uniaxial strain along zigzag. The ultimate strength decreases
uniaxial, zigzag, and equibiaxial deformations, as shown in with respect to the g-BN concentration.
Figure 5.7. At a heavy strain of 0.3, the carbon atoms form There is a correlation between the debonding of structures
“islands.” There are voids in the g-BN domain during uniaxial shown in Figures 5.5 through 5.9 and ultimate strength (Figure
armchair deformation, while the system breaks into pieces in 5.3). This indicates that the instabilities occur at debonding
the other two cases, as shown in Figure 5.7d through f. sites in the atomic structures. Therefore, the observed strain
(a) (b) (c)
ηa = 0.16 ηz = 0.16 ηb = 0.16
(d) (e) (f )
ηa = 0.30 ηz = 0.30 ηb = 0.30
(a) (b) (c)
ηa = 0.16 ηz = 0.16 ηb = 0.16
(d) (e) (f )
ηa = 0.30 ηz = 0.30 ηb = 0.30
(a) (b) (c)
ηa = 0.16 ηz = 0.16 ηb = 0.16
(d) (e) (f )
ηa = 0.30 ηz = 0.30 ηb = 0.30
FIGURE 5.8 Atomic configurations of the pure graphene in the Lagrangian strain of 0.16 (a)–(c) and 0.30 (d)–(f).
(a) (b) (c)
ηa = 0.16 ηz = 0.16 ηb = 0.16
(d) (e) (f )
ηa = 0.30 ηz = 0.30 ηb = 0.30
FIGURE 5.9 Atomic configurations of the pure g-BN in the Lagrangian strain of 0.16 (a)–(c) and 0.30 (d)–(f).
softening is due to the progressive debonding in the structure C12, as well as the Poisson ratio ν, show a rather more com-
during stretching. plicated behavior (Figure 5.10). The similar trend of in-plane
stiffness and C11 are based on the fact that C11 is dominant,
about six times bigger than C12. For the same reason, Poisson
5.7 ELASTIC CONSTANTS
ratio ν has a similar trend as C12. The increment of the Poisson
The elastic constants are essential to the continuum descrip- ratio with respect to the g-BN concentration indicates a dec-
tion of the elastic properties. The continuum responses are the rement of the network connectivity [89], consistent with the
least-square fit to the stress–strain results from the DFT cal- decrement of the density.
culations, as plotted in Figure 5.3. We then have 20 values for The effect of the g-BN concentration on the higher-order
the 14 independent elastic constants of g-BNC from DFT cal- elastic constants is shown in Figure 5.11. In an overview, the
culations. The 14 independent elastic constants of g-BNC are third- and fifth-order elastic constants are all negative, in con-
determined by least-square fit to overdetermined equations trast with a positive FOEC. From the magnitude, the longi-
[63]. The results of these 14 independent elastic constants are tudinal modes (diagonal terms) of elastic constants are much
grouped in SOEC, TOEC, FOEC, and FFOEC as listed in larger than the shear modes (off-diagonal terms) of elastic
Table 5.1. constants. Comparing the elastic constants in different orders,
Owing to the symmetry, only C11 and C12 are indepen- one can notice that the TOEC of g-BNC vary mildly with the
dent. C11 decreases linearly with respect to CBN, as plotted concentrations of g-BN. However, the fourth- and fifth-order
in the top panel of Figure 5.10. The in-plane stiffness Ys elastic constants have a more complex response to the BN
Ys can be obtained from the elastic moduli C11 and C12 as modification.
Ys = (C112
− C12
2
) / C11. The Poisson ratio ν, which is the ratio By considering the variation with respect to the BN con-
of the transverse strain to the axial strain, can be obtained centration, we found that BN modification has a large effect
from elastic moduli as ν = C11/C12. Our results of Ys, ν and on the longitudinal mode elastic constants, but is insensitive
elastic constants of the five configurations are shown in the to the shear mode elastic constants. For the TOEC, the lon-
bottom panels of Figure 5.10. Our calculated value of in- gitudinal modes C111 and C222 monotonically increase with
plane stiffness of graphene (344.6 N/m) is in good agree- respect to the g-BN concentration, as opposed to the small
ment with the experimental value (340 ± 50 N/m) [45], and wiggles in the curve of shear mode C112. The ratio of C111 to
theoretical predictions (348 N/m [46] and 335 N/m [85]). C112 is around six, same as the ratio of C11/C12. For the FOEC,
Our calculated value of g-BN (278.3 N/m) agrees with an ab
initio (GGA-PW91) prediction (267 N/m in Reference 85).
Poisson ratios ν are 0.18 and 0.23 for graphene and g-BN, in 0
agreement with 0.16 and 0.21, respectively [85]. Our results
of Ys, ν, C11, and C12 are also comparable with ab initio –800
Cijk (N/m)
C111
predictions [87] and tight-binding calculations [88] of BN 1600 C112
nanotubes. C222
We found that C11 and the in-plane stiffness decrease –2400
almost linearly as the g-BN concentration increases, while
–3200
25,000
360
66 20,000
345
C11 (N/m)
C11 (N/m)
Cijkl (N/m)
330 64 150,00
C1111
C1112
315 62 10,000 C1122
C2222
300 60 5,000
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0
CBN CBN 0
0.225 –80,000
340
Cijklm (N/m)
0.210
Y9 (N/m)
320 –160,000
C11111
V
0.195 C11112
300 –240,000 C11122
C12222
280 0.180 C22222
–320,000
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
CBN CBN CBN
FIGURE 5.10 Linear elastic constants (SOEC) C11, C12, Ys, and ν FIGURE 5.11 High-order elastic constants, TOEC, FOEC, and
as a function of g-BN concentrations CBN. FFOEC as a function of g-BN concentrations CBN.
the longitudinal modes C1111 and C2222 generally decease with 60

respect to g-BN concentrations, with disturbances around Armchair
40
CBN = 0.5. The shear mode C1122 curve has larger fluctuations
than the C1112. For the fifth-order elastic constants, the longi- 20
tudinal modes C1111 and C2222 are very sensitive to the g-BN
concentration, with the minima at CBN = 0.5. The shear modes 0
DFT 4th
C11112, C11122, and C12222 are relatively inert to the g-BN con- 2nd 5th
–20 3rd
centration. The valley points of C11111 and C22222 at CBN = 0.5
indicate the instability of g-BNC around CBN = 0.5. This is
Second P-K stress (N/m)

consistent with our previous observation of bond breaking. 60 Zigzag
As a comparison of the SOEC, our results of graphene
agree with a previous study of using a 2-atom unit cell [46], 30
as summarized in Table 5.1. However, there is considerable
0
difference in the higher-order (>2) elastic constants. This is
mainly because the primitive unit cell (2-atom unit cell) does
–30
not have the freedom to distort along the K1 mode (soft mode)
as the primitive translational symmetry is enforced [62]. Such
80 Biaxial
K1 soft mode is a precursor to a phase transition as in soft-
mode theory, or directly leads to mechanical failure. This 40
failure mechanism affects the nonlinear behavior of graphene
around the ultimate strain, which is characterized by the high- 0
order elastic constants. –40
The high-order elastic constants are strongly related to
anharmonic properties, including thermal expansion, ther- –80
–0.1 0.0 0.1 0.2
moelastic constants, and thermal conductivity. With higher- Lagrangian strain η
order elastic constants, we can easily study the pressure effect
on the second-order elastic moduli, generalized Gruneisen
FIGURE 5.12 The predicted stress–strain responses from different
parameters, and equations of states. For example, when pres- orders: second, third, fourth, and fifth order, and compared with the
sure is applied, the pressure-dependent second-order elastic DFT calculations in the three deformation directions of CBN = 0.25.
moduli can be obtained from C11, C12, C22, C111, C112, C222,
Ys, and ν [63,70,90,91]. In addition, using the higher-order
elastic continuum description, one can obtain the mechani- Our results illustrate that the monatomic layer structures
cal behaviors under various loading conditions, for example, possess different mechanical behaviors in contrast with the
under applied stresses rather than strains as demonstrated in a bulk or multilayered structures, where the SOECs are suf-
previous study [46]. Furthermore, the high-order elastic con- ficient in most cases. The SOECs are relatively easier to be
stants are important in understanding the nonlinear elasticity calculated from the strain energy curves [18,85], however,
of materials such as changes in acoustic velocities due to finite they are not sufficient. The high-order elastic constants are
strain. As a consequence, nanodevices such as nanosurface required for an accurate description of the mechanical behav-
acoustic wave sensors and nanowaveguides could be synthe- iors of monatomic layer structures since they are vulnerable
sized by introducing local strain [70]. to strain due to the geometrical confinements.
A good way to check the importance of the high-order The g-BNC heterostructures are unstable under large ten-
elastic constants is to consider the case when they are miss- sion. All stress–strain curves in the previous section show
ing. With the elastic constants, the stress–strain response can that such types of material will soften when the strain is
be predicted from the elastic theory [63]. We take the con- greater than the ultimate strain. From the viewpoint of chemi-
figuration of the g-BNC with CBN = 0.25 as an example to cal bonding, this is due to the bond breaking. This softening
demonstrate here. When we only consider the second-order behavior is determined by the high-order elastic constants.
elasticity, the stress varies with strain linearly. As illustrated The high-order elastic constants reflect the high-order nonlin-
in Figure 5.12, the linear behaviors are only valid within a ear bond strength under heavy strains. The negative values of
small strain range, about −0.02 ≤ η ≤ 0.02, in the three defor- FFOECs ensure the softening of materials under heavy strain
mation directions. With the knowledge of the elastic constants beyond ultimate strains. The salient large values of FFOECs
up to the third order, the stress–strain curve can be accurately of CBN = 0.5 g-BNC structure in Figure 5.11 make it easier to
predicted within the range of −0.06 ≤ η ≤ 0.06. Using the soften at heavy strains, indicating less stability.
elastic constants up to the fourth order, the mechanical behav- Our results of mechanical properties of g-BNC are limited
iors can be well treated up to a strain as great as 0.12. For to zero temperature due to current DFT calculations. Once the
the strains beyond 0.12, the fifth-order elastic constants are finite temperature is considered, the thermal expansions and
required for an accurate modeling. The analysis of the other dynamics will, in general, reduce the interactions between
configurations comes to the same results. atoms. As a result, the longitudinal mode elastic constants
will decrease with respect to the temperature of the system. predicted up to a strain as heavy as 0.12. For the strains
The variation of shear mode elastic constants should be more beyond 0.12, the fifth-order elastic constants are required for
complex in responding to the temperature. A thorough study an accurate modeling. The high-order elastic constants reflect
will be interesting, which is, however, beyond the scope of the high-order nonlinear bond strength under large strains.
this study. The salient large values of FFOECs of the CBN = 0.5 g-BNC
structure make it easier to soften at heavy strains, indicating
less stability. Our results may provide guidance in chemically
5.8 CONCLUSIONS
tuning the mechanical properties of these heterostructures in
In summary, we proposed an accurate approach to chemically practical applications, as well as the safe ranges of strain for
tuning the mechanics of graphene with that of BN. Using a the demanded engineering of the g-BNC.
g-BNC domain model of (B3N3)x(C6)1−x, we studied the con-
centration dependence of the mechanical properties of the
hybrid graphene/boron nitride monolayer through first-prin- ACKNOWLEDGMENTS
ciples calculations based on DFT. The nonlinear elasticity of The authors would like to acknowledge the generous financial
g-BNC was investigated by applying large deformations. The support from the Defense Threat Reduction Agency (DTRA)
continuum description of the elastic response is formulated by Grant # BRBAA08-C-2-0130 and # HDTRA1-13-1-0025.
expanding the elastic strain energy density in a Taylor series
of strain truncated after the fifth-order term. We obtained a
total of 14 nonzero independent elastic constants for up to the REFERENCES
tenth-order tensor for each examined g-BNC structure. The 1. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y.
elastic constants (including those in high orders) are useful Zhang, S. V. Dubonos, I. V. Grigorieva, and A. A. Firsov.
to develop an accurate continuum description of mechanical Electric field effect in atomically thin carbon films. Science,
behaviors. 306(5696):666, Oct 22, 2004.
2. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, M. I.
The results of the DFT calculations for the three defor-
Katsnelson, I. V. Grigorieva, S. V. Dubonos, and A. A. Firsov.
mation modes of zigzag, armchair, and biaxial revealed that Two-dimensional gas of massless Dirac fermions in graphene.
graphene, g-BN, and g-BNC monolayers have anisotropic Nature, 438(7065):197–200, Nov 10, 2005.
deformations and failure behaviors. The ultimate strength is 3. Y.-M. Lin, K. A. Jenkins, A. Valdes-Garcia, J. P. Small, D. B.
also highly anisotropic in these monolayers, and changed with Farmer, and P. Avouris. Operation of graphene transistors at
the g-BN concentrations. The ultimate strains are smaller in gigahertz frequencies. Nano Lett., 9(1):422–426, Jan 2009.
g-BNC heterogeneous monolayers than those in homogeneous 4. Y. M. Lin, C. Dimitrakopoulos, K. A. Jenkins, D. B. Farmer,
H. Y. Chiu, A. Grill, and P. Avouris. 100-GHz transistors from
monolayers, graphene and g-BN, which may be related to the
wafer-scale epitaxial graphene. Science, 327(5966):662, Feb 5,
heterogeneity of the g-BNC monolayer. The failure mecha- 2010.
nism, from the viewpoint of the bond strength was studied. In 5. L. Liao, Y.-C. Lin, M. Bao, R. Cheng, J. Bai, Y. Liu, Y.
pure graphene and the 25% g-BN system, C–C bonds along Qu, K. L. Wang, Y. Huang, and X. Duan. High-speed gra-
the loading direction break first. For the 50% g-BN system, phene transistors with a self-aligned nanowire gate. Nature,
B–N bonds along the armchair break first in armchair defor- 467(7313):305–308, Sep 16, 2010.
mation, while C–N bonds along the zigzag break first in zig- 6. M.-B. Ruben, G.-N. Cristina, G.-H. Julio, and Z. Felix. 2d
materials: To graphene and beyond. Nanoscale, 3:20–30,
zag and equibiaxial deformation. For the 75% g-BN system,
2011.
C–N bonds along the armchair break first in armchair defor- 7. K. S. Novoselov, D. Jiang, F. Schedin, T. J. Booth, V. V.
mation, while C–B bonds along the zigzag break first in zig- Khotkevich, S. V. Morozov, and A. K. Geim. Two-dimensional
zag and equibiaxial deformation. In pure g-BN system, B–N atomic crystals. Proc. Nat. Acad. Sci. USA, 102(30):10451–
bonds along the loading direction break first. 10453, Jul 26, 2005.
The results show that the in-plane stiffness as well as 8. Y. Ma, Y. Dai, M. Guo, and B. Huang. Graphene-diamond
TOECs of graphene can be linearly tuned with g-BN con- interface: Gap opening and electronic spin injection. Phys.
centration. The longitudinal modes of elastic constants of Rev. B, 85:235448, Jun 2012.
9. G. Brumfiel. Graphene gets ready for the big time. Nature,
g-BNC are much larger than the shear modes of elastic con- 458(7237):390–391, Mar 26, 2009.
stants. The TOECs of g-BNC vary mildly with the concen- 10. T. P. Kaloni, Y. C. Cheng, and U. Schwingenschloegl.
trations of g-BN. However, the fourth- and fifth-order elastic Electronic structure of superlattices of graphene and hexago-
constants have a more complex response to BN modification. nal boron nitride. J. Mater. Chem., 22(3):919–922, 2012.
BN modification has a large effect on the longitudinal mode 11. T. B. Martins, R. H. Miwa, A. J. R. da Silva, and A. Fazzio.
elastic constants, but is insensitive to the shear mode elastic Electronic and transport properties of boron-doped graphene
constants. We also find that the harmonic elastic constants nanoribbons. Phys. Rev. Lett., 98(19):196803, May 2007.
12. A. Lherbier, X. Blase, Y.-M. Niquet, F. Triozon, and S. Roche.
are only valid with a small range of −0.02 ≤ η ≤ 0.02. With
Charge transport in chemically doped 2d graphene. Phys. Rev.
the knowledge of the elastic constants up to the third order, Lett., 101(3):036808, Jul 2008.
the stress–strain curve can be accurately predicted within the 13. Q. Peng and S. De. Tunable band gaps of mono-layer hex-
range of −0.06 ≤ η ≤ 0.06. Using the elastic constants up to agonal BNC heterostructures. Physica E, 44:1662–1666, Apr
the fourth order, the mechanical behaviors can be accurately 2012.
14. J. N. Coleman, M. Lotya, A. O’Neill, S. D. Bergin, P. J. King, 32. J. Li and V. B. Shenoy. Graphene quantum dots embedded in
U. Khan et al. Two-dimensional nanosheets produced by liq- hexagonal boron nitride sheets. Appl. Phys. Lett., 98(1):013105,
uid exfoliation of layered materials. Science, 331(6017):568– 2011.
571, 2011. 33. Y. Liu, X. Wu, Y. Zhao, Xiao Cheng Zeng, and J. Yang. Half-
15. A. Nag, K. Raidongia, K. P. S. S. Hembram, R. Datta, U. V. metallicity in hybrid graphene/boron nitride nanoribbons with
Waghmare, and C. N. R. Rao. Graphene analogues of bn: dihydrogenated edges. J. Phys. Chem. C, 115(19):9442–9450,
Novel synthesis and properties. ACS Nano, 4(3):1539, Mar May 19, 2011.
2010. 34. H. Park, A. Wadehra, J. W. Wilkins, and A. H. Castro
16. C. Li, Y. Bando, C. Zhi, Y. Huang, and D. Golberg. Thickness- Neto. Magnetic states and optical properties of single-layer
dependent bending modulus of hexagonal boron nitride carbon-doped hexagonal boron nitride. Appl. Phys. Lett.,
nanosheets. Nanotechnology, 20(38):385707, Sep 23, 2009. 100(25):253115, 2012.
17. L. Song, L. Ci, H. Lu, P. B. Sorokin, C. Jin, J. Ni, A. G. 35. M. Bernardi, M. Palummo, and J. C. Grossman. Optoelectronic
Kvashnin et al. Large scale growth and characterization of properties in monolayers of hybridized graphene and hexago-
atomic hexagonal boron nitride layers. Nano Lett., 10(8):3209– nal boron nitride. Phys. Rev. Lett., 108:226805, Jun 2012.
3215, 2010. 36. Y. Ma, Y. Dai, M. Guo, C. Niu, Y. Zhu, and B. Huang.
18. M. Topsakal, E. Aktürk, and S. Ciraci. First-principles study Evidence of the existence of magnetism in pristine vx2 mono-
of two- and one-dimensional honeycomb structures of boron layers (x = s, se) and their strain-induced tunable magnetic
nitride. Phys. Rev. B, 79(11):115442, Mar 2009. properties. ACS Nano, 6(2):1695–1701, 2012.
19. K. Watanabe, T. Taniguchi, and H. Kanda. Direct-bandgap 37. Y. Ma, Y. Dai, M. Guo, C. Niu, J. Lu, and B. Huang. Electronic
properties and evidence for ultraviolet lasing of hexagonal and magnetic properties of perfect, vacancy-doped, and non-
boron nitride single crystal. Nat. Mater., 3(6):404, Jun 2004. metal adsorbed mose2, mote2 and ws2 monolayers. Phys.
20. C. Zhi, Y. Bando, C. Tang, H. Kuwahara, and D. Golberg. Chem. Chem. Phys., 13:15546–15553, 2011.
Large-scale fabrication of boron nitride nanosheets and their 38. S. Y. Kim and H. S. Park. On the utility of vacancies and tensile
utilization in polymeric composites with improved thermal strain-induced quality factor enhancement for mass sensing
and mechanical properties. Adv. Mater., 21(28):2889, Jul 7, using graphene monolayers. Nanotechnology, 21(10):105710,
2009. Mar 12, 2010.
21. M. Kawaguchi, T. Kawashima, and T. Nakajima. Syntheses 39. Y. Ma, Y. Dai, W. Wei, C. Niu, L. Yu, and B. Huang. First-
and structures of new graphite-like materials of composition principles study of the graphene@MoSe2 heterobilayers. J.
bcn(h) and bc3n(h). Chem. Mater., 8(6):1197–1201, 1996. Phys. Chem. C, 115(41):20237–20241, Oct 20, 2011.
22. K. Suenaga, C. Colliex, N. Demoncy, A. Loiseau, H. Pascard, 40. F. Guinea, M. I. Katsnelson, and A. K. Geim. Energy gaps and
and F. Willaime. Synthesis of nanoparticles and nanotubes a zero-field quantum Hall effect in graphene by strain engi-
with well-separated layers of boron nitride and carbon. neering. Nat. Phys., 6(1):30–33, Jan 2010.
Science, 278(5338):653–655, 1997. 41. Q. Peng, J. Crean, A. Dearden, C. Huang, X. Wen, S. P. A.
23. W. Q. Han, W. Mickelson, J. Cumings, and A. Zettl. Bordas, and S. De, Defect engineering of 2D monatonic-layer
Transformation of bxcynz nanotubes to pure BN nanotubes. materials. Modern Phys. Lett. B, 27:1330017, 2013.
Appl. Phys. Lett., 81(6):1110–1112, 2002. 42. Q. Peng, A. Dearden, J. Crean, L. Han, S. Liu, X. Wen, and
24. T. Kawasaki, T. Ichimura, H. Kishimoto, A. A. Akbar, T. S. De, New materials graphyne, graphdiyne, graphone, and
Ogawa, and C. Oshima. Double atomic layers of graphene/ graphane: Review of properties, synthesis, and application in
monolayer h-BN on Ni(111) studied by scanning tunneling nanotechnology. Nanotechnol. Sci. Appl., 7:1–29, 2014.
microscopy and scanning tunneling spectroscopy. Surf. Rev. 43. Z. H. Aitken and R. Huang. Effects of mismatch strain and
Lett., 9(3–4):1459–1464, Jun–Aug 2002. substrate surface corrugation on morphology of supported
25. Q. Tang, Z. Zhou, and Z. Chen. Molecular charge transfer: monolayer graphene. J. Appl. Phys., 107(12):123531, Jun 15,
A simple and effective route to engineer the band struc- 2010.
tures of BN nanosheets and nanoribbons. J. Phys. Chem. C, 44. Q. Peng, A. R. Zamiri, W. Ji, and S. De. Elastic properties
115(38):18531–18537, Sep 29, 2011. of hybrid graphene/boron nitride monolayer. Acta Mechanica,
26. Q. Peng, W. Ji, and S. De. First-principles study of the effects 223:2591–2596, 2012.
of mechanical strains on the radiation hardness of hexagonal 45. C. Lee, X. Wei, J. W. Kysar, and J. Hone. Measurement of
boron nitride monolayers. Nanoscale, 5:695–703, 2013. the elastic properties and intrinsic strength of monolayer gra-
27. L. Ci, L. Song, C. Jin, D. Jariwala, D. Wu, Y. Li, A. Srivastava phene. Science, 321(5887):385, 2008.
et al. Atomic layers of hybridized boron nitride and graphene 46. X. Wei, B. Fragneaud, C. A. Marianetti, and J. W. Kysar.
domains. Nat. Mater., 9(5):430–435, May 2010. Nonlinear elastic behavior of graphene: Ab initio calculations
28. J. da Rocha Martins and H. Chacham. Disorder and segrega- to continuum description. Phys. Rev. B, 80(20):205407, Nov
tion in b-c-n graphene-type layers and nanotubes: Tuning the 2009.
band gap. ACS Nano, 5(1):385, 2011. 47. G. Leibfried and W. Ludwig. Theory of anharmonic effects in
29. S. Bhowrnick, A. K. Singh, and B. I. Yakobson. Quantum dots crystals. Solid State Phys., 12:275–444, 1961.
and nanoroads of graphene embedded in hexagonal boron 48. R. N. Thurston and K. Brugger. Third-order elastic con-
nitride. J. Phys. Chem. C, 115(20):9889, 2011. stants + velocity of small amplitude elastic waves in homoge-
30. M. P. Levendorf, C.-J. Kim, L. Brown, P. Y. Huang, R. W. neously stressed media. Phys. Rev. A, 133(6A):1604, 1964.
Havener, D. A. Muller, and J. Park. Graphene and boron 49. K. Brugger. Thermodynamic definition of higher order elastic
nitride lateral heterostructures for atomically thin circuitry. coefficients. Phys. Rev. A, 133(6A):1611, 1964.
Nature, 488(7413):627–632, Aug 30, 2012. 50. Y. Hiki. Higher-order elastic-constants of solids. Annu. Rev.
31. Y. Ding, Y. Wang, and J. Ni. Electronic properties of graphene Mater. Sci., 11:51, 1981.
nanoribbons embedded in boron nitride sheets. Appl. Phys. 51. K. Brugger. Determination of 3rd-order elastic coefficients in
Lett., 95(12):123105, 2009. crystals. J. Appl. Phys., 36(3P1):768, 1965.
52. W. B. Gauster and M. A. Breazeale. Ultrasonic measurement 72. Q. Peng, C. Liang, W. Ji, and S. De. A first principles investi-
of the nonlinearity parameters of copper single crystals. Phys. gation of the mechanical properties of g-ZnO: The graphene-
Rev., 168:655–661, Apr 1968. like hexagonal zinc oxide monolayer. Comput. Mater. Sci.,
53. H. Kobayashi and Y. Hiki. Anharmonicity in noble metals; 68:320–324, 2013.
nonlinear elasticity in whiskers. Phys. Rev. B, 7:594–601, Jan 73. Q. Peng, C. Liang, W. Ji, and S. De. A theoretical analysis of
1973. the effect of the hydrogenation of graphene to graphane on its
54. Y. Hiki, J. F. Thomas, and A. V. Granato. Anharmonicity in mechanical properties. Phys. Chem. Chem. Phys., 15:2003–
noble metals: Some thermal properties. Phys. Rev., 153:764– 2011, 2013.
771, Jan 1967. 74. Q. Peng, X.-J. Chen, S. Liu, and S. De. Mechanical stabilities
55. J. A. Garber and A. V. Granato. Theory of the temperature and properties of graphene-like aluminum nitride predicted
dependence of second-order elastic constants in cubic materi- from first-principles calculations. RSC Adv., 3:7083–7092,
als. Phys. Rev. B, 11:3990–3997, May 1975. 2013.
56. W. P. Mason and T. B. Bateman. Relation between 3rd-order 75. Q. Peng, C. Liang, W. Ji, and S. De. A first-principles study of
elastic moduli and thermal attenuation of ultrasonic waves the mechanical properties of g-GeC. Mech. Mater., 64:135–
in nonconducting and metallic crystals. J. Acoust. Soc. Am., 141, 2013.
40(4):852, 1966. 76. Q. Peng, C. Liang, W. Ji, and S. De. Mechanical properties of
57. J. Melngailis, A. A. Maradudin, and A. Seeger. Diffraction g-GaN: A first principles study. Appl. Phys. A, 13:483–490,
of light by ultrasound in anharmonic crystals. Phys. Rev., 2013.
131:1972–1975, Sep 1963. 77. Q. Peng, X.-J. Chen, W. Ji, and S. De. Chemically tuning
58. A. D. Brailsfo. Anharmonicity contributions to dislocation mechanics of graphene by BN. Adv. Eng. Mater., 15:718–727,
drag. J. Appl. Phys., 43(4):1380, 1972. 2013.
59. M. Wuttig and T. Suzuki. Non-linear anelasticity and the 78. Q. Peng, X. Wen, and S. De. Mechanical stabilities of silicene.
martensitic-transformation. Acta Metall., 27(5):755–761, RSC Adv., 3:13772–13781, 2013.
1979. 79. Q. Peng, Z. Chen, and S. De. A density functional theory study
60. K. Fossheim, K. Kajimura, T. G. Kazyaka, R. L. Melcher, and of the mechanical properties of graphane with van der Waals
N. S. Shiren. Dynamic polarization echoes in piezoelectric corrections. Mech. Adv. Mater. Struc., 22:717–721, 2015.
powders. Phys. Rev. B, 17:964–998, Feb 1978. 80. Q. Peng and S. De. Outstanding mechanical properties of
61. F. Liu, P. Ming, and J. Li. Ab initio calculation of ideal strength monolayer MoS2 and its application in elastic energy storage.
and phonon instability of graphene under tension. Phys. Rev. Phys. Chem. Chem. Phys., 15:19427–19437, 2013.
B, 76(6):064120, Aug 2007. 81. Q. Peng and S. De, Elastic limits of silicane. Nanoscale,
62. C. A. Marianetti and H. G. Yevick. Failure mechanisms of B:12071–12079, 2014.
graphene under tension. Phys. Rev. Lett., 105:245502, 2010. 82. Q. Peng and S. De, Mechanical properties and instabilities of
63. Q. Peng, W. Ji, and S. De. Mechanical properties of the hexag- ordered graphene oxide C6O monolayers, RSC Adv., 3:24337–
onal boron nitride monolayer: Ab initio study. Comput. Mater. 24344, 2013
Sci., 56:11, 2012. 83. L. Liu, Y. P. Feng, and Z. X. Shen. Structural and electronic
64. G. Kresse and J. Hafner. Ab initio molecular dynamics for properties of h-bn. Phys. Rev. B, 68(10):104102, Sep 2003.
liquid metals. Phys. Rev. B, 47:558, 1993. 84. Y. Baskin and L. Meyer. Lattice constants of graphite at low
65. G. Kresse and J. Furthuller. Efficiency of ab-initio total temperatures. Phys. Rev., 100(2):544, 1955.
energy calculations for metals and semiconductors using a 85. M. Topsakal, S. Cahangirov, and S. Ciraci. The response of
plane-wave basis set. Comput. Mater. Sci., 6:15, 1996. mechanical and electronic properties of graphane to the elas-
66. W. Kohn and L. J. Sham. Self-consistent equations including tic strain. Appl. Phys. Lett., 96(9):091912, Mar 1, 2010.
exchange and correlation effects. Phys. Rev., 140(4A):A1133, 86. M. S. C. Mazzoni, R. W. Nunes, S. Azevedo, and H. Chacham.
Nov 1965. Electronic structure and energetics of layered structures. Phys.
67. J. Perdew, K. Burke, and M. Ernzerhof. Generalized gradi- Rev. B, 73(7):073108, 2006.
ent approximation made simple. Phys. Rev. Lett., 77:3865, 87. K. N. Kudin, G. E. Scuseria, and B. I. Yakobson. bn, and c
1996. nanoshell elasticity from ab initio computations. Phys. Rev. B,
68. P. E. Blöchl. Projector augmented-wave method. Phys. Rev. B, 64(23):235406, 2001.
50(24):17953–17979, Dec 1994. 88. E. Hernandez, C. Goze, P. Bernier, and A. Rubio. Elastic prop-
69. R. O. Jones and O. Gunnarsson. The density functional for- erties of c and bxcynz composite nanotubes. Phys. Rev. Lett.,
malism, its applications and prospects. Rev. Mod. Phys., 80(20):4502, May 18, 1998.
61(3):689–746, Jul 1989. 89. G. N. Greaves, A. L. Greer, R. S. Lakes, and T. Rouxel.
70. Q. Peng, W. Ji, and S. De. Mechanical properties of graphyne Poisson’s ratio and modern materials. Nat. Mater., 10(11):823–
monolayer: A first-principles study. Phys. Chem. Chem. Phys., 837, Nov 2011.
14:13385–13391, 2012. 90. J. F. Nye. Physical Properties of Crystals. Oxford Science
71. Q. Peng, C. Liang, W. Ji, and S. De. A first principles investi- Publications, Oxford, 1995.
gation of the mechanical properties of g-TlN. Model. Numer. 91. S. Yu Davydov. Third order elastic moduli of single layer gra-
Simul. Mater. Sci., 2:76–84, 2012. phene. Phys. Solid State, 53(3):665, Mar 2011.
6 Some Mechanical Properties of
Graphene and Their Role in Forming
Polymer Nanocomposites
Kausala Mylvaganam and Liangchi Zhang
CONTENTS
Abstract........................................................................................................................................................................................ 93
6.1 Introduction........................................................................................................................................................................ 93
6.2 Mechanics of Structural Instability of Graphene............................................................................................................... 94
6.3 Controlled Ripple Texturing of Graphene.......................................................................................................................... 95
6.4 Graphene-Based Polymer Nanocomposites........................................................................................................................ 95
6.4.1 Nanocomposites with Noncovalent Forces at the Interface.................................................................................... 96
6.4.2 Nanocomposites with Covalent Bonds between Polymer and Graphene-Based Fillers........................................ 96
6.4.3 Preservation of Graphene Properties while Forming Chemical Bonds................................................................. 97
6.5 Key Challenges in the Formation of Composites............................................................................................................... 99
6.6 Assessment of Reinforcement........................................................................................................................................... 100
6.7 Applications.......................................................................................................................................................................101
6.8 Summary and Future Scope..............................................................................................................................................101
References.................................................................................................................................................................................. 102
ABSTRACT This has brought about a significant wave of investigations for

a precise characterization of their properties [4–8]. Among
Graphene, a two-dimensional material with a large specific these nanoscale carbon materials, graphene (a single layer of
surface area, has outstanding thermal, mechanical, and elec- graphite) is a two-dimensional material with a large specific
trical properties. This chapter discusses the key material surface area and, therefore, provides a special advantage over
properties that determine the resistance of graphene to its other nanostructures. In terms of its elastic stiffness and frac-
structural instability, and the effectiveness of controlled gra- ture strength, it could be the strongest material ever obtained.
phene wrinkles in the formation of composites with polymers, It has a Young’s modulus as high as 1 TPa and a strength of
in order to make advanced multifunctional materials. This about 130 GPa [9]. Hence, it is an attractive candidate for
chapter further points out that the exceptional mechanical making advanced multifunctional materials.
properties of graphene could be utilized by forming chemi- When graphene was first recognized in 2004, it was iso-
cal bonds at the graphene–polymer interface on functional- lated from graphite via mechanical exfoliation by using a
ized graphene sheets. A focus of the chapter is to understand simple scotch tape [3,10]. The applicability of this method
the mechanism associated with the formation of such com- is limited due to lack of scalability. Since then, several
posites while maintaining the intact structure of graphene, approaches have been employed for large-scale production,
and to demonstrate how the chemical reaction helps to deal such as those involving epitaxial growth on SiC substrate [11–
with the challenging problem of restacking graphene sheets. 13], chemical vapor deposition (CVD) [14,15] (often known
Following this, the chapter discusses challenges to the forma- as bottom-up methods), and chemical exfoliation methods
tion of composites, ways of assessing reinforcement, and the [16,17] (termed as top-down methods). Capabilities of various
applications of the resultant composites. production techniques are summarized in Table 6.1. Methods
associated with oxidation of graphite followed by exfoliation
of graphite oxide to graphene oxide, and then reduction of
6.1 INTRODUCTION
graphene oxide have been more preferable in the synthesis
Since the recognition of nanoscale carbon materials, such as of large-area graphene. To prepare graphene-based compos-
buckyballs (in 1985) [1], carbon nanotubes (CNT) (in 1991) ites, these methods have been employed due to their low cost
[2], and graphene (in 2004) [3], their potential applications and high yield [18–20]. Graphene oxide has many attractions
have led to extensive research efforts because of their supe- over graphene; it is readily available in large quantities, much
rior mechanical, electrical, electronic, and thermal properties. easier to exfoliate and disperse in a polymer matrix, and it has
93
any bending resistance [9]. On the other hand, Hemmasizadeh

TABLE 6.1 et al. [21] and Duan et al. [22] considered the monolayer
Comparison of Various Graphene Production graphene as a thin plate with a nonzero bending stiffness
Techniques to develop an equivalent-continuum model for studying the
mechanical behavior of single-layered graphene sheets. To
Techniques Comments
analyze a monolayer GNS either as an elastic membrane or an
Mechanical exfoliation from Led to the first recognition of
elastic thin plate, which are the concepts in the conventional
graphite using a scotch tape [3,10] graphene in 2004, lacks scalability
Oxidation of graphite, followed by More preferable, can produce
continuum mechanics theory, “effective quantities” such as
exfoliation to graphene oxide, large-area graphene “effective thickness” and “effective Young’s modulus” need
followed by reduction [18–20] to be introduced [5–8]. However, in some cases, such equiva-
Chemical exfoliation [16,17] Versatile and scalable lent quantities may not be reliably defined at the nanoscopic
Epitaxial growth on SiC substrate Can synthesize large-area graphene, scale. As such, in principle, continuum mechanics theory is
[11–13] sublimation of silicon from SiC not applicable at the nanoscopic scale. The flowchart given
requires elevated temperatures in Figure 6.1 describes the concepts used in studying the
(>1200°C) mechanical properties.
Chemical vapor deposition of Can synthesize large-area graphene For example, Wang et al. performed a molecular dynamics
methane [14,15] simulation-based nanoindentation to characterize the proper-
ties of nanomaterials, and incorporated these results into a
continuum mechanics theory [7]. To simulate the nanoindenta-
built-in functional groups available to form a strong interface tion process, they applied a point load at the center of a simply
with a polymer matrix. However, the mechanical properties of supported circular graphene sheet of radius 16 nm using a tiny
graphene oxide are inferior to those of graphene. indenter tip of radius 0.535 nm. At a critical indentation depth
Graphene sheets exhibit intrinsic microscopic roughening of around 0.7 nm, a sudden ripple-like transverse deflection
such that the surface normal varies by several degrees. As such (i.e., wrinkling) emerged. The postrippled structure (Figure
they are not perfectly flat—ripples are formed. In other words, 6.2a) showed that the wrinkling displacement was almost zero
ripples are an intrinsic feature of a graphene sheet. However, in the vicinity of the graphene center as well as at the simply
the ripples are expected to strongly influence electronic prop- supported boundary. Further increasing the indentation depth
erties by inducing effective magnetic fields and changing local led to greater ripple-like displacements and a larger number
potentials as well as generating sites of enhanced reactivity of ripples in the circumferential direction as shown in Figure
due to the curvature-induced stress. While a rippled surface 6.2b. Clearly, the molecular dynamics (MD) simulation
caused by the structural instability (i.e., the random ripples) results showed that graphene behaves in a fundamentally dif-
could significantly degrade graphene, controlled ripple textur- ferent manner from a membrane without a bending stiffness;
ing could lead to strong graphene nanosheet (GNS) composite a membrane would wrinkle whenever a compressive stress is
formation and device design based on local strain and selec- generated. Consistent with the works on CNTs, this indicates
tive bandgap engineering. Hence, understanding the mechan- that thin elastic plate theory with a nonzero bending stiffness
ics and capturing the key factors that are responsible for the is more appropriate for treating graphene.
structural instability of graphene as well as ways of forming With a two-dimensional elastic shell model [23–25], for
controlled ripples are of primary importance. single-walled CNTs, it has been shown that the deformation
Hence, the first part of this chapter will focus on the mate-
rial properties that determine the resistance of graphene to
structural instability, and on the methods that can control Mechanical properties of graphene
ripple formation. The chapter will then discuss the means of using continuum model
harvesting the outstanding mechanical and functional proper-
ties of graphene by forming nanocomposites with polymers.
The challenges in the formation of polymer composites, ways
of assessing the reinforcement, and the applications will also Graphene as membrane without Graphene as thin plate with
be discussed. bending resistance [9] bending stiffness [21–22]
6.2 MECHANICS OF STRUCTURAL
INSTABILITY OF GRAPHENE Requires effective thickness and effective Young’s
modulus—not defined at nanoscale [5–8]
The conventional experimental methods for measuring the
mechanical properties of graphene are very challenging
because of its nanoscale dimension. Lee et al. used atomic Requires multiscale approach
force microscope nanoindentation to measure the mechanical
properties of monolayer graphene suspended over open holes FIGURE 6.1 Flowchart describing the concepts used in studying
by treating it as a membrane of a fictitious thickness without the mechanical properties of graphene.
Some Mechanical Properties of Graphene and Their Role in Forming Polymer Nanocomposites 95
(a) (b) θ (degree)

0
–1 0 150 300 450
(w0 = 0.7 nm)
–2
Deflection (10–1 nm)

–3
–4 (w0 = 0.9 nm)
–5 (w0 = 1.1 nm)

–6
(w0 = 2.1 nm)
–7
–8
–9
FIGURE 6.2 (a) Postwrinkling mode and (b) the wrinkling displacement along the circumferential direction at a radius of 11 nm of a
simply supported monolayer graphene predicted by MD simulation under indentation. (Adapted from Wang C Y, Mylvaganam K and Zhang
L C. 2009. Phys. Rev. B 80, 155445-5.)
is controlled by four stiffnesses: in-plane extension stiffness

(Kextension), in-plane torsion stiffness (Ktorsion), bending stiff-
ness (Dbending), and off-plane torsion stiffness (Dtorsion). Dbending
and Dtorsion quantify the resistance to changes of the off-plane
π-bonds due to off-plane deformation (e.g., bending), while
Kextension and Ktorsion reflect its resistance to the variation in the
in-plane deformation (e.g., in-plane stretching) [24–26]. By
treating the monolayer graphene as a 2D material, and incor-
porating the MD simulation observations and results, Wang
et al. [7] showed that the wrinkling in graphene induced by 300 K 475 K 300 K
the circumferential compression with the critical value of the
indentation depth at the onset of wrinkling is solely deter- FIGURE 6.3 In situ SEM images of a single-layer graphene
mined by the ratio of Dbending/Kextension. This is in accordance through a thermal cycle. (Adapted from Bao W Z et al. 2009. Nat.
with the numerical results based on the von Karman plate the- Nanotechnol. 4, 562–6.)
ory for circular plates [22]. Thus, the resistance of graphene
to structural instability is determined by the two key material
properties, the bending stiffness and the in-plane stiffness. periodic ripples. On annealing in a furnace in argon up to
700 K and imaging at room temperature, the graphene sam-
ples underwent dramatic morphology changes. They found
6.3 CONTROLLED RIPPLE
that the preexisting ripples almost disappeared when the
TEXTURING OF GRAPHENE temperature rose to 450–600 K, and then on cooling, ripples
Ripples are an intrinsic property of graphene and they form with larger amplitudes and longer wavelengths were formed,
due to thermal fluctuation. The curvature-induced stress in that is, annealing to a higher temperature and then cooling
ripples will generate sites of enhanced reactivity. Since the down to 300 K produced larger ripples (Figure 6.3). These
intrinsic ripples are distributed randomly over the surface, demonstrate that the formation of ripples is controllable by
reactions with other substances will take place at random thermal treatment.
positions. Thus, the superior properties of graphene can be
lost and, thereby, significantly degrade the graphene. Bao 6.4 GRAPHENE-BASED POLYMER
et al. showed that one- and two-dimensional periodically
NANOCOMPOSITES
controlled ripples in suspended graphene sheets can be cre-
ated using both spontaneously and thermally generated Similar to the difficulties in the separation of CNTs [28], the
strains [27]. They controlled ripple orientation, wavelength, dispersion of graphene sheets is also challenging due to the
and amplitude by controlling boundary conditions, and by aggregation of graphene sheets caused by the van der Waals
making use of the negative thermal expansion coefficient forces. In the literature, a variety of processing routes have
of graphene. In other words, the ripples can be controllably been reported for dispersing GNSs and graphene oxide-
produced by means of simple thermal manipulation. On sus- derived fillers into polymer matrices. Most dispersion meth-
pending graphene sheets across predefined trenches on Si/ ods produce composites that are noncovalent assemblies
SiO2 substrates and examining their morphology at differ- where the interactions between the polymer matrix and the
ent temperatures using a scanning electron microscope, Bao filler are relatively weak dispersive forces. The properties
et al. found that the graphene sheets were not flat, but with and performances of graphene–polymer composites depend
not only on the quality of graphene-based filler and polymer methods; the addition of nanomaterials increases the viscosity
matrix, but also on the dispersion of the filler, the bonding of the polymer melt significantly and makes the processing
between the filler and matrix, and the ratio of filler to matrix. more difficult.
To effectively make use of the superior mechanical In situ polymerization involves mixing of a graphene-
strength of graphene, in addition to dispersion, strong inter- based filler with a monomer, followed by polymerization.
facial strength between GNSs and polymer is necessary for This technique provides a high level of dispersion without a
efficient stress transfer from graphene to polymer matrix prior exfoliation step. It has been employed to successfully
[28,29]. This means graphene should first be functionalized. produce composites of poly(ethylene) [39] and poly(methyl
Thus, dispersion and functionalization of graphene sheets are methacrylate) (PMMA) [40,41] with enhanced properties.
of crucial importance for their end applications. A coupling For example, with loadings as low as 0.05 wt % of graphene
agent is often used to obtain good dispersion. For example, oxide-filled PMMA composites showed a large shift (over
when making a nanocomposite with high-density polyethyl- 15°C) in glass transition temperature versus neat PMMA. The
ene, Wang and Chen used a surfactant to prevent aggregation morphology of the composite can be controlled by tuning the
of GNSs and obtained an increase of 290% in the elonga- polymerization conditions. For example, in the preparation of
tion with 10 wt % GNSs [30]. Macosko et al. compared the graphene oxide/polyaniline nanocomposite, for a concentra-
properties of linear low-density polyethylene (LLDPE)–GNS tion of 0.36 mg mL −1 graphene oxide, the optimized concen-
nanocomposites using neat and functionalized LLDPE, and tration of aniline is 0.05 M; a lower concentration of aniline
concluded that the GNSs disperse better in functionalized resulted in the growth of shorter polyaniline chains, and a
LLDPE, and thereby, enhance the mechanical properties. higher concentration produced both randomly connected and
aligned polyaniline chains.
6.4.1 Nanocomposites with Noncovalent
Forces at the Interface 6.4.2 Nanocomposites with Covalent Bonds
between Polymer and Graphene-Based Fillers
Noncovalent forces at the interface do not disrupt the extended
π-conjugation on the graphene surface. Similar to the conven- Grafting of polymer chains to a surface provides a versatile
tional polymer processing, solution mixing, melt blending, tool for surface modification and functionalization. When
and in situ polymerization are applied in the fabrication of making graphene-based polymer nanocomposites, the graft-
graphene–polymer composites. ing reaction may proceed either by grafting end-functional-
Solution mixing requires both graphene and polymer to be ized polymer chains to the graphene surface (“grafting to”)
dispersed in a common solvent. The method is straightfor- or by polymerization from the functionalized graphene sur-
ward, requires no special instruments, and allows for large- face (“grafting from”). Both approaches have been used to
scale production. It has been used for incorporating graphene introduce strong chemical bonds between polymer chains
oxide-derived fillers (having good dispersibility in water or and graphene oxide. In the “grafting to” approach, polymer
polar organic solvents) into a variety of polymers such as chains are covalently bonded onto graphene by esterification
polystyrene [31], polyimides [32], polymethyl methacrylates or amidation. In the “grafting from” approach, vinyl poly-
[33,34], and so on. The graphene oxide in the composite can mer molecules such as polystyrene and polyethylene glycol
be converted to conductive reduced graphene oxide upon are covalently and stably anchored on graphene to produce
chemical or thermal reduction. In the solution-mixing meth- single-layer functionalized graphene from graphene oxide
ods, the dispersion of platelets in the composite is largely [42]. In both approaches, the reaction is between preformed
governed by the level of exfoliation of the platelets achieved functionalized polymer chains and the functional groups on
prior to mixing. Phase transfer techniques [35] and lyophili- graphene oxide. In other words, the form of graphene used is
zation methods [36] (in which the GNSs were prepared using graphene oxide, but not fully exfoliated graphene. Graphene
an environmentally friendly freeze-drying process) have been oxide has many attractions over pristine graphene. It is readily
employed to facilitate solution mixing of graphene-based obtainable in large quantities, has built-in functional groups
composites. However, the solution-mixing techniques cannot available for covalent bonding to form a strong interface, is
be applied to prepare uniform composites, as graphene has a easily exfoliated in water such that nanocomposites can read-
strong interaction with the polymer, and as a result, graphene ily be prepared with the use of water-soluble polymers, and it
and polymer would aggregate. can be chemically modified to prepare nanocomposites with
Melt blending offers a simple way of dispersing nanopar- non-water-soluble polymers. On the other hand, the individual
ticles in a polymer matrix. It was used to disperse ther- nanoplatelet of graphene oxide is often wrinkled and has sig-
mally reduced graphene oxide in polycarbonate [37] and nificantly inferior mechanical properties to graphene.
poly(ethylene-2,6-napthalate) [38]. In melt mixing, polymer The PE–GNS nanocomposites, which were prepared
melt and graphene-based fillers are mixed under high shear through in situ polymerization by carrying out the polymer-
conditions. In this, no solvent is used, and as such, it is con- ization reaction of ethylene in the presence of treated GNSs,
sidered more economical compared with solution mixing. were found to be stiffer and thermally more stable [43]. The
However, melt mixing does not provide the same level of dis- nanocomposite with ~15 wt % GNSs presented an increase
persion of the filler as solution mixing or in situ polymerization of 30°C in the degradation temperature compared to that of
neat PE. In situ polymerization is an especially effective tech- this type of reaction a continuous process. Again, during this
nique to construct a nanocomposite material where the poly- reaction, a π-bond is broken and a covalent bond is formed
mer chain is grown directly on the graphene sheet via one with a change in the hybridization of the C atom at which the
of the chain-growth polymerization methods. Depending on reaction takes place. DFT studies showed that the reaction is
the type of monomer, different chain-growth methods have exothermic. In the formation of a polymer–graphene compos-
been used to synthesize different polymers. For example, for ite, the growth of every polymer chain involves a number of
monomers with vinyl group such as ethylene and styrene, the these reactions and as such the overall reaction can be highly
free radical addition polymerization method is being used exothermic.
(Scheme I of Figure 6.4); for vinyl monomers with strong Chain termination: A growing polymer chain radical can
electronegative groups, the anionic addition polymerization terminate in different ways. It can be a chemical reaction that
method is being used (Scheme II of Figure 6.4); for olefins ceases the formation of reactive intermediates in the propa-
with electron-donating substituents and heterocycles, the cat- gation step. For example, when the growing chain combines
ionic addition polymerization method is being used (Scheme with another radical (i.e., an oxy radical used in the function-
III of Figure 6.4). alization, or a functionalized graphene radical, or another
The following section presents a detailed study of the in propagating polymer chain radical), it can effectively bring
situ free radical addition polymerization mechanism associ- the polymer chain growth to a halt. On the other hand, the
ated with the formation of PE–GNS composite. This provides reactive radical can functionalize a new graphene sheet. This
an understanding of how a strong covalent bond is formed at would terminate the growing polymer chain, but at the same
the graphene–polymer interface, how the composite forma- time it can bring the new graphene sheet to life so that a new
tion helps to disperse graphene sheets, and how the degrada- polymer chain can grow on it. Mylvaganam et al. [44] have
tion of GNSs is minimized while forming strong bonds at the examined both the possibilities through DFT calculations.
interface. Their study concluded the following:
Free radical addition polymerization of ethylene in the
presence of GNSs: The mechanism of this polymerization is 1.
Termination through combining with another radi-
characterized by three steps, chain initiation, chain propaga- cal—The combination of a growing polyethylene
tion and chain termination. chain radical with a methoxy functionalized gra-
Chain initiation: Organic peroxides or other reagents phene radical showed that the product is stable and
which decompose to give oxy radicals by thermal or photo the reaction is exothermic.
chemical reactions are used as initiators. First the oxy radical 2.
Termination by functionalizing another graphene
reacts with graphene and functionalizes the GNS. Then the sheet—DFT studies of the reaction between a grow-
resulting radical reacts with an ethylene molecule and thereby ing polymer chain and a graphene sheet showed that
initiates the polymerization reaction. this reaction is also exothermic. During this reaction,
In the functionalization of GNS, the oxy radical uses one of although the growth of the polymer chain is termi-
the electrons from a π-bond of graphene and forms a covalent nated, a new graphene sheet is functionalized at the
bond leaving the other electron on the graphene sheet. This same time. This means it can lead to another polymer
process involves breaking of a π-bond, the change in hybrid- chain growth. Moreover, this reaction demonstrates the
ization of the C atom at which the covalent bond is formed polymer chain function as spacers, thereby preventing
(both consume energy), and the formation of a –C–O– bond the restacking of graphene sheets (Figure 6.5b).
(releases energy). Thus, the overall reaction can be either
endothermic or exothermic, depending on the amount of 6.4.3 Preservation of Graphene Properties
energy consumed and released. A quantum mechanical study
while Forming Chemical Bonds
showed that this reaction is exothermic [44]. The electron left
on the graphene sheet reacts with a monomer molecule eth- In such chemically bonded composites, the C atoms of gra-
ylene. As graphene sheet allows a double-sided reaction due phene sheet at which the reaction takes place are converted
to its 2D structure, in the initiation reaction with ethylene, to nonplanar sp3 carbon atoms. This could lead to the loss of
the covalent bond preferably forms on the opposite side. This superior properties of graphene. Too many chemical bonds
process involves the hybridization change of the reacting car- at the polymer–GNS interface would lead to more defects
bon on graphene, the breakage of the ethylene π-bond, and on graphene, thereby weakening its intrinsic properties. To
the formation of the C–C bond. DFT calculations showed the preserve the intact structure of graphene, therefore, it is nec-
reaction is endothermic, and the free electron formed during essary to control the chemical bond formation with polymer.
the reaction with ethylene is localized to the free end of the The detailed study on the mechanism presented in the pre-
growing polymer chain. vious section demonstrated that during the covalent bond
Chain propagation: This is a process in which a reactive formation, some energy is consumed to change the hybrid-
intermediate is continuously regenerated during the course of ization of the graphene C atom from sp2 to sp3. On a rippled
a chemical reaction. The new radical generated in the above GNS, carbon atoms at the crest of the ripples have geometries
chain initiation reaction reacts with another ethylene mole- closer to the sp3 hybridization. Hence, when forming compos-
cule, thus producing a new radical (Figure 6.5a) and making ites, the covalent bond formation with the graphene carbon
Scheme I
Δ
R-O-O-R 2R-O.
H H H H H H H H
R O. C C R O C C. R O C C C C.
H H H H H H H H
n
(Initiation) (Propagation)
H H H H H H H H H H H H H H H H
C. .C
R O C C C C C C O R R O C C C C C C C C O R
H H H H H H H H H H H H H H H H
n m n m
(Termination)
Scheme II
H O R3 H R3 H R3 H R3
C C Bu C C O: Bu C C O C C O:
Bu Li+
H R4 Li+ H R4 H R4 Li+
H R4 n
H
C2H5 C Bu
CH3
(Termination can occur either through quenching due to trace impurities
or through the addition of water or alcohol)
Scheme III
HA H+ A–
H H H H H H H H H
H+ H C C H C C C C C C+
H C C+
H R H R H R H R
H R
n
H H H H H H H H H H H H
H C C C C C C+ A– H C C C C C C A
H R H R H R H R H R H R
n n
(Termination by combination)
FIGURE 6.4 Free radical addition (Scheme I), anionic (Scheme II), and cationic (Scheme III) polymerization schemes.
(a) (b)
1.558 
1.607 
FIGURE 6.5 QM-optimized geometry of graphene–PE chain demonstrating (a) the propagation step and (b) the termination step by func-
tionalizing another graphene sheet. (Adapted from Mylvaganam K and Zhang L C. 2013. J. Phys. Chem. C 117, 2817–23.)
atoms at the crest of the ripples would require less energy. Research,” a University of Manchester startup company,
Mylvaganam et al. [44] have demonstrated this by generat- reported that flakes of up to 1 mm2 can be obtained by this
ing ripples on a GNS through the application of compressive method but at a high cost. The most essential technological
strain and examining the reactivity of a simple radical on challenge is the difficulty of achieving controlled production
carbon atoms having different curvature along a ripple. They of large sheets.
found that when the radius of curvature is small the radi-
cal binds tightly to the graphene C atom with a shorter bond • Dispersity of graphene and graphene-based sheets in
length and higher binding energy. This demonstrates that the a polymer solution.
crests of the ripples are more vulnerable to the attack, that
is, ripples on graphene sheets are active centers for growing The low solubility of graphene oxide in nonpolar organic
chemically bonded graphene–polymer composites. As noted solvents poses a great challenge to the synthesis of graphene–
in Section 6.3, it is possible to control the ripples on GNS polymer composites in organic solutions. To solve this prob-
[27]. This means that the covalent bond formation between lem, short-time metastable dispersions of graphene oxide are
GNS and polymer can also be controlled. Hence, the struc- produced either by ultrasonication or high-speed shearing.
ture of graphene can be preserved to some extent while Modifying the graphene oxide sheets with functional mol-
forming strong chemical bonds with polymer when forming ecules would also increase the solubility in various kinds
composites. of solvents. For example, Stankovich et al. [45] functional-
Having understood the mechanism of the formation of ized graphene oxide sheets with phenyl isocyanate, mixed it
chemically bonded polymer–GNS composite, and the ways to with polystyrene solution in DMF and then reduced graphene
preserve the properties of graphene in composites, it is impor- oxide. They found that the dispersed suspension maintained
tant to know the challenges involved in the production stage very well.
and assess the reinforcement and its resistance to fracture and
fatigue crack propagation to prevent failure. • Realizing wide applications.
6.5 KEY CHALLENGES IN THE Graphene is a promising filler to improve the mechani-

cal, electrical, and thermal properties of polymers. However,
FORMATION OF COMPOSITES
many technical barriers of graphene, such as its large-scale
• Reducing the cost for large-scale production of production, structure control, dispersion in a matrix, and
single-layer pristine graphene. interfacial interaction between graphene and matrix must be
taken into account. Although these barriers had been over-
The quantity and quality of graphene directly impacts the come to a certain extent when using graphene oxide as fillers,
performance of composites and their applications. Large- their mechanical, electrical, and thermal properties are infe-
scale production would be the first step toward commercial rior to graphene. Moreover, the morphology and the proper-
graphene-based electronic devices. The cost of graphene is ties of graphene oxide are not consistent, depending on how
linked to its method of production and its quality. Pristine, they are exfoliated. As such, the properties of the resulting
defect-free ultrahigh-quality graphene is produced individu- polymer composites vary, even though they are fabricated by
ally using the micromechanical cleavage method. “Graphene the same method and with the same polymer.
6.6 ASSESSMENT OF REINFORCEMENT containing cracks to resist fracture, and fatigue is the
progressive and localized damage that occurs under
The aim of incorporating graphene into the polymer matrix cyclic loading. Thus, the material’s resistance to frac-
is to improve its mechanical properties and thermal stability ture and fatigue crack propagation are of supreme
of the matrix. The stronger interfacial interaction between the importance in preventing failure. Hence, they can
polymer matrix and graphene should have a significant effect be used to assess the degree of reinforcement of a
on its mechanical properties. Thus, it is necessary to assess composite. An epoxy polymer reinforced with only
the properties of a composite once it is made. 0.125% weight of functionalized graphene sheets
increased the fracture toughness of the pristine
1.
Through stress–strain curves. The classical way epoxy by ~65% and the fracture energy by ~115%;
of assessing reinforcements is to plot the ratio of under fatigue conditions, it drastically reduced the
the Young’s modulus of the composites (calculated rate of crack propagation in the epoxy ~25-fold [49].
using stress–strain curves) to that of the matrix All these changes can be attributed to the enhanced
against the volume fraction of filler particles (i.e., specific surface area of functionalized graphene
graphene). For example, in the work of Zhao et al. sheets.
[46], Poly(vinyl alcohol) with only 0.3% by volume 3.
Through Raman G’ band. Graphene monolayers
of reduced graphene oxide showed a large effect on have characteristic Raman spectra in which the G’
the stress–strain curve, as shown in Figure 6.6a. A band can be fitted with a single peak (in graphene
further increase in the % by volume of graphene bilayers, this band is made up of 4 peaks) [50].
oxide led to the increase in Young’s modulus and During deformation, this band shifts to a lower or
tensile strength, but to the decrease in the elonga- higher wavenumber, depending on the deformation
tion at the break point (Figure 6.6b). For example, mode. This stress sensitivity of the graphene G’
the Young’s modulus increased by a factor of 10 for band is used to monitor stress transfer in a compos-
the composite with 1.8% by volume of graphene ite. Similar to the evaluation of the level of adhe-
oxide. Polypropylene reinforced with reduced gra- sion between a carbon fiber and a polymer matrix
phene oxide also showed significant improvement through mapping the strain along the fiber, the
for a loading of only 0.1% by weight of graphene strain across the graphene monolayer is mapped to
oxide [47]. The work by Khan et al. [48] on the rein- evaluate the stress transfer from the polymer to the
forcement of polyurethane by pristine graphene also graphene. For example, in a graphene-PMMA com-
showed a large increase in the slope of the stress– posite, the evaluation of Raman spectra measured
strain curve with graphene loading (i.e., the Young’s under strain revealed an interfacial shear stress of
modulus increase by a factor of 100 for the high- ~2.3 MPa [51]. Comparing this value with the CNT
est loadings). In addition, they also found that for composites with covalent bonding at the interface
a given volume % of graphene, the reinforcement (up to 500 MPa have been predicted [52]), and those
effect decreased as the size of the graphene flake with only van der Waals interactions (2.7 MPa for
was reduced. CNT–PE composite) showed that there were weak
2.
Through fracture toughness and fatigue behavior. interactions between the graphene monolayer and
Fracture toughness describes the ability of a material PMMA.
(a) 50 (b) 60
3.0 vol%
1.8 vol%
40 0.6 vol% 50
Elongation at break (%)
Tensile strength (MPa)
200
Tensile stress (MPa)
0.3 vol%
30 0.0 vol% 40
100
20 30
10 20
0
0 10
0 50 100 150 200 250 0 1 2 3
Strain (%) Graphene (vol%)
FIGURE 6.6 (a) Stress–strain plots and (b) mechanical properties of polyvinyl alcohol–graphene oxide composite. (Adapted from Zhao X
et al. 2010. Macromolecules 43, 2357–63.)
6.7 APPLICATIONS of electrodes and electrolytes; a pseudo-capac-

itor works predominantly through fast revers-
A range of applications have already been explored. Many ible redox reactions. Conducting polymers are
of them have focused on harnessing the improvements in the typical pseudo-capacitive electrode materials.
properties of composites containing graphene. Adding graphene to the polymers can improve
the mechanical and electrochemical cycling
1. Some graphene-based polymer composites exhibit stability of conducting polymer electrodes. For
much higher electrical conductivity due to the forma- example, the graphene oxide/polyaniline com-
tion of a conducting network by graphene sheets in the posite-based supercapacitor exhibited a high
polymer matrix. Electrical conductivity of composites specific capacitance of 531 F g−1 at a current
is useful in electromagnetic interference (EMI) shield- density of 200 mA g−1 compared to 216 F g−1 of a
ing and antistatic coatings, while potentially main- polyaniline-based supercapacitor [56].
taining the transparency (due to its low percolation c. Fuel cells. These generate energy by catalytic oxi-
threshold) of the polymer. For example, incorporating dation of fuels at low temperatures. Platinum is
functionalized graphene into an epoxy matrix showed the most efficient catalyst in fuel cells. However,
21 dB shielding efficiency at 8.2 GHz for 15 wt % high cost, CO poisoning, and agglomeration
loading, which indicates that they may be used as hinder its practical application. The composites
effective lightweight EMI-shielding materials [53]. of graphene materials and nitrogen-containing
2. The mechanical reinforcement of composites at low polymers have been tested to be effective cata-
loadings of graphene oxide-derived fillers, offers lysts for oxygen reduction reaction. For example,
uses such as capacitive pressure sensors in micro- poly(diallyldimethylammonium chloride) was
electro-mechanical system (MEMS) applications. used as an electron acceptor for functionalizing
For example, the incorporation of only 8 vol % func- graphene to impart electrocatalytic activity for
tionalized graphene oxide layers into polyelectro- oxygen reduction reactions in fuel cells [57].
lyte multilayers in a layer-by-layer assembly showed d. Biocompatibility of graphene-based composites
an enhancement of elastic modulus by an order of has created a new developing research direction
magnitude (from 1.5 to 20 GPa) [54]. In addition, it in biomedical applications. These are exempli-
was reported that the multilayered nanocomposites fied below:
can be released as robust, free-standing membranes i. Bai et al. [58] demonstrated that physical
which can be suspended over large apertures and cross-linking of poly(vinyl alcohol) chains
sustain large mechanical deformations. between graphene oxide platelets formed a
3. Graphene-based nanocomposites have found appli- hydrogel, which showed that the release of
cations in various energy-related systems. The fol- vitamin B12 is controllable depending on
lowing are some examples: solution pH, and that it could be useful in
a. Lithium ion batteries (LIB). These are attractive drug delivery.
energy storage devices due to their high energy ii. Freestanding biocompatible films of
and power densities, long cycling life and good poly(oxyethylene sorbitan laurate) with
environmental compatibility. The traditional 38 wt % reduced graphene oxide could find
cathode materials of LIBs suffer from limited use in transplant devices and implants.
capacities and nonrenewable mineral resources.
Alternative polymeric cathode materials are
6.8 SUMMARY AND FUTURE SCOPE
light, environmentally friendly, mechanically
flexible and compatible, but suffer from poor In terms of elastic stiffness and fracture strength, graphene
electrical conductivity and slow redox reactions. is the strongest material obtained so far. However, a simply
The introduction of graphene as a conducting supported graphene sheet wrinkles under a central point load,
additive has improved the charge/discharge rate and such wrinkling is initiated by the circumferential com-
capability. For example, polypyrrole/reduced pressive force. The key material properties that determine the
graphene oxide composite electrode showed resistance of graphene to structural instability are the bending
74% capacity retention as the discharging rate stiffness and in-plane stiffness. Graphene and graphene oxide
increased from 1°C to 20°C [55]. have added a new dimension to the production of lightweight,
b. Supercapacitors. These are important energy low-cost, and high-performance composite materials. Studies
storage devices that can be charged/discharged had shown that good reinforcement is only found at relatively
at rapid rates. They can be classified by their low levels of graphene loading. However, there are challenges
storage mechanisms into two types: electrical in getting good dispersions of graphene sheets and exfoliation
double-layer capacitors (EDLC) and pseudo- of graphene into a single/few layered material with reasonable
capacitors. EDLCs store energy by the accumu- lattice dimensions and lower cost, and enhancing the interfa-
lation of electrostatic charges at the interfaces cial interaction between GNSs and a polymer matrix.
The outstanding properties of graphene make it attractive 14. Kim K S, Zhao Y, Jang H, Lee S Y, Kim J M, Ahn J H, Kim P,
for applications in flexible electronics. It shows great prom- Choi J Y and Hong B H. 2009. Large-scale pattern growth of
ise as a flexible touch-screen display for mobile devices. graphene films for stretchable transparent electrodes. Nature
457, 706–10.
Given the ability to control many of its characteristics by
15. Park H J, Meyer J, Roth S and Skakalova V. 2010. Growth and
an external electric field (gate voltage), it has a potential properties of few-layer graphene prepared by chemical vapor
to become an excellent material for spintronics. By using deposition. Carbon 48, 1088–94.
nanoscale electronic channels and tailoring its geometri- 16. Park S and Ruoff R S. 2009. Chemical methods for the pro-
cal symmetry, nanodiodes and transistors that are based on duction of graphenes. Nat. Nanotechnol. 4, 217–24.
single-layered device architectures have been demonstrated 17. Stankovich S, Dikin D A, Piner R D, Kohlhaas K A,
to exhibit extremely high speeds. These devices are par- Kleinhammes A, Jia Y, Wu Y, Nguyen S T and Ruoff R S.
2007. Synthesis of graphene-based nanosheets via chemical
ticularly ideal for use in high-speed electronics applications
reduction of exfoliated graphite oxide. Carbon 45, 1558–65.
in the next generation of computational and communica- 18. Lee S H, Dreyer D R, An J H, Velamakanni A, Piner R D,
tions devices, far-infrared THz detection and emission Park S, Zhu Y W, Kim S O, Bielawski C W and Ruoff R S.
equipment, ultrahigh sensitive chemical sensors, etc. Great 2010. Polymer brushes via controlled, surface-initiated atom
efforts are underway around the world to develop batteries, transfer radical polymerization (ATRP) from graphene oxide.
fuel cells, photovoltaic cells, and supercapacitors based on Macromol. Rapid Commun. 31, 281–8.
graphene. 19. Ren L L, Wang X Y, Guo S Z and Liu T X. 2011.
Functionalization of thermally reduced graphene by in situ
atom transfer radical polymerization. J. Nanopart. Res. 13,
REFERENCES 6389–96.
20. Kim H, Abdala A A and Macosko C W. 2010. Graphene/poly-
1. Kroto H W, Heath J R, Obrien S C, Curl R F and Smalley R E. mer nanocomposites. Macromolecules 43, 6515–30.
1985. C-60—Buckminsterfullerene. Nature 318, 162–3. 21. Hemmasizadeh A, Mahzoon M, Hadi E and Khandan R. 2008.
2. Iijima S. 1991. Helical microtubules of graphitic carbon. A method for developing the equivalent continuum model of a
Nature 354, 56–8. single layer graphene sheet. Thin Solid Films 516, 7636–40.
3. Novoselov K S, Geim A K, Morozov S V, Jiang D, Zhang Y, 22. Duan W H and Wang C M. 2009. Nonlinear bending and
Dubonos S V, Grigorieva I V and Firsov A A. 2004. Electric stretching of a circular graphene sheet under a central point
field effect in atomically thin carbon films. Science 306, load. Nanotechnology 20, 075702–7.
666–9. 23. Huang Y, Wu J and Hwang K C. 2006. Thickness of graphene
4. Mylvaganam K and Zhang L C. 2004. Important issues in a and single-wall carbon nanotubes. Phys. Rev. B 74, 245413–9.
molecular dynamics simulation for characterising the mechan- 24. Tu Z C and Ou-Yang Z. 2002. Single-walled and multiwalled
ical properties of carbon nanotubes. Carbon 42, 2025–32. carbon nanotubes viewed as elastic tubes with the effective
5. Vodenitcharova T, Mylvaganam K and Zhang L C. 2007. Young’s moduli dependent on layer number. Phys. Rev. B 65,
Mechanical interaction between single-walled carbon nano- 233407–14.
tubes during the formation of a bundle. J. Mater. Sci. 42, 25. Wang C Y and Zhang L C. 2008. An elastic shell model for char-
4935–41. acterizing single-walled carbon nanotubes. Nanotechnology
6. Vodenitcharova T and Zhang L C. 2003. Effective wall thick- 19, 195704–6.
ness of a single-walled carbon nanotube. Phys. Rev. B 68, 26. Odegard G M, Gates T S, Nicholson L M and Wise K E. 2002.
165401–4. Equivalent-continuum modeling of nano-structured materi-
7. Wang C Y, Mylvaganam K and Zhang L C. 2009. Wrinkling als. Compos. Sci. Technol. 62, 1869–80.
of monolayer graphene: A study by molecular dynamics and 27. Bao W Z, Miao F, Chen Z, Zhang H, Jang W Y, Dames C and
continuum plate theory. Phys. Rev. B 80, 155445–5. Lau C N. 2009. Controlled ripple texturing of suspended gra-
8. Wang C Y and Zhang L C. 2008. A critical assessment of the phene and ultrathin graphite membranes. Nat. Nanotechnol.
elastic properties and effective wall thickness of single-walled 4, 562–6.
carbon nanotubes. Nanotechnology 19, 075705–5. 28. Xiao K Q and Zhang L C. 2005. Effective separation and
9. Lee C, Wei X D, Kysar J W and Hone J. 2008. Measurement alignment of long entangled carbon nanotubes in epoxy. J.
of the elastic properties and intrinsic strength of monolayer Mater. Sci. 40, 6513–16.
graphene. Science 321, 385–8. 29. Mylvaganam K and Zhang L C. 2004. Chemical bonding in
10. Novoselov K S, Jiang D, Schedin F, Booth T J, Khotkevich polyethylene-nanotube composites: A quantum mechanics
V V, Morozov S V and Geim A K. 2005. Two-dimensional prediction. J. Phys. Chem. B 108, 5217–20.
atomic crystals. Proc. Natl. Acad. Sci. U. S. A. 102, 10451–3. 30. Wang L W and Chen G H. 2010. Dramatic improvement in
11. Emtsev K V, Bostwick A, Horn K, Jobst J, Kellogg G L, Ley mechanical properties of GNs-reinforced HDPE nanocom-
L, McChesney J L et al. 2009. Towards wafer-size graphene posites. J. Appl. Polym. Sci. 116, 2029–34.
layers by atmospheric pressure graphitization of silicon car- 31. Stankovich S, Dikin D A, Dommett G H B, Kohlhaas K M,
bide. Nat. Mater. 8, 203–7. Zimney E J, Stach E A, Piner R D, Nguyen S T and Ruoff R S.
12. Huang H, Chen W, Chen S and Wee A T S. 2008. Bottom-up 2006. Graphene-based composite materials. Nature 442, 282–6.
growth of epitaxial graphene on 6H-SiC(0001). ACS Nano 2, 32. Chen D, Zhu H and Liu T X. 2010. In situ thermal preparation
2513–18. of polyimide nanocomposite films containing functionalized
13. Robinson J, Weng X J, Trumbull K, Cavalero R, Wetherington graphene sheets. ACS Appl. Mater. Interfaces 2, 3702–8.
M, Frantz E, LaBella M, Hughes Z, Fanton M and Snyder D. 33. Das B, Prasad K E, Ramamurty U and Rao C N R. 2009.
2010. Nucleation of epitaxial graphene on SiC(0001). ACS Nano-indentation studies on polymer matrix composites rein-
Nano 4, 153–8. forced by few-layer graphene. Nanotechnology 20, 125705–5.
34. Ramanathan T, Abdala A A, Stankovich S, Dikin D A, 46. Zhao X, Zhang Q H, Chen D J and Lu P 2010 Enhanced
Herrera-Alonso M, Piner R D, Adamson D H et al. 2008. mechanical properties of graphene-based poly(vinyl alcohol)
Functionalized graphene sheets for polymer nanocomposites. composites. Macromolecules 43, 2357–63.
Nat. Nanotechnol. 3, 327–31. 47. Song P G, Cao Z H, Cai Y Z, Zhao L P, Fang Z P and Fu S Y.
35. Choi E Y, Han T H, Hong J H, Kim J E, Lee S H, Kim H W 2011. Fabrication of exfoliated graphene-based polypropylene
and Kim S O. 2010. Noncovalent functionalization of graphene nanocomposites with enhanced mechanical and thermal prop-
with end-functional polymers. J. Mater. Chem. 20, 1907–12. erties. Polymer 52, 4001–10.
36. Cao Y W, Feng J C and Wu P Y. 2010. Preparation of organi- 48. Khan U, May P, O’Neill A and Coleman J N. 2010. Development
cally dispersible graphene nanosheet powders through a of stiff, strong, yet tough composites by the addition of solvent
lyophilization method and their poly(lactic acid) composites. exfoliated graphene to polyurethane. Carbon 48, 4035–41.
Carbon 48, 3834–9. 49. Rafiee M A, Rafiee J, Srivastava I, Wang Z, Song H H, Yu
37. Kim H and Macosko C W. 2009. Processing-property rela- Z Z and Koratkar N. 2010. Fracture and fatigue in graphene
tionships of polycarbonate/graphene composites. Polymer 50, nanocomposites. Small 6, 179–83.
3797–809. 50. Ferrari A C, Meyer J C, Scardaci V, Casiraghi C, Lazzeri M,
38. Kim H and Macosko C W. 2008. Morphology and prop- Mauri F, Piscanec S et al. 2006. Raman spectrum of graphene
erties of polyester/exfoliated graphite nanocomposites. and graphene layers. Phys. Rev. Lett. 97, 187401–4.
Macromolecules 41, 3317–27. 51. Gong L, Kinloch I A, Young R J, Riaz I, Jalil R and Novoselov
39. Fim F D, Guterres J M, Basso N R S and Galland G B 2010 K S. 2010. Interfacial stress transfer in a graphene monolayer
Polyethylene/graphite nanocomposites obtained by in situ nanocomposite. Adv. Mater. 22, 2694–7.
polymerization. J. Polym. Sci. Part A: Polym. Chem. 48, 692–8. 52. Wagner H D, Lourie O, Feldman Y and Tenne R. 1998. Stress-
40. Jang J Y, Kim M S, Jeong H M and Shin C M. 2009. Graphite induced fragmentation of multiwall carbon nanotubes in a
oxide/poly(methyl methacrylate) nanocomposites prepared polymer matrix. Appl. Phys. Lett. 72, 188–90.
by a novel method utilizing macroazoinitiator. Compos. Sci. 53. Liang J, Wang Y, Huang Y, Ma Y, Liu Z, Cai F, Zhang C, Gao
Technol. 69, 186–91. H and Chen Y. 2009. Electromagnetic interference shielding
41. Potts J R, Lee S H, Alam T M, An J, Stoller M D, Piner R D of graphene/epoxy composites. Carbon 47, 922–5.
and Ruoff R S. 2011. Thermomechanical properties of chemi- 54. Kulkarni D D, Choi I, Singamaneni S and Tsukruk V V. 2010.
cally modified graphene/poly(methyl methacrylate) compos- Graphene oxide-polyelectrolyte nanomembranes. ACS Nano
ites made by in situ polymerization. Carbon 49, 2615–23. 4, 4667–76.
42. He H K and Gao C. 2010. General approach to individually 55. Yang Y, Wang C, Yue B, Gambhir S, Too C O and Wallace G
dispersed, highly soluble, and conductive graphene nanosheets G. 2012. Electrochemically synthesized polypyrrole/graphene
functionalized by nitrene chemistry. Chem. Mater. 22, 5054–64. composite film for lithium batteries. Adv. Energ. Mater. 2,
43. Fim F D, Basso N R S, Graebin A P, Azambuja D S and 266–72.
Galland G B. 2013. Thermal, electrical, and mechanical prop- 56. Wang H, Hao Q, Yang X, Lu L and Wang X. 2009. Graphene
erties of polyethylenegraphene nanocomposites obtained by in oxide doped polyaniline for supercapacitors. Electrochem.
situ polymerization. J. Appl. Polym. Sci. 128, 2630–7. Commun. 11, 1158–61.
44. Mylvaganam K and Zhang L C. 2013. In situ polymerization 57. Wang S, Yu D, Dai L, Chang D W and Baek J-B. 2011.
on graphene surfaces. J. Phys. Chem. C 117, 2817–23. Polyelectrolyte-functionalized graphene as metal-free elec-
45. Stankovich S, Piner R D, Nguyen S T and Ruoff R S. 2006. trocatalysts for oxygen reduction. ACS Nano 5, 6202–9.
Synthesis and exfoliation of isocyanate-treated graphene 58. Bai H, Li C, Wang X and Shi G. 2010. A pH-sensitive graphene
oxide nanoplatelets. Carbon 44, 3342–7. oxide composite hydrogel. Chem. Commun. 46, 2376–8.
Section II
Chemical Fabrication/Properties
7 Grain Boundaries in Chemical Vapor
Deposition-Grown Graphene
László P. Biró and Philippe Lambin
CONTENTS
Abstract...................................................................................................................................................................................... 107
7.1 Introduction...................................................................................................................................................................... 107
7.2 Brief Overview of Graphene CVD................................................................................................................................... 108
7.2.1 Nucleation and Growth......................................................................................................................................... 109
7.2.2 Growth Temperature Effects................................................................................................................................ 109
7.3 GB Model Structures.........................................................................................................................................................110
7.3.1 Regular GBs...........................................................................................................................................................110
7.3.2 Irregular GBs.........................................................................................................................................................111
7.4 Experimental Methods for Revealing the Structure of GBs in CVD Graphene...............................................................112
7.4.1 Electron Microscopy..............................................................................................................................................112
7.4.2 Low-Energy Electron Microscopy........................................................................................................................113
7.4.3 Scanning Probe Microscopy..................................................................................................................................114
7.4.3.1 AFM Investigation of GBs in CVD Graphene.......................................................................................114
7.4.3.2 STM Investigation of GBs in CVD Graphene........................................................................................114
7.4.4 Optical Microscopy...............................................................................................................................................115
7.5 Consequences of the Presence of GBs...............................................................................................................................117
7.5.1 Electrical Transport...............................................................................................................................................117
7.5.2 Thermal Transport.................................................................................................................................................119
7.5.3 Mechanical Properties.......................................................................................................................................... 120
7.5.4 Chemical Properties............................................................................................................................................. 120
7.6 Summary and Outlook..................................................................................................................................................... 121
Acknowledgments...................................................................................................................................................................... 121
References.................................................................................................................................................................................. 121
ABSTRACT experimentally produced GBs of several microns in length

could be completely identical in all of their atomic-scale
Chemical vapor deposition (CVD) is the most perspective details, but despite this, certain generalized conclusions may
method for producing large-area graphene. The produced be formulated, which may be helpful for experimentalists in
graphene sheet is a patchwork of crystallites with random ori- the interpretation of results and in planning new experiments,
entation separated by grain boundary (GB) regions. The GBs which could lead to a more systematic picture of GBs in CVD
will have a decisive role in defining the electrical transport graphene and to a better understanding of ways to improve its
properties of graphene films. Their structure may vary from electrical transport properties.
fully periodic to completely disordered depending primarily
on the synthesis temperature and substrate used for CVD. The
7.1 INTRODUCTION
chapter mainly focuses on the experimental findings on gra-
phene grown by CVD under a very wide range of experimen- The production of graphene by chemical vapor deposition
tal conditions (temperature, pressure hydrogen/hydrocarbon (CVD) on Cu foils seems to be perhaps closest to fulfill the
ratio, gas flow velocity, and substrates). The most frequently demands of practical applications [1]. However, the grown
used methods for atomic-scale characterization of GB struc- graphene films are polycrystalline, with numerous grain
tures, their possibilities and limitations, and the alterations of boundaries (GBs) and their transport properties are one to two
the GBs in CVD graphene during investigation are discussed. orders of magnitude below those of graphene films produced
The effects of GB disorder on electrical and thermal trans- by micromechanical cleavage from highly oriented pyrolytic
port are reviewed and the relatively scarce data available on graphite (HOPG) [2]. Additionally the GBs, if occurring in
the chemical properties of the GBs are summarized. GBs are nanopatterned graphene [3], may have dramatic effects on
complex-enough nanoobjects that it may be unlikely that two the performances of various graphene-based nanodevices [4].
107
100,000
Number of peer review papers in • Most cold-rolled, polycrystalline Cu foils after the
annealing step previous to the CVD can be regarded
10,000 as having a dominant (100) orientation.
• When used for growth, a direct carbon source (by
1000
evaporating elemental carbon in ultrahigh vacuum
WOS
[UHV]) [9], at each nucleation site, four graphene

100
grains nucleated and grew approximately in the four
10 equivalent 〈001〉 direction of the (100) substrate,
similar to what is reported in low-pressure CVD
1 (LPCVD) [10]. Higher growth velocities of graphene
1985 1990 1995 2000 2005 2010 2015 in the 〈001〉 direction of the Cu substrate resulted in
Years four-lobed graphene islands consisting of the four
graphene grains—having different crystallographic
FIGURE 7.1 Statistics of papers on “graphene” (squares) and “gra- orientation with respect to the substrate—joined at
phene and GBs” (circles). The black lines help guiding the eyes. Data the nucleation site.
collected from the Web of Science database (May 17, 2013), num- • At the growth temperature used for CVD growth,
bers for the year 2013, as an incomplete year were not included.
steps are formed on the Cu surface due to Cu subli-
mation; step migration is quenched under the grow-
Recently, we have reviewed the experimental data on the ing graphene layer, leading to a strongly stepped
structure in CVD-grown graphene [5] and some of the effects morphology, which will lead to nanowrinkle forma-
arising from GBs on electrical and thermal transport, and the tion even after the graphene is transferred to atomi-
scarce data on the chemical properties of the GBs. Despite the cally smooth surfaces, like freshly cleaved mica, and
fact that the attention of the graphene community focused on will impact on the transport properties of the gra-
questions related with GBs with a certain delay (mainly after phene layer [11].
2009), these days, the number of papers on GBs in graphene
is increasing at a very fast pace, Figure 7.1. On the other hand, when viewing the CVD of graphene
In this chapter, we propose to give an equilibrated presen- from the point of view of surface chemistry, it was clearly
tation of the status of the field related with graphene GBs, shown that the morphology of the growing graphene islands
briefly going through the most relevant findings of earlier depends on the pressure and composition of the gas mix-
papers, and focusing more on the new findings of the papers ture used for growth, as well [12–15]. The work reported in
published since our earlier review [5]. Reference 15 demonstrated that the partial pressure of hydro-
gen can alter the morphology of the graphene island from
dendritic (low partial pressure), through irregular (medium
7.2 BRIEF OVERVIEW OF GRAPHENE CVD
range), to perfectly hexagonal (high partial pressure) shapes.
From the point of large-scale applications, CVD on Cu [6] The CVD growth itself is a very complex process and
is the most advantageous alternative for growing large-area several methods have been developed (for recent reviews
graphene [1]. Numerous papers have been published in the see References 16 and 17), which established rather differ-
last few years on the CVD growth of graphene on transition ent values for the “optimal” parameters depending on the use
metal surfaces and several chapters within this book will of different growth pressures, flow velocities, the kind of the
focus on this issue. Therefore, it is beyond our scope to give hydrocarbon used as carbon feedstock, hydrogen/hydrocar-
an extensive review of the graphene CVD. However, a few bon mixing ratios, growth tube geometries, etc. (see Table
recent results have to be pointed out as they will be useful 7.1). This multidimensional parameter space makes it very
in the following discussions. The CVD of graphene on Cu is difficult to reach some simple and generally valid conclusions
a complex process at the crossroads of surface physics and [5]. The various CVD methods, despite the many variations in
surface chemistry. A recent review, written from the point the values of the used parameters can be roughly categorized
of view of surface physics, gives a good historical overview into three groups depending on the
on the growth of graphene on various transition metal single
crystals [7]. A few points are clearly pointed out regarding • Pressure used [12]
growth on Cu: • LPCVD (0.1–1 Torr)
• Ambient pressure CVD (APCVD)
• The solubility of carbon in Cu is very low. • Solid [6] or liquid substrate used [18,19]
• Owing to the weak interaction between graphene
and Cu, even when using a single-crystalline Cu The use of molten Cu as substrate may provide several
substrate, the graphene crystallites have random ori- advantages, such as the complete elimination of the previous
entation on the substrate. However, a weak epitaxy surface history, higher growth temperature, and the possibil-
relationship between graphene and Cu has recently ity to use surface tension for “ordering” the graphene islands
been demonstrated [8]. floating on molten Cu (see e.g., Figure 2d in Reference 19).
Grain Boundaries in Chemical Vapor Deposition-Grown Graphene 109
able to control the reciprocal orientation of the neighbor-

TABLE 7.1 ing graphene grains nucleated on the PMMA nanopatterns.
Set of Parameters Playing a Role in the CVD Production The advantage of this approach may be that the supersatura-
of Graphene on Polycrystalline Cu Foil and Having an tion needed for nucleation is provided only locally, by the
Influence on the Grains and Their Boundaries decomposition of the prepatterned PMMA disks; however, it
can only reduce, and cannot completely eliminate parasitic
Component Sensitive Parameters
nucleation. It was reported that long pregrowth annealing
Cu foil and its Thickness, surface pretreatment, annealing times (3 hat 1045°C) may facilitate suppression of parasitic
treatment temperature and environment prior to CVD,
nucleation in APCVD [25].
surface texture
According to a recent report for LPCVD, the nucleation
Gas mixture Composition, hydrocarbon(s) used, flow velocity,
temperature
probability is strongly growth temperature dependent [21].
Gas pressure <1 Torr (LPCVD), >10 Torr (APCVD)
Nucleation density is the result of competition between the
Oven temperature Heating rate, CVD temperature, cooling rate
rates of nucleus growth by adatom capture, surface diffusion
Reactor geometry Gas flow (laminar, turbulent), temperature gradient, of carbon species, and desorption of carbon adatoms. Two
sample holder nucleation density regimes can be defined: one controlled by
carbon adatom species capture at low temperatures (<870°C)
and the other, controlled by desorption at high temperatures
In summary, the shape of the growing graphene grains will (>870°C). In all cases, an important role is played by the sub-
be determined by a number of weak and competing i nteractions strate roughness [8].
between the Cu substrate, the graphene layer, and the sur- The growth of the already nucleated graphene platelets
rounding atmosphere. This may be the reason for the impres- under special conditions of LPCVD (the so-called “envelope”
sive variety of growing grain shapes reported in the l iterature. growth [14] yielding dendritic grains of up to 0.2 mm in size)
The shape of the growing grains is important for the formation proceeds by the addition of carbon resulting from the newly
of the GBs as it will determine the geometry of the interfaces decomposed hydrocarbon molecules on the Cu surface to
between the neighboring grains. As shown recently, special the edges of the platelet. Convincingly, this has been dem-
pretreatments of the Cu foil may greatly reduce the number of onstrated by using 13C labeled hydrocarbons alternating with
graphene nucleation centers, thereby increasing the grain size 12C containing hydrocarbon [14]. On the other hand, several
and making it possible to control their shapes [20]. reports support the Smoluchowski-type ripening process
by the formation of small graphitic islands diffusing on Cu
and sticking in an irreversible manner to the larger graphene
7.2.1 Nucleation and Growth
islands [26,27]. This process is based on the very strong in-
In any crystallization process, some supersaturation is needed plane C–C bond in graphene and indicates a high supersatura-
for graphene nucleation to occur [21]. If nucleation occurs tion level, leading to continuous nucleation.
only once at the beginning of the growth process, or con- In APCVD, the growth of graphene is not any more sur-
tinuously during the entire growth, it has a decisive role in face limited, the growth of several stacked layers may occur
the CVD growth of graphene. The number of nuclei can be and rotational misorientation of the overlapping layers may be
directly related with the probability to cross a GB between present [12,13].
two points situated at a certain distance from each other in
CVD graphene. This is very clearly illustrated in Figure 7.1 of
7.2.2 Growth Temperature Effects
Reference 10 for LPCVD. As Raman mapping has shown, the
interdomain regions where two graphene grains originating The three most relevant methods of producing graphene sam-
from different nuclei grow together have a higher density of ples are [28]
defects [22,23]. Therefore, the overall electrical properties of
a film will be strongly related with the number of GBs per unit • Micromechanical cleavage from HOPG
length, ultimately with the nucleation process. The number of • Epitaxial growth on SiC
nuclei may be reduced by using high growth temperature, low • CVD on transition metals
pressure, and low methane flux; however, under certain criti-
cal values for pressure and flux, it is not possible to achieve each having its characteristic temperature ranges. HOPG is
full coverage of the substrate by graphene [10] at any growth graphitized in the temperature range of 3000°C at very high
times used. pressure [29]; epitaxial growth of good-quality graphene on
The nucleation may be kept under control to a certain SiC is usually carried out in the temperature range of 1500°C
extent if seeded growth is used in atmospheric pressure [30,31], while the various CVD processes operate in the range
chemical vapor deposition (APCVD) [23]. However, for of 1000°C. The atomic movements during the formation of
practically exploiting this procedure one has to be able to GBs under these different conditions may be very different
provide a large single-crystalline graphene or a few-layer and lead to GBs of different structures [5]. During the graphi-
graphite, or, when poly(methyl methacrylate) (PMMA) tization of HOPG from a strongly disordered pyrolytic precur-
nanopatterns are used as nucleation sites [24], one has to be sor, a mosaic crystal with high perfection is obtained, with
mosaic grains of several tens of microns in size exhibiting

an angular spread of the normal to the graphitic planes in the
range of degree. It is very unlikely that during this process,
a strong disorder may be preserved in the GB regions, and
more likely it may also attain a minimum energy configura-
tion and present a regular and periodic structure [32]. So gra-
phene flakes obtained by micromechanical cleavage will also
exhibit more frequently this kind of GBs.
The graphene grown epitaxially on SiC, opposite to the Cu
substrate may interact strongly enough with the SiC that the
first layer is usually regarded as a buffer layer, its properties
being different from freestanding graphene [33,34]. Owing to
this interaction with the single-crystalline substrate and to the (2, 1)
typical growth temperatures used (1500°C), the GBs in epitax-
ial graphene on SiC will be less disordered as compared with
graphene grown by CVD in the temperature range of 1000°C.
Most frequently, the CVD on Cu is carried out using cold-
rolled Cu foils of typical thickness 25 μm. Usually, the surface
pretreatment is restricted to an annealing step at the growth
FIGURE 7.2 Atomic model of a symmetric GB aligned with the
temperature (1000°C) for tens of minutes under a reducing
(2,1) direction (long arrow). The structure was optimized with the
atmosphere, to remove oxidized Cu. This treatment will also Brenner–Tersoff potential constrained to two dimensions. (Adapted
recrystallize the cold-rolled Cu, giving it a predominant (100) from Simonis, P. et al. 2002. Surface Science 511 (1–3): 319–22.) In
texture; however, the surface morphology arising from the the dislocation model, each fused pentagon–heptagon pair lies at
cold rolling will be largely preserved, and the sublimation the core of an edge dislocation obtained by inserting an extra zigzag
of the Cu surface will lead to significant step formation and chain of atoms, here visualized by dashed lines.
migration [9]. Under these conditions, the nucleation is usu-
ally random and will result in many disordered GBs. 1. The edges of the grains to be connected have well-
defined crystallographic directions (n1,m1) and
(n2,m2) that are made parallel to each other after a
7.3 GB MODEL STRUCTURES suitable rotation. The resulting GB will be defined as
It is known since the works of Bragg and Burgers that a GB an ordered or a regular one.
in a three-dimensional crystal can be viewed as an array of 2. The edges of the grains are irregular, which leaves
dislocations [35]. This point of view also works for a two- gaps in between when the grains come close to each
dimensional (2D) crystal like graphene, as illustrated in other. The gaps must be filled in with atoms until
Figure 7.2. The GB displayed there (horizontal line at the they are closed. The resulting GB will be defined as
middle of the drawing) is a periodic arrangement of fused a disordered or an irregular one.
pentagon–heptagon pairs interconnected by one hexagonal
edge [36]. A fused pentagon–heptagon pair in graphene is
7.3.1 Regular GBs
an edge dislocation having the smallest possible Burgers vec-
tor b [37], namely one lattice parameter (a = 0.245 nm). In Figure 7.2 is an example of a regular GB that has actually been
Figure 7.2, the array of dislocations generated by the pen- observed at the surface of HOPG by STM [36]. It is a symmet-
tagon–heptagon pairs along the GB is visualized in a some- ric GB in the sense that the two grains match up after folding
what arbitrary way by the sawtooth dashed lines. Each of the figure in half along the boundary line. The crystallographic
these lines can be viewed as the result of inserting a semi- direction of the two edges obtained by cutting the figure along
infinite zigzag chain of atoms ending on a heptagon that, the boundary line can be defined by the same lattice vector
indeed, contains an extra atom compared with the hexagons. (n,m) in both GBs, the vector (2,1) in the case of Figure 7.2.
In this model, the misfit orientation of the grains is related The GB is periodic, with periodicity d = a√(n2 + m2 + nm),
to the arrangement of dislocations through Frank’s formula where a is the lattice parameter of graphene. Symmetric GBs
θ = 2 sin−1(b/2d) [38], with b the magnitude of the Burgers have been discussed widely in the literature [39–45].
vector and d the distance between two dislocation cores. A regular GB can be periodic, if the edge lattice vec-
For the structure displayed in Figure 7.2 (d = 0.651 nm, see tors (n1,m1) and (n2,m2) of the two grains to be connected
below), this expression gives θ = 21.8°. are commensurate. Commensurability is obviously realized
The dislocation model is useful to understand, for example, by choosing two lattice vectors symmetric with respect to
GB energies [39,40]. A more simple way, conceptually speak- an armchair or zigzag direction. This choice brings back to
ing, to generate a GB is to approach two pieces of graphene symmetric GBs. It is also possible to find the commensurate
and to interconnect them as best as possible. One may then lattice vectors that are not equivalent by symmetry. Couple
envision two extreme situations: (7,0) and (5,3) is an example of edge orientations leading to an
(7,0) one row of hexagons in Figure 7.2 but not so well in Figure
(a) 7.3a. The latter structure indeed alternates two adjacent 5–7
pairs with an isolated 5–7 pair.
Many regular GBs are not commensurate. An interesting
example is obtained by connecting a graphene grain with an
armchair edge to a grain with a zigzag edge, the ratio of the
periodicities of these two edges being the irrational number
√3. A zigzag–armchair interface, such as shown in Figure
7.3b, is a 30°GB. Optimized structural models connecting
armchair and zigzag nanoribbons are presented in Reference
46 along with an investigation of their electronic properties.
Models of that sort are interesting planar analogs of intra-
molecular junctions between armchair and zigzag nanotubes
(5,3)
[47]. The GB shown in Figure 7.3b is exclusively composed of
(b) interconnected 5–7 pairs and therefore presents a high density
of topological defects.
7.3.2 Irregular GBs
Whereas regular GBs are observed from time to time at the
surface of HOPG [32,36], their presence in graphene is not so
frequent. Nevertheless, graphene grown by CVD on single-
crystal Ni(111) can present line defect structures that have
been interpreted as regular GBs between two grains that have
the same crystallographic orientation but are laterally trans-
lated by half a lattice parameter along the zigzag direction
FIGURE 7.3 (a) GB obtained by connecting a graphene plane ter- [48]. The GB is realized by a succession of octagons alternat-
minated by a zigzag (7,0) edge to another half plane whose edge ing with a pair of pentagons. The good epitaxial relationship
is parallel to the (5,3) direction. The lattice vectors (7,0) and (5,3) between graphene and the Ni(111) surface allows the growth
having the same length (arrows), the GB is periodic (d = 1.72 nm). of orientational correlated grains [16].
The structure has been optimized in three dimension using Brenner–
GBs are constitutional elements of graphene grown by CVD
Tersoff potential. (b) Incommensurate GB obtained by connecting
semi-infinite zigzag and armchair graphene nanoribbons. (Adapted
on a copper foil [5,16]. When observed by STM, the GBs are
from Li, X. F. et al. 2011. The Journal of Physical Chemistry C imaged as irregular and wide protrusions (see Section 7.4.2)
115 (25): 12616–24, structural optimization based on hybrid density [49]. Rather than being a consequence of local variation of the
functional.) topography, the protrusion revealed by STM at the location
of a GB is of electronic origin [50]. The difficulty in generat-
asymmetric GB illustrated in Figure 7.3a. The misfit orienta- ing irregular GBs on the computer explains why there are not
tion is 21.8°, exactly the same as for the symmetric GB shown many structural models proposed in the literature so far [51–
in Figure 7.2. In addition to sharing this geometrical param- 53]. Figure 7.4 shows an example of such a structural model,
eter, the structures of the two GBs bear some resemblance. here corresponding to the same misfit angle as in Figure 7.2,
Both have pentagon–heptagon (5–7) pairs, well separated by 21.8°. The model has been obtained by a method described in
FIGURE 7.4 Ball-and-stick model of an irregular GB in graphene with 21.8° misfit angle. (Adapted from Vancsó, P. et al. 2013. Carbon 64:
101–10.) The structure, composed of 597 atoms, was optimized with the Brenner–Tersoff potential constrained to two dimensions.
(a)
• The atomic-scale structure of different types of GBs
is extremely relevant in determining their properties
• The GBs are extended over several microns, that is,
on a much larger scale than the atomic scale, while
their local, atomic structure may differ along their
length from place to place (they can be “meandering”)
• The graphene grains touching at the GB are just
one single-atom thick, so the energy pumped in this
system by using energetic electrons methods such
as high-resolution transmission electron micros-
copy (HRTEM) and scanning transmission electron
microscopy (STEM) may alter the examined object
during the process of examination [56,57]
(b) (c) 7.4.1 Electron Microscopy

Electron microscopy, in particular HRTEM [56], and STEM
27°
[55,58] are very useful in revealing the grain structure of a
sheet of polycrystalline graphene. The different grain orienta-
tions can be revealed, for example, in large-scale dark-field
TEM (DF-TEM) images, while the atomic-scale details may
be investigated using aberration-corrected annular dark-field
scanning transmission electron microscopy (ADF-STEM),
with a subnanometer electron beam scanned over the sample
FIGURE 7.5 TEM investigation of CVD-grown graphene. (a)
while the medium-to-high-angle scattered electrons are col-
Gray scale, dark-field image overlay depicting the shapes and lattice lected [55].
orientations of several grains extracted using several different aper- A detailed GB study using HRTEM revealed that under
ture locations and shade coding the regions corresponding to the the conditions of e-beam irradiation, GBs are predominantly
diffraction spots encircled with different shades of gray in the dif- constructed of quasi-continuous chains of pentagon–hepta-
fraction pattern on the right hand of the image. Scale bar 500 nm. gon pairs, and the GBs tend to take on meandering, locally
(b) Atomic-resolution ADF-STEM images of two graphene grains, curved configurations as in Figure 7.5 [55]. Movies taken on
(bottom left, top right) intersect with a 27° relative rotation. An ape- GB evolution on electron microscope convincingly show that,
riodic line of defects stitches the two grains together. (c) The image
once such a configuration is attained, its general structure and
from (b) with the pentagons, heptagons, and distorted hexagons
of the GB outlined. Scale bars in (b) and (c) 0.5 nm. (Reproduced
shape will be conserved despite the continuous transforma-
with permission from Huang, P.Y. et al. 2011. Nature 469 (7330): tions between several, essentially equivalent configurations.
389–92.) This is attributed to the fact that significant changes in the
atomic structure of graphene can be induced via bond rota-
tions even under 80-kV electron irradiation, as shown previ-
Reference 54. In addition to several pentagon–heptagon pairs, ously [59], whereas knock-on sputtering of three-coordinated
there is a nine-edge polygon, adjacent to a pentagon, which carbon atoms becomes significant at slightly higher accel-
contains a twofold coordinated atom. eration voltages. On the other hand, if edges are present, for
Alternatively to computer modeling, aberration-corrected example regions containing double-coordinated carbon atoms
transmission electron microscope (TEM) may attain such a edge restructuring may take place [57,60,61]. Ultimately, such
nice resolution that it becomes possible to sketch the atomic a process can possibly lead to a better grain stitching than the
structure of a GB that, by chance, crosses the imaging field one that resulted after growth. However, the fairly low-energy
of graphene (see Figure 7.5b and c). Rare images of that sort threshold of 50 eV for carbon atom displacement from the
reveals heptagon–pentagon pairs distributed along a winding edges [57] should make one cautious of the possible modifica-
path, with distorted hexagons in between [55]. tion of the as-grown GBs in the electron microscope. Another
aspect that has to be taken into account is that in the TEM
7.4 EXPERIMENTAL METHODS FOR image, the region of the GBs appears as a flattened out image,
REVEALING THE STRUCTURE OF whereas in fact the equilibrium structure is a 3D structure,
as it was calculated in free space for the GB shown in Figure
GBs IN CVD GRAPHENE
7.5 [62].
Table 7.2 lists the available experimental techniques reviewed A further aspect that has to be taken into account is that
in this section and summarizes their main advantages. for making possible the TEM investigation the graphene layer
The investigation of GBs in graphene is a somewhat chal- has to be removed from the solid surface on which it was
lenging task for the following reasons: grown and has to be transferred onto a grid. This process can
TABLE 7.2
List of Experimental Techniques Used for the Study of Grain Boundaries in Graphene Samples Grown by
CVD on a Cu Substrate
Class Technique Accessible Information
Electron microscopy HRTEM and STEM Grain structure of a suspended graphene sample
Aberration-corrected annular dark-field STEM Atomic structure of the boundary
Dark-field transmission electron microscope Grain orientations
LEEM Grain structure of graphene on its Cu substrate
DF-LEEM Grain orientations
SPM AFM Misorientation angle between adjacent grains on a
sample transferred onto an atomically flat support
STM and STS Misorientation of adjacent grains and electronic
structure of the GB in a sample on its Cu substrate
Optical microscopy Bright field microscopy Morphology of the graphene grains or islands on
the Cu substrate used for the growth
Dark-field microscopy/microscopy with polarized light Morphology of the Cu surface underneath
induce alterations of the GB regions [63]; therefore, methods the typical four-lobed star-shaped grains growing on (100)
such as scanning tunneling microscopy (STM) [49] can bring Cu surfaces are in fact composed of four crystallites, each
very valuable information to complement the data obtained of them having a different orientation with respect to the Cu
by STEM. substrate [9]. In the online supporting material of the very
The reciprocal misorientation of neighboring grains was same report, movies show the way in which the motion of the
found by TEM [55] to cover the full angular range from 0° surface steps produced on the Cu surface is quenched under
to 30°, with a somewhat more frequent occurrence of small- the growing graphene layer.
angle GBs (~10°) and large-angle GBs (~30°). However, a In agreement with the above results for graphene grown
recent examination of CVD graphene grown on Cu foil has on Cu(100)/MgO(100), it was found from LEED data that the
revealed a nonuniform distribution of the misfit angle that graphene covers the Cu surface with two preferential [10] ori-
peaked at about 19° [8]. In fact, it was found [55] that the entations with angles of 0 ± 2° and 30 ± 2° with respect to the
growth conditions may have a dramatic effect on the grain underlying Cu[011] lattice [65].
size by comparing three different CVD methods: A and B, Of course these findings raise the question of the dif-
two furnace-based CVD methods differing only in the purity ferences between intragrain GBs and intergrain GBs [5].
of the Cu foil used, while method C used a rapid thermal pro- Intragrain GBs are those formed around one single nucleation
cessor tube furnace. The slight differences between growth center, for example, within a four-lobed star-shaped grain,
methods effected significant changes in the grain size, shape, while intergrain GBs are those formed when such grains—
and crystallographic orientation of the CVD graphene. For formed around different nucleation centers—touch together
example, with growth method C, grains averaging 1–4 μm and fill the available surface, see Figure 7.6. As the two kinds
were obtained, which is an order of magnitude larger than the of GBs are formed under different supersaturation and diffu-
grains grown using method A (shown in Figure 7.5). sion conditions, very likely their atomic structure and, as a
Interesting correlation is reported between the average consequence, their physical properties will be also different.
grain size as determined from DF-TEM data and the electri- For an experimental series of scanning electron micro-
cal transport properties [64]. Clearly the small grain samples scope (SEM) images illustrating the above process, see Figure
(grain size less than 5 μm) exhibited transport properties at 7.4 in Reference 66.
1.6 K approximately one order of magnitude worse as com- Dark-field LEEM (DF-LEEM) very clearly revealed that
pared with large grain samples (grain size in the region of on Cu (111) the contrast of the DF image is quite uniform
250 μm). in the entire view, representing the single orientation of gra-
phene in the measured area, moreover, this was found con-
stant after scanning 1 mm with the electron beam [65]. On
7.4.2 Low-Energy Electron Microscopy
the other hand, the DF LEEM images of graphene/Cu(100)
Low-energy electron microscopy (LEEM) can be also used showed a marked contrast to that of graphene on Cu(111). The
to reveal the grain structure [9,65]; however no informa- as-grown graphene exhibited a clear multidomain structure
tion will be available on the structure of the GB itself. On with small domains. Under one selected diffraction condition,
the other hand, the advantage of the method is that the grain a given mosaic structure appeared, while the same diffraction
structure can be investigated right on the substrate on which condition rotated by 30° gave a different mosaic pattern [65].
the graphene layer was grown. It was shown by LEEM that This is in agreement with the TEM results discussed earlier.
These two kinds of GBs seem to have different oxidation

rates, suggesting that the type and density of structural devia-
tion from the perfect hexagonal network is different in the
two kinds of GBs. As clearly seen from Figure 7.7, the intra-
grain GBs were not oxidized. The intragrain GBs behave in
the same way like the GBs of HOPG [32], while the intergrain
GBs are oxidized.
7.4.3.2 STM Investigation of GBs in CVD Graphene

STM may prove very useful in the investigation of graphene
GBs right on the Cu substrate on which the material was
grown, without any manipulation needed, in exactly the state
in which the GBs are after growth. Additionally, STM in com-
bination with STS can equally well reveal the misorientation
of neighboring grains and the characteristic electronic struc-
FIGURE 7.6 Schematic drawing showing the intragrain GBs
already existing in the graphene grains and the intergrain GBs that
ture of the GBs, Figure 7.8 [49]. In subsequent experiments,
will be generated when the three graphene grain grow together. very similar findings were reported for CVD graphene that
was transferred from the Cu substrate to SiO2 [72].
The STM and STS investigation of CVD graphene on the
7.4.3 Scanning Probe Microscopy
Cu substrate on which it was grown [49] revealed that the
Scanning probe microscopes (SPMs) are far more wide- conductivity within the GB is about an order of magnitude
spread in laboratories than STEMs or LEEMs. Atomic force lower as compared with the defect-free graphene lattice. This
microscopy (AFM) is much more popular than STM; how- implies that from the electrical point of view, the intercon-
ever, due to the complex step structure that develops dur- nected GB network divides the graphene sample into highly
ing the CVD growth on the surface of Cu [67], topographic conductive single-crystalline islands delimited by more
AFM is less well suited for direct investigation of GBs on resistive interface regions of a few (3–5) nanometer widths.
the Cu support if no special pretreatments are used, like These findings are in agreement with transport measurements
purposeful oxidation of the Cu through the GB (discussed reporting that the intergrain conductivity (across a GB) was
in Section 7.4.4). STM on the other hand is quite well suited found to be reduced by approximately an order of magnitude
to investigate both the misorientation angles of neighboring as compared with the intragrain conductivity (no GB), when
grains as it can provide atomic-resolution information on transport through purposefully produced, individual GBs was
the grain structure, and for the study of the local electronic measured [23,73]. The GBs of CVD graphene form n-doped
structure of the GBs by scanning tunneling spectroscopy inversion channels within the overall p-doped graphene lat-
(STS) [49]. tice [49]. The origin of this phenomenon can be attributed
to self-doping, whereupon a charge transfer occurs between
7.4.3.1 AFM Investigation of GBs in CVD Graphene the defect-free graphene lattice and the nonhexagonal carbon
If graphene is transferred from the Cu substrate on which it rings making up the GB. An additional source of doping could
was grown, it is best to transfer it onto an atomically smooth, come from functional groups, or adsorbates, which preferen-
oxidation-resistant substrate like freshly cleaved mica or tially attach to the defective GBs; this will be discussed in
h-BN [68]. In this way, the corrugation of the substrate will more detail in Section 7.5.4.
have practically no influence on the topography of graphene The particular electronic structure of the intergrain GBs
or its oxidation. It is well known that local stress or defects may be associated with the way in which these GBs are
may affect the oxidation of graphene-type materials [69,70]. formed in the stage when the growth of neighboring grains
Very similar result was obtained by using CVD graphene considerably hinder the formation of a regular structure. As
[26] transferred to freshly cleaved mica substrates and sub- it was revealed recently by the systematic investigation of
jected to a mild oxidation in air at 500°C as those reported by such GBs in CVD graphene by low-temperature STM and
STEM [55]. The intergrain GBs were revealed by oxidation, STS measurements combined with structural modeling, the
while the intragrain GBs and the structurally perfect regions irregular GBs formed under these conditions are susceptible
were unaffected as confirmed by the parallel oxidation of of containing double coordinated C atoms, which may have a
micromechanically cleaved graphene (from HOPG) under characteristic electronic structure Figure 7.9 [50].
identical conditions [71]. By taking atomic-resolution AFM Comparing the spectra in Figure 7.9 with those reported for
images within each grain, it is possible to characterize the HOPG [74], one can clearly conclude that in the GBs grown by
misorientation of neighboring grains. The false-color grain CVD there appear additional characteristic features slightly
misorientation map and the corresponding grain size and mis- above the Fermi level, which is attributed to the presence of
orientation statistics are shown in Figure 7.7. C atoms in the region of the GB which has only two neigh-
It is worth to emphasize here again the differences between bors [50]. Very few such features were identified in the GBs
the intragrain and intergrain GBs already mentioned above. of HOPG grown at very high temperatures and pressures [74].
4.7
nm
1 nm
500 nm
2.8 (nm–1) 0
0° 30°
80 14
12
# of boundaries
60 10
# of grains
8
40 6
20 4
2
0 0
0.0 0.2 0.4 0.6 0.8 1.0 0 5 10 15 20 25 30
Characteristic size of grains (μm) Relative angle of grain boundary (deg)
FIGURE 7.7 AFM image of oxidized CVD graphene on mica showing the etch trenches. By comparing the Fourier transform of atomic-
resolution AFM images of individual grains, one can make a gray scale map of the crystallographic orientation of the grains relative to
a given direction (overlayed on the image showing the oxidation trenches). Two small images to the left show one such atomic-resolution
image and the Fourier transform thereof. The atomic-resolution image was recorded by contact mode AFM on the grain shown by the arrow.
Histograms at the bottom show the distribution of the grain size (left) and the relative angle of the grains forming the boundaries (right).
(Reproduced with permission from Nemes-Incze, P. et al. 2011. Applied Physics Letters 99 (2): 023104-1–023104-3.)
Tight-binding DFT calculations equally confirm that these within the 2D conductance map indicates a difference in the
carbon atoms are responsible for additional electronic states electronic structure at the particular energy of Ub = −200 mV.
close to the Fermi energy in the CVD-grown graphene GBs The different grains are revealed by the moiré-type patterns
[50]. More detailed investigation of such defects was reported that are formed by overlaying the hexagonal graphene with
recently on the surface of HOPG, as well [75]. The fact that the Cu(111). The nucleation and growth of individual plate-
HOPG is used both as sample and calibration standard in lets, which are mobile at elevated temperature, takes place.
STM since almost 30 years, and it is the first time that such Platelets stitch together with other platelets to form larger
GBs are investigated in detail on HOPG, clearly indicates that sheets resulting in different domains and domain boundaries.
they may not be very frequent in the usual HOPG samples. This is not initially apparent in the topographic data but is
On the other hand they seem to be quite common in CVD clearly observed with STS mapping. The observation of mul-
graphene. tiple moiré patterns, which results from different rotational
STM and STS measurements carried out in UHV on alignments of the graphene lattice (2.46 Å) with the under-
graphene grown on Cu(111) at 1000°C revealed that despite lying Cu(111) lattice (2.56 Å), indicates that the graphene is
the single-crystalline nature of the substrate, the synthe- weakly coupled to the substrate [8,76]. The moiré patterns are
sis on Cu(111) involves the nucleation of monolayer islands very useful in revealing differences in the orientations of the
of graphene that patch together like a quilt to form larger grains, which otherwise would not be visible in larger-scale
single-layer sheets; as a result there are numerous domain images.
boundaries that scatter electrons and can affect carrier mobil-
ity. The overall interaction between the graphene and Cu(111)
7.4.4 Optical Microscopy
substrate results in the observation of two predominant moiré
patterns [76]. The graphene platelets are clearly resolved by The main disadvantage of the methods discussed so far for
STS. The contrast between the graphene and the Cu(111) the investigation of GBs in CVD graphene is that they do
Topography Conductivity
(a) 0.5 15.0 (b)
50
+338 mV
10.0
pA/mV
nm
5.0
0
0
0 5.0 10.0 15.0 nm 0 15 nm
Doping I–V characteristics

(c) 200 (d)
10 50
dI/dV
mV
5 × 10
35
–300 300
Bias (mV)
I (nA)
0
0
–5
–10
–150 –400 –200 0 200 400
0 15 nm Bias (mV)
FIGURE 7.8 Atomic-resolution constant-current topographic STM image (100 mV, 1 nA) of a 29° graphene GB, with the boundary
region displaying a bright contrast. (b) Spatially resolved tunneling conductivity map revealing a markedly suppressed conductivity at
the GB region. (c) Spatial map of the Dirac point (minimum of dI/dV) position relative to the Fermi energy, indicating local doping. (d)
Representative individual current–voltage characteristics acquired at the GB (low) and over the unperturbed graphene lattice (steep). The
corresponding numerical derivatives (differential conductance) are shown in the inset. (Reproduced with permission from TapasztÓ, L. et al.
2012. Applied Physics Letters 100 (5): 053114.) (See original paper for color coded information.)
not allow fast evaluation of the GBs over large-area samples. by CuO will have a darker, brownish red color, while the
The methods based on optical microscopy come to fill in this regions protected by the graphene will be lighter. It is again
missing possibility. However, the optical methods will not be worth to emphasize here that the intra-GBs are not affected by
able to provide data on the atomic details of the GBs itself, or this oxidation (see, e.g., Figure 7.5 in Reference 78). Density
on the misorientation between the two graphene grains sepa- functional calculations support that the non-hexagonal rings
rated by the GB region. occurring in the GB regions are responsible for the fast pen-
The atomic-scale size of the GB region makes it impos- etration of the radicals and, interestingly, the functionaliza-
sible to distinguish the different optical properties of the GB tion of the GB by the OH groups increased the penetration
region, if any, from that of defect-free graphene. Therefore, probability of the O radical [77].
the optical methods rely on the different chemical properties Figure 7.10 briefly illustrates the opportunities offered
of the GBs, to oxidize the Cu underneath the graphene layer by using the second oxidative method to elucidate the effect
to a sufficient extent that will make the GB region observable of various growth parameters. One may note how, with the
under the optical microscope [77,78]. The first method [77] is increasing growth temperature, the size of the graphitic grains
based on somewhat sophisticated way of achieving the oxida- increases, while their number is reduced, that is, the number
tion of the Cu under graphene in the region of the GB, using of GB per unit length will be also reduced, and their shapes
O and OH radicals generated by UV irradiation in a wet envi- turn from irregular to the six-lobed star.
ronment. The second method [78] uses simple oxidation on Quite recently, an even simpler method was reported,
incompletely covered Cu substrates to reveal the morphology which does not employ oxidation, but relies on the scattering
of graphene islands by oxidizing in air at 160°C or by H2O2 the of light due to the steps that develop on the Cu surface dur-
Cu which is not covered by graphene. The Cu surface covered ing growth. This scattering is used to distinguish in dark-field
dI/dV on GB and graphene

(a) (b) 23°
dI/dV (a.u)
5 nm
(d) 25
20
–0.4 –0.2 0.0 0.2 0.4

Number of peaks
Bias voltage (V) 15
(c) Calculated average DOS

10
Density of states
0
–0.4 –0.2 0.0 0.2 0.4 –0.4 –0.2 0.0 0.2 0.4
Energy (eV) Peak position (V)
FIGURE 7.9 (a) dI/dV Plots measured at selected points shown on the topography image. (b) Upper five curves were measured on the GB
and lower four curves on the graphene, the positions of the measurement are shown by dots. (c) Statistical average of the density of states of
the six disordered GBs, calculated by DFT. (d) Histogram of the positions of LDOS peaks in 60 dI/dV spectra, measured over different posi-
tions on three distinct GBs (misorientation angles: 23°, 25.5°, 27°). (Reproduced with permission from Péter Nemes, I. et al. 2012. Carbon
64: 178–86.) (For color coded information see cited paper.)
optical microscopy the regions covered by graphene (with electric transport through the CVD graphene are at least
steps) from the regions not covered by graphene (without one order of magnitude below those of cleaved graphene [2].
steps) [79]. In fact, this method is very similar to the method Another category of property that may be strongly influenced
using polarized light for revealing graphene grains on the Cu by the presence and the kind of the GBs are the mechani-
substrate and making visible the orientation of the Cu steps cal properties. Last but not least, the chemical properties of
[67] that will originate the nanowrinkles, which in turn will the GBs may be different from those of defect-free graphene
impact the electrical transport even after the transfer of gra- under other aspects too, not only as it regards the penetra-
phene to atomically flat mica [11]. tion of OH and O radicals. In the following sections, we will
examine these effects in more detail.
7.5 CONSEQUENCES OF THE PRESENCE OF GBs
7.5.1 Electrical Transport
One could expect that the presence of GBs, in other words,
the polycrystallinity of the CVD-grown graphene is primar- Computer modeling predicts that GBs may have under sev-
ily going to affect the transport properties, local electronic eral aspects profound effects on electric transport [4,80]. The
structure, and the local chemistry. Some of these, like the propagation of charge carriers is strongly influenced by the
local electronic structure (localized states in the STS spectra), localized states that may appear on the GBs [48] and also
and the permeation of radicals (OH and O) through the GBs depends a lot on the type of the GB, regular or irregular (under
have been already discussed above. In the same time, it is well the aspect of the arrangement of nonhexagonal rings, and/or
known that most frequently the parameters characterizing containing/not containing double-coordinated C atoms) [54].
(a) individual grains. The electric transport within each of the

two grains is comparable, while the transport through the GB
is clearly reduced, but all transport is ohmic.
In a recent report on electric transport through the GBs
of CVD-grown polycrystalline graphene, it was found that
20 μm the electrical conductance may improve by one order of
magnitude for the GBs with better interdomain connectivity
Jme [85]. The grain structure of CVD graphene was first imaged
(b)
using TEM and individual domains and the GBs were then
addressed electrically in polycrystalline graphene. Such an
approach was found to be crucial, because graphene domain
structures generated during synthesis form nontrivial patterns
that are strongly dependent on the growth conditions and dif-
20 μm ficult to be predicted a priori [85]. Three different CVD meth-
ods were used: (i) method A yielded grain size D ≈ 1 μm and
T
(c)
irregular polygonal shape with good stitching between the
grains; (ii) method B yielded D ≈ 10 μm and grains of more
regular polygonal shapes, with good stitching; (iii) whereas
growth C yielded large, four-lobed, flower-like grains with
D ≈ 50 μm and frequent stitching defects. For an A-type GB,
it was found that RL–R (measured through the GB between the
20 μm
left and the right grains) exhibited an increased resistance,
particularly at gate biases near VDirac, which was attributed
to additional scattering caused by the GB. The results of the
FIGURE 7.10 Graphene grains on Cu substrate visualized by
measurement were interpreted to describe the presence of a
optical micrographs taken following the oxidation of the Cu not
covered by graphene. Effects of temperature and gas feeding rate
GB as simply an extension of the conductance channel defined
on graphene growth. The different growth conditions used before by an effective length λ associated with the GB. A different
annealing (T = growth temperature, JMe = methane feeding rate): electrical behavior was observed for analogous measurements
T(°C)/JMe (sccm) for 1.5 min: (a) 1000/1.0, (b) 1000/0.3, (c) 1020/0.3. at a GB from growth C: RL–R was considerably greater at all
(Reproduced with permission from Jia, C. et al. 2012. Scientific gate values than the values within individual grains, signify-
Reports 2 (January): 707.) ing increased scattering at the GB. The GB resistivity was
about an order of magnitude over that found for growth A.
Clearly, the properties of GBs are of extreme importance The mean λ values for A samples was 110 nm (five devices)
for practical applications of graphene as they may have dra- and it was 880 nm for C samples (six devices). The results
matic effects on the electric transport properties, as well of TEM and transport measurements on individual GBs in
[39,81,82]. For example, first-principles quantum transport CVD graphene shows that the conductance of GBs is highly
calculations showed that in the case of periodic GBs, depend- correlated with their structure [85]. These experimental data
ing on the GB structure, two distinct transport behaviors can strongly support that the character of the GB regular versus
be found—either high transparency or perfect reflection of irregular has a major influence on the electric transport across
charge carriers over remarkably large energy ranges [82]. the GB. In other words, if no proper ways are found, which
Certain special structures, like the so-called 558 line defect will allow a precise control of the formation and structure of
[48], may even exhibit important and attractive properties, the GBs, their presence may be a serious impediment in the
like valley filtering [83]. If the intrinsic structure of a GB may way of practical applications using CVD-grown graphene.
be almost transparent to electronic transport, it is worth men- Promising routes to overcome the bottleneck arising from
tioning that chemical species adsorbed on the GB (extrinsic the way in which the graphene grains are stitched together
defects) may increase the scattering of electrons substantially, seem to be the application of suitable pretreatments of the
which influence the global conductivity of the sample [84]. solid-state Cu foil [20] and the use of molten Cu [18,19] sub-
A very conclusive experiment on a purposefully produced strates for graphene CVD. The molten Cu may be supported
GB between two grains clearly shows that even in the region on SiO2, by Mo or W foil. One of the main advantages of
where straight edges of two similar hexagonal graphene crys- using molten catalyst is the complete elimination of previous
tallites have grown together, an excess of defects is found and surface history (rolling, oxidation, crystalline structure, GBs
the intergrain electric transport is reduced as compared with in the Cu foil, etc.) and another important advantage is the
the intragrain electric transport, Figure 7.11 [23]. elimination of the influence of Cu steps formation and migra-
The Raman maps in Figure 7.11a and b very convincingly tion during the CVD growth. It was found that typically the
show that graphene is continuous over the area of the two hexagonal graphene flakes (HGFs) were well dispersed on the
grown together grains and in the region of the GB the defect surface of the molten Cu, and there was no clear alignment
density is significantly higher than over most of the area of relation between different HGFs when the HGFs were not
ID (x,y) IG (x,y)
(a) (b)
3000
1000
2500
800
2000
600
1500
400
1000
200 500
2 μm 2 μm
0 0
(c) (d) 40 V 7,8

2 3 4 5 6 V 8,9
V 9,10
I (μA)
20 T = 300 K
0
7 0 20 40 60 80 100 120
1 10 9 8
5 μm V (mV)
FIGURE 7.11 Two graphene grains grown together on purpose for the investigation of GB effects on electric transport. (a) and (b)
Spectroscopic Raman mapping of graphene grains and GBs, intensity maps of the “D” (a), and “G” (b) bands, respectively, for two coalesced
graphene grains with a single GB. (c) Electronic transport across a single GB, optical image of a device with multiple electrodes (numbered
1–10) contacting two coalesced graphene grains (indicated by dashed lines). (d) Representative room-temperature I–V curves measured
within each graphene grain and across the GB. (Reproduced with permission from Yu, Q. et al. 2011. Nature Materials 10 (6): 443–9.)
fully covering the surface. As the density or coverage of HGFs of application. In most carbon materials, heat transport by
on the Cu surface increased, introducing spatial constraint acoustical phonons is much more efficient than by electrons.
of the HGFs, the HGFs became self-aligned into an ordered The small mass of the carbon atom along with strong covalent
structure with the most compact packing arrangement (Figure bonds and the large phonon mean free path are responsible
7.2d in Reference 19). The edge-to-edge alignment of HGFs for high thermal conductivity of diamond, single-walled car-
led to the formation of low-angle GBs for adjacent HGFs. bon nanotubes, and graphene. Early calculations predicted a
The combination of high levels of hydrogen flow in thermal conductivity κ of 7.4 × 103 W/mK for ideal graphene
AP-CVD with molten Cu may be extremely promising for at 300 K [88], a factor of 4 larger than the value measured
a dramatic reduction of structural disorder originating from at room temperature along the basal planes in high-purity
randomly nucleated and randomly oriented graphene platelets graphite [89]. By comparison, the experimental room-tem-
growing together. A very important advantage of the liquid perature κ values for flakes of exfoliated graphene suspended
catalytic surface is that, opposite to the solid surfaces, includ- over SiO2/Si trenches are in the range 3.08 × 103 to 5.15 × 103
ing the single-crystalline ones like Cu(111) or Cu(100), it stays W/mK [90]. For supported samples, phonon–substrate inter-
atomically smooth under the growth conditions for HGFs and actions reduce the value of κ down to 600 W/mK in the case
it allows the free movement of the HGF, so that they can self- of exfoliated graphene deposited on amorphous SiO2 [91].
align. Not to mention that large-area molten metal surfaces Phonon scattering by GBs is expected to reduce heat con-
are much cheaper to be produced than large-area single- ductivity in polycrystalline graphene. Reference 92 reports
crystalline metallic foils. the κ values, here measured at 350 K, around 2.5 × 103 and
370 W/mK for CVD graphene in the suspended geometry
and in the supported geometry, respectively. These values are
7.5.2 Thermal Transport
significantly smaller than the corresponding data obtained on
Since a few years, thermal transport in graphene has attracted exfoliated samples. GB effects can be investigated by com-
growing interest of the scientific community [86], in particu- paring the thermal properties of samples presenting different
lar for applications in the thermal management of Li batteries grain sizes. In Reference 93, the characteristic size L a of well-
[87]. In this domain, like in many others, graphene presents ordered domains, as determined by Raman spectroscopy (ID/
interesting physics, together with important perspectives IG intensity ratio), was chosen as a parameter. The thermal
conductivity of CVD graphene samples produced by differ- by a heptagon and a hexagon [45,42]. For polycrystalline gra-
ent protocols and subsequently transferred on silicon pillars phene containing several irregular GBs, microscopic calcu-
was monitored versus this parameter. The scaling law κ ~ L1a/ 3 lations predict a mechanical strength roughly half the one
fits the experimental data well [93]. When L a increases well computed for pristine graphene [53]. For recent calculation of
above the phonon mean free path in graphene, κ must saturate the strength failure through GBs, see Reference 98.
to a value governed by the quality of the grains and should no
longer depend on their size. Calculations put this critical value
7.5.4 Chemical Properties
between 100 nm [94] and several hundred nanometers [95].
Around and below that limit, not only the size of the grains As discussed earlier in relation to the oxidation processes tak-
but also the roughness of their boundaries does contribute to ing place across GBs, exploited for the investigation of GBs by
the sample thermal conductivity [86]. In addition, calculations optical microscopy, these linear defects have chemical prop-
indicate that the thermal resistance of a regular GB depends erties that deviate from that of perfect graphene. As discussed
on the misfit angle and on the symmetric (like e.g., in Figure above in relation to the oxidation of the GBs themselves
7.2) [94] versus asymmetric (like e.g., in Figure 7.3) [96] and the oxidation occurring through the GBs, the chemical
arrangement of the grains. In the latter case, interestingly, the properties of the intragrain and the intergrain GBs are differ-
boundary between zigzag and armchair nanoribbons (Figure ent. Not surprisingly, these findings show that the particular
7.3b) is predicted to exhibit thermal rectification—heat should chemical properties of these linear defects are decided by the
flow more easily from armchair to zigzag than from zigzag to kind of defects and disorder they may contain. As theoreti-
armchair [96]. cal models and calculations show the GBs may contain non-
hexagonal rings ranging from squares to nonagons and even
double-coordinated carbon atoms [51,53,71].
7.5.3 Mechanical Properties
Hydrogen is always present in the CVD growth and usu-
From the mechanical point of view, atomistic calculations ally the graphene samples grown by CVD are cooled down
based on molecular dynamics show that GBs in graphene may in an argon hydrogen mixture; therefore it is interesting to
be as strong as perfect graphene [42,45,53]. Recent experi- know if hydrogen has any preference for the GBs. Indeed, it
ments carried out on graphene grown by CVD on a Cu foil was found by model calculations that the H adsorption energy
support these theoretical predictions [63]. Measuring the elas- increases as the rings deviate from hexagons, the largest
tic constants of a monolayer CVD graphene demands to trans- adsorption energy of εi > −7 eV [51] being found for dangling
fer the monolayer onto a specially designed support, such as bonds. These results are in agreement with ab initio calcu-
Si/SiO2 containing microperforations or microtrenches [97]. lations reported recently on H adsorption on nonhexagonal
This operation must be done with care in order not to dete- rings in graphene GB regions [99]. Computed STM images
riorate the mechanical properties of the sample. When these of GBs after H adsorption indicate that the H atoms will be
conditions are met, the measured in-plane stiffness of poly- localized on nonhexagonal rings and will appear as tunneling
crystalline graphene membranes does not differ statistically current maxima in the STM images.
from that measured on single-crystal graphene, 340 N/m*. The energetics and electronic properties of substitutional
However, the mechanical strength (mechanical tension lead- B (BC) and N (NC) doping, and BN codoping in graphene
ing to fracture) of samples with average grain sizes 1–5 µm is with distinct GB defects were investigated by ab initio sim-
significantly smaller than that of single-crystal graphene, but ulations. The results indicate that a single B or N impurity
still retains 85% of the strength of a defect-free sample [63]. atom and an isolated BN pair prefer to incorporate into the
This result contrasts the earlier experiments having concluded GB region [100]. The calculated electronic properties show
to a reduction of strength by about a factor of 1/3 in polycrys- that the incorporation of a single substitutional B (N) atom
talline graphene [55]. produces a p-type (n-type) doping in all the investigated GB
Coming back to the theoretical predictions, it is important structures. However, the BN codoping introduces only small
to point out that regular and infinitely long GBs were consid- modifications of the host electronic structure.
ered in References 42, 45, whereas irregular GBs were inves- Recent statistical data on devices prepared from CVD gra-
tigated in Reference 53. According to molecular dynamics phene grown on Cu and transferred to SiO2 substrates sub-
simulations, the strength of a regular GB increases with the jected to various treatments after transfer—aimed at reducing
square of the tilt angle when it involves a periodic arrange- the water trapped between the SiO2 and the graphene layer—
ment of pentagon–heptagon pairs [42]. There is no such sim- indicated that indeed the trapped water can significantly
ple relationship between strength and tilt angle in the other degrade the electric parameters of such devices [101]. Using
cases [45]. For those GBs that involve pentagon–heptagon exfoliated (kish) graphene, it was shown experimentally that
pairs to the exclusion of any other defect, the calculations dry oxygen and the wet ambient dope graphene in different
show that the fracture starts by the breaking of a bond shared ways on an SiO2 substrate. Dry oxygen dopes graphene from
the basal plane while the ambient dopes graphene from the
* For a 2D membrane under uniaxial in-plane strain, Hooke law writes
edges or from the substrate in the vicinity of the edge. Upon
γ = Cε, where γ is the applied force per length unit (mechanical tension), vacuum annealing, doping is fully reversible in the former
ε is the resulting strain, and C is the in-plane stiffness. case and only partially reversible in the latter case. This shows
that oxygen and moisture are the primary species in the atmo- conditions of the CVD process, while the number of the GBs
sphere responsible for doping graphene and the defects—the per unit length is related to the number of nuclei from which
edge regions or the GBs—have a marked effect on graphene graphene grains start growing. Experimental data clearly show
doping [102]. The effect of moisture was confirmed recently by that in real CVD-grown samples, the GBs deviate strongly
high-resolution electron energy loss spectroscopy (HREELS), from the ones usually taken into account in theoretical mod-
which was used to study the influence of a humid environment eling. Oxidation experiments confirm that the oxidation
on the collective electronic excitations of a quasi-freestand- behavior of the intragrain and intergrain GBs is different, the
ing polycrystalline graphene monolayer supported on Pt(111) intragrain GBs very likely being closer to that kind of regular
[103]. It was found that water adsorption at room temperature GBs which are most frequently used in theoretical and model-
on graphene/Pt(111) is dissociative and it gives rise to adsor- ing approaches. The way in which the stitching of the graphene
bate fragments, these adsorbed species change the energy, the platelets will take place will impact on the electric properties
dispersion, and the lifetime of the graphene sheet plasmon. of the GBs, a more disordered GB corresponding to a more
The graphene growth was carried out in UHV and water was significant increase in resistivity and, as shown by several
distilled in the vacuum system through a leak valve [103]. STM images measurements and transport data, an increased
In another UHV experiment, it was shown quite convinc- scattering of the charge carriers in the region of the GB. On the
ingly that water molecules can penetrate through GBs in CVD other hand, the presence of the GBs may strongly enhance the
graphene at temperatures much below room temperature; the sensitivity of chemiresistors, which is clearly a positive effect.
water that crossed the GBs can be frozen under the graphene The investigation of the chemical properties of the GBs is only
layer and may induce rupture and exfoliation of graphene [104]. in its infancy, but already at this stage, it is clear that the GBs
Recently, an anisotropic hydrogen etching method of gra- exhibit an extreme sensitivity to humidity even under ambient
phene catalyzed by the copper substrate was developed. The conditions; this constitutes an inconvenience from the point
anisotropic etching is the reverse reaction of CVD graphene of view of environmental stability. The GBs may constitute
growth, and it is most efficient and selective for the develop- the route through which, despite the remarkable gas sealing
ment of zigzag edges at 800oC and 500 m Torr of hydrogen properties of cleaved graphene, various chemical moieties can
gas flow [105]. As the reaction starts from existing defect penetrate underneath the CVD-grown graphene and react with
sites, it is also suitable for revealing the grain structure of the substrate, as it was shown by the purposeful oxidation of
CVD-grown graphene. Copper played an important role in the Cu substrate for visualizing the GBs.
catalyzing the etching reaction, as no etching was observed It is very clear from the issues enumerated above that “GB
for graphene transferred to the Si/SiO2 substrate under similar engineering” will have to be a focus field if CVD graphene is
conditions. As the reaction exhibits selectivity for the zigzag going to reach industrial applications. This implies that both
edges, it may also be useful for revealing over large areas the experimental and computational methods aiming at a detailed
crystallographic orientation of graphene islands. characterization of the GBs have to be developed. Owing to
GBs were reported recently to significantly enhance the the geometrical characteristics of GBs—transversal extension
sensibility of chemiresistors in the detection of toluene (an on the atomic scale and length on several microns or tens-
electron donor) and of 1,2-dichlorobenzene (an electron accep- of-microns scale—these methods will have to be multiscale
tor) [106]. The increase in the chemical sensitivity of CVD methods, which will allow smooth blending of the atomic-
graphene as compared with exfoliated graphene is attributed scale and microscopic-scale experimental techniques and
to the presence of GBs. When the exfoliated and the CVD computational methods.
graphene were processed in ribbons of 5 μm width, the sensi-
tivity of CVD chemiresistors was enhanced into the parts per
billion (ppb) range. Clearly, the combination of edges and line
ACKNOWLEDGMENTS
defects enhances the response of the sensor, compared with This chapter in Hungary was supported by OTKA grant
edges or line defects alone [106]. K101599. LPB acknowledges the agreement between the
F.R.S.-FNRS (Belgium) and HAS (Hungary) that made pos-
sible the cooperation with PhL.
7.6 SUMMARY AND OUTLOOK
The properties of structurally perfect graphene are truly
outstanding under many aspects. On the other hand these REFERENCES
properties suffer strong modifications even from point, or 1. Bae, S., H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J.
close-to-point, size defects [107]. The effect of deviations from Balakrishnan et al. 2010. Roll-to-roll production of 30-inch gra-
the ideal structure is only amplified by the arrangement of phene films for transparent electrodes. Nature Nanotechnology
structural imperfections in linear defects like GBs. As it was 5 (August): 574–8. doi:10.1038/nnano.2010.132. http://www.
detailed earlier, the GBs themselves can be of various kinds: nature.com/doifinder/10.1038/nnano.2010.132.
2. Novoselov, K.S., A.K. Geim, S.V. Morozov, D. Jiang, Y.
regular, irregular, containing only nonhexagonal rings or con- Zhang, S.V. Dubonos, I.V. Grigorieva, and A.A. Firsov. 2004.
taining double-coordinated C atoms, as well as intragrain and Electric field effect in atomically thin carbon films. Science
intergrain GBs. The structure of the GB—and through this 306 (5696): 666–9. doi:10.1126/science.1102896. http://www.
their specific properties—are determined primarily by the ncbi.nlm.nih.gov/pubmed/15499015.
3. Biró, L.P., P. Nemes-Incze, and P. Lambin. 2012. Graphene: 16. Mattevi, C., H. Kim, and M. Chhowalla. 2011. A review of
Nanoscale processing and recent applications. Nanoscale chemical vapour deposition of graphene on copper. Journal
4 (6): 1824–39. doi:10.1039/c1nr11067e. http://www.ncbi.nlm. of Materials Chemistry 21 (10): 3324–34. doi:10.1039/
nih.gov/pubmed/22080243. c0jm02126a. http://xlink.rsc.org/?DOI=c0jm02126a.
4. Márk, G., P. Vancsó, C. Hwang, P. Lambin, and L. Biró. 2012. 17. Ago, H., Y. Ogawa, M. Tsuji, S. Mizuno, and H. Hibino. 2012.
Anisotropic dynamics of charge carriers in graphene. Physical Catalytic growth of graphene: Toward large-area single-crys-
Review B 85 (12): 125443. doi:10.1103/PhysRevB.85.125443. talline graphene. The Journal of Physical Chemistry Letters 3
http://link.aps.org/doi/10.1103/PhysRevB.85.125443. (16): 2228–36. doi:10.1021/jz3007029. http://pubs.acs.org/doi/
5. Biró, L.P. and P. Lambin. 2013. Grain boundaries in graphene abs/10.1021/jz3007029.
grown by chemical vapor deposition. New Journal of Physics 18. Wu, Y.A., Y. Fan, S. Speller, G.L. Creeth, J.T. Sadowski,
15 (3): 035024. doi:10.1088/1367-2630/15/3/035024. http:// K. He, A.W. Robertson, C.S. Allen, and J.H. Warner. 2012.
stacks.iop.org/1367-2630/15/i=3/a=035024?key=crossref. Large single crystals of graphene on melted copper using
b2237af953202d6473fc5625c18c75f0. chemical vapor deposition. ACS Nano 6 (6):
5010–17.
6. Li, X., W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner et al. doi:10.1021/nn3016629. http://www.ncbi.nlm.nih.gov/pubmed/
2009. Large-area synthesis of high-quality and uniform gra- 22617012.
phene films on copper foils. Science 324 (5932): 1312–14. 19. Geng, D., B. Wu, Y. Guo, L. Huang, Y. Xue, J. Chen, G.
doi:10.1126/science.1171245. http://www.ncbi.nlm.nih.gov/ Yu, L. Jiang, W. Hu, and Y. Liu. 2012. Uniform hexagonal
pubmed/19423775. graphene flakes and films grown on liquid copper surface.
7. Batzill, M. 2012. The surface science of graphene: Metal Proceedings of the National Academy of Sciences 109 (21):
interfaces, CVD synthesis, nanoribbons, chemical modifica- 7992–6. doi:10.1073/pnas.1200339109. http://www.pnas.org/
tions, and defects. Surface Science Reports 67 (3–4): 83–115. cgi/doi/10.1073/pnas.1200339109.
doi:10.1016/j.surfrep.2011.12.001. http://linkinghub.elsevier. 20. Yan, Z., Z. Peng, and J.M. Tour. 2014. Chemical vapor depo-
com/retrieve/pii/S0167572911000690. sition of graphene single crystals. Account of Chemical
8. Wilson, N.R., A.J. Marsden, M. Saghir, C.J. Bromley, R. Research 47 (4): 1327–37. doi: 10.1021/ar4003043. http://pubs.
Schaub, G. Costantini, T.W. White et al. 2013. Weak mismatch acs.org/doi/abs/10.1021/ar4003043
epitaxy and structural feedback in graphene growth on cop- 21. Kim, H.K., C. Mattevi, M.R. Calvo, J.C. Oberg, L. Artiglia, S.
per foil. Nano Research 6 (February): 99–112. doi: 10.1007/ Agnoli, C.F. Hirjibehedin, M. Chhowalla, and E. Saiz. 2012.
s12274-013-0285-y http://link.springer.com/article/10.1007/ Activation energy paths for graphene nucleation and growth
s12274-013-0285-y#page-1 on Cu. ACS Nano 6 (4): 3614–23. doi:10.1021/nn3008965.
9. Wofford, J.M., S. Nie, K.F. McCarty, N.C. Bartelt, and O.D. http://www.ncbi.nlm.nih.gov/pubmed/22443380.
Dubon. 2010. Graphene Islands on Cu foils: The interplay 22. Li, X., W. Cai, L. Colombo, and R.S. Ruoff. 2009. Evolution
between shape, orientation, and defects. Nano Letters 10 of graphene growth on Ni and Cu by carbon isotope labeling.
(October 27): 4890–6. doi:10.1021/nl102788f. http://www. Nano Letters 9 (12): 4268–72. doi:10.1021/nl902515k. http://
ncbi.nlm.nih.gov/pubmed/20979362. www.ncbi.nlm.nih.gov/pubmed/19711970.
10. Li, X., C.W. Magnuson, A. Venugopal, J. An, J.W. Suk, B. Han, 23. Yu, Q., L. A. Jauregui, W. Wu, R. Colby, J. Tian, Z. Su, H. Cao
M. Borysiak et al. 2010. Graphene films with large domain et al. 2011. Control and characterization of individual grains
size by a two-step chemical vapor deposition process. Nano and grain boundaries in graphene grown by chemical vapour
Letters 10 (11): 4328–34. doi:10.1021/nl101629g. http://pubs. deposition. Nature Materials 10 (6): 443–9. doi:10.1038/
acs.org/doi/abs/10.1021/nl101629g. nmat3010. http://www.ncbi.nlm.nih.gov/pubmed/21552269.
11. Ni, G.-X., Y. Zheng, S. Bae, H.R. Kim, A. Pachoud, Y.S. Kim, 24. Wu, W., L. a Jauregui, Z. Su, Z. Liu, J. Bao, Y.P. Chen, and Q.
C.-L. Tan et al. 2012. Quasi-periodic nanoripples in graphene Yu. 2011. Growth of single crystal graphene arrays by locally
grown by chemical vapor deposition and its impact on charge controlling nucleation on polycrystalline Cu using chemi-
transport. ACS Nano 6 (2): 1158–64. doi:10.1021/nn203775x. cal vapor deposition. Advanced Materials (Deerfield Beach,
http://www.ncbi.nlm.nih.gov/pubmed/22251076. Florida) 23 (42): 4898–903. doi:10.1002/adma.201102456.
12. Bhaviripudi, S., X. Jia, M.S. Dresselhaus, and J. Kong. 2010. http://www.ncbi.nlm.nih.gov/pubmed/21953626.
Role of kinetic factors in chemical vapor deposition synthesis 25. Wang, H., G. Wang, P. Bao, S. Yang, W. Zhu, X. Xie, and
of uniform large area graphene using copper catalyst. Nano W.-J. Zhang. 2012. Controllable synthesis of submillimeter
Letters 10 (10): 4128–33. doi:10.1021/nl102355e. http://www. single-crystal monolayer graphene domains on copper foils
ncbi.nlm.nih.gov/pubmed/20812667. by suppressing nucleation. Journal of the American Chemical
13. Wu, B., D. Geng, Y. Guo, L. Huang, Y. Xue, J. Zheng, J. Chen Society 134 (8): 3627–30. doi:10.1021/ja2105976. http://www.
et al. 2011. Equiangular hexagon-shape-controlled synthesis ncbi.nlm.nih.gov/pubmed/22324740.
of graphene on copper surface. Advanced Materials (Deerfield 26. Hwang, C., K. Yoo, S. Kim, E. Seo, H. Yu, and L.P. Biró.
Beach, Fla.) 23 (31): 3522–5. doi:10.1002/adma.201101746. 2011. Initial stage of graphene growth on Cu substrate.
http://www.ncbi.nlm.nih.gov/pubmed/21726004. The Journal of Physical Chemistry C 115 (October 10):
14. Li, X., C.W. Magnuson, A. Venugopal, R.M. Tromp, J.B. 22369–74. doi:10.1021/jp205980d. http://pubs.acs.org/doi/
Hannon, E.M. Vogel, L. Colombo, and R.S. Ruoff. 2011. abs/10.1021/jp205980d.
Large-area graphene single crystals grown by low-pressure 27. T N’Diaye, A., M. Engler, C. Busse, D. Wall, N. Buckanie,
chemical vapor deposition of methane on copper. Journal of F.-J. M. zu Heringdorf, R. van Gastel, B. Poelsema, and T.
the American Chemical Society 133 (9): 2816–19. doi:10.1021/ Michely. 2009. Growth of graphene on Ir(111). New Journal
ja109793s. http://www.ncbi.nlm.nih.gov/pubmed/21309560. of Physics 11 (2): 023006-1–023006-22. doi:10.1088/1367-
15. Vlassiouk, I., M. Regmi, P. Fulvio, S. Dai, P. Datskos, G. Eres, 2630/11/2/023006. http://stacks.iop.org/1367-2630/11/i=2/a=0
and S. Smirnov. 2011. Role of hydrogen in chemical vapor 23006?key=crossref.41fefc8129b7d1091023ff6f80606e2b.
deposition growth of large single-crystal graphene. ACS Nano 28. Geim, A.K. 2009. Graphene: Status and prospects. Science
5 (7): 6069–76. doi:10.1021/nn201978y. http://www.ncbi.nlm. 324 (5934): 1530–4. doi:10.1126/science.1158877. http://www.
nih.gov/pubmed/21707037. ncbi.nlm.nih.gov/pubmed/19541989.
29. Moore, A.W., A.R. Ubbelohde, and D.A. Young. 1964. 43. Carlsson, J., L. Ghiringhelli, and A. Fasolino. 2011. Theory
Stress recrystallization of pyrolytic graphite. Proceedings of and hierarchical calculations of the structure and energetics
the Royal Society of London. Series A, Mathematical and of [0001] tilt grain boundaries in graphene. Physical Review
Physical Sciences 280 (1381): 153–69. B 84 (16): 1–10. doi:10.1103/PhysRevB.84.165423. http://link.
30. Berger, C., Z. Song, T. Li, X. Li, A.Y. Ogbazghi, R. Feng, Z. Dai aps.org/doi/10.1103/PhysRevB.84.165423.
et al. 2004. Ultrathin epitaxial graphite: 2D electron gas prop- 44. Liu, T.-H., G. Gajewski, C.-W. Pao, and C.-C. Chang. 2011.
erties and a route toward graphene-based nanoelectronics. The Structure, energy, and structural transformations of graphene
Journal of Physical Chemistry B 108 (52): 19912–16. doi:10.1021/ grain boundaries from atomistic simulations. Carbon 49 (7):
jp040650f. http://pubs.acs.org/doi/abs/10.1021/jp040650f. 2306–17. doi:10.1016/j.carbon.2011.01.063. http://linkinghub.
31. Sprinkle, M., M. Ruan, Y. Hu, J. Hankinson, M. Rubio- elsevier.com/retrieve/pii/S0008622311000911.
Roy, B. Zhang, X. Wu, C. Berger, and W.A. de Heer. 2010. 45. Wei, Y., J. Wu, H. Yin, X. Shi, R. Yang, and M. Dresselhaus.
Scalable templated growth of graphene nanoribbons on SiC. 2012. The nature of strength enhancement and weakening by
Nature Nanotechnology 5 (October): 727–31. doi:10.1038/ pentagon–heptagon defects in graphene. Nature Materials 11
nnano.2010.192. http://www.nature.com/doifinder/10.1038/ (9): 759–63. doi:10.1038/nmat3370. http://www.ncbi.nlm.nih.
nnano.2010.192. gov/pubmed/22751178.
32. Červenka, J. and C. Flipse. 2009. Structural and electronic 46. Li, X.-F., L.-L. Wang, K.-Q. Chen, and Y. Luo. 2011. Design
properties of grain boundaries in graphite: Planes of peri- of graphene-nanoribbon heterojunctions from first principles.
odically distributed point defects. Physical Review B 79 (19): The Journal of Physical Chemistry C 115 (25): 12616–24.
195429-1–195429-5. doi:10.1103/PhysRevB.79.195429. http:// doi:10.1021/jp202188t. http://pubs.acs.org/doi/abs/10.1021/
link.aps.org/doi/10.1103/PhysRevB.79.195429. jp202188t.
33. Hass, J., W. A de Heer, and E.H. Conrad. 2008. The growth 47. Lambin, P., A. Fonseca, J.P. Vigneron, J.B. Nagy, and A.A.
and morphology of epitaxial multilayer graphene. Journal of Lucas. 1995. Structural and electronic properties of bent
Physics: Condensed Matter 20 (32): 323202. doi:10.1088/0953- carbon nanotubes. Chemical Physics Letters 245 (1): 85–9.
8984/20/32/323202. http://stacks.iop.org/0953-8984/20/i=32/ doi:10.1016/0009-2614(95)00961-3. http://linkinghub.elsevier.
a=323202?key=crossref.6e9ad3cbd67eb6a00befc8f257117a92. com/retrieve/pii/0009261495009613.
34. Magaud, L., F. Hiebel, F. Varchon, P. Mallet, and J.-Y. Veuillen. 48. Lahiri, J., Y. Lin, P. Bozkurt, I.I. Oleynik, and M. Batzill. 2010.
2009. How the SiC substrate impacts graphene’s atomic and An extended defect in graphene as a metallic wire. Nature
electronic structure. Physica Status Solidi (RRL)—Rapid Nanotechnology 5 (5): 326–9. doi:10.1038/nnano.2010.53.
Research Letters 3 (6): 172–4. doi:10.1002/pssr.200903145. http://www.ncbi.nlm.nih.gov/pubmed/20348912.
http://doi.wiley.com/10.1002/pssr.200903145. 49. TapasztÓ, L., P. Nemes-Incze, G. Dobrik, K.J. Yoo, C. Hwang,
35. Read, W. and W. Shockley. 1950. Dislocation models of and L.P. BirÓ. 2012. Mapping the electronic properties of indi-
crystal grain boundaries. Physical Review 78 (3): 275–89. vidual graphene grain boundaries. Applied Physics Letters
doi:10.1103/PhysRev.78.275. http://link.aps.org/doi/10.1103/ 100 (5): 053114. doi:10.1063/1.3681375. http://link.aip.org/
PhysRev.78.275. link/APPLAB/v100/i5/p053114/s1&Agg=doi
36. Simonis, P., C. Goffaux, P.A. Thiry, L.P. Biró, P. Lambin, and 50. Péter Nemes, I., P. Vancsó, Z. Osváth, G.I. Márk, X.
V. Meunier. 2002. STM Study of a grain boundary in graph- Jin, Y.-S. Kim, C. Hwang, P. Lambin, C. Chapelier,
ite. Surface Science 511 (1–3): 319–22. doi:10.1016/S0039- L.P. Biró. 2012. Carbon 64: 178–86. doi: 10.1016/j.car-
6028(02)01511-X. http://linkinghub.elsevier.com/retrieve/pii/ bon.2013.07.050, http://linkinghub.elsevier.com/retrieve/pii/
S003960280201511X. S0008622313006842.
37. Lauginie, P. and J. Conard. 1997. New growing modes for 51. Malola, S., H. Häkkinen, and P. Koskinen. 2010. Structural,
carbon: Modelization of lattice defects, structure of tubules chemical, and dynamical trends in graphene grain boundaries.
and onions. Journal of Physics and Chemistry of Solids Physical Review B 81 (16): 165447-1–165447-6. doi:10.1103/
58 (11): 1949–63. doi:10.1016/S0022-3697(97)00075-9. http:// PhysRevB.81.165447. http://link.aps.org/doi/10.1103/PhysRevB.
linkinghub.elsevier.com/retrieve/pii/S0022369797000759. 81.165447.
38. Shockley, W. and W. Read. 1949. Quantitative predictions 52. Mesaros, A., S. Papanikolaou, C. Flipse, D. Sadri, and J.
from dislocation models of crystal grain boundaries. Physical Zaanen. 2010. Electronic states of graphene grain boundaries.
Review 75 (4): 692. doi:10.1103/PhysRev.75.692. http://link. Physical Review B 82 (20): 205119-1–205119-8. doi:10.1103/
aps.org/doi/10.1103/PhysRev.75.692. PhysRevB.82.205119. http://link.aps.org/doi/10.1103/PhysRevB.
39. Yazyev, O.V. and S.G. Louie. 2010. Topological defects in gra- 82.205119.
phene: Dislocations and grain boundaries. Physical Review B 53. Kotakoski, J. and J. Meyer. 2012. Mechanical properties of
81 (19): 195420-1–195420-7. doi:10.1103/PhysRevB.81.195420. polycrystalline graphene based on a realistic atomistic model.
http://link.aps.org/doi/10.1103/PhysRevB.81.195420. Materials science; Computational physics. Physical Review B
40. Liu, Y. and B.I. Yakobson. 2010. Cones, pringles, and grain 85 (19): 195447. doi:10.1103/PhysRevB.85.195447. http://arxiv.
boundary landscapes in graphene topology. Nano Letters 10 org/abs/1203.4196.
(May 19): 2178–83. doi:10.1021/nl100988r. http://pubs.acs. 54. Vancsó, P., G.I. Márk, P. Lambin, A. Mayer, Y.-S. Kim, C.
org/doi/abs/10.1021/nl100988r. Hwang, and L.P. Biró. 2013. Electronic transport through
41. Da Silva Araújo, J., and R.W. Nunes. 2010. Complex evolution ordered and disordered graphene grain boundaries. Carbon
of the electronic structure from polycrystalline to monocrys- 64: 101–10. doi:10.1016/j.carbon.2013.07.041. http://www.
talline graphene: Generation of a new Dirac point. Physical sciencedirect.com/science/article/pii/S0008622313006751
Review B 81 (7): 073408. doi:10.1103/PhysRevB.81.073408. 55. Huang, P.Y., C.S. Ruiz-Vargas, A.M. van der Zande, W.S.
http://link.aps.org/doi/10.1103/PhysRevB.81.073408. Whitney, M.P. Levendorf, J.W. Kevek, S. Garg et al. 2011.
42. Grantab, R., V.B. Shenoy, and R.S. Ruoff. 2010. Anomalous Grains and grain boundaries in single-layer graphene atomic
strength characteristics of tilt grain boundaries in graphene. patchwork quilts. Nature 469 (7330): 389–92. doi:10.1038/
Science (New York, NY) 330 (6006): 946–8. doi:10.1126/sci- nature09718. http://www.nature.com/doifinder/10.1038/nature
ence.1196893. http://www.ncbi.nlm.nih.gov/pubmed/21071664. 09718.
56. Kurasch, S., J. Kotakoski, O.J. Lehtinen, V. Skakalova, 68. Decker, R., Y. Wang, V.W. Brar, W. Regan, H. Tsai, Q. Wu, W.
J.H. Smet, C. Krill, A.V. Krasheninnikov, and U. Kaiser. Gannett, A. Zettl, and M.F. Crommie. 2011. Local electronic
2012. Atom-by-atom observation of grain boundary properties of graphene on a BN substrate via scanning tun-
m igration in g raphene. Nano Letters 12 (6): 3168–73. neling microscopy. Nano Letters 11 (6): 2291–5. doi:10.1021/
doi:10.1021/nl301141g. http://www.ncbi.nlm.nih.gov/pubmed/ nl2005115. http://www.ncbi.nlm.nih.gov/pubmed/21553853.
22554303. 69. Solís-Fernández, P., J.I. Paredes, S. Villar-Rodil, L. Guardia,
57. Kotakoski, J., D. Santos-Cottin, and A.V. Krasheninnikov. M.J. Fernández-Merino, G. Dobrik, L.P. BirÓ, A. Martínez-
2012. Stability of graphene edges under electron beam: Alonso, and J.M.D. TascÓn. 2011. Global and local oxidation
Equilibrium energetics versus dynamic effects. ACS Nano 6 behavior of reduced graphene oxide. The Journal of Physical
(1): 671–6. doi:10.1021/nn204148h. http://www.ncbi.nlm.nih. Chemistry C 115 (16): 7956–66. doi:10.1021/jp2003744. http://
gov/pubmed/22188561. pubs.acs.org/doi/abs/10.1021/jp2003744.
58. Kim, K., Z. Lee, W. Regan, C. Kisielowski, M.F. Crommie, 70. Yamamoto, M., T.L. Einstein, M.S. Fuhrer, and W.G. Cullen.
and A. Zettl. 2011. Grain boundary mapping in polycrystalline 2012. Charge inhomogeneity determines oxidative reactivity of
graphene. ACS Nano 5 (3): 2142–6. doi:10.1021/nn1033423. graphene on substrates. ACS Nano 6 (9): 8335–41. doi:10.1021/
http://www.ncbi.nlm.nih.gov/pubmed/21280616. nn303082a. http://www.ncbi.nlm.nih.gov/pubmed/22917254.
59. Kotakoski, J., A. Krasheninnikov, U. Kaiser, and J. Meyer. 71. Nemes-Incze, P., K.J. Yoo, L. TapasztÓ, G. Dobrik, J. Lábár,
2011. From point defects in graphene to two-dimensional Z.E. Horváth, C. Hwang, and L.P. BirÓ. 2011. Revealing the
amorphous carbon. Physical Review Letters 106 (10): 105505- grain structure of graphene grown by chemical vapor depo-
1–105505-4. doi:10.1103/PhysRevLett.106.105505. http://link. sition. Applied Physics Letters 99 (2): 023104-1–023104-3.
aps.org/doi/10.1103/PhysRevLett.106.105505. doi:10.1063/1.3610941. http://link.aip.org/link/APPLAB/v99/
60. Wang, Z., Y.G. Zhou, J. Bang, M.P. Prange, S.B. Zhang, and i2/p023104/s1&Agg=doi.
F. Gao. 2012. Modification of defect structures in graphene 72. Koepke, J.C., J.D. Wood, D. Estrada, Z.-Y. Ong, K.T. He, E.
by electron irradiation: Ab initio molecular dynamics simula- Pop, and J.W. Lyding. 2013. Atomic-scale evidence for poten-
tions. The Journal of Physical Chemistry C 116 (30): 16070–9. tial barriers and strong carrier scattering at graphene grain
doi:10.1021/jp303905u. http://pubs.acs.org/doi/abs/10.1021/jp boundaries: A scanning tunneling microscopy study. ACS
303905u. Nano 7 (1): 75–86. doi:10.1021/nn302064p. http://www.ncbi.
61. Santana, A., A.M. Popov, and E. Bichoutskaia. 2013. Stability nlm.nih.gov/pubmed/23237026.
and dynamics of vacancy in graphene flakes: Edge effects. 73. Jauregui, L.A., H. Cao, W. Wu, Q. Yu, and Y.P. Chen. 2011.
Chemical Physics Letters 557 (February): 80–7. doi:10.1016/j. Electronic properties of grains and grain boundaries in gra-
cplett.2012.11.077. http://linkinghub.elsevier.com/retrieve/pii/ phene grown by chemical vapor deposition. Solid State
S0009261412013838. Communications 151 (16): 1100–4. doi:10.1016/j.ssc. 2011.
62. Yakobson, B.I. and F. Ding. 2011. Observational geology 05.023. http://linkinghub.elsevier.com/retrieve/pii/S003810981
of graphene, at the nanoscale. ACS Nano 5 (3): 1569–74. 1002596.
doi:10.1021/nn200832y. http://www.ncbi.nlm.nih.gov/pubmed/ 74. Červenka, J., M.I. Katsnelson, and C.F.J. Flipse. 2009.
21417493. Room-temperature ferromagnetism in graphite driven by
63. Lee, G.-H., R.C. Cooper, S.J. An, S. Lee, A. van der Zande, N. two-dimensional networks of point defects. Nature Physics
Petrone, A.G. Hammerberg et al. 2013. High-strength chem- 5 (11): 840–4. doi:10.1038/nphys1399. http://www.nature.com/
ical-vapor-deposited graphene and grain boundaries. Science doifinder/10.1038/nphys1399.
340 (6136): 1073–6. doi:10.1126/science.1235126. http://www. 75. Feng, L., X. Lin, L. Meng, J.-C. Nie, J. Ni, and L. He. 2012.
sciencemag.org/cgi/doi/10.1126/science.1235126. Flat bands near Fermi level of topological line defects
64. Petrone, N., C.R. Dean, I. Meric, A.M. van der Zande, P.Y. on graphite. Applied Physics Letters 101 (11): 113113.
Huang, L. Wang, D. Muller, K.L. Shepard, and J. Hone. 2012. doi:10.1063/1.4752441. http://link.aip.org/link/APPLAB/v101/
Chemical vapor deposition-derived graphene with electrical i11/p113113/s1&Agg=doi.
performance of exfoliated graphene. Nano Letters 12 (6): 76. Gao, L., J.R. Guest, and N.P. Guisinger. 2010. Epitaxial gra-
2751–6. doi:10.1021/nl204481s. http://www.ncbi.nlm.nih.gov/ phene on Cu(111). Nano Letters 10 (9): 3512–16. doi:10.1021/
pubmed/22582828. nl1016706. http://www.ncbi.nlm.nih.gov/pubmed/20677798.
65. Ogawa, Y., B. Hu, C.M. Orofeo, M. Tsuji, K. Ikeda, S. Mizuno, 77. Duong, D.L., G.H. Han, S.M. Lee, F. Gunes, E.S. Kim, S.T.
H. Hibino, and H. Ago. 2012. Domain structure and boundary Kim, H. Kim et al. 2012. Probing graphene grain bound-
in single-layer graphene grown on Cu(111) and Cu(100) films. aries with optical microscopy. Nature 490 (7419): 235–9.
The Journal of Physical Chemistry Letters 3 (2): 219–26. doi:10.1038/nature11562. http://www.nature.com/doifinder/
doi:10.1021/jz2015555. http://pubs.acs.org/doi/abs/10.1021/jz 10.1038/nature11562.
2015555. 78. Jia, C., J. Jiang, L. Gan, and X. Guo. 2012. Direct optical
66. Rasool, H.I., E.B. Song, M. Mecklenburg, B.C. Regan, Kang characterization of graphene growth and domains on growth
L. Wang, B.H. Weiller, and J.K. Gimzewski. 2011. Atomic- substrates. Scientific Reports 2 (January): 707. doi:10.1038/
scale characterization of graphene grown on copper (100) sin- srep00707. http://www.pubmedcentral.nih.gov/articlerender.
gle crystals. Journal of the American Chemical Society 133 fcgi?artid=3463814&tool=pmcentrez&rendertype=abstract.
(32): 12536–43. doi:10.1021/ja200245p. http://www.ncbi.nlm. 79. Kong, X., H. Ji, R.D. Piner, H. Li, C.W. Magnuson, C. Tan,
nih.gov/pubmed/21732685. A. Ismach, H. Chou, and R.S. Ruoff. 2013. Applied Physics
67. Kertész, K., A.A. Koós, A.T. Murdock, Z. Vértesy, P. Nemes- Letters 43: 043119. doi: 10.1063/1.4816752. http://dx.doi.
Incze, P.J. Szabó, Z.E. Horváth et al. 2012. Polarized light org/10.1063/1.4816752.
microscopy of chemical-vapor-deposition-grown graphene on 80. Márk, G.I., P. Vancsó, P. Lambin, C. Hwang, and L.P. Biró.
copper polarized light microscopy of chemical-vapor-deposi- 2012. Forming electronic waveguides from graphene grain
tion-grown graphene on copper. Applied Physics Letters 100: boundaries. Journal of Nanophotonics 6: 061718-1–061718-6.
213103-1–213103-5. doi:10.1063/1.4719205. doi:10.1117/1.
81. Kumar, S.B. and J. Guo. 2012. Strain-induced conductance 94. Bagri, A., S.-P. Kim, R.S. Ruoff, and V.B. Shenoy. 2011.
modulation in graphene grain boundary. Nano Letters 12 (3): Thermal transport across twin grain boundaries in polycrystal-
1362–6. doi:10.1021/nl203968j. http://www.ncbi.nlm.nih.gov/ line graphene from nonequilibrium molecular dynamics sim-
pubmed/22324382. ulations. Nano Letters 11 (August 24): 3917–21. doi: 10.1021/
82. Yazyev, O.V. and S.G. Louie. 2010. Electronic transport in nl202118d. http://pubs.acs.org/doi/abs/10.1021/nl202118d
polycrystalline graphene. Nature Materials 9 (10): 806–9. 95. Serov, A.Y., Z.-Y. Ong, and E. Pop. 2013. Effect of grain
doi:10.1038/nmat2830. http://www.ncbi.nlm.nih.gov/pubmed/ boundaries on thermal transport in graphene. Applied Physics
20729847. Letters 102 (January 23): 033104.1–5 doi: 10.1063/1.4776667.
83. Gunlycke, D. and C.T. White. 2011. Graphene valley filter http://apl.aip.org/resource/1/applab/v102/i3/p033104_s1
using a line defect. Physical Review Letters 106 (13): 136806. 96. Caoa, H.-Y., H. Xianga, X.-G. Gonga. 2012. Unexpected large
doi:10.1103/PhysRevLett.106.136806. http://link.aps.org/doi/ tthermal rectification in asymmetric grain boundary of gra-
10.1103/PhysRevLett.106.136806. phene. Solid State Communications 152 (19): 1807–10. doi:
84. Zhang, H., G. Lee, C. Gong, L. Colombo, and K. Cho. 2014. 10.1016/j.ssc.2012.07.013. http://www.sciencedirect.com/sci-
Grain boundary effect on electrical transport properties of ence/a rticle/pii/S0038109812004310
graphene. Journal of Physical Chemistry C 118 (5): 2338–43. 97. Lee, C., X. Wei, J.W. Kysar, and J. Hone. 2008. Measurement
doi: 10.1021/jp411464w. http://pubs.acs.org/doi/abs/10.1021/ of the elastic properties and intrinsic strength of mono-
jp411464w layer graphene. Science (New York, NY) 321 (5887): 385–8.
85. Tsen, A.W., L. Brown, M.P. Levendorf, F. Ghahari, P.Y. Huang, doi:10.1126/science.1157996. http://www.ncbi.nlm.nih.gov/
R.W. Havener, C.S. Ruiz-Vargas, D. A. Muller, P. Kim, and J. pubmed/18635798.
Park. 2012. Tailoring electrical transport across grain bound- 98. Hana, J., S. Ryua, D. Sohnb, and S. Ima. 2014. Mechanical
aries in polycrystalline graphene. Science 336 (6085): 1143–6. strength characteristics of asymmetric tilt grain boundar-
doi:10.1126/science.1218948. http://www.sciencemag.org/cgi/ ies in graphene. Carbon 68 (March): 250–7. doi: 10.1016/j.
doi/10.1126/science.1218948. carbon.2013.10.085. http://www.sciencedirect.com/science/
86. Balandin, A.A. 2011. Thermal properties of graphene and article/pii/S0008622313010476
nanostructured carbon materials. Nature Materials 10 (July 99. Brito, W.H., R. Kagimura, and R.H. Miwa. 2011. Hydrogenated
22): 569–81. doi:10.1038/nmat3064. http://www.nature.com/ grain boundaries in graphene. Applied Physics Letters 98 (21):
nmat/journal/v10/n8/full/nmat3064.html 213107-1–213107-3. doi:10.1063/1.3592578. http://link.aip.
87. Goli, P., S. Legedza, A. Dhar, R. Salgado, J. Renteria, and A.A. org/link/APPLAB/v98/i21/p213107/s1&Agg=doi.
Balandin. 2014. Graphene-enhanced hybrid phase change 100. Brito, W., R. Kagimura, and R. Miwa. 2012. B and N doping in
materials for thermal management of Li-ion batteries. Journal graphene ruled by grain boundary defects. Physical Review B
of Power Sources 248 (February 15): 37–43. doi: 10.1016/j. 85 (3): 035404-1–035404-6. doi:10.1103/PhysRevB.85.035404.
jpowsour.2013.08.135. http://www.sciencedirect.com/science/ http://link.aps.org/doi/10.1103/PhysRevB.85.035404.
article/pii/S0378775313014857 101. Chan, J., A. Venugopal, A. Pirkle, S. McDonnell, D. Hinojos,
88. Berber, S., Y.-K. Kwon, and D. Tománek. 2000. Unusually high C.W. Magnuson, R.S. Ruoff, L. Colombo, R.M. Wallace, and
thermal conductivity of carbon nanotubes. Physical Review E.M. Vogel. 2012. Reducing extrinsic performance-limiting
Letters 84 (May 15): 4613–6. doi: 10.1103/PhysRevLett.84.4613. factors in graphene grown by chemical vapor deposition. ACS
http://link.aps.org/doi/10.1103/PhysRevLett.84.4613 Nano 6 (4): 3224–9. doi:10.1021/nn300107f. http://www.ncbi.
89. Nihira, T. and T. Iwata. 1975. Thermal resistivity changes nlm.nih.gov/pubmed/22390298.
in electron-irradiated pyrolytic graphite. Japanese Journal 102. Yang, Y. and R. Murali. 2011. Binding mechanisms of molec-
of Applied Physics 14 (August 5): 1099–104. doi:10.1143/ ular oxygen and moisture to graphene. Applied Physics Letters
JJAP.14.1099. http://jjap.jsap.jp/link?JJAP/14/1099/ 98 (9): 093116-1–093116-3. doi:10.1063/1.3562317. http://link.
90. Ghosh, S., I. Calizo, D. Teweldebrhan, E.P. Pokatilov, D.L. aip.org/link/APPLAB/v98/i9/p093116/s1&Agg=doi.
Nika, A.A. Balandin, W. Bao, F. Miao, and C.N. Lau. 2008. 103. Politano, A., A.R. Marino, and G. Chiarello. 2012. Effects of a
Extremely high thermal conductivity of graphene: Prospects humid environment on the sheet plasmon resonance in epitax-
for thermal management applications in nanoelectronic ial graphene. Physical Review B 86 (8): 085420. doi:10.1103/
circuits. Applied Physics Letters 92 (Apr 16): 151911.1–3. PhysRevB.86.085420. http://link.aps.org/doi/10.1103/PhysRevB.
doi:10.1063/1.2907977. http://apl.aip.org/resource/1/applab/v92/ 86.085420.
i15/p151911_s1 104. Feng, X., S. Maier, and M. Salmeron. 2012. Water splits epi-
91. Seol, J.H., I. Jo, A.L. Moore, L. Lindsay, Z.H. Aitken, M.T. taxial graphene and intercalates. Journal of the American
Pettes, X. Li et al. 2010. Two-dimensional phonon trans- Chemical Society 134 (12): 5662–8. doi:10.1021/ja3003809.
port in supported graphene. Science 328 (April 9): 213–16. http://www.ncbi.nlm.nih.gov/pubmed/22400993.
doi: 10.1126/science.1184014. http://www.sciencemag.org/ 105. Zhang, Y., Z. Li, P. Kim, L. Zhang, and C. Zhou. 2012.
content/328/5975/213 Anisotropic hydrogen etching of chemical vapor deposited
92. Cai, W., A.L. Moore, Y. Zhu, X. Li, S. Chen, L. Shi, and R.S. graphene. ACS Nano 6 (1): 126–32. doi:10.1021/nn202996r.
Ruoff. 2010. Thermal transport in suspended and supported http://www.ncbi.nlm.nih.gov/pubmed/22010852.
monolayer graphene grown by chemical vapor deposition. 106. Salehi-Khojin, A., D. Estrada, K.Y. Lin, M.-H. Bae, F. Xiong,
Nano Letters 10 (April 20): 1645–51. doi: 10.1021/nl9041966. E. Pop, and R.I. Masel. 2012. Polycrystalline graphene rib-
http://pubs.acs.org/doi/abs/10.1021/nl9041966 bons as chemiresistors. Advanced Materials (Deerfield Beach,
93. Vlassiouk, I., S. Smirnov, I. Ivanov, P.F. Fulvio, S. Dai, H. Florida) 24 (1): 53–7, 52. doi:10.1002/adma.201102663. http://
Meyer, M. Chi, D. Hensley, P. Datskos, and N.V. Lavrik. 2011. www.ncbi.nlm.nih.gov/pubmed/22113971.
Electrical and thermal conductivity of low temperature CVD 107. Banhart, F., J. Kotakoski, and A.V. Krasheninnikov. 2011.
graphene: The effect of disorder. Nanotechnology 22 (May Structural defects in graphene. ACS Nano 5 (1): 26–41.
25): 275716.1–9. doi: 10.1088/0957-4484/22/27/275716. http:// doi:10.1021/nn102598m. http://www.ncbi.nlm.nih.gov/pubmed/
iopscience.iop.org/0957 21090760.
8 Graphene-Based Biological
and Chemical Sensors
Shirui Guo, Wei Wang, Cengiz S. Ozkan, and Mihrimah Ozkan
CONTENTS
Abstract...................................................................................................................................................................................... 127
8.1 Introduction...................................................................................................................................................................... 127
8.2 Graphene-Based Electronic Sensors................................................................................................................................. 128
8.2.1 Synthesis of Graphene.......................................................................................................................................... 128
8.2.1.1 Chemical Vapor Deposition................................................................................................................... 128
8.2.1.2 Chemical Exfoliation............................................................................................................................. 130
8.2.1.3 Other Methods....................................................................................................................................... 132
8.2.2 Electronic Structure of Graphene......................................................................................................................... 132
8.2.3 GFET Sensor........................................................................................................................................................ 133
8.2.4 Chemical Functionalization.................................................................................................................................. 133
8.2.5 GFET Gas Sensor................................................................................................................................................. 139
8.2.6 GFET Biosensor................................................................................................................................................... 140
8.2.6.1 DNA Sensor........................................................................................................................................... 140
8.2.6.2 pH Sensor................................................................................................................................................141
8.2.6.3 Graphene Immunosensor........................................................................................................................141
8.2.7 Graphene Electrochemical Glucose Sensor.......................................................................................................... 142
8.3 Concluding Remarks.........................................................................................................................................................143
References...................................................................................................................................................................................143
ABSTRACT or interpreted by an observer or by an electronic instrument.

The input of the physical quantity can be anything available
Graphene, a single atomic sheet and two-dimensional car- in our daily life, such as mass, temperature, luminous inten-
bonaceous material, has attracted great attention for sensor sity, electrical field, force, pressure, etc. The ultimate goal of
applications due to its intrinsic electrical and physical prop- sensing is to achieve a level of sensitivity to resolve individual
erties. Many studies have shown the high sensitivity of gra- quanta of entity.1 For biological and chemical sensor, which is
phene-based sensors resulting from the intrinsic properties of a type of commonly used sensor in our daily life, the entity
graphene, in applications such as pH, DNA, and vapor sen- can be atoms or molecules.
sors, among others. In this chapter, we focus on the different The first use of graphite can be traced back to 4th mil-
types of graphene devices for biomolecule and gas sensing, lennium B.C. when Mariţa culture applied it as a ceramic
which are relevant to biological and chemical detection. We paint to decorate pottery in southeastern Europe.2 The band
will discuss the different working principles for correspond- theory of graphite has been studied from 1947 when P. R.
ing sensors. Functionalization of graphene with different Wallace used tight binding approximation method, assuming
chemicals via various methods is critical to fabricate sensors conduction only happens in one layer which is the graphene
with high selectivity. The sensitivity dependence of graphene layer.3 Graphene, the two-dimensional (2D) monolayer of car-
charge transport on electron-withdrawing or electron-donat- bon atoms tightly packed into a honeycomb lattice, is a basic
ing chemical groups leads to their applications as ultrasensi- building block for graphite and other graphitic materials of
tive chemical sensors and biosensors. We summarized typical all other dimensionalities.4 While Landau and Peierls argued
examples of functionalization with p or n doping effect and that strictly 2D crystals were thermodynamically unstable
hopefully this will motivate readers to digest and explore and could not exist,5 2D graphene oxide (GO) sheets actu-
more functional molecules for sensing applications. ally have been synthesized by Benjamin C. Brodie in 18596
and further modified by Hummers and Offeman in 1957.7
However, until 2004, the first experimental discovery of gra-
8.1 INTRODUCTION
phene was reported by A. K. Geim and K. S. Novoselov.8
Sensor by definition is a converter that measures the input of a Many remarkable properties of graphene make it a ris-
physical quantity and coverts it into a signal, which can be read ing star for researches and applications in various areas.
127
Ballistic transport of electrons along the atomically thin

layer, along with mobilities exceeding 15,000 cm 2 V−1 s −1 Mechanical exfoliation
(research,
and an ambipolar field effect make graphene a particularly prototyping)
CVD
good candidate for the next round of semiconductor devices.9
(coating, bio, transparent
Apart from the extremely high carrier mobility, graphene
conductive layers,
absorbs only 2% of incoming light, independent of wave- electronics,
lengths across the visible spectrum.10 These properties make photonics)
graphene an excellent candidate for photovoltaic devices.11
Quality
Transparent conducting films have been made from gra-
phene for flexible displays,12 touch screens, and solar cells.13
High strength and flexibility of graphene enable bendable SiC
devices which are useful in electromechanical systems.14 (electronics, Molecular
The large and planar surface area of a single sheet and low RF transistors) assembly
electrical noise of graphene field effect transistor (GFET) (nanoelectronics)
allow the application in optoelectronic devices and sensors Liquid-phase exfoliation
possible.15 A cutoff frequency of 100 GHz for a gate length (coating, composites,
inks, energy storage,
of 240-nm GFETs has been achieved, which is higher bio, transparent conductive layers)
than state-of-the-art silicon transistors of the same gate
Price (for mass production)
length enabling application for digital data transmission.16
The vast possibilities of graphene and the rapid advance-
ments in research are propelling this new material toward FIGURE 8.1 There are several methods of mass production of
novel achievements that are already beginning to influence graphene, which allow a wide choice in terms of size, quality, and
price for any particular application. (Adapted from K. Novoselov et
industry.17
al. Nature 2012, 490, 192–200.)
Gas sensors and label-free biosensors with high minia-
turization and integration have attracted intense research
interest since they could potentially make advanced low-
cost molecular diagnostics routinely available. Given the
high sensitivity nature of graphene and various function- TABLE 8.1
alizations with ease, graphene-based sensors have been Graphene Synthesis and Preparation
used for various sensing applications. Many factors have Synthesis
demonstrated the lab-on-chip realization with graphene, Routes Methods Name Approach
such as the wafer scale synthesis, high quality with single
Bottom up Chemical vapor Using transition metal as catalyst to
layer or double layer control, controllable funtionaliza- deposition decompose carbon source. Typical
tion, and thousands of circuits’ production in one batch of catalysts: Ru,19 Ir,20 Co,21 Re,22 Pt,21a,23
photolithography. Pd,21a Cu,24 Fe21b,25, and Ni.21b
Epitaxial Decompose SiC at high temperature, for
graphene example, 1000–1500°C.26
8.2 GRAPHENE-BASED ELECTRONIC SENSORS Top down Chemical Using strong oxidant to exfoliate
exfoliation graphite7,27,26 or using small molecules
8.2.1 Synthesis of Graphene to exfoliate graphite: ethanol,
For mass production of electronic sensor devices, large-scale tetrahydrofuran,28 N-methyl-pyrrol-
syntheses of high-quality graphene are necessary. Currently, idone,29 chloroform, isopropanol,
there are several major approaches to achieve low-cost, large- acetone, dimethylformamide.28
Micromechanical Using Scotch tape to exfoliate graphite.8
scale production with high quality.
exfoliation
As shown in Figure 8.1, the roadmap of graphene devel-
opment, large-area epitaxial graphene (chemical vapor depo-
sition [CVD], and thermal decomposition of silicon carbide
[SiC]) which has comparable quality as mechanically exfoli- nickel and copper foils and reagent grade platinum foils.30
ated graphene is perfect for mass production of high-quality The diffraction patterns of the obtained carbon film indicate
electronic devices with lower price. Chemically exfoliated the presence of crystalline carbon which are graphene lay-
graphene films with very low mass production price and ers. In 1969, Robertson reported the synthesis of graphite/
abundant chemical groups are perfect candidates for electro- few-layer graphene by decomposing methane over nickel.31
chemical electrodes (Table 8.1). Different transition metals require different decomposing
temperatures. Pd, Ru, Ir, Ni, and Cu have been utilized for
8.2.1.1 Chemical Vapor Deposition growing graphite materials. However, due to the limita-
In 1961, few-layer graphene has been grown by pyrolyze tion of characterization techniques, graphene layers were
pure carbon suboxide (C3O2) at 713°C with electrolytic not thoroughly characterized until 2004, when the first
Graphene-Based Biological and Chemical Sensors 129
experimental discovery of graphene was reported by A. K. determine the deposition mechanism, which ultimately defines
Geim and K. S. Novoselov.8 the morphology and thickness of the graphene films. Among
In 2006, Somani et al. utilized nickel sheets to grow few- all the transition metals, copper foil is of particular interest for
layer graphene, and the number of layers was estimated to be growing single-layer graphene. The phase diagram of Ni and
35 layers with high-resolution transmission electronic micros- C reveals that the solubility of carbon in nickel at high tem-
copy.19 This approach opened venues for graphene synthesis, perature (above 800°C) forms a solid solution and lowering
although several issues such as thickness, control, and unifor- the temperature decreases the solubility, allowing carbon to
mity had to be solved. Then nanometer-thick (1–2 nm) gra- diffuse out of Ni (Figure 8.2a).
phene films grown on nickel foil (hundred micrometer) were It is well known from metallurgical studies that the for-
achieved by several groups through the CVD approach.20 The mation of metastable Ni3C phase promotes the precipitation
diffusion and segregation of carbon impurities from the bulk of carbon out of Ni. Carbon preferentially precipitates out at
to the surface during the annealing and cooling stages induces the grain boundaries of polycrystalline Ni substrates so that
the formation of graphite on nickel. At the same time, gra- the thickness of the graphite at the grain boundaries is sub-
phene has been grown with various catalysts including Ru,21 stantially larger than within the grains. Thus, the number of
Ir,22 Co,23 Re,24 Pt,23a,25 Pd,23a Cu,32 and Fe.23b,33 Revisiting the graphene layers shows significant difference along the surface
formation mechanism of carbon layer on different transition of Ni. By additional etching of the as-grown graphene films
metals provides chances to produce high-quality graphene. with H2 at 1000°C, single-layer graphene was obtained on Ni
The carbon solubility in the metal and the growth conditions foil.23b Co and Fe show similar catalytic behaviors as can be
(a) Atomic percent carbon (b) Atomic percent carbon

0102030 40 50 60 70 80 90 100 0 2 4 6 8 10 12 14 16 18
4000 1800
3500 1600 L
L 1495 L + C (graphite)
1400
3000 1320 C
(αCo)
Temperature (°C)
Temperature (°C)
1200 0.9 2.6

L + graphite (αCo) + graphite
2500 1121°C
(C, graphite)
1000
2000
800
1500 1326.5 C 600

1455
0.6 422 C
1000 400
(Ni) (Ni) + graphite (εCo)
500 200
0 10 20 30 40 50 60 70 80 90 100 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Ni Weight percent carbon C Co Weight percent carbon C
Atomic percent carbon Atomic percent carbon

(c) (d)
0 0.04 0.08 0.12 0.16 0.2 0.24
0 5 10 15 20 2400
1536 2200
L 2000 L 1150
1392 L + (C)
L + C(graphite) 1100 C
γFe + L 1800 1084.77
Temperature (°C)
Temperature (°C)
2.09 1050 L + (Cu)

(γFe) 1154 C 4.3 1600
L + (C) 950 (Cu) (Cu) + (C)
911 γFe + Fe3C (or graphite) 1400
850
800 740 C 1200 0
1100 C 4 8 12 16
0.6
1084.87 C content (at ppm)
1000 (Cu) 0.0076
αFe αFe + Fe3C
(Cu) + (C)
800
450
0 1 2 3 4 5 6 600
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.05
Fe Weight percent carbon C Cu Weight percent carbon C
FIGURE 8.2 Binary phase diagrams of transition metals and carbon: (a) Ni–C; (b) Co–C; (c) Fe–C; (d) Cu–C. The low carbon solubility
in Cu of 0.008 wt.% at 1084°C as reported is highlighted in the inset of panel (d) for the temperature and composition of interest for graphene
growth. (Adapted from C. Mattevi, H. Kim, M. Chhowalla, Journal of Materials Chemistry 2011, 21, 3324−3334.)
(a) (b)
8 inch
After
heating
Before
heating
39 inch
FIGURE 8.3 Photographs of roll-based production of graphene films. (a) Copper foil wrapping around a 7.5-inch quartz tube to be inserted
into an 8-inch quartz reactor. The lower image shows the stage in which the copper foil reacts with CH4 and H2 gases at high temperatures.
(b) Roll-to-roll transfer of graphene films from a thermal release tape to a PET film at 120°C.
concluded from the phase diagrams (Figure 8.2b and c). In con- Mobility of CVD graphene varies from 300 to
trast, copper does not form any carbide phases (Figure 8.2d), 5000 cm2 V−1 s−1, which is much lower than pristine graphene
which make a good candidate for perfect graphene single- obtained by micromechanical exfoliation of highly oriented
layer growth.32 From atomic orbital point of view, we can also pyrolytic graphite. The major hurdle is graphene domain
arrive at the same conclusion. Carbon forms on metal surface boundaries where carbon–carbon could not form perfect sp2–
through forming intermediates or solid solutions with metal. sp2 bond resulting in the degradation of electronic properties.
The bond strength between metal and carbon highly depends Large-area single-graphene crystals are feasible for high-
on the d orbital structure of metal atoms. Fe has asymmetrical quality device fabrication and they have been grown with
distribution of electrons in the d-shell {[Ar]3d64s2}, leading to copper as the catalyst. Qinkai Yu et al. devised an approach
mutual repulsion which may explain its higher affinity toward using prepatterned growth graphene seeds to control gra-
carbon with respect to Co {[Ar]3d74s2}, Ni {[Ar]3d84s2}, and phene nucleation, opening a route toward scalable fabrication
Cu where the 3d shell is progressively filled, suggesting less of single-crystal graphene devices without the grain boundar-
reactive configurations. Copper has a filled 3d-electron shell ies (Figure 8.4a).35
{[Ar]3d104s1}, the most stable configuration (along with the Recently, Yan et al. reported 4.5 mm2 of one single gra-
half filling 3d5) because the electron distribution is symmetri- phene crystal grown on copper foil. The study shows that
cal which minimizes reciprocal repulsions. As a result, Cu long-time annealing and electrochemical polishing improves
can form only weak bonds with carbon via charge transfer the quality of the copper foil. Optimized methane/hydrogen
from the p electrons in the sp2-hybridized carbon to the empty ratio introduces less active sites for the single-graphene crys-
4s states of copper.34 The 3d7 and 3d8 orbitals of Co and Ni are tal leading to large-area growth of the single-graphene crystal
between the most unstable electronic configuration (Fe) and (Figures 8.4b and 8.5).27
the most stable one (Cu).
With the privileges of copper foil, large-scale production 8.2.1.2 Chemical Exfoliation
of high-quality graphene has been achieved. Sukang Bae Graphene produced from chemical exfoliation of graphite
et al. reported roll-to-roll synthesis of 30-inch graphene films is also attractive compared with CVD graphene. With such
for transparent electrodes (Figure 8.3).12 an approach, large quantity and facile fabrication without
(a) (b)
2.3 mm
10 µm 1 mm
FIGURE 8.4 (a) SEM image of a typical graphene grain array grown from an array of seed crystals, with a relatively short growth time
(15 min).36 (b) Synthesis of large-size (~2.3 mm) single-crystal graphene monolayer domains using the cold-plasma chemical vapor deposi-
tion (CP-CVD) system. Typical SEM image of as-produced graphene domains on Cu. (Adapted from D. R. Dreyer et al. Chemical Society
Reviews 2010, 39, 228–240.)
MFC
Gas lines MFC Cu foil Exhaust
MFC
CVD furnace
Graphene forms on
Methane
Cu foil during cooling
H2
Ar
Graphene film
Transferred on Etching of Cu foil

desired substrate in FeCl3 aq. sol.
FIGURE 8.5 Schematic illustration of CVD synthesis of single-layer graphene with copper foil. Transfer of single-layer graphene consists
of coated PMMA layer, etched bottom catalyst layer, rinsing graphene layer, and transfer to desired substrates.
high-temperature processing could be realized in any chemis- presence of an oxidizing agent. In 1859, the first GO was
try laboratory. The obtained graphene sheets could be chemi- obtained by Brodie who utilized potassium chlorate and
cally functionalized, dispersed in polymer matrices, and nitric acid to treat graphite repeatedly.6 In 1959, Hummers
deoxygenated to yield novel composites, making it a good and Offeman demonstrated a less hazardous and more
candidate for various applications.38 Moreover, graphene film efficient method for graphite oxidation, which involves a
or composite could be easily prepared according to Figure 8.6. mixture of sodium nitrate, potassium permanganate, and
Synthesis of graphite oxide (GO) can be achieved by concentrated sulfuric acid.7 The method was later modified
placing graphite in one or more concentrated acids in the and utilized to produce high-quality GO and reduced gra-
phene oxide (rGO) sheets.39
Single-layer GO sheet produced from approaches men-
Graphite tioned above is rich in different functional groups leading
it to be a good candidate for further functionalization. The
dominant chemical structures present on the surface of GO
Graphite are tertiary alcohols and ethers with very low quantities of
oxide carboxylic acid at the periphery of the GO (Figure 8.7).40
Different approaches have been devised to obtain graphene
sheets by reducing GO sheets, including thermal reduction
GO (approximately 30% of the mass of the GO is lost during the
dispersion reduction process, leaving behind vacancies and topological
GO-host defects throughout the plane of the rGO platelet),41 chemical
solution reduction (N2H4,42 NaBH426), and photoreduction.43 Among
all these methods, N2H4 has been deemed as the most efficient
GO film rGO rGO-host yet most toxic one. By combining NaBH4 reducing and ther-
dispersion solution mal annealing approaches, high quality graphene with impu-
rities of less than 0.5 wt.% of sulfur and nitrogen in the final
rGO film rGO composite film product has been obtained. This method is particularly effec-
tive in the restoration of the pi-conjugated structure, and leads
to highly soluble and conductive graphene materials.26b The
amphiphilic property enables GO unique role as surfactant. It
has been utilized to disperse carbon nanotubes (CNTs). The
interaction between GO and CNT not only provides means to
separate CNTs but also individual GO sheets, which increases
the effective active area for electrochemical reaction. More
FIGURE 8.6 Process scheme for fabricating rGO-based thin
films. The schematic illustrations show the structure of the mate- than 12 mg/mL of SWCNT dispersed in water without sur-
rial at each stage of the process. The gray and the orange sheets factant assistance has been achieved. The dispersion dem-
represent not oxidized and oxidized graphene sheets, respectively. onstrates high stability and permeability as ink material to
(Adapted from V. C. Tung et al. Nano Letters 2009, 9, 1949–1955.) prepare paper electrodes and it is an effective ink material for
CO2H
CO2H
HO2C
CO2H
HO2C
HO2C
CO2H
HO2C
CO2H
= –OH = –O–
OH
OH
OH O
OH OH
O HO O OH
O O
O OH O HO
OH OH
O O
OH O OH O
HO
OH OH O
O
OH HO HO OH
HO O HO
HO O O
FIGURE 8.7 Variations of the Lerf–Klinowski model indicating ambiguity regarding the presence or absence of carboxylic acids on the
periphery of the basal plane of the graphitic platelets of GO. (Adapted from D. R. Dreyer et al. Chemical Society Reviews 2010, 39, 228–240.)
preparing paper electrodes. The approach provides a binder- contains two equivalent atoms with a nearest distance of
free ink-printing method for large-scale fabrication of elec- 1.42 Å (Figure 8.8a). This crystal structure results in the
trochemical energy storage devices without other additives, touching between bonding (π) and antibonding (π*) orbitals
binders, or additional current collectors (Scheme 8.1).26c
OH OH
8.2.1.3 Other Methods OH OH
The first visible graphene single layer was obtained by O
O OH
micromechanical exfoliation of highly ordered pyrolytic O
OH OH
graphite (HOPG) in 2004, when the Nobel laureates, Geim O
OH HO OH
and Novoselov used Scotch tape to peel graphene layer and HO OH O

to transfer onto Si/SiO2 substrate.8 Later on, epitaxial gra- O
OH
phene was obtained by thermal decomposition of SiC at GO/CNT ink for supercapacitor
1000–1500°C.44
Paper
8.2.2 Electronic Structure of Graphene
Ink
In the graphene structure, each carbon atom is bonded to
three other carbon atoms according to sp2 hybridization. The
GO/CNT paper
hexagonal network of carbon atoms are connected by a strong
covalent sp2–sp2 σ–σ in-plane bonding. Owing to an addi-
tional 2pz orbital π−π bonding, which is oriented perpendicu- SCHEME 8.1 GO-SWNT noncovalent interactions, and preparation
larly to the atomic plane, the delocalized electrons appear. of supercapacitor electrodes with GO/SWNT ink and paper. (From
They are responsible for electrical conductivity of graph- S. Guo et al. Journal of Materials Research 2013, 28(7), 918–926.
ite which is highly anisotropic.20 The unit cell of graphene Copyright Materials Research Society 2013.)
(a) (c) Area: Sh

Β1
Α1
Α Β
Α2 Β dc–c = 0.1421 nm
2
(b) (d)
π*2
π*
π*1
FIGURE 8.9 Schematic representation of the arrangement of car-
E=0 → E=0 → bon atoms within the graphene layer.
k k
π1
π
The theoretical surface area could be calculated from the
π2
following process using one hexagon unit cell (Figure 8.9):
The surface area of one hexagon is
FIGURE 8.8 Crystal structures of graphene single layer (a) and 3

s = 3d 2C − C = 5.246 × 10 −20 m 2 (8.1)
AB-stacked bilayer (c). The unit cell of single-layer graphene and 2
bilayer graphene consist of 2 and 4 atoms in each unit cell, respec-
tively. The hopping between nearest neighbor atom sites A and B where s corresponds to two carbon atoms whose weight is, taking
resulting in the linear dispersive electronic band structure (b). The into account the atomic weight of carbon (MC = 12.01 g/mol)
interlayer coupling for bilayer graphene cannot be ignored as the and the Avogadro number (NA = 6.023 × 1023 mol−1),
hopping between A 1 and B 2 affected the electronic band structure
and resulting in two parabolic electronic bands (d). (From D. Zhan 2MC
et al. Advanced Materials 2012, 24, 4055–4069.) WC = = 3.988 × 10 −23 g (8.2)
NA
Combining Equations 8.1 and 8.2 gives the specific surface

at each energy valleys K and K′ when considering hopping area of one side of a graphene sheet (Sgraphene):
only occurs between the nearest neighbor atomic sites.45 The
valence and conduction bands touch at the Brillouin zone cor- Sgraphene = s/WC = 1315 m2/g.
ners thus making graphene a zero-bandgap semiconductor.
The carriers’ speed, called as Fermi velocity, is as high as Thus, considering both sides of the graphene sheet, the spe-
106 m/s (~1/300 c, where c is the velocity of light), and thus cific surface area should be 2630 m2/g, which further demon-
the electrical transport properties of graphene are remark- strates the potential application of GFETs as sensor devices.
ably ascribed to such photon-like relativistic Dirac fermions.
Particularly, the gapless graphene displays ambipolar electric
field effect with ballistic property presenting extremely high 8.2.4 Chemical Functionalization
mobility at room temperature as observed experimentally.46 The tuning of the Fermi level in graphene is an important fac-
Furthermore, graphene can be doped by electrons or holes tor in determining the successful operation of the field effect
with tunable charge concentration, and the maximum can transistor sensors or resistive-type sensors. Chemical doping
exceed ~1013 cm2.47 or charge transfer doping is an effective approach to achieve
the Fermi level tuning goals. p-Type doping drives the Dirac
points of graphene above the Fermi level and n-type doping
8.2.3 GFET Sensor
drives the Dirac points below the Fermi level. Or we can say
The unique electronic property of graphene enables GFET n-type or p-type doping causes a shift of the Fermi level in
extraordinary for the new generation of electronic devices. the graphene from the Dirac point into the conduction band
Unlike the conventional semiconductors, graphene serves or valence band. Many approaches have been applied for such
as the channel that will facilitate functionalization of the purposes including molecules, nanoparticles, and polymers.
channel with various molecules, elements, and polymers Depending on the presence of electron-donating or elec-
providing more chances for sensor device applications. The tron-withdrawing functional groups, n-type or p-type doping
2D nature of graphene allows total exposure of all of its sur- effect could be achieved through molecule doping approach.
face carbon atoms to the adsorbed molecules, providing the The presence of positive or negative charges on the surface
high surface area, and it is inherently a low-noise material of nanoparticles could be utilized to achieve n-type or p-type
due to the quality of its crystal lattice, which leads to screen doping through charge transfer mechanism. Considering vari-
charge fluctuations more than one-dimensional systems such ous gases demonstrate different doping effects (Table 8.2),
as CNTs. GFET should be applied effortlessly.
TABLE 8.2
Summary of Doping Effects with Molecules, Polymers, Elements, and Nanoparticles Tuning Electronic Structure
of Graphene
Molecules Molecular Structure or Functional Groups (Monomer) Doping Effect (n-Type or p-Type)
p-Toluenesulfonic acid O n-Type36
OH
S

Tetrafluoro-tetracyanoquinodimethane N N p-Type37
(F4-TCNQ)
F F
F F
N N
4-Bromobenzenediazonium tetrafluoroborate Br p-Type48
F
F B– F
N+
F
N
poly(ethylene imine) n-Type48
HN
1,5-Naphthalenediamine NH2 n-Type49
NH2
9,10-Dimethylanthracene n-Type49
9,10-Dibromoanthracene Br p-Type49
Br
(Continued )
TABLE 8.2 (Continued )

of Graphene
Tetrasodium 1,3,6,8-pyrenetetrasulfonic acid Na+ p-Type49
O–
O S O
O
–O
O
Na+ S
S Na+
O O–
O
O S O
O–
Na+
Water H2O p-Type8
Ammonia NH3 n-Type46b,50
Nitrogen dioxide NO2 p-Type46b
Sulfur dioxide SO2 p-Type51
Benzoyl peroxide n-Type52

Tetracyanoethylene (TCNE) N p-Type53
N
(tridecafluoro-1,1,2,2-tetrahydrooctyl) F F F F F F p-Type54
trichlorosilane
Cl F
Si
Cl F
Cl F F F F F
Ethanol n-Type8
OH
Nitrophenyl diazonium tetrafluoroborate O n-Type55
N+
–O
Azidotrimethyltin n-Type56
Sn
:N
(Continued )

of Graphene
2,4,6-Tris(dimethylamino)-1,3,5-triazine n-Type57
N
N N
N N N
p-Tert-butylcalix[4]arene p-Type58
OH
OH HO
OH
4-Amino-2,2,6,6-tetramethyl-1-piperidinyloxy n-Type59
H2N
(4-amino-TEMPO
N
O•
CF4 CF p-Type9b
Aminopropyltriethoxysilane n-Type60
Si NH2
O
O
Hexamethyldisilazane Neither n-type nor p-type61
N
H
Si Si
(Continued )

of Graphene
Hydrogen silasesquioxane (HSQ) (a) (b) n-Type62
R R
R Si O Si R
O O Si O Si
Si Si R O O
R O R O O
O O Si O Si
R R
R Si O Si R Si O Si
O O R
O O
O O
Si O Si
R Si O Si R
R R
Ladder type T8 cube
(c) R R
O
Si Si
O
R O O
Si Si
O O
R
R
O O
O Si O
Si R
O
Si Si
O
R R
Cage representation
Pyrene-1-sulfonic acid OH n-Type63
O S O
3,4,9,10-perylenetetracarboxylic diimide O p-Type63
NH
NH
O
1-Aminopyrene NH2 n-Type64
(Continued )

of Graphene
1-Nitropyrene O O– p-Type64
N+
1-Pyrenecarboxylic acid O OH p-Type64
1-Pyrenesulfonic acid OH p-Type64
O S O
Poly(styrene-b-4-vinylpyridine) PS block PVP block p or n type9b
CH2
CH2
CH n CH m
N
TiO2 nanoparticles GO and TiO2 Electron transfer from TiO2 to
graphene65
TiO2 thin film CVD graphene n-Type66
Fe3O4, TiO2, CdSe nanoparticles Ligands: trioctylphosphine oxide, oleic acid n-Type67
Potassium K n-Type68
Phosphorus P n-Type69
Silver Ag n-Type70
Gold Au p-Type71
Copper Cu n-Type70
Bismuth Bi n-Type71b
Antimony Sb n-Type71b
Nitrogen N n-Type69
For n-type (p-type) doping, the electrons have to be released Figure 8.10 shows the device fabricated with GO film for
into (extracted out of) the graphene layer. All of the examples detecting hydrogen cyanide (HCN). Doping effects as well as
(molecules, polymers, nanoparticles, and elements) shown in charge transfer enable GFET detection of various molecules.
Table 8.2 demonstrate such ability to tune the electronic struc- Graphene synthesized from reducing GO, exfoliating HOPG,
ture of graphene. However, there are still ambiguity regarding growing with CVD, and thermal reduction from SiC have
certain functional groups which are needed to be studied fur- been utilized for such applications. Various detection limits
ther. For example, with the same functional group, p-toluene- have been achieved as summarized in Table 8.2.
sulfonic acid shows n-type doping whereas 1-pyrenesulfonic Despite the advancements shown earlier in Table 8.2,
acid shows p-type doping. Note that even elements in the same there are still several challenges that need to be overcome. (1)
group may display different doping effects to graphene devices Selectivity: Graphene is exquisitely sensitive to the chemical
(Ag and Au).70 This is related to the nature of elements’ elec- environment and hence is easily affected by a range of differ-
tron negativity. The higher the electron negativity, the higher ent gas species and mixtures (Table 8.3). Therefore, specifically
the electron-withdrawing property of the element, indicating identification of gases components is challenging. For example,
p-type doping effect. Moreover, to precisely manipulate the exposure to either NO2 or O2 causes a similar increase in gra-
electronic structure, one or more molecules are still needed. phene conductance. Moreover, it is impossible to detect single
gas for the gas mixture. For example, a mixture that contains
p-type doping molecules such as NO2 and an n-type doping mol-
8.2.5 GFET Gas Sensor
ecules such as NH3 may not create any net change in graphene’s
The report of an individual molecule detection with conductivity. Future research needs to focus on functionalizing
graphene46b intrigues the GFET sensor development. Working graphene with capture agents that will enable the specific bind-
principles of most GFET sensors are based on the change in ing of target gases to the graphene surface. (2) Reversibility:
electrical conductance with exposure of gases. For example, The thermal energy at room temperature is typically not
(a) (b)
1
ΔG/G0 response, ppth
GO film
0.1 MDL (as 3σ): 70 ppb
300 μm 100 1000

SiO2
Concentration HCN:N2 (ppb)
FIGURE 8.10 (a) Optical image of electrically isolated GO device with interdigitated metal contacts. (b) Response of an rGO sensor to
different concentrations of HCN as a chemical-warfare agent. (Adapted from J. T. Robinson et al. Nano Letters 2008, 8, 3137–3140.)
TABLE 8.3
Graphene FET for Various Gas Sensors
Sensor Material Gases or Vapor Detected Detection Limit Year
Mechanically exfoliated graphene, shaped by lithography; H2O, NH3, nonanal, octanoic acid, Depended on vapor and presence of 200973
measurements made with and without PMMA resist trimethylamine photoresist: for example, 5 ppm for
nonanal with resist; >30 ppm without
Hydrazine-rGO with variable levels of reduction NO2, NH3, dinitrotoluene (DNT) <5 ppm NO2 and NH3; 28 ppb DNT 200974
GO reduced chemically or thermally or by both approaches H2O <0.1 Torr of H2O vapor 200875
Low-temperature thermally rGO NO2 ~2 ppm 200976
nanoscale flakes of abraded graphite NO2 <60 ppb 200777
Hydrazine-rGO with variable levels of reduction HCN, chloroethylethyl sulfide (CEES), 70 ppb HCN; 0.5 ppb CEES; 5 ppb 200872
Dimethylmethylphosphonate DMMP; 0.1 ppb DNT
(DMMP), DNT
Mechanically exfoliated graphene, shaped by lithography NO2, H2O, I2, NH3, CO, ethanol ~1 ppb 200746b
CVD graphene, Pt H2 25 ppm 201078
Epitaxial graphene, Pt H2 1% 201179
rGO, Pt H2 50 ppm 201380
CVD graphene NH3 30 mTorr 201181
enough to overcome the activation energy needed for molecu- single-walled CNTs for device applications including reso-
lar desorption for gases such as NO2 and NH3 on the surface of nant tunneling diodes, FETs, and biochemical sensors.86
graphene. Typically, it needs high temperature, ultraviolet (UV) The sensors based on FETs are promising to revolution-
exposure, or vacuum for gas desorption to clean the graphene ize bioanalytical research by offering direct, real-time, highly
surface and expel the chemisorbed molecules. Such treatments specific, ultrasensitive, and label-free detection of the desired
usually result in irreversible reaction between molecules and biomolecule.87 Although there are plenty of achievements in
graphene devices. Still innovative approaches to functionalize CNT88 and nanowire-based FET devices,87a,89 there are still a
graphene surfaces are needed to control the binding energy of lot of limitations including nonspecific binding,87c instability
target molecules to the graphene surface. (3) Stability: Practical of the device,90 and high cost that limits future applications.
usage of sensors requires high stability regardless of the change Graphene-based DNA FET sensor provides bright future for
in environmental conditions such as moisture, temperature, such device applications. We worked on such device applica-
residual charge buildup, or contamination. This creates addition by functionalizing graphene with the noncovalent method
tional difficulties for reliable and repeatable sensing. (4) Cost: by immersing the devices in a 2 mg/mL 1-pyrenebutanoic
To be competitive with commercial sensors, graphene-based acid succinimidyl ester in anhydrous methanol solution for
devices must be mass producible at low cost. Therefore top- 2 h. The functionalization method keeps the high conductiv-
down methods such as exfoliation of GO could be used to mass ity of the CVD graphene as well as introduces functional end
produce graphene nanosheets at low cost. Moreover, large- for ssDNA immobilization (Figure 8.11).
scale synthesis with CVD could substantially lower the cost of Other methods have also been applied for DNA detection.
graphene-based chemical sensors. (5) Types of graphene: As Mohanty et al. applied GO for sensing DNA and studied the
mentioned in the graphene synthesis part, different approaches interface of bacterial and GO.84a The shift of Dirac point was
could give different types of graphene with desirable functional
groups. While CVD graphene provides large-area thin films for
(a)
high-quality field-effect transistor (FET), rGO could also pro-
vide films with more accessible functional groups. At the same
time, graphene foam formed by rGO or synthesized via nickel
S D
foam will definitely endow greater potential for GFET sensors. G
(b)
8.2.6 GFET Biosensor Noncovalent bond
For biosensors, ionic screening of the charged impurities on

graphene plays an important role. NaF has been systematically
Immobilization of
studied as an example. Counterions (Na+) in the solution could probe DNA
accumulate on the graphene surface to neutralize the impu-
rity charges underneath the graphene, which result in orders
of magnitude, increase of mobility and decrease of minimum Hybridization with
conductivity.15b Such effects could be utilized to monitor ionic complementary DNA
concentration change during biological interactions.

Another interesting mechanism is called gating effect which
is related to the electronic property of graphene mentioned in (c)
Section 8.2.4. Typically, positively or negatively charged bio-
molecules will act as gating modulators which will drive Fermi
level in the graphene from the Dirac point into the conduction
Graphene
band or valence band. Deoxyribonucleic acid (DNA) gating
effect has been well studied through Raman and theoretical
Substrate
modeling.82 With such an effect, single-stranded DNA (ssDNA)
could be identified with a low detection limit of 3 nM.83
With all these mechanisms, GFET has been utilized for
detecting various biological materials, such as DNA,83,84 cells,85 20 µm
bacterium,84a and proteins.84a Incorporation of nanoparticles or
functionalization of graphene with polymers and selective func-
tional groups will definitely improve its sensitivity and selectivity.
FIGURE 8.11 (a) Schematic illustration of graphene device. (b)
8.2.6.1 DNA Sensor Schematic illustration of the graphene sensor concept: modifying
graphene with noncovalent bond and its binding of DNA strand. (c)
DNA and peptide nucleic acids (PNAs) that have base Confocal laser scanning microscope image of graphene after immo-
sequences that offer specificity are attractive assembly linkers bilization with c-DNA. The probe DNA sequence is (5′-3′) amine-
for bottom-up nanofabrication. Engineered ssDNA sequences AAC-TGC-CAG-CCT-ATG-TCC-AA; complementary DNA
are employed in the nanoarchitectures of end-functionalized sequence is (5′-3′)FAM-TTG-GAC-ATA-GGC-TGG-CAG-TT.
(b) 0.48 0.4 y = 0.0978x–0.9686

0.2 2
R = 0.9973
0.46 0.0
pH 12 –0.2
Conductivity (mS)
0.44 –0.4
–0.6
–0.8
0.42 pH 10 0 2 4 6 8 10 12 14
pH units
0.40 pH 7
0.38
(a) 0.36 pH 4
de pH 2
ec tro 0.34
e el
p-gat ting –1.0 –0.5 0.0 0.5 1.0
To ula
Ins
SiO 2
Drain + Solution (c) 0.66 y = 0.0988x–0.8808
rce R2 = 0.9978
Sou 0.64
0.62 pH 12
Conductivity (mS)
Epitaxia pH 10
l graphen
e 0.60 2 4 6 8 10 12
pH units
0.58 pH 7
0.56 pH 4
0.54 pH 2
0.52
0.50
Increasing pH
0.48
–1.0 –0.5 0.0 0.5 1.0
Gate voltage (V)
FIGURE 8.12 (a) Schematic illustration of a solution-gated graphene field effect transistor (SGFET). (b) Conductivity versus gate potential
plot of SGFET with 1–2 graphene layers. (c) Conductivity versus gate potential plot of SGFET with 3–4 graphene layers. (From P. K. Ang
et al. Journal of the American Chemical Society 2008, 130, 14392–14393.)
explained by the negative charge gating effect of DNA. Further based on aptamer-modified immunoglobulin E (IgE) shows a
Lin et al. verified this effect by Raman spectroscopy and deeply dependence of the drain current variation on the IgE concen-
understood the mechanism by theoretical modeling.91 tration and dissociation constant was estimated to be 47 nM,
which is comparable to previously reported best results for
8.2.6.2 pH Sensor CNTs.95 Gong et al. reported graphene resistance sensor
Theoretically, ions may be specifically bound at the inner detection of the cancer biomarker prostate-specific antigen.
Helmhotz plane at the graphene/electrolyte interface during The detection limit could reach 0.08 ng/mL. In device fabri-
the charging process. The ideally polarizable graphene/elec- cation, graphene sheets synthesized from the thermal expan-
trolyte interface allows capacitive charging of the surface by sion method with different concentrations was dropped onto
hydroxonium (H3O+) or hydroxyl (OH−) ions, which plays a a confined polycarbonate, which serves as the resistance
similar role as the gate potential in changing the Fermi level sensor.
via modulating the channel conductance by doping “holes”
or “electrons,” repectively (Figure 8.12a).92 Many GFET pH
sensors based on such mechanism have been reported.92,93 In
such an application, a linear relationship between the pH and
elec
Drai
Source
electrod
the Dirac point was observed, which proves the a vailability

trod
n
of GFET for application in solution systems (Figure 8.12b

e
and c). Besides the FET sensor, some graphene-based elec-

e
trochemical sensors may work for pH-sensing applications, Graphene

for example, glucose oxidase (GOD)–graphene– chitosan SiO2
nanocomposite-based electrochemical sensor used for glu- Si sub.
cose detection can also be pH responsive.94
8.2.6.3 Graphene Immunosensor

A
8.2.6.3.1 Recent Advances in Functional
Graphene Biosensors Current Bias
Similar with the DNA and pH sensor, immunosensors have
also been developed (Figure 8.13). Label-free immunosensor FIGURE 8.13 Schematic illustration of immunosensor setup.
8.2.7 Graphene Electrochemical Glucose Sensor 6.0×10–5 32

b
24 c
Different from FET-based biological and chemical sensors, elec-
ΔI/µA
16 d
trochemical sensors produce electrical signals via electrochemi- 4.0×10–5
cal reaction. As shown in Figure 8.14, a typical electrochemical 8 e
sensor consists of three major components; they are sensing/ 0 f
0 3 6 9 12 15 18
2.0×10–5
I/A
working electrode, counterelectrode, and the electrolyte. A stable Cglucose/mM a
and constant external driving voltage is required to improve the
performance of a sensor, and a reference electrode is necessary 0.0
to be introduced. Additional optional components also have a
gas-permeable membrane (also called hydrophobic membrane);
besides mechanical protection to the sensor, the membrane also –2.0×10–5
functions as a filter to remove unwanted particulates. For gra- 0.1 0.0 –0.1 –0.2 –0.3 –0.4 –0.5 –0.6 –0.7
phene electrochemical sensors, the working/sensing electrodes E/V vs. Ag/AgCl
are made from graphene or graphene-based composites.
Compared with graphene-based solid-state biological and
FIGURE 8.15 Cyclic voltammograms of GOD–graphene–chito-
chemical sensors, graphene-based electrochemical sensors have san/GCE in PBS with 0.1 M KCl at a scan rate of 100 mV s−1 in the
a few merits: (i) easiness of fabrication; (ii) low-power consump- presence of different concentrations of glucose N2 saturated without
tion; (iii) long life expectancy; (iv) very good selectivity to target glucose, (b) O2 saturated without glucose, and (c through f) with
molecules; and (v) minimally affected by changes in pressure. glucose of 3.0, 5.0, 8.0, and 10.0 mM. (Adapted from X. Kang et al.
Glucose biosensor has been developed and used in glucose- Biosensors and Bioelectronics 2009, 25, 901–905.)
monitoring and diabetes control devices for decades since
Clark and Lyons proposed the concept of glucose enzyme elec-
trodes in 1962 at the Cincinnati Children’s Hospital.96 Their
first glucose-sensing measurement was based on the monitor- (a) (b)
ing of the oxygen consumed by the enzyme-catalyzed reaction 1000 1715
3100 1.2 nm
Glucose + O2 glucose oxidase gluconic acid + H 2O2 (8.3)

Development of novel glucose sensors are still being pur-
sued for the properties of noninvasive and continuous moni-
C O 200 nm
tors. C. Shan et al. reported the polyvinylpyrrolidone-protected
graphene/polyethylenimine-functionalized glucose oxidase (c) O O
(GOD)–graphene–chitosan nanocomposite, and this type of + H2N–GOX NHS
graphene composite helps the immobilization of the enzyme OH EDC NH–GOX
which helps to keep its bioactivity. The biosensor exhibits a wider
linearity range from 0.08 to 12 mM glucose with a detection (d) 200
limit of 0.02 mM and higher sensitivity (37.93 µAmM−1 cm−2)
compared with other nanostructured supports.94 This compos- 160
ite also shows to be pH responsive (Figure 8.15). 120
I (μA/cm2)
Dai et al. prepared a biocompatible glucose biosensor based 200

150
I (µA/cm2)
on GOD-immobilized GO electrode which exhibits a broad 80 100

linear range of up to 28 mM mm−2 glucose with a sensitivity 50
40
of 8.045 mA cm−2 M−1 (Figure 8.16).97 The glucose-sensing 0
0 5 10 15 20 25 30
(Glucose) (mM)
0
Capillary diffusion barrier 0 600 1200 1800 2400 3000 3600 4200
Hydrophobic membrane
t(s)
Sensing electrode
FIGURE 8.16 (a) Schematic representation of the molecular struc-
ture of a GO sheet, (b) AFM micrograph of the GO sheets, and (c)
Electrolyte schematic immobilization of GOx into GO sheets via peptide bonds
Reference electrode between the amine groups of GOx and the carboxylic acid of GO. (b,
inset) FTIR spectrum of GO. (d) Typical amperometric responses at
the GOx covalently immobilized GO electrode (black line) and the
Counter electrode pristine GO electrode (dark gray line) in 0.4 MPBS (pH 7.4) with
the successive injection of 0, 5, 10, 15, 20, 25, and 30 mM glucose,
respectively. Calibration curve obtained for glucose determination
FIGURE 8.14 Schematic illustration of basic setup for electro- at 0.4 V versus Ag/AgCl in 0.4 MPBS (pH 7.4). (Adapted from Y.
chemcial sensor. Liu et al. Langmuir 2010, 26, 6158–6160.)
TABLE 8.4
Summarized Performances for Graphene-Based Glucose Sensors
Response Potential Linear Range Detection
Sensor Material Time (s) (V) (Up to mM) Sensitivity Limit Year
Enzymes immobilized exfoliated graphite <5 0.7 6 14.17 µA/(mM cm2) 10 µM 200798
nanoplatelets
GOD–graphene–chitosan N.A. ~0.5 12 37.93 μA mM−1 cm−2 0.02 mM 200994
GOD/Pt/functional graphene sheets/chitosan N.A. N.A. N.A. N.A. 0.6 μM 200999
(GOD/Pt/FGS/chitosan)
Polyvinylpyrrolidone-protected graphene/ N.A. −0.49 14 N.A. N.A. 2009100
polyethylenimine-functionalized ionic liquid/GOD
Gold decorated functionalized graphene sheets N.A. 0.8 30 N.A. 1 μM 2010101
Covalently linked GOx–GO enzyme electrode N.A. 0.4 28 mM ⋅ mm −2 −2
8.045 mA ⋅ cm ⋅ M −1 N.A. 201097
Carboxyl-modified graphene oxide (GO***COOH) N.A 0.2 20 μM N.A. 1 μM 2010102
N-doped graphene N.A. −0.15 1.1 N.A. 10 µM 2010103
Pd nanoparticle–graphene hybrids N.A. 0.4 10 μM to 5 mM N.A. 1 μM 2011104
(R = 0.998)
Cu nanoparticle-modified graphene sheets <2 0.5 4.5 N.A. 0.5 μM (signal/ 201275
noise = 3)
Pt/graphene/TiO2 N.A 0.7 0.1–8 mM 0.94 μA ⋅ mM−1 ⋅ cm−2 0.1 mM 2014105
CuO/graphene-modified screen-printed carbon N.A 0.7 0.122 μM to 2367 μA ⋅ mM−1 ⋅ cm−2 34.3 nM 2013106
electrode (CuOG-SPCE) 0.5 mM (S/N = 3)
performances of the various graphene-based materials sys- 7. W. S. Hummers Jr, R. E. Offeman, Journal of the American
tems are summarized in Table 8.4. Chemical Society 1958, 80, 1339–1339.
8. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y.
Zhang, S. V. Dubonos, I. V. Grigorieva, A. A. Firsov, Science
8.3 CONCLUDING REMARKS 2004, 306, 666–669.
9. (a) M. J. Allen, V. C. Tung, L. Gomez, Z. Xu, L. M. Chen,
Emerging as a true 2D crystal, graphene shows remarkable K. S. Nelson, C. W. Zhou, R. B. Kaner, Y. Yang, Advanced
sensitivity and detection limit for chemical and biological Materials 2009, 21, 2098–2102; (b) S. Guo, M. Ghazinejad,
sensing due to its exceptional physicochemical properties X. Qin, H. Sun, W. Wang, F. Zaera, C. S. Ozkan, M. Ozkan,
including excellent conductivity, high mechanical strength, Small 2012, 8, 1073.
large surface area, and ease of functionalization. The 2D 10. Z. Y. Chen, S. Berciaud, C. Nuckolls, T. F. Heinz, L. E. Brus,
Acs Nano 2010, 4, 2964–2968.
nature and tunable surface chemistry make the graphene lay-
11. (a) L. M. Dai, Chemsuschem 2010, 3, 797–799; (b) N. L.
ers very sensitive molecules and chemical agents which tend Yang, J. Zhai, D. Wang, Y. S. Chen, L. Jiang, Acs Nano 2010,
to modulate its electrical properties through surface modifi- 4, 887–894; (c) C. X. Guo, H. B. Yang, Z. M. Sheng, Z. S.
cation. The two types of sensors, that is, solid-state sensors Lu, Q. L. Song, C. M. Li, Angewandte Chemie-International
which are primarily based on the changes of their electrical Edition 2010, 49, 3014–3017; (d) H. X. Chang, X. J. Lv, H.
conductivity and electrochemical sensors, which are also Zhang, J. H. Li, Electrochemistry Communications 2010,
based on electrochemical reaction, have been successfully 12, 483–487.
12. S. Bae, H. Kim, Y. Lee, X. F. Xu, J. S. Park, Y. Zheng, J.
developed. We expect that gas and/or biological sensors with
Balakrishnan et al., Nature Nanotechnology 2010, 5, 574–578.
higher sensitivity and lower detection limit will be developed 13. (a) J. K. Wassei, R. B. Kaner, Materials Today 2010, 13, 52–59;
in the near future. (b) V. C. Tung, L. M. Chen, M. J. Allen, J. K. Wassei, K. Nelson,
R. B. Kaner, Y. Yang, Nano Letters 2009, 9, 1949–1955; (c) S.
S. Li, K. H. Tu, C. C. Lin, C. W. Chen, M. Chhowalla, Acs
REFERENCES Nano 2010, 4, 3169–3174; (d) G. Eda, M. Chhowalla, Advanced
1. F. Schedin, A. Geim, S. Morozov, E. Hill, P. Blake, M. Materials 2010, 22, 2392–2415; (e) S. De, P. J. King, M. Lotya,
Katsnelson, K. Novoselov, Nature Materials 2007, 6, 652–655. A. O’Neill, E. M. Doherty, Y. Hernandez, G. S. Duesberg, J. N.
2. J. Boardman, The Cambridge Ancient History, 3. Coleman, Small 2010, 6, 458–464.
3. P. R. Wallace, Physics Review 1947, 71, 476–476. 14. H. Chen, M. B. Muller, K. J. Gilmore, G. G. Wallace, D. Li,
4. A. K. Geim, K. S. Novoselov, Nature Materials 2007, 6, Advanced Materials 2008, 20, 3557–+.
183–191. 15. (a) L. Zhu, L. Luo, Z. Wang, Biosensors and Bioelectronics
5. (a) R. Peierls, Annales De l’Institut Henri Poincare 1935, 5, 2012, 35, 507–511; (b) F. Chen, J. L. Xia, N. J. Tao, Nano
177–222; (b) L. Landau, Z. Sowjetunion 1937, 11, 26–35. Letters 2009, 9, 1621–1625; (c) Y. Y. Shao, J. Wang, H. Wu, J.
6. B. C. Brodie, Philosophical Transactions of the Royal Society Liu, I. A. Aksay, Y. H. Lin, Electroanal 2010, 22, 1027–1036;
of London 1859, 149, 249–259. (d) S.-R. Guo, J. Lin, M. Penchev, E. Yengel, M. Ghazinejad, C.
S. Ozkan, M. Ozkan, Journal of Nanoscience Nanotechnology 114, 3811–3816; (c) Y. W. Zhu, M. D. Stoller, W. W. Cai, A.
2011, 11, 5258–5263; (e) G. X. Zhu, Y. J. Liu, Z. Xu, T. A. Jiang, Velamakanni, R. D. Piner, D. Chen, R. S. Ruoff, Acs Nano
C. Zhang, X. Li, G. Qi, Chemphyschem 2010, 11, 2432–2437; (f) 2010, 4, 1227–1233; (d) M. D. Stoller, S. J. Park, Y. W. Zhu,
C.-L. Zhang, Y.-X. Yuan, S.-M. Zhang, Y.-H. Wang, Z.-H. Liu, J. H. An, R. S. Ruoff, Nano Letters 2008, 8, 3498–3502.
Angewandte Chemie International Edition 2011, 50, 6851– 40. D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff, Chemical
6854, S6851/6851–S6851/6856; (g) Y. Wang, Z. Li, J. Wang, J. Society Reviews 2010, 39, 228–240.
Li, Y. Lin, Trends in Biotechnology 2011, 29, 205–212. 41. H. C. Schniepp, J. L. Li, M. J. McAllister, H. Sai, M. Herrera-
16. Y. M. Lin, C. Dimitrakopoulos, K. A. Jenkins, D. B. Farmer, Alonso, D. H. Adamson, R. K. Prud’homme, R. Car, D. A.
H. Y. Chiu, A. Grill, P. Avouris, Science 2010, 327, 662–662. Saville, I. A. Aksay, Journal of Physical Chemistry B 2006,
17. N. O. Weiss, H. Zhou, L. Liao, Y. Liu, S. Jiang, Y. Huang, X. 110, 8535–8539.
Duan, Advanced Materials 2012, 24, 5782–5825. 42. S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas,
18. K. Novoselov, V. Fal, L. Colombo, P. Gellert, M. Schwab, K. A. Kleinhammes, Y. Jia, Y. Wu, S. T. Nguyen, R. S. Ruoff,
Kim, Nature 2012, 490, 192–200. Carbon 2007, 45, 1558–1565.
19. P. R. Somani, S. P. Somani, M. Umeno, Chemical Physics 43. L. J. Cote, R. Cruz-Silva, J. X. Huang, Journal of the American
Letters 2006, 430, 56–59. Chemical Society 2009, 131, 11027–11032.
20. A. N. Obraztsov, E. A. Obraztsova, A. V. Tyurnina, A. A. 44. C. Berger, Z. M. Song, X. B. Li, X. S. Wu, N. Brown, C. Naud,
Zolotukhin, Carbon 2007, 45, 2017–2021. D. Mayou, T. B. Li, J. Hass, A. N. Marchenkov, E. H. Conrad,
21. P. W. Sutter, J. I. Flege, E. A. Sutter, Nature Materials 2008, P. N. First, W. A. de Heer, Science 2006, 312, 1191–1196.
7, 406–411. 45. D. Zhan, J. X. Yan, L. F. Lai, Z. H. Ni, L. Liu, Z. X. Shen,
22. J. Coraux, A. T. N’Diaye, C. Busse, T. Michely, Nano Letters Advanced Materials 2012, 24, 4055–4069.
2008, 8, 565–570. 46. (a) Y. B. Zhang, Y. W. Tan, H. L. Stormer, P. Kim, Nature 2005,
23. (a) J. C. Hamilton, J. M. Blakely, Surface Science 1980, 91, 438, 201–204; (b) F. Schedin, A. K. Geim, S. V. Morozov,
199–217; (b) Y. Yao, C.-P. Wong, Carbon 2012, 50, 5203–5209. E. W. Hill, P. Blake, M. I. Katsnelson, K. S. Novoselov, Nature
24. A. Y. Tontegode, E. Rut’kov, Physics-Uspekhi 1993, 36, Materials 2007, 6, 652–655.
1053–1067. 47. A. Das, S. Pisana, B. Chakraborty, S. Piscanec, S. K. Saha, U.
25. T. Land, T. Michely, R. Behm, J. Hemminger, G. Comsa, V. Waghmare, K. S. Novoselov et al., Nature Nanotechnology
Surface Science 1992, 264, 261–270. 2008, 3, 210–215.
26. (a) H. J. Shin, K. K. Kim, A. Benayad, S. M. Yoon, H. K. 48. D. B. Farmer, R. Golizadeh-Mojarad, V. Perebeinos, Y. M.
Park, I. S. Jung, M. H. Jin, H. K. Jeong, J. M. Kim, J. Y. Choi, Lin, G. S. Tulevski, J. C. Tsang, P. Avouris, Nano Letters
Advanced Functioanal Materials 2009, 19, 1987–1992; (b) W. 2009, 9, 388–392.
Gao, L. B. Alemany, L. J. Ci, P. M. Ajayan, Nature Chemistry 49. X. C. Dong, D. L. Fu, W. J. Fang, Y. M. Shi, P. Chen, L. J. Li,
2009, 1, 403–408; (c) S. Guo, W. Wang, C. S. Ozkan, M. Small 2009, 5, 1422–1426.
Ozkan, Journal of Materials Research 2013, 28(7), 918–926. 50. Y. C. Lin, C. Y. Lin, P. W. Chiu, Applied Physics Letters 2010,
27. Z. Yan, J. Lin, Z. Peng, Z. Sun, Y. Zhu, L. Li, C. Xiang, E. L. 96, 133110.
Samuel, C. Kittrell, J. M. Tour, ACS Nano 2012, 6, 9110. 51. Y. J. Ren, C. F. Zhu, W. W. Cai, H. F. Li, H. X. Ji, I. Kholmanov,
28. S. Park, J. H. An, I. W. Jung, R. D. Piner, S. J. An, X. S. Li, A. Y. P. Wu, R. D. Piner, R. S. Ruoff, Applied Physics Letters
Velamakanni, R. S. Ruoff, Nano Letters 2009, 9, 1593–1597. 2012, 100, 163114.
29. A. O’Neill, U. Khan, P. N. Nirmalraj, J. Boland, J. N. Coleman, 52. H. T. Liu, S. M. Ryu, Z. Y. Chen, M. L. Steigerwald,
Journal of Physical Chemistry C 2011, 115, 5422–5428. C. Nuckolls, L. E. Brus, Journal of the American Chemical
30. B. C. Banerjee, P. L. Walker, T. J. Hirt, Nature 1961, 192, 450. Society 2009, 131, 17099–17101.
31. S. D. Robertson, Nature 1969, 221, 1044–1046. 53. Y. H. Lu, W. Chen, Y. P. Feng, P. M. He, Journal of Physical
32. X. S. Li, W. W. Cai, J. H. An, S. Kim, J. Nah, D. X. Yang, R. Chemistry B 2009, 113, 2–5.
Piner et al., Science 2009, 324, 1312–1314. 54. B. Lee, Y. Chen, F. Duerr, D. Mastrogiovanni, E. Garfunkel,
33. (a) N. A. Vinogradov, A. A. Zakharov, V. Kocevski, J. Rusz, E. Y. Andrei, V. Podzorov, Nano Letters 2010, 10, 2427–2432.
K. A. Simonov, O. Eriksson, A. Mikkelsen et al., Physical 55. E. Bekyarova, M. E. Itkis, P. Ramesh, C. Berger, M. Sprinkle,
Review Letters 2012, 109, 026101/026101–026101/026105; (b) W. A. de Heer, R. C. Haddon, Journal of the Americal
D. Kondo, K. Yagi, N. Harada, M. Sato, M. Nihei, S. Sato, Chemical Society 2009, 131, 1336–1337.
N. Yokoyama, in MRS Proceedings, Vol. 1205, Cambridge 56. J. Choi, S. N. Yang, K. J. Kim, H. Lee, S. Kim, Current
University Press, UK 2009. Applied Physics 2011, 11, 1307–1310.
34. C. Mattevi, H. Kim, M. Chhowalla, Journal of Materials 57. S. Park, S. Yang, H. Lim, H. Lee, Bulletin of the Korean
Chemistry 2011, 21, 3324–3334. Chemical Society 2011, 32, 2199–2202.
35. Q. K. Yu, L. A. Jauregui, W. Wu, R. Colby, J. F. Tian, Z. H. Su, 58. S. Park, S. Yang, K. J. Kim, K. No, H. Lee, Bulletin of the
H. L. Cao et al., Nature Materials 2011, 10, 443–449. Korean Chemical Society 2010, 31, 2809–2812.
36. A. K. Singh, M. W. Iqbal, V. K. Singh, M. Z. Iqbal, J. H. Lee, 59. J. Choi, H. Lee, K. J. Kim, B. Kim, S. Kim, Journal of
S. H. Chun, K. Shin, J. Eom, Journal of Materials Chemistry Physical Chemistry Letters 2010, 1, 505–509.
2012, 22, 15168–15174. 60. J. Park, S. B. Jo, Y. J. Yu, Y. Kim, W. Yang, W. H. Lee, H. H.
37. W. Chen, S. Chen, D. C. Qi, X. Y. Gao, A. T. S. Wee, Journal Kim et al., Advanced Materials 2012, 24, 407–411.
of the American Chemical Society 2007, 129, 10418–10422. 61. (a) B. N. Szafranek, D. Schall, M. Otto, D. Neumaier, H. Kurz,
38. D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H. B. Nano Letters 2011, 11, 2640–2643; (b) M. Lafkioti, B. Krauss,
Dommett, G. Evmenenko, S. T. Nguyen, R. S. Ruoff, Nature T. Lohmann, U. Zschieschang, H. Klauk, K. von Klitzing,
2007, 448, 457–460. J. H. Smet, Nano Letters 2010, 10, 1149–1153.
39. (a) J. Y. Luo, L. J. Cote, V. C. Tung, A. T. L. Tan, P. E. Goins, 62. K. Brenner, R. Murali, Applied Physics Letters 2010, 96,
J. S. Wu, J. X. Huang, Journal of the American Chemical 063104.
Society 2010, 132, 17667–17669; (b) Q. Yang, X. J. Pan, F. 63. Q. Su, S. P. Pang, V. Alijani, C. Li, X. L. Feng, K. Mullen,
Huang, K. C. Li, Journal of Physical Chemistry C 2010, Advanced Materials 2009, 21, 3191–3195.
64. J. Lee, E. Hwang, E. Lee, S. Seo, H. Lee, Chemistry—A et al., Science 2003, 302, 1545–1548; (e) X. Wang, F. Liu, G.
European Journal 2012, 18, 5155–5159. T. S. Andavan, X. Y. Jing, K. Singh, V. R. Yazdanpanah, N.
65. I. V. Lightcap, T. H. Kosel, P. V. Kamat, Nano Letters 2010, Bruque et al., Small 2006, 2, 1356–1365.
10, 577–583. 87. (a) J. Hahm, C. M. Lieber, Nano Letters 2004, 4, 51–54;
66. P. H. Ho, Y. C. Yeh, D. Y. Wang, S. S. Li, H. A. Chen, Y. H. (b) M. Curreli, R. Zhang, F. N. Ishikawa, H. K. Chang, R.
Chung, C. C. Lin, W. H. Wang, C. W. Chen, Acs Nano 2012, J. Cote, C. Zhou, M. E. Thompson, IEEE Transactions on
6, 6215–6221. Nanotechnology 2008, 7, 651–667; (c) M. T. Martinez, Y. C.
67. D. Q. Wang, X. F. Liu, L. He, Y. D. Yin, D. Wu, J. Shi, Nano Tseng, N. Ormategui, I. Loinaz, R. Eritja, J. Bokor, Nano
Letters 2010, 10, 4989–4993. Letters 2009, 9, 530–536.
68. J. H. Chen, C. Jang, S. Adam, M. S. Fuhrer, E. D. Williams, 88. (a) X. C. Dong, C. M. Lau, A. Lohani, S. G. Mhaisalkar, J.
M. Ishigami, Nature Physics 2008, 4, 377–381. Kasim, Z. X. Shen, X. N. Ho, J. A. Rogers, L. J. Li, Advanced
69. S. Some, J. Kim, K. Lee, A. Kulkarni, Y. Yoon, S. Lee, T. Materials 2008, 20, 2389; (b) K. Maehashi, K. Matsumoto,
Kim, H. Lee, Advanced Materials 2012, 24, 5481–5486. Sensors 2009, 9, 5368–5378; (c) J. Kang, J. Lee, T. H. Kim, J.
70. Y. J. Ren, S. S. Chen, W. W. Cai, Y. W. Zhu, C. F. Zhu, R. S. Park, M. J. Seong, S. Hong, Nanotechnology 2008, 19, 4; (d)
Ruoff, Applied Physics Letters 2010, 97, 3. K. Balasubramanian, M. Burghard, Analytical Bioanalytical
71. (a) R. S. Sundaram, C. Gomez-Navarro, E. J. H. Lee, M. Chemistry 2006, 385, 452–468.
Burghard, K. Kern, Applied Physics Letters 2009, 95, 3; (b) I. 89. (a) X. Wang, C. S. Ozkan, Nano Letters 2008, 8, 398–404; (b)
Gierz, C. Riedl, U. Starke, C. R. Ast, K. Kern, Nano Letters J. Wang, Biosens Bioelectron 2006, 21, 1887–1892.
2008, 8, 4603–4607. 90. Y. Yamamoto, K. Maehashi, Y. Ohno, K. Matsumoto, Sensors
72. J. T. Robinson, F. K. Perkins, E. S. Snow, Z. Q. Wei, P. E. and Materials 2009, 21, 351–361.
Sheehan, Nano Letters 2008, 8, 3137–3140. 91. J. Lin, D. Teweldebrhan, K. Ashraf, G. Liu, X. Jing, Z. Yan, R.
73. Y. P. Dan, Y. Lu, N. J. Kybert, Z. T. Luo, A. T. C. Johnson, Li et al., Small 2010, 6, 1150–1155.
Nano Letters 2009, 9, 1472–1475. 92. (a) P. K. Ang, W. Chen, A. T. S. Wee, K. P. Loh, Journal of
74. J. D. Fowler, M. J. Allen, V. C. Tung, Y. Yang, R. B. Kaner, B. the American Chemical Society 2008, 130, 14392–14393; (b)
H. Weiller, Acs Nano 2009, 3, 301–306. P. K. Ang, W. Chen, A. T. S. Wee, K. P. Loh, Journal of the
75. I. Jung, D. Dikin, S. Park, W. Cai, S. L. Mielke, R. S. Ruoff, American Chemical Society 2008, 130, 14392–14393.
Journal of Physical Chemistry C 2008, 112, 20264–20268. 93. (a) Y. Ohno, K. Maehashi, Y. Yamashiro, K. Matsumoto,
76. (a) G. H. Lu, L. E. Ocola, J. H. Chen, Applied Physics Letters Nano Letters 2009, 9, 3318–3322; (b) C. X. Lim, H. Y. Hoh,
2009, 94; (b) G. H. Lu, L. E. Ocola, J. H. Chen, Nanotechnology P. K. Ang, K. P. Loh, Analytical Chemistry (Washington,
2009, 20. DC, U. S.) 2010, 82, 7387–7393; (c) L. Ren, T. Liu, J. Guo,
77. M. Qazi, T. Vogt, G. Koley, Applied Physics Letters 2007, 91, S. Guo, X. Wang, W. Wang, Nanotechnology 2010, 21,
233101. 335701/335701–335701/335707; (d) N. Lei, P. Li, W. Xue,
78. W. Wu, Z. H. Liu, L. A. Jauregui, Q. K. Yu, R. Pillai, H. L. J. Xu, Measurement Science and Technology 2011, 22,
Cao, J. M. Bao, Y. P. Chen, S. S. Pei, Sensors and Actuators 107002/107001–107002/107006.
B—Chemical 2010, 150, 296–300. 94. X. Kang, J. Wang, H. Wu, I. A. Aksay, J. Liu, Y. Lin,
79. B. H. Chu, C. F. Lo, J. Nicolosi, C. Y. Chang, V. Chen, Biosensors and Bioelectronics 2009, 25, 901–905.
W. Strupinski, S. J. Pearton, F. Ren, Sensors and Actuators 95. Y. Ohno, K. Maehashi, K. Matsumoto, Journal of the
B—Chemical 2011, 157, 500–503. American Chemical Society 2010, 132, 18012–18013.
80. P. A. Pandey, N. R. Wilson, J. A. Covington, Sensors and 96. L. C. Clark, C. Lyons, Annals of the New York Academy of
Actuators B—Chemical 2013, 183, 478–487. Sciences 1962, 102, 29–45.
81. J. Lin, J. B. Zhong, J. R. Kyle, M. Penchev, M. Ozkan, C. S. 97. Y. Liu, D. Yu, C. Zeng, Z. Miao, L. Dai, Langmuir 2010, 26,
Ozkan, Nanotechnology 2011, 22. 6158–6160.
82. J. A. Lin, D. Teweldebrhan, K. Ashraf, G. X. Liu, X. Y. Jing, 98. J. Lu, L. T. Drzal, R. M. Worden, I. Lee, Chemistry of
Z. Yan, R. Li, M. Ozkan, R. K. Lake, A. A. Balandin, C. S. Materials 2007, 19, 6240–6246.
Ozkan, Small 2010, 6, 1150–1155. 99. H. Wu, J. Wang, X. Kang, C. Wang, D. Wang, J. Liu, I. A.
83. S. R. Guo, J. lin, M. Penchev, E. Yengel, C. S. Ozkan, M. Aksay, Y. Lin, Talanta 2009, 80, 403–406.
Ozkan, Journal of Nanoscience and Nanotechnology 2010. 100. C. Shan, H. Yang, J. Song, D. Han, A. Ivaska, L. Niu,
84. (a) N. Mohanty, V. Berry, Nano Letters 2008, 8, 4469– Analytical Chemistry 2009, 81, 2378–2382.
4476; (b) X. C. Dong, Y. M. Shi, W. Huang, P. Chen, L. J. 101. T. T. Baby, S. S. J. Aravind, T. Arockiadoss, R. B. Rakhi, S.
Li, Advanced Materials 2010, 22, 1649–1653; (c) D. Li, S. P. Ramaprabhu, Sensors and Actuators B: Chemical 2010, 145,
Song, C. H. Fan, Accounts of Chemical Research 2010, 43, 71–77.
631–641; (d) Z. Y. Yin, Q. Y. He, X. Huang, J. Zhang, S. X. 102. Y. Song, K. Qu, C. Zhao, J. Ren, X. Qu, Advanced Materials
Wu, P. Chen, G. Lu et al., Nanoscale 2012, 4, 293–297. 2010, 22, 2206–2210.
85. T. Cohen-Karni, Q. Qing, Q. Li, Y. Fang, C. M. Lieber, Nano 103. Y. Wang, Y. Shao, D. W. Matson, J. Li, Y. Lin, ACS Nano
Letters 2010, 10, 1098–1102. 2010, 4, 1790–1798.
86. (a) Y. Xu, P. E. Pehrsson, L. W. Chen, R. Zhang, W. Zhao, 104. L.-M. Lu, H.-B. Li, F. Qu, X.-B. Zhang, G.-L. Shen, R.-Q. Yu,
Journal of Physical Chemistry C 2007, 111, 8638–8643; (b) R. Biosensors and Bioelectronics 2011, 26, 3500–3504.
R. Johnson, A. T. C. Johnson, M. L. Klein, Nano Letters 2008, 105. C. X. Feng, G. Q. Xu, H. P. Liu, J. Lv, Z. X. Zheng,
8, 69–75; (c) M. Zheng, A. Jagota, E. D. Semke, B. A. Diner, Y. C. Wu, Journal of the Electrochemical Society 2014, 161,
R. S. Mclean, S. R. Lustig, R. E. Richardson, N. G. Tassi, B1–B8.
Nature Materials 2003, 2, 338–342; (d) M. Zheng, A. Jagota, 106. C. L. Sun, W. L. Cheng, T. K. Hsu, C. W. Chang, J. L. Chang, J.
M. S. Strano, A. P. Santos, P. Barone, S. G. Chou, B. A. Diner M. Zen, Electrochemistry Communications 2013, 30, 91–94.
9 Printed Graphene-Based
Electrochemical Sensors
A. Tuantranont, A. Wisitsoraat, C. Sriprachuabwong,
C. Karuwan, P. Pasakon, and D. Phokharatkul
CONTENTS
Abstract.......................................................................................................................................................................................147
9.1 Introduction.......................................................................................................................................................................147
9.2 Graphene-Based Electrochemical Sensors....................................................................................................................... 149
9.2.1 Graphene-Based Electrochemical Chemical Sensors.......................................................................................... 149
9.2.2 Graphene-Based Electrochemical Biosensors...................................................................................................... 151
9.2.2.1 Enzyme-Based Electrochemical Biosensors......................................................................................... 151
9.2.2.2 Immuno-Based Electrochemical Biosensors......................................................................................... 153
9.3 Printed Graphene-Based Electrochemical Sensors.......................................................................................................... 154
9.4 Concluding Remarks........................................................................................................................................................ 156
Acknowledgment....................................................................................................................................................................... 156
References.................................................................................................................................................................................. 156
ABSTRACT distinctively produced from graphene paste prepared by mix-

ing chemically reduced graphene oxide powder with cellulose
Graphene is a novel material for electrochemical sensors due acetate binder. The reported printed graphene-based sensors
to its excellent electrical and chemical properties and printing exhibit promising electrochemical properties with high sen-
technology has recently emerged as a low-cost and practical sitivity and good reliability suitable for commercial applica-
method for fabrication of flexible and disposable electronic tions. The printed graphene-based electrochemical sensor
devices. A combination of both technologies is highly poten- technology is rapidly growing and will earn great attention
tial for producing low-cost and disposable sensors. In this in the near future due to increasing demand of low-cost and
chapter, printed graphene-based electrochemical sensor tech- disposable electrochemical sensors.
nology is succinctly introduced. This technology generally
involves the preparation of conductive graphene-based ink or
9.1 INTRODUCTION
paste, deployment of printing process and the evaluation of
printed graphene-based materials for electrochemical detec- Graphene is a nanocarbon material with unique two-dimen-
tion. First, various types of graphene-based electrochemi- sional honeycomb lattice structure exhibiting exceptional
cal sensors including chemical sensors, enzyme-based, and physical, chemical, and electronic properties including huge
immuno-based biosensors are extensively discussed. Most specific surface area [1,2], excellent electrical/thermal con-
studies report that graphene exhibits efficient electron transfer ductivity [3–9], and biochemical compatibility [10–12]. It
and direct electrochemistry of enzymes, leading to high elec- is therefore, a potential candidate for various sensing and
trochemical sensitivity and specificity toward various kinds energy storage device applications such as chemical/biosen-
of important analytes including redox couples, metal ions, sors [13–24], supercapacitors [25–28], batteries [29–33], and
oxidizing/reducing agents, drug constituents, small biomol- fuel cells [34–36]. In particular, electrochemical sensing and
ecules, DNA, and cells. However, most reported graphene- energy storage devices are the platforms that can take most
based electrochemical sensors are fabricated by nonprinted advantages of graphene’s excellent physical, electrical, and
deposition methods such as drop/dip coating, which suffer chemical properties. Moreover, graphene is more promising
from poor material control and low productivity. Recently, a for commercial applications than other carbon-based nano-
few printed graphene-based electrochemical sensors includ- materials including carbon nanotubes (CNTs), its closest
ing inkjet-printed graphene–PEDOT:PSS and screen-printed counterpart, due to its superior electrochemical performances
graphene electrodes have been demonstrated. The graphene– [16,37,38], low hazardous properties [10–12], low production
PEDOT/PSS electrode is fabricated by inkjet printing of cost, and ease of processing [37–40] (Table 9.1).
graphene conductive ink prepared by one-step electrolytic Graphene-based electrochemical sensors have been
exfoliation from graphite electrode onto plastic or paper extensively developed by various syntheses and processing
substrate whereas the screen-printed graphene electrode is methods, which may broadly be divided into nonsolution
147
TABLE 9.1
Summary of Recent Research Studies on Graphene-Based Electrochemical Chemical Sensors
Main
Category Key Issues Important Findings References
Electrochemical sensing • Determination of electrochemical • Graphene exhibits large electrochemical specific surface [13–24]
properties of graphene specific surface area, electronic/ionic area, fast HET, high electrocatalytic activities for a variety
conductivity, electron- transfer rate/ of redox reactions, wide working potential windows, low
coefficient, and electrocatalytic electrochemical resistance, and stable electrochemistry.
activity. • Electrochemical properties of graphene are mainly attributed
• Correlation of structural/material, and to structural defects, density of defective edge site, and
electrochemical properties. chemical functional groups on graphene surface.
• Electrochemical sensing mechanisms.
Preparation of graphene- • Nonsolution methods of graphene- • Solution-route methods are generally more preferred due to [37–40]
based electrochemical based electrode fabrication. its compatibility with electrochemical systems, better
sensors • Solution methods of graphene-based electrochemical sensing properties, relatively low production
electrode fabrication. cost and suitability for graphene functionalization, and
• Modification and functionalization of large-scale production of graphene/polymer composites but
graphene for electrochemical sensing. tend to suffer from poor reproducibility compared with
• Processing and large-scale nonsolution methods.
manufacturing of graphene-based • Printing approaches is one of the most promising methods
electrochemical sensors. for production of graphene-based electrochemical devices
due to low cost, high production rate, medium resolution,
and high reproducibility.
Electrochemical sensing • Evaluation and optimization of • Graphene and its composite have been demonstrated to [49–55]
performances of electrochemical sensing performances provide high electrochemical detection performances for
graphene-based such as sensitivity, detection limit, various groups of analytes including redox couples, metal
electrochemical sensors and dynamic ranges toward various ions, oxidizing/reducing agents, drug constituents, and small
groups of analytes for graphene-based biomolecules.
materials. • Graphene-based materials are also highly promising for
• New applications of graphene-based biomedical applications for detection of DNA, cell, and
electrochemical sensors particularly other biological entities.
in life sciences.
and solution-route processes. Nonsolution methods include electrochemical sensors are summarized in Table 9.1. These
mechanical cleavage, chemical vapor deposition (CVD), epi- research issues are partly introduced in this section and will
taxial growth via ultrahigh vacuum graphitation, and other be further explained in subsequent sections, while the research
vapor-phase techniques, whereas most other routes are solu- findings will be further elaborated in the later sections.
tion based including chemical oxidation/reduction, function- Printing techniques including inkjet printing, screen print-
alization of graphene oxide (GO), liquid-phase mechanical ing, flexography, gravure, and offset lithography are well
exfoliation, electrolytic exfoliation, mechanical exfoliation known industrial processes for defining patterns on a substrate.
with intercalation by small molecules, hydro/solvothermal Their key advantages including low cost, high production rate
synthesis, total organic synthesis, and other solution tech- and medium resolution have motivated the development of
niques. Between the two approaches, solution-route meth- printed electronics. Printed electronics employ various print-
ods are generally more preferred for electrochemical sensing ing methods to print multilayer functional materials to form
due to its compatibility with electrochemical systems and electronic devices such as transistors and sensors on a sub-
suitability for large-scale production [41,42]. In addition, strate that is frequently flexible and disposable. Printed devices
chemically produced graphene offers better electrochemical will enable low-cost low-performance electronics for appli-
properties due to its numerous structural defects [43–45] and cations such as smart radio frequency identification (RFID)
functional groups [46,47], which are highly advantageous for for contactless identification in trade and transport, organic
electrochemical reactions [24,48] as well as for the formation light-emitting diode (OLED) as smart label on package in
of functionalized graphene and graphene/polymer compos- consumer products and other flexible electronic applications
ites [49–55]. Moreover, they can inexpensively produce gra- such as solar cells on vehicle roofs, electronic paper, and flexi-
phene nanosheets dispersion in aqueous solution, which can ble sensors [56]. Each printed device requires various types of
directly be further processed by various methods such as spin solution-based materials including organic semiconductors,
coating and printing to fabricate electrochemical devices. inorganic semiconductors, and metallic conductors to form
Recent research issues and findings related to graphene-based their active and passive components. Advanced functional ink
Printed Graphene-Based Electrochemical Sensors 149
materials are the keys to the success of printed electronics. and boron-doped diamond electrodes. For instance, the poten-
Graphene-based ink is a promising functional material that tial window of graphene for 0.1 M PBS buffer solution (pH
can serve as an active sensing material, conductive electrode, 7.0) is ~2.5 V. Moreover, alternating current (AC) impedance
or semiconductor. measurement reveals that graphene has significantly lower
In this chapter, we will focus on the development of printed charge-transfer resistance than other carbon-based materi-
graphene-based electrochemical sensors, which covers the als including CNTs, which can be attributed to its very high
characteristics and performances of graphene-based elec- electrical conductivity and electron mobility [24,60–62].
trochemical sensors and the fabrication and recent results of Furthermore, electrochemistry of graphene is stable over
printed graphene-based electrochemical devices. First, elec- a long period of time due to its strong basal plane structure
trocatalytic properties of graphene and detection performance [67]. Graphene has been widely applied for detection of a vari-
of various kinds of graphene-based electrochemical chemi- ety of analytes. The graphene-based electrochemical sensor
cal/biosensors are extensively described. Next, preparations may broadly be classified based on structural application into
of graphene ink materials via solution-processing methods chemical sensors and biosensors, which are extensively dis-
for printed graphene sensors are explained and the charac- cussed in the following sections.
teristics of printed graphene-based electrochemical sensors
are reported. More comprehensive details in each part can be
9.2.1 Graphene-Based Electrochemical
referred to other recent review articles devoted to each issue
[16,46,57,58]. Lastly, remarks on the present results and the Chemical Sensors
future research direction for printed graphene-based electro- Graphene-based electrochemical chemical sensor is inferred
chemical sensors are given. as the sensor that uses graphene as an active electrochemical
working electrode and may contain other materials but does
9.2 GRAPHENE-BASED not comprise any biological receptor. It may be used to detect
ELECTROCHEMICAL SENSORS conventional chemical or biochemical analytes via electron
transfer at redox reaction as illustrated in Figure 9.1. The per-
The interest of graphene as an electrochemical sensing mate- formance of graphene-based electrochemical sensors is con-
rial arises from its unique electrocatalytic properties. In addi- trolled by the way that graphene is synthesized and fabricated
tion, graphene exhibits a more consistent electrochemistry since electrochemical activity of graphene is driven by edge-
compared with other carbon materials such as CNTs due to plane defect sites and specific surface area, which depends
its absence of metallic impurities [16,59]. The key electro- primarily on the synthesis and fabrication method. Chemical-
chemical properties of graphene include fast heterogeneous route syntheses are shown to be highly suitable for fabrication
electron transfer (HET), high electrocatalytic activities for a of the graphene-based electrochemical sensor since it can pro-
variety of redox reactions, wide working potential windows, duce defective graphene structures exhibiting efficient elec-
low electrochemical resistance, and stable electrochemistry trochemistry in large scale at low cost. Graphene prepared by
[24,60–62]. Electron-transfer properties of graphene have various methods has been applied for electrochemical detec-
been evaluated from cyclic voltammetry studies with stan- tion of diverse groups of analytes including redox couples,
dard redox couples including [Ru(NH3)6]3+/2+, [Fe(CN)6]3–/4−, metal ions, oxidizing/reducing agents, drug constituents,
and Fe3+/2+, which can provide defect-insensitive, surface-sen- and small biomolecules. In the case of redox couples, elec-
sitive, and combined surface/oxide-sensitive electrocatalytic trochemically reduced graphene oxide (RGO) has shown an
activities, respectively [60]. The results show that graphene electrochemical detection performance for [Fe(CN)6]3−/4− and
exhibits low redox-peak separations (ΔEp) for [Fe(CN)6]3−/4− [Ru(NH3)6]3+/2+ superior to chemically and thermally reduced
(~65 mV at 10 mV/s) and [Ru(NH3)6]3+/2+ (60–65 mV at
100 mV/s), which indicate a fast HET on graphene electrode
for single-electron reactions and high electrocatalytic activi- Potentiostat
ties toward these redox reactions [63,64]. In addition, further Working
calculations show that graphene has considerably higher electrode Reference Counter
direct electron-transfer rate constants (k0) for [Ru(NH3)6]3+/2+ e– electrode electrode
Graphene
(0.18 cm/s), [Fe(CN)6]3−/4− (0.49 cm/s), and Fe3+/2+ than other
Matrix or Analyte
carbon-based materials. Further studies reveal that the fast
supporting
electrochemistry of graphene sheets is mainly attributed to conductive Redox
edge-plane defect sites in graphene and not the basal plane, substrate reaction
which has nearly zero electron-transfer rate [65]. The high Products
electrochemical activity of defective graphene could be asso-
ciated with its unique electronic structure having high-density
electronic states over a wide energy range [60,63,66].
In addition, graphene can operate over a wide electrochem-
ical potential window comparable with other carbon-based FIGURE 9.1 Basic structure and concept of graphene-based
electrode materials including graphite, glassy carbon (GC), electrochemical chemical sensors.
ones because electrochemical reduction may induce higher (PC) detection and demonstrated very high sensitivity and
density of active edge-plane defects [68–71]. In addition, low detection limit [85,102]. Direct detection of various bio-
nitrogen doping in graphene by thermal doping can provide molecules including β-nicotinamide adenine dinucleotide
further improvement in electrochemical performance toward (NADH), bovine serum albumin (BSA), dopamine (DA),
[Fe(CN)6]3−/4− redox couple [16]. serotonin (ST), and deoxynucleic acid (DNA) using graphene-
Hydrogen peroxide (H2O2) is an important oxidizing agent based electrodes have been widely examined.
that is a key metabolism by-product in animals and its detection NADH is a biomolecule of high interest due to its very
is of great interest. Chemically reduced (CR)-GO-modified important coenzyme function with dehydrogenases in cell
GC electrode has recently been demonstrated to exhibit high metabolic redox reactions of biological substrates such as lac-
sensitivity, wide dynamic range, and lower redox potentials tate, alcohol, or glucose [40,103]. It has been demonstrated
(0.20/0.10 V) for H2O2 detection, which are superior to graph- that NADH detection performances particularly detection
ite/GC and unmodified GC electrodes [24]. Moreover, H2O2 potential and surface fouling of GC and edge-plane pyrolytic
sensitivity and selectivity of the GO-based material can be graphite electrodes (EPPGEs) are substantially improved
considerably improved by combining GO and catalytic metal upon modification with CR-GO [63,77,97,104,105]. From
oxide such as MnO2 [72]. Gold nanoparticle (NP)/chitosan ab initio molecular dynamics NADH adsorption simulation
and silver NPs are other useful additives for improving elec- and high-resolution x-ray photoelectron spectroscopy and
trocatalytical activity of graphene toward H2O2 and O2 [15,73]. electrochemical activity toward NADH are enhanced by the
Thus, graphene-based electrodes are promising candidates for increased NAD+ adsorption due to high-density oxygen-con-
various electroanalytical applications involving H2O2 sensing taining groups such as carboxylic on graphene edge-plane
such as food analysis, environmental monitoring, and enzy- defect sites present in CR-GO [103,106]. In another report,
matic electrochemical biosensors. Furthermore, CR-GO has graphene noncovalently functionalized with methylene green
also shown useful electrochemical performances for hydra- (MG) exhibits a very low NADH oxidation potential of 0.14 V
zine, an important reducing agent, with relatively low overpo- and high selectivity, which may be attributed to large effec-
tential and good detection selectivity [74,75]. tive surface area due to excellent graphene dispersion [98].
Metal ion detectors are of great interest for the environ- BSA, another biomolecule for unspecific binding encapsula-
mental analysis and graphene-based electrochemical sensors tion, is commonly found in the biological system and is also of
have recently been employed for the detection of various high interest. Recently, BSA detection based on positive-ion
metal ions including Pb2+, Cd2+, Na+, K+, Ca2+, Mg2+, and induction in phthalate solution and detection using graphene–
Hg2+. RGO-incorporated Nafion layer has been studied for ISFET electrochemical sensor has been developed and dem-
the detection of Pb2+ and Cd2+ and demonstrated high sensi- onstrated to offer high detection sensitivity [81].
tivity as well as good selectivity for Pb2+ and Cd2+ [76–78]. Neurotransmitters including dopamine (DA) and serotonin
In addition, study results show that the graphene–Nafion (ST) for central nervous, renal, hormonal, and cardiovascular
electrode gives superior sensitivity and detection limits for systems have also been widely investigated using graphene-
both analytes to Nafion film-modified bismuth electrode based electrochemical sensors. For instance, multilayer gra-
and ordered mesoporous carbon-coated GCE [79,80]. Huge phene nanoflake films (MGNFs) based electrode synthesized
specific surface area and electrical conductivity of graphene by catalyst-free microwave plasma-enhanced chemical vapor
can contribute to strongly enhanced adsorbed target ions, deposition (PECVD) has shown excellent DA detection per-
elevated charge transfer, and reduced fouling effect of non- formance with high sensitivity, low detection limit (0.17 μM),
conductive Nafion, which is required as an ion-selective filter and good selectivity with well-separated oxidation peaks
for these heavy metals [76,77]. Moreover, graphene has been from two important interfering species including ascorbic
applied for the detection of other ions including H+ (pH sen- acid (AA) and uric acid (UA) [64,107]. Similarly, graphene
sor) [81], Na+, K+, Ca2+, Mg2+, and Hg2+ [82,83] based on ion- prepared by chemical-route process has demonstrated a wide
sensitive (IS) field effect transistor (FET) platform, which is detection range of 5 mM–200 mM, high sensitivity and good
another electrochemical sensor type that utilizes graphene– selectivity against AA and ST [84,88]. The DA detection
FET structure with external reference electrode acting as a performance of graphene-based electrodes are superior to
gate that induces charges via electrochemical reaction from CNT-based ones, which may be explained by higher edge-
ions, resulting in the modulation of electrical potential and plane sites/defects, better electron transport to substrate via
drain current of graphene channel. High sensitivity and low graphene nanoconnector [64], larger sp2-like plane surface
detection limit has been achieved from the reported gra- area, and stronger π–π stacking interaction of DA and gra-
phene–ISFET sensors. phene [88,108].
Biomedical and biochemical analytes including drug con- DNA sensors have been demonstrated based on direct
stituents and small biomolecules are widely studied with DNA oxidation by graphene electrode [24]. CR-GO-modified
graphene-based electrochemical sensors. In the group of drug GC electrode has demonstrated a unique DNA detection
constituents, chemically RGO has shown excellent sensitiv- property not present in conventional electrodes with an abil-
ity as well as low detection limit for chlorpromazine (CPZ) ity to detect four free DNA bases including guanine (G), ade-
and ascorbic acid (AA) detections [86,87] whereas ther- nine (A), thymine (T), and cytosine (C) simultaneously, with
mally reduced graphene has been applied for paracetamol four well-separated oxidation peaks measured by differential
pulse voltammetry (DPV). In addition, the amounts of four some carbon–metal oxide nanocomposites, having efficient
DNA bases in single-stranded (ss)-DNAs and double-stranded electron-transfer characteristics [40,114–117].
(ds)-DNAs can be concurrently perceived by the graphene- Direct electrochemistry of graphene prepared by different
based electrode with no pretreatment process, which enables methods on enzymes has recently been extensively studied
label-free detection of a single-nucleotide polymorphism [68]. Glucose oxidase (GOD) is the most widely used enzyme
(SNP) for short nonhybridized oligomers with a definite model for direct electrochemistry of various electrode materi-
sequence. CR-GO’s antifouling properties and fast electron- als and graphene as depicted in Figure 9.2 [14,24,118–121]. The
transfer kinetics, resulting from the large number of edge- study results indicate that graphene exhibits efficient direct
plane defects and many active functional groups, are believed electrochemistry for GOD since graphene in GOD induces
to contribute to specific reactivity between the DNA species a pair of well-defined reversible redox cyclic voltammetric
and the CR-GO [106]. Moreover, DNA detection performance (CV) peaks of flavin adenine dinucleotide (FAD), the redox
of graphene is demonstrated to be further enhanced by gold active center in GOD, which would be absent if graphene is
NP decoration using electrochemical synthesis and the syn- not present. In addition, low redox peak-to-peak separation
ergistic electron-transfer effect between graphene and gold of ~69 mV, close-to-ideal FAD/FADH2 formal potential (the
NPs is believed to be a possible mechanism [100]. Similarly, average potential of cathodic and anodic peaks) of ~0.43 V
functionalization of graphene with ionic-liquid (IL) material versus Ag/AgCl, high electron-transfer-rate constant (ks)
is reported to be another effective means to enhance detec- of ~2.8 s/1, close to unity ratio of cathodic to anodic peak
tion performance of DNA bases and DA due to the synergistic current and the linear current density relationship with scan
effect of high ionic conductivity of IL and high graphene con- rates are obtained, indicating reversible and surface-limited
ductivity as well as due to the ability to form well dispersed redox process of GOD on graphene electrodes [14,24,118–
graphene in IL [101]. The key data of high-performance gra- 122]. Furthermore, the direct electrochemistry of GOD on
phene-based electrochemical chemical sensors are summa- graphene-based electrodes are repeatable, stable with 95%
rized in Table 9.2. retention over 1 week storage and considerably much better
than most of the reported CNT-based electrodes due to its
high enzyme loading on its large surface area [123–125].
9.2.2 Graphene-Based Electrochemical Biosensors
The efficient direct electrochemistry of graphene sug-
Graphene-based electrochemical biosensor is the electro- gests that enzymatic graphene-based biosensor would be
chemical sensor that integrated biological receptors such effective, however graphene preparation, surface function-
as enzyme, antibody, and complementary ssDNA with the alization, and enzyme immobilization methods play very
graphene-based sensing material. It may be divided into two important roles since the performances considerably depend
main types including enzyme-based and immuno-based elec- on the amount of enzyme loading and final effective sur-
trochemical biosensors. face area. Functionalization with polyethylenimine–poly-
vinylpyrrolidone and GOD enzyme immobilization with IL
9.2.2.1 Enzyme-Based Electrochemical Biosensors on CR-GO is reported to be a successful means to achieve
Enzyme-based electrochemical biosensors employ immobi- high-performance glucose sensor with a wide dynamic range
lized enzyme on electrode surface as the receptor to interact (2–14 mM), good sensitivity and high stability [118]. In addi-
with target biomolecules and generate electroactive products tion, coimmobilization of alcohol dehydrogenase (ADH)
that produce electrochemical responses as schematically and GOD on CR-GO [24] results in substantially improved
illustrated for graphene in Figure 9.2. Oxidase is the most glucose-sensing performances with high sensitivity, very fast
widely used enzyme for electrochemical biosensors that pro- response (9 s response time), and low detection limit of 2 μM
duce H2O2 as the main electroactive by-product. As illustrated (S/N ~3), which is superior to several other carbon-based
in the previous section, graphene-based electrochemical is electrodes such as the CNTs [126–128], highly ordered meso-
effective for H2O2 sensing and it should thus be potential for porous carbon [129], carbon nanofibers [130], and exfoliated
enzyme-based biosensing. However, the high detection sensi- graphite nanoplatelets [120]. Moreover, GOD immobiliza-
tivity of the electrode for the enzymatic product may not lead tion with chitosan is reported to be another effective means
to an efficient enzyme-based electrochemical sensor if there to yield efficient glucose biosensors with excellent sensitivity
is no electron transfer between the active center of the enzyme and long-term stability, which can be attributed to improved
and the electrode. In common electrodes, direct electron trans- graphene dispersion and enzyme immobilization capacity by
fer can hardly occur since electron from the electrode cannot chitosan [14].
reach the deep hydrophobic cavity of the enzyme’s active cen- Graphene/metal NPs system is another promising approach
ter molecules [109]. In such cases, the electron transfer can be for enzymatic biosensing. GOD/graphene/PtNPs/chitosan glu-
induced via intermediate electron mediators, which can make cose biosensor has been demonstrated with enhanced sensi-
the overall reaction quite slow. It is therefore desirable to have tivity and a low glucose detection limit of 0.6 μM, which may
direct electron transfer between the electrode and the enzyme be explained by the enlarged surface area and conductivity of
active center. This process is called direct electrochemistry graphene–metal NPs because the NPs act as spacers prevent-
of enzyme [110–113] and it can be achieved with nanostruc- ing restacking or agglomeration of graphene sheets [15,119].
tured electrode materials including CNTs, metal NPs, and Since the direct detection of H2O2, the oxidase by-product,
TABLE 9.2
Summary of High-Performance Graphene-Based Electrochemical Chemical Sensors
Minimum Reported
Active Material Synthesis Method Analytes Measurement Concentration References
RGO CO–CR Fe(CN)63−/4− CV, LSV ~mM [24]
RGO CO–CR Ru(NH3)63+/2+, Fe(CN)63−/4− CV ~mM [84]
RGO CO–CR Ru(NH3)63+/2+, Fe(CN)63−/4−, Fe3+/2+ CV ~mM [63]
RGO CO–TR PC, AA CV, SWV ~10 nM, 10 μM [85]
RGO CO–TR Fe(CN)63−/4−, CPZ CV, DPV ~0.1 mM, ~10 nM [86]
RGO CO–TR Fe(CN)63−/4−, AA CV, I versus t ~1 mM, ~0.1 μM [87]
RGO CO–CR, CO–ER Fe(CN)63−/4− CV ~mM [68]
N-doped RGO CO–TR Fe(CN)63−/4− CV ~mM [16]
RGO–Nafion CO–CR Pb2+, Cd2+ DPASV ~30 nM [76,77]
G ME H+ ISFET: I versus t ~0.01 pH [81]
G ME Na+ ISFET: I versus t ~nM [82]
RGO–CaM/MT-II CO–TR Na+, K+, Ca2+, Mg2+, Hg2+ ISFET: I versus t ~nM [83]
RGO–AuNPs–Chit CO–CR H2O2, O2, Glucose CV ~0.1 mM [15]
~0.1 mM
RGO, RGO/ Nafion, CO–CR OPH CV and I versus t ~μM, [18]
and fatigued RGON ~0.1 μM, ~0.1 μM
RGO CO–CR DNA base, DA, AA, UA, AP, H2O2 DPV ~10 μg/mL, ~mM [24]
~10 μM
RGO CO–CR DA DPV, I versus t ~μM [88]
RGO CO–CR DA, AA, ST CV, DPV ~mM [84]
G MWPECVD DA, AA, UA CV ~mM [64]
CD-RGO CO–CR DA, UA, AP, rutin, AP, TS, TP, NP, TD CV, DPV ~20 nM [89]
RGO–Nafion CO–CR Aloe emodin CV, DPV ~2 nM [90]
RGO–Chit CO–CR 4-Aminophenol CV, DPV ~60 nM [91]
FeTSPc-RGO CO–CR Isoniazid acid, UA CV, DPV ~0.6 nM [92]
RGO CO–CR, CO–ER NADH CV ~mM [68]
G ME BSA ISFET: I versus t ~10 nM [81]
RGO CO–CR Rutin CV, DPSAV ~20 nM [93]
RGO–Chit–PAC CO–CR Rutin CV, DPV ~0.6 nM [94]
G-PEDOT:PSS EE SAL, H2O2, NADH, Fe(CN)63−/4− CV ~0.1 μM [95,96]
RGO CO–CR H2O2, NADH CV ~0.1 mM, ~mM [97]
RGO-MG CO–CR NADH CV ~mM [98]
RGO CO–CR H2O2 CV, DPV ~0.05 μM [99]
RGO–AuNPs CO–ER DNA CV ~μM [100]
IL-RGO CO–CR DNA bases, DA CV ~50 nM, ~50 nM [101]
GO–MnO2 CO H2O2 CV and I versus t ~μM [72]
Note: CO = chemical oxidation, CR = chemical reduction, TR = thermal reduction, ER = electrochemical reduction, ME = mechanical exfoliation,
EE = electrolytic exfoliation, CV = cyclic voltammetry, LSV = linear sweep voltametry, SWV = square wave voltametry, DPSAV = differential
pulse anodic stripping voltammetry, PC = paracetamol, AP = acetaminophen, TS = tyrosine, TP = tryptophan, NP = norepinephrine, TD = thio-
ridazine, CPZ = chlorpromazine, AA = ascorbic acid, TMA = trimethylamine, DNT = dinitrotoulene, CD = cyclodextrin, HCN = hydrogen
cyanide, PAC = poly(amidoamine), CEES = chloroethyl ethyl sulfide, DMMP = dimethyl methyl phosphonate, CaM = calmodulin,
MT-II = metallothionein, FeTSPc = iron-tetrasulfophthalocyanine, R = resistance, I = current, f = resonance frequency.
is not highly sensitive at low working potential, its biologi- dodecyl benzene sulfonate functionalized graphene sheets
cal enzyme, horseradish peroxidase (HRP), is often used to are self-assembled in aqueous solution into novel hierarchi-
increase the sensitivity and some graphene-based enzymatic cal nanostructures by means of electrostatic attraction [23].
H2O2 sensors have been developed [131]. For instance, a Moreover, Au NP/sulfonated graphene/HRP/chitosan-modi-
graphene-nanoplatelet H2O2 biosensor is directly fabricated fied GC electrode has been prepared by coimmobilization and
from graphite with the aid of 1,3,6,8-pyrenetetrasulfonic acid electrodeposition for ultrasensitive H2O2 sensing [134]. Apart
(TPA) [132] and HRP [133]. In addition, HRP and sodium from GOD and HRP, dehydrogenase that produces NADH
rate. Thus, graphene electrode with the effective immobiliza-

Potentiostat
tion method is a promising route to an efficient electrochemi-
Working cal immunosensor.
electrode Reference Counter Various approaches have been employed to efficiently
electrode electrode
Graphene immobilize the different kinds of antibody on the graphene
surface. First, NPs of metals such as gold are reported to be
Enzyme
e- Analyte
effective for enhancing immobilization of enzymes and anti-
Direct
electro- bodies of various biosystems such as the human IgG [139]
Enzymatic
chemistry reaction and the cancer biomarker, alpha-fetoprotein (AFP) [140] on
Matrix or
Products
the graphene surface, which result in highly sensitive elec-
supporting trochemical immunosensors. An extension of this approach
conductive
substrate
employs a nanolabel with chitosan-enwrapped graphene as a
core and multi-Au NPs as a shell for the conjugation of HRP-
anti-carcinoembryonic antigen (CEA), which is used as AB2
FIGURE 9.2 Basic structure and concept of enzyme-based gra- for the electrochemical immunosensor that exhibits high CEA
phene electrochemical sensors.
detection performances owing to huge binding surface area
provided by the graphene–gold NP nanolabel [141].
as by-product is another widely used enzyme for biosensing. Conversely, functionalized graphene on screen-printed
Ethanol-dehydrogenase biosensor-based immobilized ADH– carbon electrode (SPCE) with effectively immobilized AB1
graphene electrodes have been demonstrated to give fast etha- of AFP has been used in conjunction with multi-enzyme-
nol response, wide dynamic range, and low detection limit labeled carbon nanospheres (CNSs) to yield an ultrasensi-
superior to ADH–graphite and ADH immobilized GC elec- tive AFP immunosensor with a linear dynamic range of
trodes [24]. Excellent catalytic activity of graphene toward the 0.05–6 ng/mL and a low detection limit of 0.02 ng/mL [142].
NADH and the effective transport of ethanol and the NADH The ultrahigh performances can be attributed to signal ampli-
products through enzyme immobilized graphene matrices are fication via multiple binding events of HRP–AB2-CNSs on
believed to be responsible for the observed performance [135]. effectively immobilized graphene surface. Similarly, gra-
phene-modified electrode is combined with labels based on
9.2.2.2 Immuno-Based Electrochemical Biosensors IL, mesoporous silicas SBA-15, and HRP to achieve ultra-
Immuno-based electrochemical biosensor employs anti- sensitive electrochemical immunsensors for the detection
body conjugated with enzyme as the biological receptor of breast cancer susceptibility gene (BRCA1) [143]. In addi-
and detects the change of electrochemical activity due to tion, similar multienzyme amplification strategy has been
the specific binding reaction of an antibody and an antigen. applied for ultrasensitive detection of phosphoylated p53 at
Electrochemical immunosensor is commonly implemented Ser392 (phospho-p53392) but using GO with bioconjugates of
based on a sandwich scheme, in which a primary antibody HRP and p53392 antibody (p53392-AB2) as a nanolabel. The
(AB1) is immobilized onto the electrode surface; the antigen GO-based nanolabels will be captured onto the electrode sur-
(AG) then specifically binds to the AB1 and another second- face via a cascaded immunoreaction, generating an amplified
ary antibody (AB2) labeled with an enzyme finally attaches to electrocatalytic signal via the decrease of enzymatic oxidiza-
the other side of the AG as depicted in Figure 9.3 [136–138]. tion of thionine due to the occurrence of H2O2 [144].
The performances of electrochemical immunosensors depend In another approach, AB1 of a prostate-specific antigen
primarily on the antibody loading capacity and electrode is covalently conjugated via succinimidyl ester bonding to
properties such as specific surface area and electron-transfer 1-pyrenebutanoic acid organic molecules, which are noncova-
lently bound to graphene via π–π interaction while AB2 con-
jugated with HRP enzyme is covalently bound to the amino
Potentiostat group of thionine mediator, which is physically adsorbed on
Working graphene. This approach yields a highly sensitive immuno-
electrode Reference Counter sensor with regeneration ability due to strong binding of AB1
electrode electrode
Graphene on graphene and is promising for advanced biological sens-
Antigen ing applications [145]. Most electrochemical immunosensors
AB1 (analyte)
-
Enzyme require enzyme labeling, which results in complicated assay
ee
– Immuno-
Enzymatic procedure, and thus a label-free method is preferred in many
reaction
Matrix or Products practical applications. Recently, a label-free electrochemical
AB2 immunosensor for AFP detection is developed by immobiliz-
supporting
conductive ing anti-AFP on GCE modified with graphene and thionine
substrate and demonstrates high sensitivity, wide dynamic range (0.05–
2.00 ng/mL), and low detection limit (5.77 pg/mL, S/N = 3)
FIGURE 9.3 Basic structure and concept of immuno-based [146]. The key data of high-performance graphene-based
graphene electrochemical sensors. electrochemical biosensors are listed in Table 9.3.
TABLE 9.3
Summary of High-Performance Graphene-Based Electrochemical Biosensors
Synthesis Minimum Reported
Active Material Method Analytes Measurement Concentration References
RGO–Chit–GOx CO–TR Glucose CV ~10 μM [14]
RGO–AuNPs–Chit–GOx CO–CR H2O2, O2, and glucose CV ~0.1 mM [15]
~0.1 mM
RGO, RGO/ Nafion, and fatigued RGON CO–CR OPH CV and I versus t ~μM, [18]
~0.1 μM, ~0.1 μM
SDBS- functionalized RGO–HRP CO–CR H2O2 I versus t ~10 μM [23]
RGO–ADH CO–CR NADH, ethanol DPV, I versus t ~10 μM, [24]
~0.1 mM
RGO–GOx CO–CR Glucose I versus t ~μM [24]
PtNPs–RGO/Chit–GOx CO–CR Glucose CV and I versus t ~0.1 μM [119]
RGO–GOx CO–ER Glucose CV ~10 μM [121]
G-TPA–HRP ME-SL H2O2 CV and I versus t ~0.1 μM [131]
Sulfonated graphene–chitosan–HRP CO–CR H2O2 CV and I versus t ~1.7 μM [134]
SDBS–graphene–HRP CO–CR H2O2 CV and I versus t ~0.1 μM [23]
ssDNA–graphene–HRP CO–CR H2O2 CV and I versus t ~38.5 μM [147]
Au-NPs-lCs-RGO–A-H-IgG CO–CR H-IgG CV and I versus t ~ng/mL [139]
Au–NPs–RGO–A-AFP CO–CR AFP CV and I versus t ~ng/mL [140]
AuNPs–Chit–RGO–HRP–A-CEA CO–CR CEA CV and I versus t ~ng/mL [141]
CNS–RGO–HRP–A-AFP CO–CR AFP CV and I versus t ~ng/mL [142]
RGO–IL–MPS–HRP–A-BRCA1 CO–CR BRCA1 CV and I versus t ~ng/mL [143]
TN–GO–HRP–A-p53392 CO–CR p53392 CV and I versus t ~ng/mL [144]
RGO–PBT–TN–HRP–A-PSA CO–CR PSA CV and I versus t ~ng/mL [145]
RGO–TN–A-AFP CO–CR AFP CV and I versus t ~ng/mL [146]
Note: IL = ionic liquid, NPs = nanoparticles, DPV = differential pulse voltammetry, ME-SL =
mechanical exfoliation in solution,
MWPECVD = microwave plasma CVD, Au-NPs = gold nanoparticles, CNS = carbon nanosphere, ionic liquid, MPS = mesoporous sili-
cas SBA-15, TPA = tetrasodium1,3,6,8-pyrenetetrasulfonic acid, Chit = chitosan, DA = dopamine, AA = ascorbic acid, UA = uric acid,
AP = acetaminophen, OPH = organophosphate, A- = anti-, H-IgG = human immunoglobin, AFP = alpha-fetoprotein, CEA = carcinoem-
bryonic antigen, BRCA1 = breast cancer susceptibility gene, p53392 = phosphoylated p53 at Ser392, PBT = 1-pyrenebutanoic acid,
PSA = prostate-specific antigen, TCC = thiol-containing compound, TN = thionine, IgE = immunoglobulin E, BSA = bovine serum albu-
min, PL = photoluminescence, CL = chemiluminscence.
9.3 PRINTED GRAPHENE-BASED
ELECTROCHEMICAL SENSORS
Most reported graphene-based electrochemical chemical sen- Contact of
sors are fabricated by various common deposition or modifica- working electrode
tion methods including drop-coating, dipping reaction, spray Insulating
coating, and electrodeposition. These processes suffer from layer
poor control of deposition materials and inability to directly Reference Counter
form electrode patterns leading to poor reproducibility when electrode electrode
produced in large scale. Printing methods, which produce pat-
Graphene-based
terned material by means of inked-type transfer processes, working electrode
are potential solutions to these shortcomings and are suitable
for mass production of fully integrated electrochemical sen-
sors at low cost. A typical design for the graphene-based elec-
trochemical sensors may be depicted as illustrated in Figure Substrate
9.4. It comprises multiprinted layers of working, reference,
and counterelectrodes on an insulating substrate, which can
be printed by various printing technology. Printing technol-
ogy may be classified into sheet-based and roll-to-roll-based FIGURE 9.4 Typical design of fully integrated printed graphene
schemes. Sheet-based approaches including screen, inkjet, electrochemical sensors.
aerosol jet, stamping, nanoimprinting, and dot-matrix print- inclusion considerably enhances electrochemical responses
ing are serial printing processes, which are suitable for toward the analytes, with 10% being an optimum graphene
low-volume and good-precision work. Roll-based schemes concentration. The oxidation signals for H2O2, NADH, and
including gravure, offset, flexographic, electrophotographic, K4Fe(CN)6 of the SPGE with the optimal graphene concen-
and laser printing are based on continuous roll-to-roll trans- tration are found to be ~2.0, ~1.8, and ~1.7 times higher than
fer processes, which allow rapid mass production over a large those of a screen-printed carbon paste electrode, respectively.
area with fair pattern resolution [148]. Printing method is gen- SPGCE can be fabricated at relatively lower cost than SPGE.
erally selected based on requirements of printed layers, prop- Thus, SPGCE is a promising alternative for low-cost and
erties of printed materials as well as economic and technical advanced electrochemical sensing applications.
aspects of printed products. Subsequently, inkjet printing has been used to modify
Few printing processes have recently been applied for the SCPEs with graphene–poly (3,4 ethylenedioxythiophene):po
fabrication of graphene-based electrochemical sensors, which lystyrene-sulfonic acid (PEDOT:PSS) [95,96]. Inkjet printing,
may be based on graphene only or composite of graphene and which produces patterned layers by controlling droplets of ink
other materials such as conductive polymer, which is widely onto a substrate through a nozzle by either thermal or piezo-
used due to its printability, mechanical flexibility, and suitable electric actuators, is presently the most widely used method
electrochemical functionalities [149]. First, screen printing in fabrication printed devices due to its ability to use easy-to-
has been applied for the fabrication of graphene-based elec- make low-viscosity inks, flexibility, and versatility [157–161].
trochemical sensors [150]. Screen printing is a well-developed However, it suffers from low throughput, limited resolution
printing method that forms a printed layer by pressing a vis- (~50 µm), poor thickness homogeneity, and inability to use
cous ink or paste through a patterned screen using a squeegee. high-viscosity inks such as organic dielectrics or high-density
The patterned screen is generally fabricated using a photo- inks such as metal pastes [158]. The graphene–PEDOT:PSS
sensitive screen coating and standard photolithographic pro- ink has been produced by one-step electrolytic exfoliation.
cess. The resolution of screen printing is primarily limited by The electrolytic exfoliation is conducted by immersing two
screen making precision and particle size of the paste mate- graphite rods in PEDOT:PSS electrolyte with an applied
rial to be around 30–50 μm. The resolution is acceptable for potential of 5–12 V between them for 8–40 h. Graphite oxi-
electrochemical sensor fabrication whereas the thickness can dation takes place on the positive electrode (anode) where the
be well controlled. It is thus attractive for the fabrication of graphene layers are exfoliated as black dispersion in the elec-
low-cost electrochemical sensors. However, the application of trolyte. Large agglomerates can be separated from the dis-
this method for sensor fabrication requires the development persed product by centrifugation. The size and concentration
of high-viscosity (typically > 1000 cP) screen-printing paste of graphene are found to increase with applied voltage and
[151]. To make high-viscosity inks, polymeric binders must be time. The main advantages of the method include the ability to
added but binders adversely affect the properties of printed directly form stable graphene nanosheets in conducting poly-
materials [152]. mer solutions, low cost, and ease of large-scale production.
Nevertheless, the viscous graphene ink for screen printing The produced graphene-dispersed PEDOT:PSS conduc-
has successfully been formulated by thorough mixing of RGO tive ink has been inkjet printed with varying numbers of
powder and cellulose acetate binder dissolved in cyclohexane printed layers on flexible substrates such as PVC and paper
solvent [150]. The RGO powder is prepared by chemical oxida- to modify SPCE [95,96]. The inkjet-printed graphene–
tion of graphite powder, ultrasonication, and chemical reduc- PEDOT:PSS SCPEs have been applied for electrochemi-
tion using hydrazine [153,154] based on Hummers’ method cal detection of various biochemical analytes, including
[155]. The graphene paste is screen printed on polyvinyl chlo- H2O2, NADH, [Fe(CN)6]3−/4−, and salbutamol. The results
ride (PVC) substrate and baked at 70°C for 30 min to form demonstrated that the electrochemical responses for H2O2,
the screen-printed graphene electrode (SPGE). SPGE is then NADH, and [Fe(CN)6]3−/4− of PEDOT:PSS-modified and GP–
employed for simultaneous determination of ascorbic acid PEDOT:PSS-modified SPCEs are 2–4 and 3–13 times higher
(AA), DA, and uric acid (UA) by differential pulse voltam- than those of unmodified SPCE, respectively. Moreover, the
metry. SPGEs exhibit three well-defined and fully resolved electrochemical responses for salbutamol of PEDOT:PSS-
anodic peaks that allow their sensitive and selective deter- modified and GP-PEDOT:PSS-modified SPCEs are 30 and
minations in real biological samples. In addition, it has wide 150 times higher than that of unmodified SPCE. In addition,
linear dynamic ranges for AA, DA, and UA of 4.0–4500 μM, the optimum number of inkjet-printed layer is reported to
0.5–2000 μM, and 0.8–2500 μM, respectively, and corre- be ~5. Moreover, excellent analytical features with relatively
sponding detection limits (S/N = 3) of 0.95 μM, 0.12 μM, wide dynamic ranges, high sensitivities, low detection limits,
and 0.20 μM, respectively. Moreover, SPGNE possesses high and good reproducibility have been demonstrated. Thus, the
determination reproducibility and stability, which can be inkjet printing is a promising method for practical electro-
attributed to the well-controlled screen printing process. More chemical sensor fabrications.
recently, screen-printed graphene–carbon paste electrode In another development, graphene electrode has been pre-
(SPGCE) has been fabricated by simple mixing of graphene pared by inkjet printing of GO ink and thermal reduction
prepared by electrolytically exfoliated graphene into com- [162]. Single-layer and few-layer GO solutions are produced
mercial SPCE [156]. It has been demonstrated that graphene by sonication of the GO powders produced by Hummer’s
method in deionized water for 2 h and filtration to remove ACKNOWLEDGMENT

large-size particulates. The aqueous GO inks are then ink-
jet printed with varying numbers of printed layers on two Adisorn Tuantranont and Anurat Wisitsoraat acknowledge
polymeric substrates including poly(ethylene-terephthalate) Thailand Research Fund (TRF) and all financial support
(PET) and polyimide (PI) at room temperature. Prior to print- from the National Science and Technology Development
ing, the substrates are cleaned in acetone/ethanol and then Agency (NSTDA) and from the National Research Council of
surface treated to improve wetting ability by dip coating in Thailand (NRCT).
PSS and polyethyleneimine (PEI) solutions, respectively. The
GO films are then thermally reduced under Ar/H2 ambient REFERENCES
at 200–400°C for 3 h. To improve the electrochemical per-
1. Geim, A. K., Novoselov, K. S. The rise of graphene. Nature
formance of the graphene electrode, ferrocene-functionalized Materials 2007, 6 (3), 183–191.
reduced graphene oxide (Fc-RGO) is modified on the printed 2. Park, S., Ruoff, R. S. Chemical methods for the production of
electrode by drop casting. Fc-RGO is prepared by sonicating graphenes. Nature Nanotechnology 2009, 4 (4), 217–224.
RGO prepared by chemical oxidation/reduction in Fc solution, 3. Service, R. F. Carbon sheets an atom thick give rise to
which is a mixture of water and dimethylformamide (DMF), graphene dreams. Science 2009, 324 (5929), 875–876 + 877.
for 2 days. It is then centrifuged, washed, and redispersed in 4. Kim, K. S., Zhao, Y., Jang, H., Lee, S. Y., Kim, J. M., Ahn, J. H.,
DMF. The modified electrode has been applied for electro- Kim, P., Choi, J. Y., Hong, B. H. Large-scale pattern growth of
graphene films for stretchable transparent e lectrodes. Nature
chemical detection of H2O2. The results show that Fc-RGO 2009, 457 (7230), 706–710.
exhibits high catalytic activity to H2O2 comparable with some 5. Bae, S., Kim, H., Lee, Y., Xu, X., Park, J. S., Zheng, Y.,
enzymatic H2O2 sensors and much better performance than Balakrishnan, J. et al. Roll-to-roll production of 30-inch gra-
the unmodified graphene electrode. phene films for transparent electrodes. Nature Nanotechnology
2010, 5 (8), 574–578.
6. Eda, G., Lin, Y. Y., Miller, S., Chen, C. W., Su, W. F.,
9.4 CONCLUDING REMARKS Chhowalla, M. Transparent and conducting electrodes for
organic electronics from reduced graphene oxide. Applied
In conclusion, graphene’s unusual structural, electrical, and Physics Letters 2008, 92 (23), 233305 (2551).
chemical properties make it highly attractive for electrochemi- 7. Mattevi, C., Eda, G., Agnoli, S., Miller, S., Mkhoyan, K. A.,
cal sensing applications. Graphene-based electrochemical sen- Celik, O., Mastrogiovanni, D., Cranozzi, C., Carfunkel, E.,
sors may similarly be divided into two main types including Chhowalla, M. Evolution of electrical, chemical, and struc-
chemical and biosensors based on the absence and presence tural properties of transparent and conducting chemically
derived craphene thin films. Advanced Functional Materials
of biological receptors, respectively. Functionalized RGOs
2009, 19 (16), 2577–2583.
prepared by solution-route processes have shown superior per- 8. Balandin, A. A., Ghosh, S., Bao, W., Calizo, I., Teweldebrhan,
formance compared with most other carbon-based electrodes D., Miao, F., Lau, C. N. Superior thermal conductivity of
in direct electrochemistry as well as the enzyme-based and single-layer graphene. Nano Letters 2008, 8 (3), 902–907.
immuno-based electrochemistry due to its very high electron- 9. Yu, A., Ramesh, P., Sun, X., Bekyarova, E., Itkis, M. E.,
transfer rate, edge-plane defect, its excellent direct electro- Haddon, R. C. Enhanced thermal conductivity in a hybrid
chemical oxidation of small biomolecules, and direct enzyme graphite nanoplatelet—Carbon nanotube filler for epoxy com-
posites. Advanced Materials 2008, 20 (24), 4740–4744.
electrochemistry. Graphene electrochemical sensors for detec-
10. Park, S., Mohanty, N., Suk, J. W., Nagaraja, A., An, J.,
tion of a wide range of nonbiological and biological analytes Piner, R. D., Cai, W., Dreyer, D. R., Berry, V., Ruoff, R. S.
including redox couples, metal ions, oxidizing/reducing agents, Biocompatible, robust free-standing paper composed of a
drug constituents, small biomolecules, DNA, and cells have TWEEN/graphene composite. Advanced Materials 2010, 22
been reported with high sensitivity and specificity. However, (15), 1736–1740.
most reported graphene-based electrochemical sensors are fab- 11. Wang, K., Ruan, J., Song, H., Zhang, J., Wo, Y., Guo, S., Cui,
ricated based on nonprinted deposition or modification meth- D. Biocompatibility of graphene oxide. Nanoscale Research
Letters 2010, 6 (1), 1–8.
ods, which suffer from poor control of deposition materials
12. Yan, X., Chen, J., Yang, J., Xue, Q., Miele, P. Fabrication of free-
and inability to directly form electrode patterns leading to poor standing, electrochemically active, and biocompatible graphene
reproducibility when produced in large scale. Printing methods oxide–polyaniline and graphene–polyaniline hybrid papers.
including sheet-based and roll-based techniques are promis- ACS Applied Materials and Interfaces 2010, 2 (9), 2521–2529.
ing for economical large-scale fabrication of graphene-based 13. Ang, P. K., Chen, W., Wee, A. T. S., Kian, P. L. Solution-gated
electrochemical sensors since it can directly process graphene epitaxial graphene as pH sensor. Journal of the American
ink produced by solution-based approaches. Nevertheless, the Chemical Society 2008, 130 (44), 14392–14393.
development of printed graphene-based electrochemical sen- 14. Kang, X., Wang, J., Wu, H., Aksay, I. A., Liu, J., Lin, Y.
Glucose Oxidase–graphene–chitosan modified electrode for
sor is still at early stage with few reports that apply inkjet and direct electrochemistry and glucose sensing. Biosensors and
screen printing for the fabrication of graphene-based sensors. Bioelectronics 2009, 25 (4), 901–905.
The research in this area is rapidly growing and will earn great 15. Shan, C., Yang, H., Han, D., Zhang, Q., Ivaska, A., Niu, L.
attention in the near future due to increasing demand of low- Graphene/AuNPs/chitosan nanocomposites film for glucose bio-
cost and disposable electrochemical sensors. sensing. Biosensors and Bioelectronics 2009, 25 (5), 1070–1074.
16. Shao, Y., Wang, J., Wu, H., Liu, J., Aksay, I. A., Lin, Y. 34. Seger, B., Kamat, P. V. Electrocatalytically active graphene–
Graphene based electrochemical sensors and biosensors: A platinum nanocomposites. role of 2-D carbon support in pem
review. Electroanalysis 2010, 22 (10), 1027–1036. fuel cells. Journal of Physical Chemistry C 2009, 113 (19),
17. Baby, T. T., Aravind, S. S. J., Arockiadoss, T., Rakhi, R. B., 7990–7995.
Ramaprabhu, S. Metal decorated graphene nanosheets as 35. Shao, Y., Liu, J., Wang, Y., Lin, Y. Novel catalyst support
immobilization matrix for amperometric glucose biosensor. materials for PEM fuel cells: Current status and future pros-
Sensors and Actuators, B: Chemical 2010, 145 (1), 71–77. pects. Journal of Materials Chemistry 2009, 19 (1), 46–59.
18. Choi, B. G., Park, H., Park, T. J., Yang, M. H., Kim, J. S., Jang, S. 36. Shao, Y., Sui, J., Yin, G., Gao, Y. Nitrogen-doped carbon
Y., Heo, N. S., Lee, S. Y., Kong, J., Hong, W. H. Solution chem- nanostructures and their composites as catalytic materials for
istry of self-assembled graphene nanohybrids for high-perfor- proton exchange membrane fuel cell. Applied Catalysis B:
mance flexible biosensors. ACS Nano 2010, 4 (5), 2910–2918. Environmental 2008, 79 (1), 89–99.
19. Dong, H., Gao, W., Yan, F., Ji, H., Ju, H. Fluorescence reso- 37. Pumera, M., Ambrosi, A., Bonanni, A., Chng, E. L. K., Poh,
nance energy transfer between quantum dots and graphene H. L. Graphene for electrochemical sensing and biosensing.
oxide for sensing biomolecules. Analytical Chemistry 2010, TrAC—Trends in Analytical Chemistry 2011, 29 (9), 954–965.
82 (13), 5511–5517. 38. Ratinac, K. R., Yang, W., Gooding, J. J., Thordarson, P., Braet,
20. Feng, L., Chen, Y., Ren, J., Qu, X. A graphene functionalized F. Graphene and related materials in electrochemical sensing.
electrochemical aptasensor for selective label-free detection Electroanalysis 2011, 23 (4), 803–826.
of cancer cells. Biomaterials 2011, 32 (11), 2930–2937. 39. Balasubramanian, K., Burghard, M. Biosensors based on
21. Lu, C. H., Yang, H. H., Zhu, C. L., Chen, X., Chen, G. N. carbon nanotubes. Analytical and Bioanalytical Chemistry
A graphene platform for sensing biomolecules. Angewandte 2006, 385 (3), 452–468.
Chemie—International Edition 2009, 48 (26), 4785–4787. 40. Wang, J. Carbon-nanotube based electrochemical biosensors:
22. Mohanty, N., Berry, V. Graphene-based single-bacterium res- A review. Electroanalysis 2005, 17 (1), 7–14.
olution biodevice and DNA transistor: Interfacing graphene 41. Joung, D., Chunder, A., Zhai, L., Khondaker, S. I. High yield
derivatives with nanoscale and microscale biocomponents. fabrication of chemically reduced graphene oxide field effect
Nano Letters 2008, 8 (12), 4469–4476. transistors by dielectrophoresis. Nanotechnology 2010, 21 (16),
23. Zeng, Q., Cheng, J., Tang, L., Liu, X., Liu, Y., Li, J., Jiang, 165202.
J. Self-assembled graphene–enzyme hierarchical nanostruc- 42. Eda, G., Chhowalla, M. Chemically derived graphene oxide:
tures for electrochemical biosensing. Advanced Functional Towards large-area thin-film electronics and optoelectronics.
Materials 2010, 20 (19), 3366–3372. Advanced Materials 2010, 22 (22), 2392–2415.
24. Zhou, M., Zhai, Y., Dong, S. Electrochemical sensing and 43. Chen, J. H., Cullen, W. G., Jang, C., Fuhrer, M. S., Williams,
biosensing platform based on chemically reduced graphene E. D. Defect scattering in graphene. Physical Review Letters
oxide. Analytical Chemistry 2009, 81 (14), 5603–5613. 2009, 102 (23), 236805.
25. Stoller, M. D., Park, S., Yanwu, Z., An, J., Ruoff, R. S. 44. Meyer, J. C., Kisielowski, C., Erni, R., Rossell, M. D.,
Graphene-based ultracapacitors. Nano Letters 2008, 8 (10), Crommie, M. F., Zettl, A. Direct imaging of lattice atoms
3498–3502. and topological defects in graphene membranes. Nano Letters
26. Vivekchand, S. R. C., Rout, C. S., Subrahmanyam, K. S., 2008, 8 (11), 3582–3586.
Govindaraj, A., Rao, C. N. R. Graphene-based electrochemi- 45. Bagri, A., Mattevi, C., Acik, M., Chabal, Y. J., Chhowalla,
cal supercapacitors. Journal of Chemical Sciences 2008, 120 M., Shenoy, V. B. Structural evolution during the reduction of
(1), 9–13. chemically derived graphene oxide. Nature Chemistry 2010, 2
27. Wang, H., Hao, Q., Yang, X., Lu, L., Wang, X. Graphene oxide (7), 581–587.
doped polyaniline for supercapacitors. Electrochemistry 46. Singh, V., Joung, D., Zhai, L., Das, S., Khondaker, S. I.,
Communications 2009, 11 (6), 1158–1161. Seal, S. Graphene based materials: Past, present, and future.
28. Wang, Y., Shi, Z., Huang, Y., Ma, Y., Wang, C., Chen, M., Progress in Materials Science 2011, 56 (8), 1178–1271.
Chen, Y. Supercapacitor devices based on graphene materials. 47. Szabo, T., Berkesi, O., Forgo, P., Josepovits, K., Sanakis,
Journal of Physical Chemistry C 2009, 113 (30), 13103–13107. Y., Petridis, D., Dekany, I. Evolution of surface functional
29. Guo, P., Song, H., Chen, X. Electrochemical performance groups in a series of progressively oxidized graphite oxides.
of graphene nanosheets as anode material for lithium-ion Chemistry of Materials 2006, 18 (11), 2740–2749.
batteries. Electrochemistry Communications 2009, 11 (6), 48. Fowler, J. D., Allen, M. J., Tung, V. C., Yang, Y., Kaner, R. B.,
1320–1324. Weiller, B. H. Practical chemical sensors from chemically
30. Liang, M., Zhi, L. Graphene-based electrode materials derived graphene. ACS Nano 2009, 3 (2), 301–306.
for rechargeable lithium batteries. Journal of Materials 49. Kou, R., Shao, Y., Wang, D., Engelhard, M. H., Kwak, J. H.,
Chemistry 2009, 19 (33), 5871–5878. Wang, J., Viswanathan, V. V., Wang, C., Lin, Y., Wang, Y.,
31. Wang, C., Li, D., Too, C. O., Wallace, G. G. Electrochemical Aksay, I. A., Liu, J. Enhanced activity and stability of Pt cata-
properties of graphene paper electrodes used in lithium batter- lysts on functionalized graphene sheets for electrocatalytic
ies. Chemistry of Materials 2009, 21 (13), 2604–2606. oxygen reduction. Electrochemistry Communications 2009,
32. Yang, S., Cui, G., Pang, S., Cao, Q., Kolb, U., Feng, X., 11 (5), 954–957.
Maier, J., Müllen, K. Fabrication of cobalt and cobalt oxide/ 50. Qi, X., Pu, K. Y., Zhou, X., Li, H., Liu, B., Boey, F., Huang,
graphene composites: Towards high-performance anode W., Zhang, H. Conjugated-polyelectrolyte-functionalized
materials for lithium ion batteries. ChemSusChem 2009, 3 reduced graphene oxide with excellent solubility and stability
(2), 236–239. in polar solvents. Small 2010, 6 (5), 663–669.
33. Yoo, E. J., Kim, J., Hosono, E., Zhou, H. S., Kudo, T., Honma, 51. Ramanathan, T., Abdala, A. A., Stankovich, S., Dikin, D.
I. Large reversible Li storage of graphene nanosheet families A., Herrera-Alonso, M., Piner, R. D., Adamson, D. H. et al.
for use in rechargeable lithium ion batteries. Nano Letters Functionalized graphene sheets for polymer nanocomposites.
2008, 8 (8), 2277–2282. Nature Nanotechnology 2008, 3 (6), 327–331.
52. Schniepp, H. C., Li, J. L., McAllister, M. J., Sai, H., Herrera- 69. Guo, H. L., Wang, X. F., Qian, Q. Y., Wang, F. B., Xia, X. H. A
Alonson, M., Adamson, D. H., Prud’homme, R. K., Car, R., green approach to the synthesis of graphene nanosheets. ACS
Seville, D. A., Aksay, I. A. Functionalized single graphene Nano 2009, 3 (9), 2653–2659.
sheets derived from splitting graphite oxide. Journal of 70. Ramesha, G. K., Sampath, N. S. Electrochemical reduction of
Physical Chemistry B 2006, 110 (17), 8535–8539. oriented graphene oxide films: An in situ Raman spectroelec-
53. Veca, L. M., Lu, F., Meziani, M. J., Cao, L., Zhang, P., Qi, trochemical study. Journal of Physical Chemistry C 2009, 113
G., Qu, L., Shrestha, M., Sun, Y. P. Polymer functionaliza- (19), 7985–7989.
tion and solubilization of carbon nanosheets. Chemical 71. Shao, Y., Wang, J., Engelhard, M., Wang, C., Lin, Y. Facile
Communications 2009, (18), 2565–2567. and controllable electrochemical reduction of graphene oxide
54. Xu, L. Q., Yang, W. J., Neoh, K. G., Kang, E. T., Fu, G. D. and its applications. Journal of Materials Chemistry 2009, 20
Dopamine-induced reduction and functionalization of gra- (4), 743–748.
phene oxide nanosheets. Macromolecules 2010, 43 (20), 72. Li, L., Du, Z., Liu, S., Hao, Q., Wang, Y., Li, Q., Wang, T.
8336–8339. A novel nonenzymatic hydrogen peroxide sensor based on
55. Xu, Y., Liu, Z., Zhang, X., Wang, Y., Tian, J., Huang, Y., Ma, MnO2/graphene oxide nanocomposite. Talanta 2010, 82 (5),
Y., Chen, Y. A graphene hybrid material covalently function- 1637–1641.
alized with porphyrin: Synthesis and optical limiting prop- 73. Zhan, B., Liu, C., Shi, H., Li, C., Wang, L., Huang, W., Dong,
erty. Advanced Materials 2009, 21 (12), 1275–1279. X. A hydrogen peroxide electrochemical sensor based on
56. Allen, M., Lee, C., Ahn, B., Kololuoma, T., Shin, K., silver nanoparticles decorated three-dimensional graphene.
Ko, S. R2R gravure and inkjet printed RF resonant tag. Applied Physics Letters 2014, 104 (24).
Microelectronic Engineering 2011, 88 (11), 3293–3299. 74. Dreyer, D. R., Park, S., Bielawski, C. W., Ruoff, R. S. The
57. Allen, M. J., Tung, V. C., Kaner, R. B. Honeycomb carbon: A chemistry of graphene oxide. Chemical Society Reviews 2010,
review of graphene. Chemical Reviews 2009, 110 (1), 132–145. 39 (1), 228–240.
58. Chang, J., Ge, T., Sanchez-Sinencio, E. In Midwest Symposium 75. Wang, Y., Wan, Y., Zhang, D. Reduced graphene sheets
on Circuits and Systems, 2012, pp. 582–585. modified glassy carbon electrode for electrocatalytic oxi-
59. Martin, P. Electrochemistry of gaphene: New horizons for dation of hydrazine in alkaline media. Electrochemistry
sensing and energy storage. Chemical Record 2009, 9 (4), Communications 2010, 12 (2), 187–190.
211–223. 76. Li, J., Guo, S., Zhai, Y., Wang, E. Nafion–graphene nano-
60. McCreery, R. L. Advanced carbon electrode materials for composite film as enhanced sensing platform for ultra-
molecular electrochemistry. Chemical Reviews 2008, 108 (7), sensitive determination of cadmium. Electrochemistry
2646–2687. Communications 2009, 11 (5), 1085–1088.
61. Jia, J., Kato, D., Kurita, R., Sato, Y., Maruyama, K., Suzuki, 77. Li, J., Guo, S., Zhai, Y., Wang, E. High-sensitivity determination
K., Hirono, S., Ando, T., Niwa, O. Structure and electro- of lead and cadmium based on the Nafion–graphene composite
chemical properties of carbon films prepared by an electron film. Analytica Chimica Acta 2009, 649 (2), 196–201.
cyclotron resonance sputtering method. Analytical Chemistry 78. Liu, G., Chen, J., Hou, X., Huang, W. A highly-sensitive elec-
2007, 79 (1), 98–105. trochemical sensor for the simultaneous detection of Cd2+
62. Niwa, O., Jia, J., Sato, Y., Kato, D., Kurita, R., Maruyama, K., and Pb2+ using liquid phase-exfoliated graphene. Analytical
Suzuki, K., Hirono, S. Electrochemical performance of ang- Methods 2014, 6 (15), 5760–5765.
strom level flat sputtered carbon film consisting of sp2 and 79. Kefala, G., Economou, A., Voulgaropoulos, A. A study of
sp3 mixed bonds. Journal of the American Chemical Society Nafion-coated bismuth-film electrodes for the determination
2006, 128 (22), 7144–7145. of trace metals by anodic stripping voltammetry. Analyst
63. Tang, L., Wang, Y., Li, Y., Feng, H., Lu, J., Li, J. Preparation, 2004, 129 (11), 1082–1090.
structure, and electrochemical properties of reduced graphene 80. Zhu, L., Tian, C., Yang, R., Zhai, J. Anodic stripping voltam-
sheet films. Advanced Functional Materials 2009, 19 (17), metric determination of lead in tap water at an ordered mesopo-
2782–2789. rous carbon/nafion composite film electrode. Electroanalysis
64. Shang, N. G., Papakonstantinou, P., McMullan, M., Chu, 2008, 20 (5), 527–533.
M., Stamboulis, A., Potenza, A., Dhesi, S. S., Marchetto, H. 81. Ohno, Y., Maehashi, K., Matsumoto, K. Chemical and bio-
Catalyst-free efficient growth, orientation and biosensing logical sensing applications based on graphene field-effect
properties of multilayer graphene nanoflake films with sharp transistors. Biosensors and Bioelectronics 2010, 26 (4),
edge planes. Advanced Functional Materials 2008, 18 (21), 1727–1730.
3506–3514. 82. Sofue, Y., Ohno, Y., Maehashi, K., Inoue, K., Matsumoto, K.
65. Davies, T. J., Hyde, M. E., Compton, R. G. Nanotrench arrays Highly sensitive electrical detection of sodium ions based on
reveal insight into graphite electrochemistry. Angewandte graphene field-effect transistors. Japanese Journal of Applied
Chemie International Edition in English 2005, 44, 5251. Physics 2011, 50 (6 Part 2), 06GE07.
66. Fischer, A. E., Show, Y., Swain, G. M. Electrochemical 83. Sudibya, H. G., He, Q., Zhang, H., Chen, P. Electrical detection
performance of diamond thin-film electrodes from differ- of metal ions using field-effect transistors based on micropat-
ent commercial sources. Analytical Chemistry 2004, 76 (9), terned reduced graphene oxide films. ACS Nano 2011, 5 (3),
2553–2560. 1990–1994.
67. Ambrosi, A., Pumera, M. Nanographite impurities domi- 84. Alwarappan, S., Erdem, A., Liu, C., Li, C. Z. Probing the elec-
nate electrochemistry of carbon nanotubes. Chemistry—A trochemical properties of graphene nanosheets for biosensing
European Journal 2010, 16 (36), 10946–10949. applications. Journal of Physical Chemistry C 2009, 113 (20),
68. Wang, J., Yang, S., Guo, D., Yu, P., Li, D., Ye, J., Mao, L. 8853–8857.
Comparative studies on electrochemical activity of gra- 85. Kang, X., Wang, J., Wu, H., Liu, J., Aksay, I. A., Lin, Y. A
phene nanosheets and carbon nanotubes. Electrochemistry graphene-based electrochemical sensor for sensitive detection
Communications 2009, 11 (10), 1892–1895. of paracetamol. Talanta 2010, 81 (3), 754–759.
86. Parvin, M. H. Graphene paste electrode for detection of chlor- 100. Du, M., Yang, T., Jiao, K. Immobilization-free direct elec-
promazine. Electrochemistry Communications 2011, 13 (4), trochemical detection for DNA specific sequences based on
366–369. electrochemically converted gold nanoparticles/graphene
87. Li, F., Li, J., Feng, Y., Yang, L., Du, Z. Electrochemical behav- composite film. Journal of Materials Chemistry 2010, 20 (41),
ior of graphene doped carbon paste electrode and its applica- 9253–9260.
tion for sensitive determination of ascorbic acid. Sensors and 101. Du, Y., Guo, S., Dong, S., Wang, E. An integrated sens-
Actuators, B: Chemical 2011, 157 (1), 110–114. ing system for detection of DNA using new parallel-motif
88. Wang, Y., Li, Y., Tang, L., Lu, J., Li, J. Application of gra- DNA triplex system and graphene-mesoporous silica–
phene-modified electrode for selective detection of dopamine. gold nanoparticle hybrids. Biomaterials 2011, 32 (33),
Electrochemistry Communications 2009, 11 (4), 889–892. 8584–8592.
89. Guo, Y., Guo, S., Ren, J., Zhai, Y., Dong, S., Wang, E. 102. Li, J., Liu, J., Tan, G., Jiang, J., Peng, S., Deng, M., Qian, D.,
Cyclodextrin functionalized graphene nanosheets with high Feng, Y., Liu, Y. High-sensitivity paracetamol sensor based
supramolecular recognition capability: Synthesis and host— on Pd/graphene oxide nanocomposite as an enhanced elec-
Guest inclusion for enhanced electrochemical performance. trochemical sensing platform. Biosensors and Bioelectronics
ACS Nano 2010, 4 (7), 4001–4010. 2014, 54, 468–475.
90. Li, J., Chen, J., Zhang, X. L., Lu, C. H., Yang, H. H. A novel 103. Pumera, M., Scipioni, R., Iwai, H., Ohno, T., Miyahara, Y.,
sensitive detection platform for antitumor herbal drug aloe- Boero, M. A mechanism of adsorption of b-nicotinamide
emodin based on the graphene modified electrode. Talanta adenine dinucleotide on graphene sheets: Experiment and

2010, 83 (2), 553–558. theory. Chemistry—A European Journal 2009, 15 (41),
91. Yin, H., Ma, Q., Zhou, Y., Ai, S., Zhu, L. Electrochemical 10851–10856.
behavior and voltammetric determination of 4-aminophe- 104. Musameh, M., Wang, J., Merkoci, A., Lin, Y. Low-potential
nol based on graphene–chitosan composite film modified stable NADH detection at carbon-nanotube-modified glassy
glassy carbon electrode. Electrochimica Acta 2010, 55 (23), carbon electrodes. Electrochemistry Communications 2002,
7102–7108. 4 (10), 743–746.
92. Li, N., Zhu, M., Qu, M., Gao, X., Li, X., Zhang, W., Zhang, 105. Valentini, F., Amine, A., Orlanducci, S., Terranova, M. L.,
J., Ye, J. Iron–tetrasulfophthalocyanine functionalized gra- Palleschi, G. Carbon nanotube purification: Preparation
phene nanosheets: Attractive hybrid nanomaterials for elec- and characterization of carbon nanotube paste electrodes.
trocatalysis and electroanalysis. Journal of Electroanalytical Analytical Chemistry 2003, 75 (20), 5413–5421.
Chemistry 2011, 651 (1), 12–18. 106. Banks, C. E., Compton, R. G. Exploring the electrocatalytic
93. Du, H., Ye, J., Zhang, J., Huang, X., Yu, C. Graphene nanosheets sites of carbon nanotubes for NADH detection: An edge plane
modified glassy carbon electrode as a highly sensitive and pyrolytic graphite electrode study. Analyst 2005, 130 (9),
selective voltammetric sensor for rutin. Electroanalysis 2010, 1232–1239.
22 (20), 2399–2406. 107. Pakapongpan, S., Mensing, J. P., Phokharatkul, D., Lomas, T.,
94. Yin, H., Zhou, Y., Cui, L., Liu, T., Ju, P., Zhu, L., Ai, S. Tuantranont, A. Highly selective electrochemical sensor for
Sensitive voltammetric determination of rutin in pharmaceu- ascorbic acid based on a novel hybrid graphene–copper phtha-
ticals, human serum, and traditional Chinese medicines using locyanine–polyaniline nanocomposites. Electrochimica Acta
a glassy carbon electrode coated with graphene nanosheets, 2014, 133, 294–301.
chitosan, and a poly (amido amine) dendrimer. Microchimica 108. Liao, C., Zhang, M., Niu, L., Zheng, Z., Yan, F. Organic
Acta 2011, 173 (3–4), 337–345. electrochemical transistors with graphene-modified gate

95. Karuwan, C., Sriprachuabwong, C., Wisitsoraat, electrodes for highly sensitive and selective dopamine

A., Phokharatkul, D., Sritongkham, P., Tuantranont, A. sensors. Journal of Materials Chemistry B 2014, 2 (2),
Inkjet-printed graphene–poly(3,4-ethylenedioxythiophene): 191–200.
poly(styrene-sulfonate) modified on screen printed carbon 109. Ghindilis, A. L., Atanasov, P., Wilkins, E. Enzyme-catalyzed
electrode for electrochemical sensing of salbutamol. Sensors direct electron transfer: Fundamentals and analytical applica-
and Actuators, B: Chemical 2012, 161 (1), 549–555. tions. Electroanalysis 1997, 9 (9), 661–674.
96. Sriprachuabwong, C., Karuwan, C., Wisitsorrat, A., 110. Leger, C., Bertrand, P. Direct electrochemistry of redox
Phokharatkul, D., Lomas, T., Sritongkham, P., Tuantranont, enzymes as a tool for mechanistic studies. Chemical Reviews
A. Inkjet-printed graphene–PEDOT:PSS modified screen 2008, 108 (7), 2379–2438.
printed carbon electrode for biochemical sensing. Journal of 111. Armstrong, F. A., Hill, H. A. O., Walton, N. J. Direct electro-
Materials Chemistry 2012, 22 (12), 5478–5485. chemistry of redox proteins. Accounts of Chemical Research
97. Lin, W. J., Liao, C. S., Jhang, J. H., Tsai, Y. C. Graphene 1988, 21 (11), 407–413.
modified basal and edge plane pyrolytic graphite elec- 112. Zhang, W., Li, G. Third-generation biosensors based on the
trodes for electrocatalytic oxidation of hydrogen peroxide direct electron transfer of proteins. Analytical Sciences 2004,
and b-nicotinamide adenine dinucleotide. Electrochemistry 20 (4), 603–609.
Communications 2009, 11 (11), 2153–2156. 113. Wu, Y., Hu, S. Biosensors based on direct electron transfer in
98. Liu, H., Gao, J., Xue, M. Q., Zhu, N., Zhang, M. N., Cao, T. redox proteins. Microchimica Acta 2007, 159 (1–2), 1–17.
B. Processing of graphene for electrochemical application: 114. Yao, Y. L., Shiu, K. K. Direct electrochemistry of glucose
Noncovalently functionalize graphene sheets with water- oxidase at carbon nanotube–gold colloid modified elec-
soluble electroactive methylene green. Langmuir 2009, 25, trode with poly(diallyldimethylammonium chloride) coating.
12006–12010. Electroanalysis 2008, 20 (14), 1542–1548.
99. Thevenot, D. R., Toth, K., Durst, R. A., Wilson, G. S. 115. Katz, E., Willner, I., Angew integrated nanoparticle–
Electrochemical biosensors: Recommended definitions and biomolecule hybrid systems: Synthesis, properties, and appli-
classification. Pure and Applied Chemistry 1999, 71 (12), cations. Angewandte Chemie International Edition 2004, 43,
2333–2348. 6042–6108.
116. Sarma, A. K., Vatsyayan, P., Goswami, P., Minteer, S. D. 133. Lu, Q., Dong, X., Li, L. J., Hu, X. Direct electrochemistry-
Recent advances in material science for developing enzyme based hydrogen peroxide biosensor formed from single-layer
electrodes. Biosensors and Bioelectronics 2009, 24 (8), graphene nanoplatelet–enzyme composite film. Talanta 2010,
2313–2322. 82 (4), 1344–1348.
117. Dai, Z. H., Ni, J., Huang, X. H., Lu, G. F., Bao, J. C. Direct 134. Zhou, K., Zhu, Y., Yang, X., Luo, J., Li, C., Luan, S. A novel
electrochemistry of glucose oxidase immobilized on a hexag- hydrogen peroxide biosensor based on Au–graphene–HRP–
onal mesoporous silica-MCM-41 matrix. Bioelectrochemistry chitosan biocomposites. Electrochimica Acta 2010, 55 (9),
2007, 70 (2), 250–256. 3055–3060.
118. Shan, C., Yang, H., Song, J., Han, D., Ivaska, A., Niu, L. 135. Chen, H., Müller, M. B., Gilmore, K. J., Wallace, G. G., Li,
Direct electrochemistry of glucose oxidase and biosensing for D. Mechanically strong, electrically conductive, and biocom-
glucose based on graphene. Analytical Chemistry 2009, 81 patible graphene paper. Advanced Materials 2008, 20 (18),
(6), 2378–2382. 3557–3561.
119. Wu, H., Wang, J., Kang, X., Wang, C., Wang, D., Liu, J., Aksay, 136. Cui, H. F., Ye, J. S., Zhang, W. D., Li, C. M., Luong, J. H.
I. A., Lin, Y. Glucose biosensor based on immobilization T., Sheu, F. S. Selective and sensitive electrochemical detec-
of glucose oxidase in platinum nanoparticles/graphene/ tion of glucose in neutral solution using platinum–lead alloy
chitosan nanocomposite film. Talanta 2009, 80 (1), 403–406. nanoparticle/carbon nanotube nanocomposites. Analytica
120. Lu, J., Drzal, L. T., Worden, R. M., Lee, I. Simple fabrication Chimica Acta 2007, 594 (2), 175–183.
of a highly sensitive glucose biosensor using enzymes immo- 137. Fan, A., Lau, C., Lu, J. Magnetic bead-based chemilumines-
bilized in exfoliated graphite nanoplatelets nafion membrane. cent metal immunoassay with a colloidal gold label. Analytical
Chemistry of Materials 2007, 19 (25), 6240–6246. Chemistry 2005, 77 (10), 3238–3242.
121. Wang, Z., Zhou, X., Zhang, J., Boey, F., Zhang, H. Direct 138. Papkovsky, D. B., O’Riordan, T. C., Guilbault, G. G. An immu-
electrochemical reduction of single-layer graphene oxide and nosensor based on the glucose oxidase label and optical oxygen
subsequent functionalization with glucose oxidase. Journal of detection. Analytical Chemistry 1999, 71 (8), 1568–1573.
Physical Chemistry C 2009, 113 (32), 14071–14075. 139. Wang, G., Huang, H., Zhang, G., Zhang, X., Fang, B., Wang,
122. Wisitsoraat, A., Pakapongpan, S., Sriprachuabwong, C., L. Gold nanoparticles/l-cysteine/graphene composite based
Phokharatkul, D., Sritongkham, P., Lomas, T., Tuantranont, immobilization strategy for an electrochemical immunosen-
A. Graphene–PEDOT:PSS on screen printed carbon elec- sor. Analytical Methods 2010, 2 (11), 1692–1697.
trode for enzymatic biosensing. Journal of Electroanalytical 140. Su, B., Tang, J., Huang, J., Yang, H., Qiu, B., Chen, G., Tang,
Chemistry 2013, 704, 208–213. D. Graphene and nanogold-functionalized immunosens-
123. Guiseppi-Elie, A., Lei, C., Baughman, R. H. Direct elec- ing interface with enhanced sensitivity for one-step electro-
tron transfer of glucose oxidase on carbon nanotubes. chemical immunoassay of alpha-fetoprotein in human serum.
Nanotechnology 2002, 13 (5), 559–564. Electroanalysis 2010, 22 (22), 2720–2728.
124. Deng, C., Chen, J., Chen, X., Xiao, C., Nie, L., Yao, S. Direct 141. Zhong, Z., Wu, W., Wang, D., Shan, J., Qing, Y., Zhang, Z.
electrochemistry of glucose oxidase and biosensing for glu- Nanogold-enwrapped graphene nanocomposites as trace
cose based on boron-doped carbon nanotubes modified elec- labels for sensitivity enhancement of electrochemical immu-
trode. Biosensors and Bioelectronics 2008, 23 (8), 1272–1277. nosensors in clinical immunoassays: Carcinoembryonic anti-
125. Cai, C., Chen, J. Direct electron transfer of glucose oxidase gen as a model. Biosensors and Bioelectronics 2010, 25 (10),
promoted by carbon nanotubes. Analytical Biochemistry 2379–2383.
2004, 332 (1), 75–83. 142. Du, D., Zou, Z., Shin, Y., Wang, J., Wu, H., Engelhard, M.
126. Liu, G., Lin, Y. Amperometric glucose biosensor based H., Liu, J., Aksay, L. A., Lin, Y. Sensitive immunosensor for
on self-assembling glucose oxidase on carbon nanotubes. cancer biomarker based on dual signal amplification strategy
Electrochemistry Communications 2006, 8 (2), 251–256. of graphene sheets and multienzyme functionalized carbon
127. Lin, Y., Lu, F., Tu, Y., Ren, Z. Glucose biosensors based on nanospheres. Analytical Chemistry 2010, 82 (7), 2989–2995.
carbon nanotube nanoelectrode ensembles. Nano Letters 143. Cai, Y., Li, H., Du, B., Yang, M., Li, Y., Wu, D., Zhao, Y.,
2004, 4 (2), 191–195. Dai, Y., Wei, Q. Ultrasensitive electrochemical immunoassay
128. Rubianes, M. D., Rivas, G. A. Carbon nanotubes paste for BRCA1 using BMIMaBF4-coated SBA-15 as labels and
electrode. Electrochemistry Communications 2003, 5 (8), functionalized graphene as enhancer. Biomaterials 2011, 32
689–694. (8), 2117–2123.
129. Zhou, M., Shang, L., Li, B., Huang, L., Dong, S. Highly 144. Du, D., Wang, L., Shao, Y., Wang, J., Engelhard, M. H., Lin,
ordered mesoporous carbons as electrode material for the Y. Functionalized graphene oxide as a nanocarrier in a mul-
construction of electrochemical dehydrogenase- and oxidase- tienzyme labeling amplification strategy for ultrasensitive
based biosensors. Biosensors and Bioelectronics 2008, 24 (3), electrochemical immunoassay of phosphorylated p53 (s392).
442–447. Analytical Chemistry 2011, 83 (3), 746–752.
130. Wu, L., Zhang, X., Ju, H. Amperometric glucose sensor based 145. Yang, M., Javadi, A., Li, H., Gong, S. Ultrasensitive immu-
on catalytic reduction of dissolved oxygen at soluble car- nosensor for the detection of cancer biomarker based on
bon nanofiber. Biosensors and Bioelectronics 2007, 23 (4), graphene sheet. Biosensors and Bioelectronics 2010, 26 (2),
479–484. 560–565.
131. Mao, S., Yu, K., Lu, G., Chen, J. Highly sensitive protein sen- 146. Wei, Q., Mao, K., Wu, D., Dai, Y., Yang, J., Du, B., Yang,
sor based on thermally-reduced graphene oxide field-effect M., Li, H. A novel label-free electrochemical immunosensor
transistor. Nano Research 2011, 4 (10), 921–930. based on graphene and thionine nanocomposite. Sensors and
132. Dong, X., Shi, Y., Zhao, Y., Chen, D., Ye, J., Yao, Y., Gao, Actuators, B: Chemical 2010, 149 (1), 314–318.
F., Ni, Z., Yu, T., Shen, Z., Huang, Y., Chen, P., Li, L. J. 147. Zhang, Q., Qiao, Y., Hao, F., Zhang, L., Wu, S., Li, Y., Li,
Symmetry breaking of graphene monolayers by molecular J., Song, X. M. Fabrication of a biocompatible and conduc-
decoration. Physical Review Letters 2009, 102 (13), 135501. tive platform based on a singlestranded DNA/graphene
nanocomposite for direct electrochemistry and electroca- 155. Hummers Jr, W. S., Offeman, R. E. Preparation of graphitic
talysis. Chemistry—A European Journal 2011, 16 (27), oxide. Journal of the American Chemical Society 1958, 80
8133–8139. (6), 1339.
148. Lee, C., Kang, H., Kim, H., Nguyen, H. A. D., Shin, K. Quality 156. Karuwan, C., Wisitsoraat, A., Phokharatkul, D., Sriprach-
control with matching technology in roll to roll printed elec- uabwong, C., Lomas, T., Nacapricha, D., Tuantranont, A. A
tronics. Journal of Mechanical Science and Technology 2010, disposable screen printed graphene–carbon paste electrode
24 (1), 315–318. and its application in electrochemical sensing. RSC Advances
149. Bollstrom, R., Tobjork, D., Dolietis, P., Salminen, P., Preston, 2013, 3 (48), 25792–25799.
J., Osterbacka, R., Toivakka, M. Printability of functional inks 157. Lee, M. J. Inkjet printing technology still in progress. Journal
on multilayer curtain coated paper. Chemical Engineering of the Korean Ceramic Society 2011, 48 (6), 543–548.
and Processing: Process Intensification 2011, 68, 13–20. 158. Mandal, S., Purohit, G., Katiyar, M., in Materials Science
150. Ping, J., Wu, J., Wang, Y., Ying, Y. Simultaneous determina- Forum, 2013, pp. 250–274.
tion of ascorbic acid, dopamine and uric acid using high-per- 159. Rida, A., Yang, L., Vyas, R., Tentzeris, M. M. Conductive
formance screen-printed graphene electrode. Biosensors and inkjet-printed antennas on flexible low-cost paper-based sub-
Bioelectronics 2012, 34 (1), 70–76. strates for RFID and WSN applications. IEEE Antennas and
151. Kim, K. S., Myung, W. R., Jung, S. B. Adhesion character- Propagation Magazine 2009, 51 (3), 13–23.
istics of silver tracks screen-printed on polyimide with an 160. Subramanian, V., Chang, P. C., Lee, J. B., Molesa, S. E.,
environmental reliability test. Journal of Nanoscience and Volkman, S. K. Printed organic transistors for ultra-low-cost
Nanotechnology 2012, 12 (7), 5769–5773. RFID applications. IEEE Transactions on Components and
152. Akhavan, V., Schroder, K., Pope, D., Farnsworth, S. Packaging Technologies 2005, 28 (4), 742–747.
Processing thick-film screen printed Metalon® CuO reduction 161. Teichler, A., Perelaer, J., Schubert, U. S. Inkjet printing of
ink with Pulse Forge® tools. Coating International 2013, 46 organic electronics-comparison of deposition techniques and
(3), 14–17. state-of-the-art developments. Journal of Materials Chemistry
153. Parades, J. I., Villar-Rodil, S., Martínez-Alonso, A., Tascón, C 2013, 1 (10), 1910–1925.
J. M. D. Graphene oxide dispersions in organic solvents. 162. Huang, L., Huang, Y., Liang, J., Wan, X., Chen, Y. Graphene-
Langmuir 2008, 24 (19), 10560–10564. based conducting inks for direct inkjet printing of flex-
154. Eda, G., Fanchini, G., Chhowalla, M. Large-area ultrathin films ible conductive patterns and their applications in electric
of reduced graphene oxide as a transparent and flexible elec- circuits and chemical sensors. Nano Research 2011, 4 (7),
tronic material. Nature Nanotechnology 2008, 3 (5), 270–274. 675–684.
10 Chemical Routes to Graphene
Quantum Dots
Photoluminescence Mechanism
and Novel Applications
Shoujun Zhu, Xiaohuan Zhao, Yubin Song, Bo Li, Junhu Zhang, and Bai Yang
CONTENTS
Abstract...................................................................................................................................................................................... 163
10.1 Introduction...................................................................................................................................................................... 163
10.2 Origin of PL in GQDs...................................................................................................................................................... 164
10.3 PL Quenching for GQDs.................................................................................................................................................. 167
10.4 Is There Real Up-Conversion PL from GQDs?.................................................................................................................170
10.5 Novel Applications............................................................................................................................................................171
10.5.1 Biomedical Applications........................................................................................................................................171
10.5.2 Optoelectronic Devices......................................................................................................................................... 172
10.5.3 Sensor....................................................................................................................................................................173
10.5.4 Assembly Composite or Nano-Hybrid...................................................................................................................174
10.6 Problem and Outlook.........................................................................................................................................................175
Acknowledgments.......................................................................................................................................................................176
References...................................................................................................................................................................................176
ABSTRACT nanodiamond, have been well known since the past many
years. To make these kinds of materials fluorescent, the size
Original graphene is a kind of zero-band semiconductor which and their surface chemical groups should be carefully modu-
cannot easily serve as a fluorescent material. To open up the lated. The as-prepared fluorescent carbon materials always
bandgap and facilitate its applications, researchers usually consist of sp2 and sp3 carbon, oxygen/nitrogen-based groups,
choose to change the surface chemical groups of original gra- and postmodified chemical groups. Up to now, many kinds
phene or convert it to 0D graphene quantum dots (GQDs). GQDs, of fluorescent carbon-based nanomaterials have been syn-
labeled as a new sort of fluorescent carbon-based material, have thesized, including carbon dots [1–3], fluorescent CNT [4],
attracted increasing attention for their good biocompatibility, graphene oxide (GO) [5–7], graphene quantum dots (GQDs)
photoelectric and optical properties, as well as convenient sur- [8–15], and so on. In this chapter, GQDs prepared by chemical
face modification. For these reasons, use of GQDs is promis- synthetic strategies are chosen as the main object of discus-
ing in numerous areas: biomedical applications, optoelectronic sion. In detail, the synthesis of GQDs can be divided into top-
devices, sensors, assembly composites, and so on. Although the down nanocutting method and bottom-up organic approaches.
photoluminescence mechanism of GQDs can be explained from The top-down nanocutting method generally includes cutting
the angle of size, surface chemical groups, and doping atoms, different carbon resources such as GO, carbon fiber, CNT,
there is no universal agreement on the specific photolumines- fullerene, and graphite electrode. The bottom-up organic
cence mechanism. In this chapter, GQDs prepared by chemical approaches always contain carbonization of carbohydrate,
synthetic strategies, which mainly feature the cutting of differ- synthesis or self-assembly of polycyclic aromatic hydrocar-
ent carbon resources and carbonization from small molecules, bons. Owing to the special chemical composition of GQDs,
are fully discussed. Besides that, the popular explanations of the they possess good biocompatibility, photoelectric, and optical
GQDs’ photoluminescence mechanism, new types of GQDs, properties, as well as convenient surface modification, which
and the novel applications of GQDs are summarized below. provide GQDs’ tunable fluorescence, physiological solubil-
ity and stability, pH sensitivity, low toxicity, and increasing
10.1 INTRODUCTION quantum yield compared with the original one. Though the
Carbon nanomaterials, such as graphite, fullerenes, car- up-conversional photoluminescence (PL) have been reported
bon nanotubes (CNT), carbon nanofibers, graphene, and by several groups, we found that it should have been further
163
investigated and proved in GQDs. As a result, GQDs promise domains, the e–h pairs can be photogenerated and perform
to be used in numerous areas: assembly composites, biomedi- radiative emission, as shown in Figure 10.1. If the reduction
cal applications, optoelectronic devices, and so on. is excessive, interconnectivity between the preexisting sp2
This review will focus on the origin of PL in GQDs, domains and the new small sp2 domains will be improved,
PL-quenching mechanism, a brief description of the validity thereby, it will facilitate the hopping of excitons to a nonra-
of the up-conversion of PL in GQDs as well as their most diative recombination center, which performs as high con-
recent scientific and technological applications. Owing to the ductivity property instead of PL behavior. Furthermore, the
increasing reports about GQDs, we apologize to the research- new-formed sp2 domains in the reduced GO are small enough
ers whose latest reports may have been left out. to give a blue emission, and the bandwidth is also narrower
than the former GO with wide red emission. Besides Chen
and Chhowalla’s groups, other groups have also investigated
10.2 ORIGIN OF PL IN GQDs
the fluorescence of GO and reduced GO (r-GO) [17–20].
To explain the PL mechanism of GQDs, the PL behavior of For example, Luo et al., proposed that the bond distortions
chemically derived GO should be introduced first because of may contribute to the fluorescence of GO and r-GO [18].
the similar chemical structures. GO contains oxygen-based Gokus et al., have observed visible luminescence in the oxy-
functional groups either on the basal plane or at the edges. gen plasma-treated graphene, and attribute the emission to
As a result, the 2–3 nm aromatic sp2 domains are surrounded CO-related localized electronic states at the oxidation sites
by linearly aligned epoxy and the hydroxyl-bond sp3 C–O [19]. Furthermore, Galande et al., have studied the pH-depen-
matrix. In such a structure of GO, the fluorescent property dent fluorescence of GO and suggested the emission of qua-
is determined by the π states of the sp2 sites. The π and π* simolecular fluorophores in such kinds of materials [20]. They
electronic levels of the sp2 clusters influenced by the bandgap found that the excited state of the fluorophore species proton-
of σ and σ* states of the sp3 matrix are strongly confined. ated in acidic media, which makes the PL spectra different in
Radiative recombination of electron–hole (e–h) pairs in such acidic and basic solutions. This kind of quasimolecular fluo-
sp2 clusters can arouse the fluorescence. Because of the wide rophores is caused by the carboxylic acid groups that are elec-
size distribution of sp2 domains existing in GO, the values of tronically coupled with the surrounding graphene core sheets.
the bandgaps of different sizes of sp2 domains range widely, The steady-state measurements are not enough to clarify
leading to the wide PL emission spectrum from visible to the PL of GO-based materials because of their low resolu-
near-infrared (NIR) [16]. tion. The time-resolved fluorescence (TRF) measurements
However, the sp2 domain should not be valued without are pretty efficient to elucidate the origin of fluorescence in
considering the surrounding sp3 matrix and other sp2 configu- carbon-based materials [21]. There are three common tech-
rations, which also contribute to the photoluminescent prop- niques used in TRF spectroscopy: the stroboscopic technique
erties. Chen and Chhowalla’s groups investigated the tunable (strobe), the time-correlated, single photon counting tech-
PL in GO, and they found that the PL of GO shift to shorter nique, and the frequency modulation or phase shift technique
wavelength after moderate reduction [7]. During the reduc- (phase). The first two are time-domain techniques, while the
tion, the preexisting sp2 domains remain unchanged while the last one is a frequency-domain technique. The time-domain
new small sp2 domains are formed. In this kind of small sp2 techniques are direct techniques. They measure fluorescence
(a) (b)
π* π*
Original GO Reduced GO
PL / a.u.
PL / a.u.
Excitation
Excitation
400 500 600 700 400 500 600 700

λ / nm λ / nm
π π
Reduction
FIGURE 10.1 Proposed PL emission mechanisms of (a) the predominant “red emission” in GO from disorder-induced localized states.
(b) The predominant “blue emission” in rGO from confined cluster states. (C. T. Chien et al.: Tunable photoluminescence from graphene
oxide, Angew. Chem. Int. Ed. 2012. 51. 6662. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
Chemical Routes to Graphene Quantum Dots 165
decay curves (i.e., fluorescence intensity as a function of time) fluorescence results of GO could be connected to understand
directly, and the researchers have full advantage of seeing these emissions in GQDs. Although considerable develop-
the physical mechanism in the course of the experiments. ment in the preparation of GQDs has been witnessed in the
Frequently, a qualitative judgment about a particular mecha- last three years, researchers are still undergoing a tough time
nism can be made by examining raw decay data and a proper to figure out a clear and reasonable PL mechanism. Following
fitting function can be thus selected. Several groups have are several important statements reported about the PL mech-
studied the fluorescence mechanism of GO using such kinds anism of GQDs (Table 10.1).
of technology. Shang et al. built the GO model according to Pan et al., suggested that blue PL of hydrothermally GQDs
experimental data, and investigated the fluorescence behav- might be attributed to free zigzag sites with a carbine-like trip-
ior utilizing TRF measurements with femtosecond resolution let ground state described as σ1π1 [38]. Under acidic conditions,
[22]. Combined with the related electronic structures based the free zigzag sites of the GQDs are protonated, forming a
on density functional theory (DFT), the e–h recombination reversible complex between the zigzag site and H+, and lead-
from conduction band bottom and nearly localized electronic ing to the breaking of the emissive triple carbene state and the
states to wide-range valance band was proved. In their point PL quenching (Figure 10.2a). To the contrary, the PL recovers
of view, the functionalized groups, such as oxidized carbon because the free zigzag sites are restored under alkaline con-
atoms (C–O, C=O, and O=C–OH) and nonoxidized –C=C–, dition. They also studied the PL of GQDs by single-particle
codetermine the fluorescence of GO-based materials. spectroscopic, and found that the PL of the GQDs was not
Similar to GO and r-GO, GQDs with defect, oxygen affected by the sizes [39]. Lin and Zhang prepared GQDs with
groups and functional groups on the surface, are another kind size about 20 nm from exfoliating and disintegrating CNTs or
of fluorescent carbon-based materials. They also have exci- graphite flakes [23]. The obvious single layer and clear zig-
tation dependence and pH-dependent emission. The current zag edge were proved in AFM and bright-field high-resolution
TABLE 10.1
Reported GQDs Related to PL Mechanism
Reference Synthetic Methods Starting Materials Size (nm) Height (nm) QY (%) PL Color Possible PL Mechanism
[17,18] Hydrothermal GO 5–13 1–2 6.9 Blue Triple carbene at the zigzag edges
No size-dependent PL
[23] K intercalation MCNTs 20 <1 6.53 Blue Triple carbene at the zigzag edges
[24] Acidic oxidation Carbon fiber 1–4 0.4–2 – Blue to yellow Triple carbene at the zigzag edges
[25] Amino- GO 2.5 1.13 19–29 Blue to yellow Resonance of amine moieties and
hydrothermal graphene core
[26] Acidic oxidation GO 1–5 1.07 – Blue Charge transfer between amino
and graphene core
[27] Acidic oxidation and GO 3–9 1.1, 1.2 2.5, 16.4 Blue and The changes of amino and NO3
amino- green
hydrothermal
[28] Amino- GO 5–7 – 16, 29 Blue and Interstate transition from π*–π to
hydrothermal green n–π due to amino groups
connecting
[29,30] Solvothermal and GO 5.3 1.2 11.4 Green Both intrinsic state and molecule
modification state
[31] Exfoliation in Graphite 4 0.5–2.5 – Blue and GOQDs: defect states with
organic solvent or nanoparticles green oxygenous functional groups;
Hummers method GQDs: intrinsic states in the
high-crystalline structure
[32] Modified Graphene 3.8, 3.7 – – Green Edge state
solvothermal nanoplatelets
[33] Acidic oxidation and Graphite powder 2.5–3.2 0.6–0.8 1–7 Blue to green Chemical groups affected the
modification carbene-like site
[34] Solvothermal and GO 2–5.5 0.6–2 – Blue Pyrazole rings at the edge
modification
[35] Acidic oxidation and Graphite powder 10 – 1.7–17.6 Blue Molecule state at the surface of
modification GQDs
[36] Microwave-assisted Glucose 2.4 – 5–15 Blue High-crystalline carbon structures
hydrothermal
[37] Hydrothermal Fructose and 5.4 5.2 6.8 Blue Additional energy levels by Cl
hydrochloric acid connecting
(a) (b)
Hydroth-
ermal σ
deoxidiz- 4 ev δE = 0.57 eV
ation
π
LUMO
3.43 eV
O H O H
H
+ + +
H+
OH OH– OH
COO– + + + COOH
H H H
O
O HOMO
(c)
29 Green emission
NH2
LUMO
300 500 700
I/arb. unit
Wavelength/nm
E = –1.73 E = –1.55
Blue emission E = –4.42 E = –3.99
HOMO
300 500 700

Wavelength/nm I II
1600 1700 1800 1900
m/z
FIGURE 10.2 (a) Models of the GQDs in acidic and alkali media, the two models can be converted depending on pH. (D. Pan et al.:
Hydrothermal route for cutting graphene sheets into blue-luminescent graphene quantum dots, Adv. Mater. 2010. 22. 734. Copyright Wiley-
VCH Verlag GmbH & Co. KGaA. Reproduced with permission.) (b) Emission mechanism of zigzag site in GQDs. (L. Lin, S. Zhang,
Creating high yield water soluble luminescent graphene quantum dots via exfoliating and disintegrating carbon nanotubes and graphite
flakes, Chem. Commun. 48, 2012, 10177. Reproduced by permission of The Royal Society of Chemistry.) (c) Comparison of MALDI-
TOF mass spectra of GQDs with green and blue emission, DFT calculated HOMO and LUMO orbital. (H. Tetsuka et al.: Optically tun-
able amino-functionalized graphene quantum dots, Adv. Mater. 2012. 24. 5333–5338. Copyright Wiley-VCH Verlag GmbH & Co. KGaA.
Reproduced with permission.)
TEM. The newly opened band gap arises from the triple car- of amine moieties and graphene core [25]. For the GQDs
benes at the zigzag edges, corresponding to the transitions with the same sizes, the emission wavelength increased with
from the highest occupied molecular orbital (HOMO) to σ the contents of amine groups (Figure 10.2c). The GQDs
and π orbital of lowest unoccupied molecular orbital (LUMO) with smaller size were observed to be violet. Furthermore,
in triple carbenes (Figure 10.2b). These two kinds of GQDs the GQDs possess high quantum yield because of decreased
with size of 9.6 nm and 20 nm, respectively, possess similar carboxylic and epoxide groups which act as the nonradiative
absorption and emission behaviors, which proved that the PL e–h recombination center. Combined with the experimental
mechanism was determined by the triple carbene at the zigzag results and ab initio calculations, the primary amine at the
edges instead of the quantum confinement effects. Although edge of GQDs have higher HOMO orbitals than those that are
the emission colors of GQDs were changed from blue, green, hydrogen-terminated, due to the strong orbital interaction with
and yellow in the work by Peng et al., it should be noted that the –NH2 groups. The resonance feature between the delo-
the applied exciting wavelengths are 318 nm, 331 nm, and calized π orbital and the molecule orbital in the –NH2 group
429 nm, respectively. So they are not multicolored under the results in a narrowing of the optical band gap. Such amino-
same exciting wavelength. Actually, the PL mechanism of as- contained GQDs have also been reported by other groups. Jin
prepared GQDs also derives from the triple carbene at the zig- et al. reported that the functionalized GQDs exhibit a red shift
zag edges [24]. The GQDs prepared by similar methods also in the PL emission spectrum, compared with the preexisting
possesses the analogical PL mechanism [40–44]. GQDs (the PL emissions of the amine-functionalized GQDs
In the work of Tetsuka et al., the ammonia-assisted hydro- also shifted with changes in the pH due to the protonation and
thermal method was used, the as-prepared GQDs are edge- deprotonation of the functional groups) [26]. Calculations with
terminated by a primary amine, allowing the electronic DFT illustrated that PL shifts resulted from charge transfers
structure to be modified with the effective orbital resonance between the functional groups and GQDs, which can tune the
(a) (b) 3.0 GQDs. In their other related work, the PL of GQDs was
2.8 changed from green emission to blue emission by tuning the
Band gap (eV)

2.6 surface chemistry [30]. In the modified GQDs, the connected
2.4 N=1 N=3 N=5 alkylamines transferred the –COOH and epoxy of GQDs into
2.2 –CONHR and –CNHR, both of which can reduce the non-
2.0 radiative recombination induced by the –COOH and epoxy
1.8 groups, and transfer GQDs from defect-state emission into
0 1 2 3 4 5 6 intrinsic-state emission. In reduction method, the carbonyl,
Number of attached–NH2 groups epoxy, and amino moieties were changed into –OH groups,
(c) which suppress nonradiative process and further enhance
0.02 (d) HOMO LUMO
0.01 NH2
integrity of π conjugated system (also reduce the defects). As
Charge
0.00 a result, intrinsic-state emission plays the leading role instead

GQD of defect-state emission (Figure 10.4b). The blue- and green-
–0.01 0 1 2 3 4 5 6
–0.02
GQD-(NH2)1
color emissions of GQDs and GOQDs were also prepared by
Number of attached–NH2 groups Seo et al. They have revealed that the green luminescence of
(e) 0 LUMO
GOQDs originates from defect states with oxygenous func-
–1
HOMO tional groups, whereas the blue luminescence of GQDs is
–2 dominated by intrinsic states in the high-crystalline structure.
The PL mechanism of solvothermally synthesized GQDs was
Energy (eV)
–3
NH2
–4 also investigated by other groups [31]. For example, Lingam
LUMO
–5 (–5.075)
et al., found the evidence for edge-state PL in solvothermally
–6 synthesized GQDs. If the edge of GQDs was destroyed or dis-
HOMO
–7 (–7.038)
appeared, the PL intensity decreased [32].
–8 Some other standpoints, such as molecule state, surface
0 1 2 3 4 5 6
passivation, are also notable for understanding the PL mecha-
Number of attached–NH2 groups
nism of GQDs. A reasonable explanation of GQDs’ PL mech-
anism will be of great benefit not only in guiding synthesis
FIGURE 10.3 (a) Schematics of isolated sp2 clusters with amino of graphene materials with multicolors, improving output and
functional groups within the sp3 carbon matrix including defects. enhancing quantum yields, but also in expanding great poten-
(b) Band gap change of GQD-(NH2)n as function of the number of tial in nanoelectronics, photovoltaic (PV) devices, biodetec-
attached NH2 groups (gray, C; white, H; and dark gray, N atom).
tion, as well as disease diagnostics/therapeutics.
(c) The evolution of average Mulliken charge for amino groups as
a function of the number of attached NH2 groups. (d) HOMO and
LUMO isosurface of NH2, GQD, GQD-(NH1)1. (e) HOMO and 10.3 PL QUENCHING FOR GQDs
LUMO energy levels of GQDs-(NH2)n. (Reprinted with permission
from S. H. Jin et al., Tuning the photoluminescence of graphene According to the discussion on the origin of PL in GQDs,
quantum dots through the charge transfer effect of functional groups, the chemical structure plays an important role in determining
ACS Nano 7, 1239. Copyright 2013 American Chemical Society.) the fluorescence of GQDs. First, the rigid structure of polycy-
clic aromatic or graphene reduces the vibration and rotational
band gap of the GQDs. The band gap of the GQDs decreases energy level, leading the excited electrons to give off the radi-
to 2.254 eV, when a GQD is functionalized by one amino, and ative emission. Second, the as-prepared GQDs always contain
the band gap will gradually decrease with the increasing num- the electron-rich groups, such as the –NH2, –OH, –N(CH3)2,
ber of –NH2 groups (Figure 10.3). and N doping, which can further enhance the PL of GQDs.
The mechanism of GQDs emission was investigated with Finally, the surface passivation could be the PL origin of some
femtosecond transient absorption spectroscopy and femtosec- GQDs. No matter what the fluorescence mechanism is, as
ond TRF dynamics measured by a fluorescence up-conversion also with other kinds of fluorescent materials, the PL quench-
technique, as well as a nanosecond time-correlated single- ing behavior also exists in GQDs which is very important in
photon-counting technique in Wang’s and Yang’s groups [29]. theory and in practical applications. A variety of processes
They found that two independent molecule-like states and can result in GQDs’ PL quenching, such as collision, energy/
a dark intrinsic state existed in solvothermally synthesized electron transfer and complex formation. In conclusion, PL
GQDs, as shown in Figure 10.4a. The intrinsic state is attrib- of GQDs is very sensitive to the surroundings, especially the
uted to a quantum confinement effect, and its PL locates at existence of a quencher, such as ferric ion, π-based molecules,
around 470 nm with a dominant short lifetime with low pho- and a strong oxidant. For example, the chloride ion is famous
toluminescence quantum yield. Another two irrelevant mole- for its ability to quench quinine fluorescence, and it is also
cule-like states (around 320 nm and 400 nm) provides a blue an ideal quencher for surface-passivated GQDs [21]. On the
fluorescence with peak at around 430 nm, and green fluores- other hand, this kind of property of GQDs can also contrib-
cence with a peak at around 530 nm. The three kinds of emis- ute to investigate the PL mechanism of GQDs and develop
sion states constitute the fascinating PL of green-fluorescence quencher-based sensors [45–50]. Furthermore, quenching
(a)
ESA 540 nm ~ 2.3 ev ESA 2 560 nm ~2.2 eV

No resonance ESA 1 495 nm ~2.5 eV
energy transfer
ABS ABS PL 470 nm

320 nm PL 430 nm 400 nm 0.2–1.6 ps
3.9 ev 4 ns PL 530 nm 3.1 ev
7.7 ns
Molecule-like Intrinsic
Molecule-like state (2) state (1) state (3)
(b)
10,000 GQDs (green emission) 10,000 GQDs (blue emission)
m-GQDs (green emission) m-GQDs (blue emission)
r-GQDs (green emission) r-GQDs (blue emission)
IRF IRF
8.1 ns 1000
Intensity
Intensity
1000 4.4 ns 2.7 ns 3.9 ns
3.5 ns 2.1 ns
100
100
10 15 20 25 30 35 40 45 10 15 20 25 30 35 40 45
Decay time (ns) Decay time (ns)
FIGURE 10.4 (a) The three kinds of emission states constitute the fascinating PL of green-fluorescence GQDs proved by the ultrafast
spectroscopy. (L. Wang et al.: Unraveling bright molecule-like state and dark intrinsic state in green-fluorescence graphene quantum dots
via ultrafast spectroscopy, Adv. Opt. Mater. 2013. 1. 264. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
(b) The PL emission tuned by the surface chemistry methods. (S. Zhu et al.: Surface chemistry routes to modulate the photoluminescence of
graphene quantum dots: From fluorescence mechanism to up-conversion bioimaging applications, Adv. Funct. Mater. 2012. 22. 4732–4740.
Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
and dequenching upon interaction with a specific molecular strongly suppress the fluorescence emission by the FRET
biological target is the basis for activation of optical contrast from the GQDs donor to the irradiative TNT acceptor through
agents for molecular imaging. intermolecular polar–polar interactions at spatial proximity.
The PL quenching mechanism mainly includes fluores- Ran et al., reported that the fluorescence of GQDs could be
cence resonance energy transfer (FRET), electron transfer, quenched by the Ag nanoparticle decorating [47]. Generally,
third dynamic quenching exciplex, and static quenching. The the noble metal clusters, or nanoparticles-supported carbon
typical PL quenching behaviors based on GQDs were listed materials have shown high stability through hybridization
in Table 10.2. Zhao et al., used graphene as an acceptor, and between the clusters and sp2 dangling bonds at the defect
mouse antihuman immunoglobulin G (mIgG)-conjugated sites of carbon materials. The fluorescence of GQDs can
GQDs as donors to achieve fluorescence quenching [45]. be quenched to a certain degree by these metal clusters or
When adding graphene to the GQDs covalently functional- nanoparticles due to the transfer processes. Furthermore, the
ized mIgG solution, the π–π stacking interaction between biothiols in the Ag nanoparticle-decorated GQDs solution can
graphene and GQDs will induce the FRET to facilitate the PL further quench GQDs fluorescence by acting as reductants
quenching (Figure 10.5a). Certainly, the addition of human and bridging the adjoining Ag nanoparticles with each other
IgG would bind the mIgG due to the specific antibody–antigen (Figure 10.6a). In a similar work, Liu et al., found that the PL
interaction, which effectively increases the distance between of GQDs could be quenched with the presence of Fe3+. The
mIgG/GQDs and graphene surface. Thus, the FRET process quenching may be primarily caused by the effective electron
will be hindered, and the fluorescence will be restored. In transfer that occurs between GQDs and Fe3+ [48]. Actually,
another π–π stacking interaction system between GQDs and it may be because of the special coordination interaction
2,4,6-trinitrotoluene (TNT) [46], the fluorescence quenching between Fe3+ ions and the phenolic hydroxy groups of these
was achieved by FRET and electron transfer (Figure 10.5b). kinds of materials that the fluorescence of the carbon-based
In detail, the resultant TNT bound at the GQDs surface can materials could be quenched by Fe3+ ions, which has been
TABLE 10.2
Reported GQDs Related to PL Quenching
Synthetic Method, Starting
Reference Material QY (%) PL Color Quenching Behavior Goal Detection “Limit of Detection”
[45] Hydrothermal, GO 7.2 Blue FRET: on–off Human IgG 0.2 μg/mL
[46] Acidic oxidation, GO – Blue FRET and electron transfer TNT 0.495 μg/mL
[47] Microwave, GO – Blue Electron transfer +
Ag and biothiols 3.5 nM
[48] Heated, citric acid and 33.6 Blue Electron transfer Fe3+ 22 μM
glutathione
[49] Heated, citric acid 8.9 Blue PL center degraded Chlorine 0.05 μM
[50] Hydrothermal, GO – Blue Energy-transfer or Phosphate (Pi) 0.1 μM
electron-transfer
[51] Microwave, GO 11.7, 22.9 Green, blue Electrochemiluminescent Cd2+ 150 nM
[52] Hydrothermal, GO 11.4 Green Aggregated complex Trypsin 33 ng/mL
[53] Hydrothermal, GO – Blue Aggregated complex Glucose –
[54] Hydrothermal, GO – Blue Electrochemiluminescence Adenosine triphosphate 0.0015 nM
[55] Acid treated, polycyclic 2.2–11.7 Green Electron transfer Fe3+ 5 nM
hydrocarbons
[56] Acidic oxidation, carbon fiber – Blue Electron transfer Ni2+ –
[57] Acidic oxidation, GO – Blue Electron transfer Cu2+ 0.2 μM
widely used for the detection of Fe3+ ions, or colored reac-

tions, in traditional organic chemistry. Fluorescence quench-
(a) ing may be contributed to nonradiative electron transfer that
involves partial electron transfer from the excited state to the
Gr LRET d orbital of Fe3+. However, the special coordination interaction
between Fe3+ ions and the phenolic hydroxy groups of GQDs
could be destroyed by the other more affined ligand, such as
Gr Gr
phosphate ions. As a result, phosphate-containing molecules
GQDs mlgG Human lgG acted as complexing agents for Fe3+, and the quenched fluo-
(b)
Exc Exc rescence of the GQDs/Fe3+ can easily recover in the presence
hv of phosphate-containing molecules.
TNT Dong et al., found that the PL of GQDs could be quenched
Quenching by the strongly oxidizing free chlorine, which can destroy the
hv surface states of GQDs and cause the PL center to degrade
[49]. First, the original GQDs should be surface-passivated,
1 2 probably by the incompletely carbonized citric acid. The
LUMO authors ascribed the PL of GQDs to the surface state, there-
fore, it is reasonable to believe that the chlorine quenches the
HOMO
GQDs TNT GQDs TNT PL of GQDs by affecting their surface states (Figure 10.6b).
Resonance energy transfer Charge transfer In order to further prove his deduction, some other oxidants
with stronger oxidation capability than free chlorine can also
FIGURE 10.5 (a) Schematic illustration of a universal immune- quench the PL of GQDs, while other reactive oxygen spe-
sensing strategy based on regulation of the interaction between cies except ⋅OH have nearly no effect on the PL signal of the
graphene and GQDs. (H. Zhao et al., A universal immunosens- GQDs. To further understand this kind of quenching mecha-
ing strategy based on regulation of the interaction between gra- nism of GQDs, the surface state-based fluorescence in GQDs
phene and graphene quantum dots, Chem. Commun. 49, 2013, 234. should also be noted. In Li’s and Zhu’s works [29,51], the PL
Reproduced by permission of The Royal Society of Chemistry.) (b) quantum yield of GQDs was enhanced by NaBH4 reduction.
Schematic of the FRET-based GQDs sensor for the detection of In Shen’s work, the PL intensity was improved after polyeth-
TNT, quenching mechanism through 1 is resonance energy transfer
ylene glycol (PEG) passivation [58]. These phenomena also
from the GQDs donor to TNT acceptor and 2 is charge transfer from
the excited GQDs to TNT. (Reprinted from Talanta, 101, L. Fan explained that the surface states may play an important role in
et al., Fluorescence resonance energy transfer quenching at the sur- the PL mechanism of GQDs.
face of graphene quantum dots for ultrasensitive detection of TNT, The pH is also an important factor to affect the PL of
192. Copyright 2012, with permission from Elsevier.) GQDs. In these kinds of GQDs, the zigzag sites play a leading
(a) medium, the GQDs self-assembled into J-type aggregation due

to the π–π interactions. At 1.0 mg mL−1 GQDs concentration,
the emission maximum shifts to 525 nm. Meanwhile, the exci-
tation maximum also red-shifted from 230 nm to 480 nm. The
aggregation behavior was also proved by absorption spectra.
Blue graphene
quantum dots
Ag+ Ag nanoparticle Biothiol At low concentration, GQDs exist in a monomeric form and a
(b) Strong typical absorption at 230 nm was observed, which is assigned
Fluorescent GQDs Quenched
FL
Non-fluorescent GQDs
FL to the π–π* transition of aromatic sp2 domains. The increase in
concentration of GQDs results in a red shift of the absorption
Strongly oxidizing free chlorine HCIO maximum, and a new absorption at 370 nm appears due to the
destroys the passivated surface self-assembled J-type aggregation. The emission transformation
Cl2
of GQDs
CIO– of GQDs allows the development of novel sensor systems by tak-
ing advantage of the self-assembled luminescent aggregation.
(c) 700
10.4 IS THERE REAL UP-CONVERSION
600 A B C
0
500 PL FROM GQDs?
Ex (nm)
1250
400
300 2500 Many groups have reported the up-conversion properties of
200 3750 fluorescent carbon-based materials. For instance, Sun’s group
700 5000 has reported the multiphoton bioimaging of carbon dots with
600 D E F
6250 pulsed infrared laser excitation [60]. The two-photon excited
Ex (nm)
500
400 7500 emission has also been observed in GO nanoparticles [61].
300 8750 The up-conversion PL (UCPL) is a phenomenon in which the
200 10000 PL emission wavelength is shorter than its excitation wave-
200 300 400 500 600 700 200 300 400 500 600 700 200 300 400 500 600 700
Em (nm) Em (nm) Em (nm) length. The UCPL can be obtained by the following several
mechanisms: two-photon absorption, anti-stokes PL (ASPL)
and auger process/thermal effect. It should be noted that an
FIGURE 10.6 (a) Schematic illustration of the mechanism
intermediate state is essential for the carriers to perform up-
of Ag+ and biothiol detection based on GQDs. (X. Ran et al.,
Ag Nanoparticle-decorated graphene quantum dots for label-free,
conversion and the excitation light must be high-powered and
rapid and sensitive detection of Ag+ and biothiols, Chem. Commun. coherent from a laser. Recently, the up-conversion phenome-
49, 2013, 1079. Reproduced by permission of The Royal Society of non seems to have been observed in the prepared GQDs under
Chemistry.) (b) PL quenching of GQDs induced by destroying the the excitation of a xenon lamp [58,62–64]. In their experi-
surface passivation. (Reprinted with permission from Y. Dong et al., ments, the UCPL measurement is conducted under a mono-
Graphene quantum dot as a green and facile sensor for free chlorine chromatic light from a xenon lamp passing through grating
in drinking water, Anal. Chem. 84, 8378. Copyright 2012 American as an excitation source. With an awareness that the efficiency
Chemical Society.) (c) 2D-fluorescence topographical map of GQDs of UCPL produced by excited light is very small for low-
aqueous solution at different concentration (mg mL −1): (A) 0.01,
intensity excitations, the so-called “UCPL” in GQDs could be
(B) 0.05, (C) 0.1, (D) 0.2, (E) 0.33, (F) 1.0. (S. Chen et al., Unusual
emission transformation of graphene quantum dots induced by
due to the excitation of second-order diffraction light (wave-
self-assembled aggregation, Chem. Commun. 48, 2012, 7637. length λ/2), which coexists in the selected first-order light
Reproduced by permission of The Royal Society of Chemistry.) (wavelength λ) from the monochromators in the fluorescence
spectrophotometer. As a result, Tan et al., commented that
role in the PL-quenching mechanism. Pan et al. suggested that there could be mistakes in explaining normal luminescence
the blue PL of the hydrothermally synthesized GQDs might excited by the second-order diffraction light (wavelength λ/2)
be attributed to free zigzag sites with a carbine-like triplet as the evidence of reported GQDs’ UCPL behavior. To make
ground state described as σ1π1 [38]. Under acidic conditions, it clear, we reinvestigated the UCPL behavior of solvother-
the free zigzag sites of the GQDs are protonated, forming a mally synthesized GQDs. In the Shimadzu RF-5301 PC spec-
reversible complex between the zigzag sites and H+, leading trophotometer, the emission spectra could be obtained under
the emissive triple carbene state broken and PL quenched. On wavelength-tunable excitation. In this circumstance, the UCPL
the contrary, the PL recovers because the free zigzag sites are seemed to have been detected under the 600–900 nm excita-
restored under alkaline condition. Similar behaviors have also tions (Figure 10.7a and b). However, if a cutoff filter (filter
been observed in other kinds of GQDs, which proves these U340 for parts of the Shimadzu RF-5301 PC spectrophotom-
kinds of GQDs have similar zigzag sites [40–44]. eter) is put in the excitation channel (between the excitation
Chen et al., investigated an unusual emission transforma- source and the quartz pool to eliminate the light at 400 nm
tion of GQDs induced by self-assembled J-type aggregation as (800/2) from excitation source), almost no UCPL would be
shown in Figure 10.6c [59]. At low concentration, the maximum detected, as shown in Figure 10.7c. The second-order diffrac-
excitation and emission wavelengths are 230 nm and 420 nm, tion (λ/2) usually exists in the selected excitation light due
respectively. At high concentration and in a strong acidic to imitated ability to exclude second-order diffraction (λ/2)
(a) (b)
400 320 nm excitation
360 nm excitation 150 600 nm excitation
400 nm excitation 700 nm excitation
300 440 nm excitation 800 nm excitation
480 nm excitation 120 900 nm excitation
Intensity
Intensity
90
200
60
100
30
0 0
300 400 500 600 700 800 300 400 500 600 700 800
Wavelength (nm) Wavelength (nm)
(c) (d)
1000 50
400 nm excitation
800 nm excitation 40
800 without filter 740 nm without filter
740 nm with filter
800 nm excitation
30
Intensity
Intensity
600 with filter
400 20
200 10
0 0
400 500 600 700 800 400 500 600 700 800
Wavelength (nm) Wavelength (nm)
FIGURE 10.7 The “up-conversion” PL of GQDs should be reinvestigated using efficient methods. (a) The excitation-dependent PL of sol-
vothermally GQDs. (b) The “up-conversion” PL in GQDs excited at 600–900 nm light. (c) The intensity of “up-conversion” PL decreased
when using 800 nm excitation with filter. (d) The commercial filters may lose effect in special wavelength section.
of diffraction gratings used in the monochromator of fluores- On the basis of the frequently reported works, and the
cence spectrophotometer. Sometimes, the designed filter can above discussion on the UCPL of carbon dots and GQDs, it is
hardly cut off all the involved wavelengths, so the “UCPL” quite important to establish a proper characterization system
was observed in such a system (Figure 10.7d). to investigate these kinds of properties [65,66]. Is there a real
For the reasons given above, a xenon lamp in commer- up-conversion photoluminescence in GQDs? If the UCPL is
cial fluorescence spectrophotometer is no longer a suitable artificial, or the intensity of the UCPL is too low to utilize, the
source for UCPL characterizations. Instead of a xenon lamp, reported works would mislead many researchers to investi-
a high-powered and coherent laser is more qualified as an gate its up-conversion property and put in more effort to seek
exciting light. Shen et al. observed the green color emission related applications. Furthermore, are there any differences
of the hydrothermally synthesized GQDs under 980 nm laser among the GQDs prepared under different methods? In other
exciting [58]. They speculated on the UCPL of GQDs to the words, why do some of the GQDs have the UCPL properties
ASPL mechanism. The energy levels of σ and π orbits were while others don’t?
provided by the carbene ground state of GQDs zigzag sites,
and the carbene ground state is related to the energy gap (ΔE)
10.5 NOVEL APPLICATIONS
between σ and π orbits. When a bunch of low-energy photons
excite the electrons in the π orbital to high-energy state, the With their superior fluorescence, physiological stability, pH
UCPL is emitted when the electrons jump back to the low- sensitivity, fine biocompatibility, and low toxicity, GQDs have
energy σ orbital. The calculated ΔE (1.1 eV) coincides with offered substantial applications in biomedical applications,
the Hoffmann’s rule (<1.5 eV), which proved the possibility optoelectronic devices, sensors, and assembly nanocompos-
of the suggested mechanism. In the work by Zhuo et al., a ites. The rapid development in synthesizing GQDs with con-
two-photon UCPL mechanism was suggested [62]. Because trollable sizes, tailorable chemical structures, and clear PL
the intermediate state is essential for this kind of UCPL, the mechanism, will promote their applications [67–71].
excitation light should be high-powered and coherent from
a laser. Although Zhu et al., detected the “UCPL” color in
10.5.1 Biomedical Applications
GQDs/living cell system under the 808 nm laser excitation
[30], the observed emission color may not only be due to the In this section, some selected applications of fluorescent
true UCPL of GQDs, but also because of the certain substance GQDs are described [72–78]. The selection criteria are their
in a complicated cell microenvironment. Thus, the true “two- recent publications, and their potential for future development
photon excited emission” in bioimaging should be carefully of scientific and technological applications. The optical prop-
investigated in detail. erties and carrying capacity are the primary keys for GQDs
to be applied in practical use. The bright PL, good biocom- cells by causing oxidative stress (Figure 10.8c) [80]. The cell
patibility, and low cytotoxicity render GQDs applicable in death induced by photoexcited GQD displayed morphologi-
biological applications, such as bioimaging [79], autophagy- cal and/or biochemical characteristics both in apoptosis and
inducing photodynamics [80], and DNA cleavage [81]. autophagy. The results indicated potential usefulness of GQD
Bioimaging-based applications of diagnostics in vitro/ in photodynamic therapy. It should be noticed that this kind
in vivo must be nontoxic and biocompatible. In the past sev- of photodynamic therapy is not the same with the reported
eral years, metal-based QDs bioimaging methodologies have GO, highly absorbing NIR for photothermal therapy. This
appeared, together with toxicity concerns for their intrinsi- kind of issue needs to be further exploited in detail. Actually,
cally toxic elements like cadmium. Compared with metal- the electron at excited state can contribute to both the radia-
based QDs, GQDs are constituted by an intrinsically nontoxic tive emission and thermogenesis. The reactive oxygen species
element, carbon, which makes them particularly useful and generated from the blue light was another transformed energy,
promising bioanalytical tools (Figure 10.8a). More and more which was used in photodynamic therapy.
groups began to exploit GQDs’ bioimaging applications as a To understand DNA scission, repair, and signal transduc-
fluorescent marker, which is shown in Figure 10.8b [72–79]. tion, the exploration of efficient DNA intercalators can never
Future development should be focused on multifunctions be ignored. Zhou et al., investigated another interesting appli-
besides the bioimaging, such as drug delivery/imaging system. cation of GQDs in DNA cleavage system [81]. They found
Carbon-based materials have been used as thermal therapy that, by using GQDs and Cu2+, about 90% supercoiled DNA
to kill cancer cells [61]. Markovic et al., reported that elec- was converted into nicked DNA, while only about 59% super-
trochemically produced GQDs irradiating blue light (470 nm, coiled DNA was cleaved with the same amount of large-sized
1 W) generate reactive oxygen species, including singlet oxy- GO and Cu2+. They speculated that the as-prepared GQDs
gen. The reactive oxygen species can kill U251 human glioma with smaller lateral size will be better than micrometer-sized
GO sheets as an intercalator to DNA molecules. Therefore,
the efficiency of GQDs is higher than that of GO in DNA
(a) (b) cleavage under the same condition. This result suggests the
120
potential applications of GQDs can be further extended in
Cell viability (%)
100
80 biological and medical research.
60
40
20 10.5.2 Optoelectronic Devices
0
0 10 20 50 100 200 400 According to theoretical calculations and experiment results,
Amount of GQDs added (μg) 100 px
cutting bulk graphene into nanoribbons and quantum dots or
(c)
chemical/physical treatment can tune the HOMO and lowest
Non-irradiated
unoccupied molecular orbital (LUMO). At the same time, it is

Crystal violet (% of control)
100
important to design the device’s structure for PV conversion
80 according to the energy level of the materials, which makes the
60 controllable energy level crucial to their application. Of late,
40 GQDs have attracted increasing attention for their perspectives
Irradiated
of energy conversion, especially PV conversion [82–85]. In

20 Control
GQD the original works done by Qu’s group, the bulk heterojunc-
0 tion polymer solar cell was based on an active layer of poly(3-
0 1 3 5 7.5 10
hexylthiophene) and GQDs, as shown in Figure 10.9a [82]. The
Irradiation time (min)
LUMO level of the electrochemically synthesized GQDs is
estimated to be 4.2–4.4 eV, roughly lying between the LUMO
FIGURE 10.8 (a) Effect of solvothermally GQDs on MG-63
cells viability. (S. Zhu et al., Strongly green-photoluminescent gra-
of P3HT and the work function of Al. The power conversion
phene quantum dots for bioimaging applications, Chem. Commun. efficiency achieved was 1.28%. Gupta et al., reported a con-
47, 2011, 6858. Reproduced by permission of The Royal Society jugated polymer connected with GQDs acceptors exhibiting a
of Chemistry.) (b) The overlay high contrast image of a nucleo- significant enhancement in organic PV [83]. In their work, the
lus stained with dark gray DAPI and GQDs (light gray) staining. aniline (ANI) was functionalized with hydrothermally synthe-
(Reprinted with permission from J. Peng et al., Graphene quantum sized GQDs first. After that, the typical device structure (ITO/
dots derived from carbon fibers, Nano Lett. 12, 844. Copyright 2012 PEDOT:PSS/P3HT:ANI-GQDs/LiF/Al and ITO/PEDOT:PSS/
American Chemical Society.) (c) The left image showed U251 cells P3HT:ANI-GSs/LiF/Al) was prepared. According to the
were incubated with control solution or GQD (200 mg/mL) and
results, the device with 1 wt% ANI-GQDs in P3HT showed the
exposed to blue light (470 nm,1 W) for the indicated times. After
24 h, cell viability was determined by crystal violet. The right image
highest conversion efficiency. Besides that, the highest power
showed the cell morphology under the 470 nm irradiation or not. conversion efficiency (5.27%) for dye-sensitized solar cells was
(Reprinted from Biomaterials, 33, Z. M. Markovic et al., Graphene obtained by polypyrrole-doped GQDs devices [85].
quantum dots as autophagy-inducing photodynamic agents, 7084. Besides the PV devices, fluorescent GQDs can also be
Copyright 2012, with permission from Elsevier.) applied in light emitting diodes (LED) because of their high
(a)
ITO P3HT GQDs Al
e–1
e–1
AI 3.5 eV 4.2 eV
P3HT:GQDs hv 4.3 eV
4.4 eV
PEDOT:PSS 4.8 eV
5.2 eV
ITO
h+1
(b) 10 7×104
Blue LED, CIE(0.242, 0.156)
Current density (mA/mm2)
2.1 eV
6×104 With GQD converter, CIE(0.282, 0.373)
8 3.55 eV
Intensity (a.u.)
ANI- 4.9 eV 5×104
4.7 eV 5.3 eV
GQD Au
6 ITO MEH 4×104
(D) PPV 5.38 eV
(C) 3×104
4 (B) 2×104
2 1×104
(A) 0
0 400 500 600 700 800
0 10 20 30
Wavelength (nm)
Bias (V)
FIGURE 10.9 (a) Schematic and energy band diagrams of the GQDs-based solar cell. (Reprinted from Adv. Mater, 23, Y. Li et al., An
electrochemical avenue to green-luminescent graphene quantum dots as potential electron-acceptors for photovoltaics, 776. Copyright 2011,
with permission from Elsevier.) (b) Measured current density of MEH-PPV with MB-GQDs (A) 0%, (B) 0.5%, (C) 1%, (D) 3% as a function
of the applied voltage. The inset was the device structure. (Reprinted with permission from V. Gupta et al., Luminscent graphene quantum
dots for organic photovoltaic devices, J. Am. Chem. Soc. 133, 9960. Copyright 2011 American Chemical Society.) (c) Luminescence spectra
of the blue LED with and without GQDs coating. Left inset: photographs of the GQDs coated LED without (top) and with applied voltage
(bottom). Right inset: the CIE chromaticity coordinates for the illuminating blue LED with and without GQDs coating. (Reprinted with
permission from L. Tang et al., Deep ultraviolet photoluminescence of water-soluble self-passivated graphene quantum dots, ACS Nano 6,
5102. Copyright 2012 American Chemical Society.)
electron mobility and the strong luminescence [86,64–66]. shifted from (0.242, 0.156) to (0.282, 0.373), demonstrating
The first GQD-based LED was prepared by the utilization that the GQDs are capable of converting blue light to white
of poly(2-methoxy-5-(2-ethylhexyloxy)-1,4-phenylenevinyl- light. By tuning the growth parameters of the GQDs, or using
ene) (MEH-PPV) and methylene blue functionalized GQDs different approaches, the dream white LEDs with CIE (0.33,
(MB-GQDs) of different concentrations [83]. The device 0.33) for solid-state-lighting will be within reach [88]. Besides
structure was ITO/PEDOT:PSS/MEHPPV:MB-GQDs/LiF/ LED, GQDs were also successfully applied as electrode mate-
Al. The MB-GQDs dispersed in the MEHPPV offers more rials and supercapacitors [89,90].
electrical transport paths, which results in an enhancement
of charge injection and the following increase in the carrier
10.5.3 Sensor
density (Figure 10.9b). As a result, only lower than usual
turn-on voltage was required to achieve the high efficiency. The PL-quenching mechanism was analyzed in detail in
Tang et al., applied glucose-derived GQDs to color convert- Section 10.3, which also presented some important applica-
ing layer of a blue emitting diode [87]. In his work, a com- tions of GQDs. Herein, two typical cases are going to dis-
mercially available blue LED was coated with GQDs solution played to further illustrate the importance of GQDs’ property.
first. After that, with the evaporation of the solvent in the Li et al., reported that GQDs were found to possess a
elevated temperature, a layer of the GQDs was formed on self-assembled aggregation tendency, and their fluorescence-
the surface of the LED (Figure 10.9c). The blue light inten- quenching property with the presence of cytochrome c (Cyt c),
sity of the GQDs-coated LED weakens while a broad band which can be disturbed by trypsin, resulting in restoration of
peak at 510 nm appeared, and the Commission International GQDs-based fluorescence as shown in Figure 10.10a and b
d’Eclairage (CIE) chromaticity coordinates the blue LED [52]. Cyt c is a well known electron transfer (ET) protein,
(a) groups, as well as doped nitrogen, which had coordination

ability with metal ions. Hence, they could form a Cd2+ –GQDs
Cyt c Trypsin complex with their effective coordination groups for metal
ions, and the complex could further induce the aggregation of
GQDs, leading to the decrease of corresponding ECL emis-
sion. Because of their novel PL and low toxicity properties, an
(b) 400 (c) increasing number of groups were focusing on exploiting the
(1) GQDs
(3) sensor applications-based on GQDs [91–94].
300 e
PL intensity
e GQDs
200 e GQDs
(2) e 10.5.4 Assembly Composite or Nano-Hybrid
ECL
100 e
e SO4•–
SO42– Qu’s groups have presented the organized assembly from 0D
0
400 500 600 S2O82– functional GQDs to 1D nanotube arrays through the electro-
Wavelength (nm) phoresis deposition of GQDs in a nanoporous anodized alu-
minum oxide template [95]. The as-prepared GQD-nanotubes
FIGURE 10.10 (a) Schematic illustration of the fluorescent had a diameter of ca. 200–300 nm, remaining a freestand-
biosensor for trypsin based on GQDs. (b) PL emission spectra ing array structure. On the basis of the unique porous nano-
of GQDs (60 μg/mL) at different conditions: (1) GQDs in PBS tube architecture of GQDs, the GQD-nanotubes provided
buffer; (2) GQDs + Cyt c (1000 μg/mL); (3) GQDs + Cyt c + tryp- an efficient charge transfer pathway between the target mol-
sin (300 μg/mL). λex = 320 nm. (X. Li et al., Self-assembled gra-
ecules and the GQDs, and thus produced a strong surface-
phene quantum dots induced by cytochrome c: a novel biosensor
for trypsin with remarkable fluorescence enhancement, Nanoscale enhanced Raman scattering (SERS) effect exceeding that on
5, 2013, 7776. Reproduced by permission of The Royal Society of flat graphene sheets. On the basis of this unique property, the
Chemistry.) (c) Schematic illustration of the ECL mechanism of applications of this kind of metal-free platform for efficient
GQDs. (Reprinted from Adv. Funct. Mater, 22, L. Li et al., A fac- SERS were exploited by the utilization of GQD-nanotubes.
ile microwave avenue to electrochemiluminescent two-color gra- In the following work, they reported functional microspheres
phene quantum dots, 2971. Copyright 2012, with permission from assembled by GQDs through the water/oil (W/O) emulsion
Elsevier.) template [96]. They started with a homogeneous mixture of
GQDs and aqueous ammonia, followed by the formations
of the W/O emulsion and assembly of GQDs, finally, after
which is able to be coupled to opposite-charged QDs effi- removal of water and separation of GQD spheres (GQDSs)
ciently through electrostatic interactions, and act as an effi- from oil. The formation of GQDSs with the W/O emulsion
cient FL quencher for QDs. Besides that, Fe3+ in Cyt c can template provides a facile way to endow the GQDSs with
also quench FL of GQDs through ET for the special coordi- new functions. The as-prepared GQDSs have a diameter of
nate interaction between Fe3+ and phenolic hydroxyl groups ca. 0.5–3 µm with a fairly smooth surface. TEM investigation
of GQDs. When trypsin is present, Cyt c can be cleaved on reveals that the as-prepared GQDSs are solid, and the densi-
the c-terminal side of arginine and lysine residues forming fied GQDs with a size of less than 5 nm are observed around
smaller fragments, and Fe3+ in Cyt c can be reduced into Fe2+ the spheres. For example, magnetic Fe3O4 nanoparticles have
with the aid of digestive enzyme. So the presence of trypsin been intercalated into the GQDSs by a mixture of GQDs with
in the system can result in the restoration of FL. Moreover, Fe3O4 nanoparticles during GQDSs synthesization.
the end of the cleaved peptide bonds of Cyt c were lysine Besides the outstanding PL-based properties, GQDs are
and arginine residues, both of which can reduce GQDs, and also a kind of carbon material that have a high refractive
the reducing GQDs subsequently increase the fluorescence. index (n = 2.6–3) and low density. Therefore, GQDs could
Furthermore, the trypsin in the sensor showed high selectivity be potential candidates for high refractive index nanocom-
compared with other kinds of proteins, such as bovine serum posites because of their small size for hybridization without
albumin, lysozym, papain, and pepsin. Li et al., observed that aggregations. Thus, it is of great significance to introduce
GQDs showed an intense ECL emission at −1.45 V with an GQDs into a polymer matrix to prepare transparent, high
onset potential at about –0.9 V, which was about nine times refractive index nanocomposites. Yang’s groups reported
higher than the background signal [50]. GQDs were applied the preparation of transparent nanocomposite films of
in the determination of Cd2+ for their intense ECL (Figure GQDs-poly(N,N′-dimethylacrylamide) (GQDs-PDMAA)
10.10c) with the presence of K2S2O8 as a necessary coreactant with a high refractive index via a facile in situ polymer-
for electrochemiluminescence (ECL) signal [51]. Specifically, ization method (Figure 10.11a) [97]. The GQDs with a
the metal ions (20 mM) were added to a mixture of pH 7.4 diameter of 3–5 nm have been successfully immobilized
tris-HCl buffer solution (TBS) (0.05 M), 0.1 M K2S2O8 and in the polymer matrix and are homogeneously dispersed in
20 ppm GQDs. The ECL quenching could be induced by the PDMAA. The GQDs-PDMAA nanocomposite films exhibit
interaction between metal ions and GQDs. To be specific, good transparency in the visible range. The refractive index
the obtained GQDs were covalently decorated with oxygen- of the transparent GNPs-PDMAA nanocomposite films is
containing functional groups, such as hydroxyl and carboxyl in the range of 1.516 to 1.976 at 632.8 nm, which linearly
(a) O H O (b)
OH OH
O O
HC CH2
C O H C N C CH CH 2.8
OH O 3 2
C
OH OH O C O H CH3
OH O O 2.6 GNPs-PDMAA
OH N GNPs-P(DMAA-St)
C O C N C CH CH
O OH n
H3C CH3 CH3
OH OH
O O
2.4
Refractive index
GO DMAA C N C CH CH2
GNPs CH3 2.2
2.0
Glass substrate
GO Solvothermal 1.8
DMAA method 1) UV-curing
1.6
2) Thermal-curing
1.4
Transparent 0 10 20 30 40 50 60 70 80 90 100
Cloudy solution Colorless
teflon autoclave transparent solution GNPs-PDMAA film GNPs content (wt%)
FIGURE 10.11 (a) Preparative scheme for transparent GQDs–PDMAA nanocomposite films via in situ polymerization route. (b) Refractive
index variation of PDMAA and the nanocomposite films with different weight content of GQDs. (G. Zhang et al., Embedding graphene
nanoparticles into poly(N,N′-dimethylacrylamine) to prepare transparent nanocomposite films with high refractive index, J. Mater. Chem
22, 2012, 21218. Reproduced by permission of The Royal Society of Chemistry.)
increases with the content of GQDs from 0 to 50 wt%. It In other words, all the factors that could affect the HOMO,
is also found that the transparent nanocomposite films can LUMO, and bandgap of GQDs can influence the PL properties.
be prepared from a mixture of N,N′-dimethylacrylamide The optical properties were determined either by the combina-
monomers and styrene monomers at an equal weight ratio tion or competition of these factors. Because of the diversity and
in the UV/thermal dual-cure process. With 50 wt% contents complexity of the different methods preparing GQDs, it was dif-
of GQDs in the nanocomposite film, the refractive index ficult to point out a general PL mechanism in GQDs system.
reached 2.058 (Figure 10.11b). However, it should be noted that nearly all the reported GQDs
possessed oxygen-containing groups, such as hydroxyl, epoxy,
carbonyl, and carboxylic acid groups. These groups contributed
10.6 PROBLEM AND OUTLOOK to the molecule-like PL behavior instead of quantum effect-
The GQDs discussed in this chapter are referred to as a like. The influence of single or multilayer in GQDs is another
kind of uncertain material because of their uncertain group important issue in understanding the PL mechanism. Lots of the
containing chemical structures. Fortunately, these inherent reported GQDs are not single layer actually, so the comparison
functionalized groups provide the opportunities for tailoring between single layer and multilayer GQDs are eagerly desired.
their chemical structures as well as their optical and opto- Besides the authoritative explanation on the PL emission
electronic properties. However, stepwise organic synthesis mechanism, the following issues also need effective solutions
methods can obtain the GQDs with precise size, certain in order to further develop the application of GQDs. First, it is
surface groups, and controllable defects, which is extremely still a great challenge to find out an efficient method for mass
important in understanding the PL mechanism as a model production of high-quality GQDs. The existing approaches are
system. However, the expensive initial reactants, time-con- not efficient enough to obtain GQDs with high quantum yield,
suming procedures, and environment polluting intermedium multiemission colors under single excitation wavelength, and
made them only suitable in theoretical research. The “nano” accurate chemical structure. Second, synthesis of new kinds of
routes, which contained “top-down” nano-cutting methods GQDs with unconventional structure is a burgeoning new area
from cheap carbon resources, and “bottom-up” organic car- and has caught the attention of researchers. For example, the
bonization approaches from carbohydrate substances, are studies on N-doped GQDs [98] and ultrathin graphitic-phase
easy to implement, easy to get raw materials, and efficient C3N4 nanosheets [99] have been reported to possess remarkable
to obtain GQDs for practical applications. For the reasons PL properties. Third, the present characterization methods are
given above, it is extremely important to promote the devel- not powerful enough to clarify the physical/chemical proper-
opment of this kind of material, which could be a promising ties of GQDs, and the “up-conversion” PL should be investi-
candidate to take the place of the organic dyes or inorganic gated in detail. Other important characterization methods such
quantum dots in relative fields. as matrix-assisted laser desorption/tonization time of flight
The widely observed PL emission in GQDs may be a result mass spectrometry is potent in figuring out the accurate chemi-
of quantum size effect, triple carbenes at the zigzag edges, con- cal structure of GQDs (Figure 10.12). Fourth, unprecedented
nected chemical groups, surface defects and heteroatom doping. diverse functions and novel applications based on GQDs will be
(a)
Peaks shift due to the
changed chemical groups
r-GQDs
r-GQDs
Intensity (a.u.)
Intensity (a.u.)
m-GQDs
m-GQDs
24 GQDs
GQDs
1000 1500 2000 2500 3000 3500 4000 1000 1250 1500 1750 2000
M/Z M/Z
(b)
+2H
–C5H4 +2H
+3H
–C5H4
–C3H5
ΔM = 74 m/z = 1429 +2H
m/z = 1355 ΔM = 74
ΔM = 74 –C5H4
m/z = 1503 ΔM = 74
m/z = 1577
m/z = 1281
1300 1400 1500 1600

m/z
FIGURE 10.12 MALDI-TOF MS spectra of GQDs, m-GQDs, and r-GQDs. The molecular weights were observed from ca. 700–3500
with 24 interval (two carbons). (S. Zhu et al.: Surface chemistry routes to modulate the photoluminescence of graphene quantum dots:
From fluorescence mechanism to up-conversion bioimaging applications, Adv. Funct. Mater. 2012. 22. 4732–4740. Copyright Wiley-VCH
Verlag GmbH & Co. KGaA. Reproduced with permission.) (b) Possible stable structure for each molecular structure that corresponds to the
observed MALDI-TOF mass pattern. (H. Tetsuka et al.: Optically tunable amino-functionalized graphene quantum dots, Adv. Mater. 2012.
24. 5333–5338. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
exploited in the fields of energy, environment, biology, and so 5. K. P. Loh, Q. Bao, G. Eda, M. Chhowalla, Nat. Chem. 2010, 2,
on [100]. Finally, reasonable theoretical simulations are eagerly 1015.
required to further explain the PL mechanism of GQDs. Owing 6. K. Yang, L. Feng, X. Shi, Z. Liu, Chem. Soc. Rev. 42, 2013,
530.
to the increasing reports about GQDs, some of latest reports
7. C. T. Chien, S. S. Li, W. J. Lai, Y. C. Yeh, H. A. Chen, I. S.
give very important results for PL mechanism and application Chen, L. C. Chen et al. Angew. Chem. Int. Ed. 51, 2012, 6662.
of GQDs, which cannot be introduced in detail. 8. L.S. Li, X. Yan, J. Phys. Chem. Lett. 1, 2010, 2572.
9. J. Shen, Y. Zhu, X. Yang, C. Li, Chem. Commun. 48, 2012,
3686.
ACKNOWLEDGMENTS 10. S. Zhu, S. Tang, J. Zhang, B. Yang, Chem. Commun. 48, 2012,
4527.
This work was supported by the National Science Foundation 11. Z. Zhang, J. Zhang, N. Chen, L. Qu, Energy Environ. Sci. 5,
of China (91123031, 20921003, 50973039) and the National 2012, 8869.
Basic Research Program of China (2012CB933800). 12. L. Li, G. Wu, G. Yang, J. Peng, J. Zhao, J.J. Zhu, Nanoscale 5,
2013, 4015.
13. X. Zhou, S. Guo, J. Zhang, Chem. Phys. Chem 14, 2013, 2627.
REFERENCES
14. M. Bacon, S. J. Bradley, T. Nann, Part. Part. Syst. Char. 4,
1. L. Cao, M. J. Meziani, S. Sahu, Y. P. Sun, Acc. Chem. Res. 46, 2014, 415.
2013, 171. 15. L. Lin, M. Rong, F. Luo, D. Chen, Y. Wang, X. Chen, Trac-
2. S. N. Baker, G. A. Baker, Angew. Chem. Int. Ed. 49, 2010, Trend. Anal. Chem. 54, 2014, 83.
6726. 16. G. Eda, Y. Y. Lin, C. Mattevi, H. Yamaguchi, H. A. Chen, I.
3. H. Li, Z. Kang, Y. Liu, S. T. Lee, J. Mater. Chem. 22, 2012, S. Chen, C. W. Chen and M. Chhowalla, Adv. Mater. 22, 2010,
24230. 505.
4. K. Welsher, Z. Liu, S. P. Sherlock, J. T. Robinson, Z. Chen, D. 17. Q. Mei, K. Zhang, G. Guan, B. Liu, S. Wang, Z. Zhang, Chem.
Daranciang, H. Dai, Nat. Nanotechnol. 4, 2009, 773. Commun. 46, 2010, 7319.
18. Z. Luo, P. M. Vora, E. J. Mele, A. T. C. Johnson, and J. M. 51. L. Li, J. Ji, R. Fei, C. Wang, Q. Lu, J. Zhang, L. Jiang, J. Zhu,
Kikkawa, Appl. Phys. Lett. 94, 2009, 111909. Adv. Funct. Mater. 22, 2012, 2971.
19. T. Gokus, R. R. Nair, A. Bonetti, M. Böhmler, A. Lombardo, 52. X. Li, S. Zhu, B. Xu, K. Ma, J. Zhang, B. Yang, W. Tian,
K. S. Novoselov, A. K. Geim, A. C. Ferrari and A. Hartschuh, Nanoscale 5, 2013, 7776.
ACS Nano 3, 2009, 3963. 53. Y. Li, L. Zhang, J. Huang, R. Liang, J. Qiu, Chem. Commun.
20. C. Galande, A. D. Mohite, A. V. Naumov, W. Gao, L. Ci, A. 49, 2013, 5180.
Ajayan, H. Gao, A. Srivastava, R. B. Weisman, P. M. Ajayan, 54. J. Lu, M. Yan, L. Ge, S. Ge, S. Wang, J. Yan, J. Yu, Biosensor.
Sci. Rep. 1, 2011, 85. Bioelectron. 47, 2013, 271.
21. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd 55. L. Zhou, J .Geng, B. Liu, Part. Part. Syst. Charact. 30, 2013, 1086.
ed., Springer, New York, 2006. 56. H. Huang, L. Liao, X. Xu, M. Zou, F. Liu, N. Li, Talanta, 117,
22. J. Shang, L. Ma, J. Li, W. Ai, T. Yu, G.G. Gurzadyan, Sci. Rep. 2013, 152.
2, 2012, 792. 57. F. Wang, Z. Gu, W. Lei, W. Wang, X. Xia, Q. Hao, Sensor.
23. L. Lin, S. Zhang, Chem. Commun. 48, 2012, 10177. Actuat. B-Chem. 190, 2014, 516.
24. J. Peng, W. Gao, B. K. Gupta, Z. Liu, R. R. Aburto, L. Ge, 58. J. Shen, Y. Zhu, C. Chen, X. Yang, C. Li, Chem. Commun. 47,
L. Song et al. Nano Lett. 12, 2012, 844. 2011, 2580.
25. H. Tetsuka, R. Asahi, A. Nagoya, K. Okamoto, I. Tajima, R. 59. S. Chen, J. Liu, M. Chen, X. Chen, J. Wang, Chem. Commun.
Ohta, A. Okamoto, Adv. Mater. 24, 2012, 5333–5338. 48, 2012, 7637.
26. S. H. Jin, D. H. Kim, G. H. Jun, S. H. Hong, S. Jeon, ACS Nano 60. L. Cao, X. Wang, M. J. Meziani, F. Lu, H. Wang, P.G. Luo, Y.
7, 2013, 1239. Lin et al. J. Am. Chem. Soc. 129, 2007, 11318.
27. H. Sun, N. Gao, L. Wu, J. Ren, W. Wei, X. Qu, Chem. Eur. J. 61. J. Li, H. Bao, X. Hou, L. Sun, X. Wang, M. Gu, Angew. Chem.
19, 2013, 13362. Int. Ed. 51, 2012, 1830.
28. G. S. Kumar, R. Roy, D. Sen, U. K. Ghorai, R. Thapa, N. 62. S. Zhao, M. Shao, S. T. Lee, ACS Nano 6, 2012, 1059.
Mazumder, S. Saha, K. K. Chattopadhyay, Nanoscale 6, 2014, 63. S. Zhu, J. Zhang, X. Liu, B. Li, X. Wang, S. Tang, Q. Meng
3384. et al. RSC Adv. 2, 2012, 2717.
29. L. Wang, S. Zhu, H. Wang, Y. Wang, Y. Hao, J. Zhang, Q. 64. M. Li, W. Wu, W. Ren, H. Cheng, N. Tang, W. Zhong, Y. Du,
Chen, Y. Zhang, W. Han, B. Yang, H. Sun, Adv. Opt. Mater. 1, Appl. Phys. Lett. 101, 2012, 103107.
2013, 264. 65. D. Tan, S. Zhou, J. Qiu, ACS Nano 6, 2012, 6530.
30. S. Zhu, J. Zhang, S. Tang, C. Qiao, L. Wang, H. Wang, X. Liu 66. M. Shao, S. Zhuo, ACS Nano 6, 2012, 6532.
et al. Adv. Funct. Mater. 22, 2012, 4732–4740. 67. C. Zhang, Y. Liu, X. Xiong, L. Peng, L. Gan, C. Chen, H. Xu,
31. F. Liu, M. H. Jang, H. D. Ha, J. Kim, Y. Cho, T. S. Seo, Adv. Org. Lett. 14, 2012, 5912.
Mater. 27, 2013, 3657. 68. J. Lee, K. Kim, W. I. Park, B. H. Kim, J. H. Park, T. H. Kim,
32. K. Lingam, R. Podila, H. Qian, S. Serkiz, A. M. Rao, Adv. S. Bong et al. Nano Lett. 12, 2012, 6078.
Funct. Mater. 23, 2013, 5062. 69. R. Liu, D. Wu, X. Feng, K. Müllen, J. Am. Chem. Soc. 133,
33. P. Luo, Z. Ji, C. Li, G. Shi, Nanoscale, 5, 2013, 7361. 2011, 15221.
34. Y. Feng, J. Zhao, X. Yan, F. Tang, Q. Xue, Carbon 66, 2014, 334. 70. Y. Dong, J. Shao, C. Chen, H. Li, R. Wang, Y. Chi, X. Lin, G.
35. Z. Qian, J. Ma, X. Shan, L. Shao, J. Zhou, J. Chen, H. Feng, Chen, Carbon 50, 2012, 4738.
RSC Adv. 3, 2013, 14571. 71. M. Nurunnabi, Z. Khatun, K. M. Huh, S. Y. Park, D. Y. Lee,
36. L. Tang, R. Ji, X. Li, K. S. Teng, S. P. Lau, Part. Part. Syst. K. J. Cho, Y. Lee, 7, 2013, 6858.
Charact. 30, 2013, 523. 72. S. Zhu, J. Zhang, C. Qiao, S. Tang, Y. Li, W. Yuan, B. Li et al.
37. X. Li, S. Lau, L. Tang, R. Ji, R. Yang, J. Mater. Chem. C 1, Chem. Commun. 47, 2011, 6858.
2013, 7308. 73. M. Zhang, L. Bai, W. Shang, W. Xie, H. Ma, Y. Fu, D. Fang
38. D. Pan, J. Zhang, Z. Li, M. Wu, Adv. Mater. 22, 2010, 734. et al. J. Mater. Chem. 22, 2012, 7461.
39. Q. Xu, Q. Zhou, Z. Hua, Q. Xue, C. Zhang, X. Wang, D. Pan, 74. D. Pan, L. Guo, J. Zhang, C. Xi, Q. Xue, H. Huang, J. Li et al.
M. Xiao, ACS Nano 12, 2013, 10654. J. Mater. Chem. 22, 2012, 3314.
40. S. Kim, S. W. Hwang, M. K. Kim, D. Y. Shin, D. H. Shin, C. 75. Y. Dong, C. Chen, X. Zheng, L. Gao, Z. Cui, H. Yang, C. Guo,
O. Kim et al. ACS Nano 6, 2012, 8203. Y. Chi, C. Ming Li, J. Mater. Chem. 22, 2012, 8764.
41. F. Yang, M. Zhao, B. Zheng, D. Xiao, L. Wu, Y. Guo, J. Mater. 76. F. Jiang, D. Chen, R. Li, Y. Wang, G. Zhang, S. Li, J. Zheng
Chem. 22, 2012, 25471. et al. Nanoscale 5, 2013, 1137.
42. D.B. Shinde, V.K. Pillai, Chem. Eur. J. 18, 2012, 12522. 77. C. Hu, Y. Liu, Y. Yang, J. Cui, Z. Huang, Y. Wang, L. Yang, H.
43. L. Minati, S. Torrengo, D. Maniglio, C. Migliaresi, G. Wang, Y. Xiao, J. Rong, J. Mater. Chem. B 1, 2013, 39.
Speranza, Mater. Chem. Phys. 137, 2012, 12. 78. J. Zhao, G. Chen, L. Zhu and G. Li, Electrochem. Commun.
44. M. Xie, Y. Su, X. Lu, Y. Zhang, Z. Yang, Y. Zhang, Mater. 2011, 13, 31.
Lett. 93, 2013, 161. 79. H. Sun, L. Wu, N. Gao, J. Ren, X. Qu, ACS Appl. Mater.
45. H. Zhao, Y. Chang, M. Liu, S. Gao, H. Yu, X. Quan, Chem. Interfaces 5, 2013, 1174.
Commun. 49, 2013, 234. 80. Z. M. Markovic, B. Z. Ristic, K. M. Arsikin, D. G. Klisic, L.
46. L. Fan, Y. Hu, X. Wang, L. Zhang, F. Li, D. Han, Z. Li, Q. M. Harhaji-Trajkovic, B. M. Todorovic-Markovic, D. P. Kepic
Zhang, Z. Wang, L. Niu, Talanta 101, 2012, 192. et al. Biomaterials 33, 2012, 7084.
47. X. Ran, H. Sun, F. Pu, J. Ren, X. Qu, Chem. Commun. 49, 81. X. Zhou, Y. Zhang, C. Wang, X. Wu, Y. Yang, B. Zheng, H.
2013, 1079. Wu, S. Guo, J. Zhang, ACS Nano 6, 2012, 6592.
48. J. Liu, X. Zhang, Z. Cong, Z. Chen, H. Yang, G. Chen, 82. Y. Li, Y. Hu, Y. Zhao, G. Shi, L. Deng, Y. Hou, L. Qu, Adv.
Nanoscale 5, 2013, 1810. Mater. 23, 2011, 776.
49. Y. Dong, G. Li, N. Zhou, R. Wang, Y. Chi, G. Chen, Anal. 83. V. Gupta, N. Chaudhary, R. Srivastava, G. D. Sharma, R.
Chem. 84, 2012, 8378. Bhardwaj, S. Chand, J. Am. Chem. Soc. 133, 2011, 9960.
50. J. Bai, L. Zhang, R. Liang, J. Qiu, Chem. Eur. J. 19, 2013, 84. M. Dutta, S. Sarkar, T. Ghosh, D. Basak, J. Phys. Chem. C
3822. 116, 2012, 20127.
85. L. Chen, C. Guo, Q. Zhang, Y. Lei, J. Xie, S. Ee, G. Guai, 93. H. Razmi, R. M. Rezaei, Biosens. Bioelectron. 41, 2013, 498.
Q. Song, C. Li, ACS Appl. Mater. Interfaces 5, 2013, 2047. 94. H. Chakraborti, S. Sinha, S. Ghosh, S. K. Pal, Mater. Lett. 97,
86. Z. Gan, X. Wu, G. Zhou, J. Shen, P. K. Chu, Adv. Opt. Mater. 2013, 78.
1, 2013, 554. 95. H. Cheng, Y. Zhao, Y. Fan, X. Xie, L. Qu, G. Shi, ACS Nano
87. L. Tang, R. Ji, X. Cao, J. Lin, H. Jiang, X. Li, K.S. Teng, C.M. 6, 2012, 2237.
Luk, S. Zeng, J. Hao, S.P. Lau, ACS Nano 6, 2012, 5102. 96. Y. Ding, H. Cheng, C. Zhou, Y. Fan, J. Zhu, H. Shao, L. Qu,
88. C. M. Luk, L. B. Tang, W. F. Zhang, S. F. Yu, K. S. Tengb, S. Nanotechnology 23, 2012, 255605.
P. Lau, J. Mater. Chem. 22, 2012, 22378. 97. G. Zhang, H. Zhang, X. Zhang, S. Zhu, L. Zhang, Q. Meng, M.
89. W. Liu, Y. Feng, X. Yan, J. Chen, Q. Xue, Adv. Funct. Mater. Wang, Y. Li, B. Yang, J. Mater. Chem. 22, 2012, 21218.
23, 2013, 4111. 98. Y. Li, Y. Zhao, H. Cheng, Y. Hu, G. Shi, L. Dai, L. Qu, J. Am.
90. Y. Hu, Y. Zhao, G. Lu, N. Chen, Z. Zhang, H. Li, H. Shao, L. Chem. Soc. 134, 2012, 15.
Qu, Nanotechnology 24, 2013, 195401. 99. X. Zhang, X. Xie, H. Wang, J. Zhang, B. Pan, Y. Xie, J. Am.
91. X. Wang, L. Chen, X. Su, S. Ai, Biosens. Bioelectron. 47, Chem. Soc. 135, 2013, 18.
2013, 171. 100. X. Chen, X. Zhou, T. Han, J. Wu, J. Zhang, S. Guo, ACS Nano
92. Y. Liu, P. Wu, ACS Appl. Mater. Interfaces 5, 2013, 3362. 7, 2013, 531.
11 Electrochemical Biosensors and
Biofuel Cells Based on Graphene
and Graphene Derivatives
Jaroslav Filip, Tomas Bertok, and Jan Tkac
CONTENTS
Abstract...................................................................................................................................................................................... 179
11.1 Introduction...................................................................................................................................................................... 179
11.2 Biosensors......................................................................................................................................................................... 180
11.2.1 Enzymatic Biosensor.............................................................................................................................................181
11.2.1.1 Amperometric Devices...........................................................................................................................181
11.2.1.2 Enzymatic FET-Based Biosensors......................................................................................................... 183
11.2.2 DNA Biosensors................................................................................................................................................... 184
11.2.2.1 Amperometric Devices.......................................................................................................................... 184
11.2.2.2 Impedimetric Devices............................................................................................................................ 186
11.2.2.3 FET Devices.......................................................................................................................................... 186
11.2.3 Immunosensors..................................................................................................................................................... 188
11.2.3.1 Amperometric Devices.......................................................................................................................... 188
11.2.3.2 FET Devices.......................................................................................................................................... 190
11.2.3.3 Impedimetric Devices............................................................................................................................ 190
11.2.4 Aptamer-Based Sensors.........................................................................................................................................191
11.2.4.1 Amperometric Devices...........................................................................................................................191
11.2.4.2 FET and Impedimetric Devices............................................................................................................. 192
11.2.5 Lectin–Glycan-Based Biosensors......................................................................................................................... 193
11.2.6 Other Graphene-Based Biosensors....................................................................................................................... 194
11.3 Biofuel Cells..................................................................................................................................................................... 194
11.3.1 Microbial BFCs.................................................................................................................................................... 194
11.3.2 Enzymatic Biofuel Cells....................................................................................................................................... 196
11.4 Summary, Outlook, and Perspective................................................................................................................................ 197
References.................................................................................................................................................................................. 197
ABSTRACT detection platform based on amperometric, impedimetric,

or field-effect transistor-based sensing will be discussed for
The most attention, from all carbonaceous nanomaterials, every group of biorecognition elements. Particular attention
is currently being paid to graphene—2D, one-atom-thick, will be paid to novel trends in immobilization techniques and
highly ordered carbon sheets with excellent electrochemical, signal amplifying methods (e.g., using metal nanoparticles or
thermal, optical, and mechanical properties. From an applica- electrochemically active labels). Finally, immobilization of
tion point of view, a good biocompatibility of graphene and its bioelectrocatalysts or microbial cells on graphene-based elec-
derivatives is highly appreciated and increasingly exploited for trode interfaces for construction of enzymatic and microbial
integration of biorecognition elements and bioelectrocatalysts biofuel cells will be covered in a separate part of the chapter.
in order to construct biosensors and biofuel cells, respectively.
First, in this chapter, various bioconjugation techniques
utilized for immobilization of the main groups of biorecogni-
11.1 INTRODUCTION
tion elements, that is, enzymes, antigens/antibodies, nucleic
acids, aptamers, and lectins/glycans on graphene-based mate- In the last few decades, the possibilities and new ways to
rials will be covered with a focus on construction of electro- employ biocatalysts in electrochemistry have been stud-
chemical biosensors. Second, the choice of an electrochemical ied with ever-increasing intensity. In the construction of
179
electrochemical biosensors, an interaction of enzymes with intensively studied and reviewed (see, e.g., Dreyer et al. 2010;
specific substrates, antigens with antibodies, nucleotides with Paredes et al. 2011, and references therein). Nevertheless, all
their complementary counterparts, aptamers with proteins, forms of rGO have in common a high hydrophobicity and a
and lectins with glycans are used to determine or quantify tendency to aggregate, what complicates handling of rGO
biological analytes of interest. The importance of biosensors or graphene solutions in a simple and a reproducible way, a
increases currently with their practical applicability to detect feature important for preparation of biosensors and BFCs.
not only biologically important molecules [e.g., glucose and Besides simple preparation of the rGO dispersions in non-
nicotinamide adenine dinuclotide (NADH)], but also patho- polar solvents or polymers, other methods are relying on
logically relevant biomarkers with a direct correlation to modification of a GO surface with a partially hydrophilic
development and/or progression of certain diseases (e.g., can- “cover,” for example, electroactive compounds (Guo et al.
cer and rheumatoid arthritis). 2011), polymers (Choi et al. 2011), or other films (Zhou et al.
Biofuel cells (BFCs) are devices where the chemical 2010). Consequent reduction of the protected GO results in
energy of a substrate is turned into electricity utilizing bio- still hydrophilic, yet reduced, that is, conductive, nanomate-
catalysts. Just like a conventional fuel cell, a BFC consists of rial. Additionally, modification of GO with metal nanopar-
a conductively connected pair of electrodes—an anode and a ticles exhibits similar antiaggregation effect (Feng et al. 2011).
cathode—where a coupled catalyzed oxidation of a supplied In this chapter, application of graphene-based materials is
(bio)fuel and a reduction of a depolarizator are carried out. summarized for construction of BFCs and electrochemical
It is obvious that the specificity of the biocatalysts toward biosensors, that is, devices employing integration of biocata-
one substrate is not a prerequisite for BFCs, rather it can be lysts/biorecognition elements with graphene. For biosensors,
a limiting factor due to an incomplete utilization of a sub- only amperometric, impedimetric and field-effect transistor
strate. More complete and effective oxidation of the biofuel is (FET)-based devices combined with biorecognition elements,
reached when whole microbial cells or multiple enzyme cas- such as enzymes, antigens/antibodies, nucleotides, aptamers,
cades are employed as biocatalysts instead of a single enzyme. and glycans/lectins are covered. Graphene-based BFCs are
Effectivity of carbon nanomaterials for the immobiliza- divided into microbial and enzymatic ones for better conve-
tion of biocatalysts and biorecognition molecules on electrode nience of a reader.
interfaces has been recognized and explored very intensively
in the past years. Especially, an enhancement of an immo-
11.2 BIOSENSORS
bilization surface area and very good electron conductivity
of carbon nanotubes (CNTs) was employed for the construc- The first introduction of an enzyme as a biorecognition part
tion of many robust bioelectronic devices (Feng and Ji 2011). of an electrochemical biosensor was reported in 1962, when
Recently, graphene—material consisting of separated, one- Clark et al. described an electrode detecting a change in
atom-thick planar sheets of sp2 hybridized carbon atoms— oxygen concentration caused by an enzymatic glucose oxi-
has been recognized as an even better electron conductor than dation (Clark and Lyons 1962). This indirect amperometric
CNTs. Besides graphene’s conductivity and excellent optical, response is an operational base of the first-generation biosen-
thermal, electronic, and mechanical properties (reviewed, sors, while in the second-generation devices the enzyme’s
e.g., in Castro Neto et al. 2009; Soldano et al. 2010), another cofactor is regenerated via oxidation and reduction of addi-
important feature of this material is that it can be readily pre- tional substances—electron mediators (a mediated electron
pared much cheaper than CNTs, most often by oxidation of transfer, MET). In the third-generation biosensors, electrons
graphite in strong mineral acids with a subsequent exfoliation are transported directly between enzymes and the electrode
of single sheets by an ultrasound, what leads to a material (a direct electron transfer, DET) (Figure 11.1a). An ampero-
termed graphene oxide (GO). In GO, conjugated sp2 hybrid- metric response provided by enzymatic oxidation/reduction
ized C─C bonds, that is, in-plane electron conductivity, is of substrates is also utilized for construction of label-based
partially disturbed. On the other side, GO provides many affinity biosensors. For example, in a so-called “sandwich
surface oxygen moieties (carboxyls, hydroxyls, epoxy groups, assay,” an enzyme-conjugated secondary biorecognition ele-
etc.), which enhance its solubility in aqueous solutions and ment (tracer probe) is anchored to the biosensor surface via
allow a simple surface modification of GO. Very often, a par- analyte molecules bound to an immobilized primary capture
tial restoration of in-plane conductivity is desired in order to molecule. The read-out signal is then generated by an enzy-
gain the best features of the constructed device. For this pur- matic reaction and its level is dependent on the concentration
pose, GO is reduced to obtain a material termed reduced GO of the analyte (Figure 11.1b). An amperometric detection of
(rGO) to distinguish it from a chemically pristine graphene nonenzymatic labels, for example, guanine, in so-called bio-
obtained either by a physical exfoliation of graphite or by the barcode deoxyribonucleic acid (DNA) assays (Wang et al.
chemical vapor deposition (CVD) method (Singh et al. 2011). 2012), or inorganic electrochemical probes other than inor-
There are three common ways to prepare rGO: thermal reduc- ganic mediators (Wu et al. 2013), is another way of employing
tion (TrGO), chemical reduction (CrGO), and electrochemi- voltamperometric methods in affinity biosensing.
cal reduction (ErGO) with a diverse range of properties of Unlike amperometry, electrochemical impedance spectros-
resulting rGO influenced by the choice of a preparation pro- copy (EIS) does not rely on redox transformation of analytes
tocol. Such dependence of rGO on a reduction protocol was or labels, but on changed interfacial properties of a biosensor
Electrochemical Biosensors and Biofuel Cells Based on Graphene and Graphene Derivatives 181
(a) (b) Gate

electrode
Tracer probe
Potentiostat
Drain Source
1st generation 2nd generation 3rd generation Channel
Analyte Mediator Secondary capture
Electrochemical Analyte
Electron transfer
transformation Primary capture FIGURE 11.3 A schematic depiction of a field effect transistor-
Tracer substrate
based (bio)sensor device.
FIGURE 11.1 (a) Schematic depiction of enzymatic biosensor

generations with analyte (spheres) and mediator (triangles) electro- see Figure 11.2a), it is possible to obtain so-called charged
chemical transformation. (b) A schematic depiction of a sandwich transfer resistance (RCT) of the surface. This will change upon
assay with the tracer probe loaded with an enzyme with its substrate an immobilization/attachment of additional molecules on the
represented by hexagons. surface in a pattern depicted in Figure 11.2b. Thus, the attach-
ment of analyzed complementary DNA strands, antibodies or
after biorecognition took place by biasing the electrode with aptamer-bound proteins can be detected and evaluated.
voltage altering with small amplitude in an optional frequency By stretching a layer of graphene between two electrodes
window. The measurement is performed typically in the pres- labeled as “drain” and “source,” a semiconductor-like device
ence of [Fe(CN)6]−3/−4 probe electrochemically transformed by is obtained whose conductivity pattern, that is, changing
the applied bias, and the resultant current is used to deter- between p- and n-type conductivity, is controlled by a poten-
mine an impedance of the given electrode surface for each tial applied to an additional “gate” electrode (Novoselov et al.
frequency. Data obtained are evaluated in a form of a Nyquist 2004). Such a configuration (see Figure 11.3) is called field-
plot, where an imaginary part of the impedance is plotted effect transistor (FET), and graphene was found to be a very
against its real counterpart (Figure 11.2a). Assuming that the appropriate material for their fabrication because of the pos-
modified electrode surface can be expressed and the Nyquist sibility to prepare single-layer sheets containing minimum
plot fitted as an electrochemical double layer with capaci- defects. Under the set parameters, that is, the gate voltage, the
tance, resistance, and diffusivity (Randles equivalent circuit, resultant drain–source current varies with the changing char-
acteristics of the environment it is placed in. If such changes
are induced by interaction of an immobilized biorecognition
(a) CDL RS—Solution resistance element with its analyte, the device is recognized as an FET-
RS RCT—Charge transfer resistance based biosensor.
RCT W CDL—Double layer capacitance
W—Diffusion-related element
–Zimg (Ω)
11.2.1 Enzymatic Biosensor
11.2.1.1 Amperometric Devices
W
RS RCT The most enzymatic amperometric biosensors were con-
structed using glucose oxidase (GOx), a flavin adenine dinu-
Zreal (Ω)
cleotide (FAD)-dependent oxidoreductase oxidizing glucose
(b) Redox probe according to the following equation:
Analyte
Glucose + O2 GOx
 → gluconic acid + H 2O2 (11.1)

In the simplest approach, mixtures of hydrophobic rGO

sheets and a volatile dispersing agent or a proper polymer
matrix (chitosan, Nafion®, etc.) have been used for fabrication
of an electrode interface and for subsequent GOx physisorp-
tion. In such an arrangement, rGO enhances a current response
FIGURE 11.2 (a) A typical Nyquist plot with a solution resis-
resulting from surface reduction of O2 or H2O2 as indirect
tance RS, a charge transfer resistance RCT, and a diffusion element
W indicated. Inset—A Randles equivalent circuit. (b) A schematic markers of GOx activity. While in a former case decreasing
depiction of an EIS biosensor operation with redox probe being elec- response with an increasing analyte concentration is observed
trochemically transformed on the electrode surface only in absence (Kang et al. 2009), in the latter one the response increases with
of an analyte. a higher analyte concentration (Wu et al. 2009; Baby et al.
2010; Piao et al. 2014). Composition and characteristics of GOx although an ultimate detection limit (LOD) of 3 pM has been
biosensors using these detection mechanisms are listed in Table obtained on an rGO interface doped with Au nanoparticles
11.1. Electrochemical detection of oxygen or hydrogen peroxide and CdS-CdTe quantum dots highly promoting H2O2 reduc-
have also been used in biosensors employing other oxidases, tion rate (Zhiguo et al. 2011).
for example, cholesterol oxidase (Aravind et al. 2011), oxalate Besides the ability of an rGO-based composite to detect
oxidase (Devi et al. 2013), and xanthine oxidase (Zhang et al. O2 or H2O2, it is also important to focus on the integration
2010). Such versatility of oxidases was used in the development of a high amount of the enzyme onto the interface, what has
of a miniaturized biosensor array. It consisted of four adjacent been achieved by entrapment of GOx into the applied polymer
planar electrodes printed with graphene ink, each one incubated matrix (Kang et al. 2009), or by immobilization of the enzyme
with a different oxidase. By reading amperometric responses within a film prepared by a layer-by-layer deposition of rGO
related to oxidation of enzymatically generated H2O2, simul- sheets altering with the pyrene-modified GOx (Liu et al. 2013).
taneous enzymatic detection of glucose, lactate, xanthine, and Instead of oxygen, some other electron shuttles such as
cholesterol was possible (Labroo and Cui 2014). ferrocene derivatives (Qiu et al. 2011; Mani et al. 2013) or
A diverse range of metal nanoparticles attached to or syn- hydroquinone (Zhou et al. 2010) can be used as mediators,
thesized on rGO surface are reported to increase sensitivity either coimmobilized with the electrode interface (Qiu et al.
for amperometric sensing of either O2 or H2O2 (Wu et al. 2011) or dissolved in an electrolyte solution, although the lat-
2009; Baby et al. 2010; Shan et al. 2010; Wang et al. 2011). ter arrangement represents a lower biocompatibility of the
Metallophthalocyanines (Zhang et al. 2013) and nonmetallic device. MET-based biosensors, on the other hand, allow glu-
porphyrins (Zhang et al. 2011) have also been tested in the cose detection to be performed at a more favorable potential
conjugation with rGO, with reported sensitivities of biosen- of 300 mV versus Ag/AgCl (Qiu et al. 2011), compared with
sors being approximately twofold higher compared with GOx a typically applied potential of—400 mV applied for GOx-
covalently immobilized (Liang et al. 2013), or immobilized biosensors based on a H2O2 reduction.
in chitosan matrix on an ErGO (Cui et al. 2013), and twofold Recently, even a direct, mediatorless electrochemical
lower compared with GOx adsorbed on TiO2 (Luo et al. 2013) transformation of an FADH cofactor in GOx (Choi et al. 2011)
or Cu nanoparticles-loaded rGO (Huang et al. 2013). Quite and in sarcosine oxidase (Zhou et al. 2012) has been reported.
surprisingly, a physisorption of GOx directly on an rGO sur- In the former case the enzyme has been adsorbed on the poly-
face provided the sensitivity of a biosensor close to the afore- mer-protected rGO, though in the latter case the DET was fur-
mentioned metallic nanoparticle-based biosensors (see Table ther promoted by an integration of Ag nanoparticles. Electron
11.1). A very high sensitivity was obtained for GOx adsorbed exchange between a cofactor of GOx and rGO-based elec-
on rGO coated with Au-Pd nanoparticles (Yang et al. 2011), trode was also achieved with the help of polyaniline (PANI), a
TABLE 11.1
Operational Characteristics and Detection Systems of Different Graphene-Based Enzymatic Glucose
Biosensors Using GOx
LOD Sensitivity
(μM) (μA mM−1 cm−2) Detection Biosensor Arrangement Reference
0.6 N/A H2O2 Thermally exfoliated graphite/CHI/PtNP Wu et al. (2009)
1 N/A H2O2 Thermally exfoliated graphite/Naf/Pt-AuNP Baby et al. (2010)
180 0.55a O2 PvP-protected CrGO/AuNP Shan et al. (2010)
N/A N/A H2O2
1 22.8 H2O2 ErGO Ping et al. (2011)
700 1.76 O2 GO/CdSNP Wang et al. (2011)
6.9 267 O2 ErGO/Au-PdNP/Naf Yang et al. (2011)
0.2 17 O2 ErGO/porphyrins Zhang et al. (2011)
3 pM 5.8 H2O2 GO/AuNP + CdTe-CdSNP/AuNP/CHI Zhiguo et al. (2011)
N/A 6.2 DET TrGO/TiO2NP/CHI Jang et al. (2012)
1.7 6.8 O2 ErGO/GOx dispersed in CHI Cui et al. (2013)
5 34 O2 CrGO/Naf/CuNP Huang et al. (2013)
20 7 O2 ErGO-COOH/GOx covalently attached Liang et al. (2013)
50 17.5 O2 (G/AB-G/TSCuPC)n/GOx/Naf Zhang et al. (2013)
0.6 N/A DET rGO/PANI/AuNP/GOx Xu et al. (2014)
N/A 3.5 DET rGO/GOx/Naf Zhang et al. (2014)
4.8 29.7 DET CrGO/MWCNT/PDDA-AuNP/GOx Yu et al. (2014)
1 24 H2O2 CrGO/Pd-PtNP/GOx Hossain and Park (2014)
a Measured in a blood sample.

conductive polymer (Xu et al. 2014). The same composite com- Immobilized tyrosinase (Liu et al. 2012) and laccase (Lac)
prising rGO, PANI and Au nanoparticles in chitosan matrix (Oliveira et al. 2013) were reported to catalyze oxidation
was used in a glucose biosensor based on an electrochemical of phenolic substrates with products being further electro-
detection of FAD+. The cofactor is reported to be reduced on chemically transformed on electrode surfaces containing gra-
electrode surface, via DET, while being continuously reoxi- phene as an enhancer of the indirect amperometric response.
dized with oxygen. Thus, a reduction current is generated Recently, Zhou et al. has reported that even direct reoxidation
related to DET of FAD cofactor. Once glucose is added, this of the active sites of Lac attached to an rGO surface is pos-
reduction current decreased because glucose reduced FADH2 sible and can be employed for detection of phenol (Zhou et al.
instead of the electrode surface (Kong et al. 2014). This DET- 2013). For ascorbate oxidase, the direct Cu-active site reoxi-
based system of glucose detection was recently described for dation on the interface prepared by simultaneous electro-
GOx immobilized on different rGO-containing nanocom- chemical reduction of GO and electrodeposition of PEDOT
posites (Palanisamy et al. 2014; Palanisamy et al. 2014; Yu has been reported (Lu et al. 2013). Simple physisorption of
et al. 2014). DET of GOx was reported on rGO surfaces with Lac (Świetlikowska et al. 2013) or bilirubin oxidase (Filip
a decreased amount of surface oxygen groups, while higher et al. 2013) to ErGO electrodes turned out to be an effective
concentration of oxygen groups was favorable for H2O2 oxida- way for “a cathodic” DET, that is, for detection of dissolved
tion (Zhang et al. 2014). O2 by its reduction to water. Lac-catalyzed O2 reduction was
Hemoglobin (Hb) is a metalloprotein widely tested for used also for determination of human erythrocytes activity
integration with different nanomaterials. It was employed with rGO/mediator/Lac interface constituting an effective
for sensitive nitrate or nitric oxide biosensing after being system for detection of changes in O2 concentration related to
immobilized on rGO sheets coated with room temperature the activity of tested tissue cells (Wu et al. 2010).
ionic liquid or Au nanoparticles (Liu et al. 2010; Wen et al. Acetylcholine esterase (AcE) performs transformation of
2012; Jiang et al. 2014). These nanomaterials enhanced DET a supplied acetylthiocholine into thiocholine, which can be
between Hb active site and an electrode. Direct NO2− enzy- easily amperometrically detected with graphene-based elec-
matic electrochemical transformation was also observed trodes to evaluate the enzyme’s activity. This was employed
for myoglobin (Mb) integrated into a GO/Nafion composite in the construction of AcE-based biosensors detecting trace
(Guo et al. 2012). Similarly, trichloracetic acid can be directly amounts of carbamate pesticides that inhibit an activity of
reduced by a heme moiety of Hb (Sun et al. 2013), horseradish AcE. Pesticides could be detected down to sub-pM concen-
peroxidase (HRP) (Sun et al. 2012) and Mb (Ruan et al. 2012), tration using AcE immobilized on metal nanoparticles-dec-
what has been applied for fabrication of biosensors detecting orated graphene (Liu et al. 2014; Wang et al. 2014) and with
this analyte. More often, however, peroxidase-like activity LOD below 1 ppb on a porous rGO/chitosan composite (Li
of Hb and Mb, and peroxidase activity of HRP and catalase et al. 2013) and on CrGO/chitosan-AuNP composite with
have been employed for H2O2 biosensors’ construction with covalently immobilized AcE (Liu et al. 2011).
pristine (Xu et al. 2010; Guo et al. 2012; Zhang et al. 2012) or To complete a list of oxidoreductases used for construction
metallic nanoparticles grafted (Zhou et al. 2010; Huang et al. of enzymatic graphene-based biosensor, NADH-dependent
2011; Feng et al. 2012; Liu et al. 2014; Song et al. 2014) rGO as alcohol dehydrogenase (ADH) should be mentioned. Pristine
the components promoting effective electron exchange rates. rGO is reported to effectively regenerate NADH at decreased
It has been revealed that a direct integration of some overpotential (Guo et al. 2011). Thus, a simple rGO/ADH
enzymes with GO can induce destabilization of their struc- ethanol biosensor has been fabricated with further overpoten-
tures and activity loss (Jin et al. 2012). On the other hand, tial decrease reached by immobilization of ADH on polythio-
modification of rGO by DNA strands improved an immobili- nine-coated rGO sheets (Li et al. 2013). Nevertheless, NAD+
zation environment for HRP (Zhang et al. 2010), while an incu- needed to be supplied to the electrolytes for all ADH-based
bation of the interface with ferrocene resulted in an enhanced biosensors. Similarly, lactate oxidase adsorbed on photocata-
HRP’s heme regeneration (Deng et al. 2013). Grafting of lytically reduced GO was applied for preparation of a lactate
rGO with positively charged polymers, for example, poly-l- biosensor with hydroxymethylferrocene as a supplied electron
lysine (Wang et al. 2013) or poly(3,4-ethylenedioxythiophene) mediator (Casero et al. 2014).
(PEDOT) has also provided an effective immobilization
matrix. Furthermore, the conductivity of PEDOT allowed the 11.2.1.2 Enzymatic FET-Based Biosensors
fabrication of the cytochrome c-based H2O2 biosensor with GOx activity, according to Equation 11.1, leads to a pH drop
excellent features such as detection limit of 0.25 μM and a caused by the generated gluconic acid. This was employed
linear response between 0.5 and 400 μM (Wang et al. 2012). in glucose FET-based biosensors (Huang et al. 2010; Kwak
Zhang et al. reported an efficient DET of HRP immobi- et al. 2012), where a conductance of enzyme-modified gra-
lized on partially reduced GO (24 h incubation with ascor- phene channels changed with changed pH. Subsequently, a
bic acid) not only for H2O2 reduction, but also for a direct measured source–drain current increased proportionally with
oxidation of phenol in the presence of H2O2 (Zhang et al. increased glucose concentration. In these devices, the CVD-
2012). Oxidation of phenolic compounds is, however, more grown graphene channel was modified with pyrenebutanoic
effectively performed in the presence of O2 by enzymes acid succinimidyl ester to covalently attach GOx (see Figure
with copper atoms as active sites (“multicopper oxidases”). 11.4). Even though operational characteristics of such devices
Pt G nanowalls”-based genosensor was revealed with a detection

Electrolyte Vg
limit of 9.4 zM for dsDNA (Akhavan et al. 2012), and an inte-
PDMS
Ag paint gration of metallic nanoparticles (Du et al. 2010; Fan et al.
well
2011) or conductive polymers (Scott et al. 2010) is an addi-
tional way for a signal enhancement. The direct DNA elec-
D trochemistry was employed for a determination of (C + A)/
S (T + G) ratio in a ssDNA (Fan et al. 2010), for a hybridization-
Graphene PSE GOD based detection of hepatitis B virus (Muti et al. 2011) and as
PET substrate Vds a signal enhancer in biobarcode DNA-based biosensors (vide
infra). Furthermore, an effective protection of DNA and RNA
FIGURE 11.4 A scheme of glucose oxidase (GOD)-based FET
strands by GO sheets against an enzymatic cleavage was
biosensor using CVD-grown graphene. (Reprinted from Biosens. described, what is an important feature for analysis of com-
Bioelectron, 37, Kwak, Y. H. et al., Flexible glucose sensor using plex biological samples (Lei et al. 2011; Cui et al. 2013).
CVD-grown graphene-based field effect transistor, 82–87, Copyright Quite often, the detection relied on electrochemical trans-
2015, with permission from Elsevier.) formation of secondary electrochemical probes. Typically,
upon hybridization of ssDNA capture probe attached to a
were not excellent, FET glucose biosensors can be effectively surface with target/analyte ssDNA, dsDNA is formed and
miniaturized. the amount of detectable intercalated electrochemical probe
In a similar way, glutamic dehydrogenase was employed increased (so-called “signal on” detection, see Figure 11.5a).
in a glutamate FET biosensor, where enzymatically generated Alternatively, the measured signal decreased due to a much
NH3+ ions cause more pronounced change in the graphene lower affinity of the probe to the formed dsDNA than to a cap-
channel conductance (Huang et al. 2010). In a simplified ture ssDNA (Figure 11.5b) in a “signal off” detection (Yang,
arrangement, lactic acid was detected by a CVD-graphene/ Guo et al. 2013). A list of such different electrochemical geno-
lactate oxidase resistance biosensor with LOD as low as sensors is given in Table 11.2.
80 nM, and with a linear response up to 20 μM. Furthermore, Also, a “signal on” arrangement was described, where the
by fixing of all three parts of the device on a poly(ethylene amperometric signal is detected only after the target DNA is
terephthalate) substrate, and due to the mechanical resil- hybridized with a capture probe, and the formed dsDNA is
ience of graphene, the device could be constructed as a flex- released from the electrode surface leaving it electrocatalyti-
ible sheet (Labroo and Cui 2013). rGO has been even used cally active (Yang et al. 2012). For such an approach, an inte-
for FET fabrication exploring the advantage of sensitiv- gration of rGO was proved to enhance electrochemistry of an
ity of rGO’s surface functional groups toward H+ released applied electrochemical probe (Sun et al. 2011) with further
upon enzymatic (AcE) hydrolysis of the analyte (acetylcho-
line) (Sohn et al. 2013). The same principle has been used
for detection of l-alanine aminotransferase. Here, a graphene
channel was loaded with the enzyme substrate (l-alanine and (a)
α-ketoglutarate); once the enzyme was present in the sample,
the substrate was utilized in the enzymatic reaction with a
detectable pH change (Her et al. 2013).
11.2.2 DNA Biosensors
11.2.2.1 Amperometric Devices
(b)
One of the first reports on nucleotides being directly electro-
chemically detected on the rGO-modified electrode comes
from Wu et al. (2005). Later, in a work of Zhou et al. (2009),
not only all four nucleotides were shown to generate distinc-
tive DPV peaks on a CrGO-modified electrode, but also a
superiority of rGO over graphite or a pristine carbon electrode
for the single-stranded DNA (ssDNA) and double-stranded Capture ssDNA Redox probe transformation
Complementary ssDNA/analyte Electron transfer
DNA (dsDNA) detection was revealed. Interfacial electro-
Redox probe
chemistry of single nucleotides has been revealed to be more
effective on the surface prepared from ErGO compared with
FIGURE 11.5 (a) A “signal on” configuration of amperomet-
CrGO (Du et al. 2010). A superiority of GO over rGOs in
ric genosensors with a redox probe amount increasing with more
the detection of a single nucleotide mismatch was reported dsDNA formed, that is, with increased concentration of target probe.
for the biosensor quantifying a total guanine amount after (b) A “signal off” configuration of amperometric genosensors where
DNA hybridization took place on the electrode surface (Loo a redox probe is released from an electrode surface after a hybridiza-
et al. 2013b). An excellent sensitivity of a “reduced graphene tion of capture and target probes.
TABLE 11.2
Characteristics of Different DNA Biosensors with Electrochemical Probes Intercalated into dsDNA
Immobilization Electrode Intercalated
Interface Electrochemical Probe LOD Linear Range Reference
GO/PANI nanowires Daunomycin 0.32 pM 2.1 pM–2.1 μM Bo et al. (2011)
GP/PB/CHI Daunomycin 16 pM 21 pM–21 nM Bo et al. (2011)
G/Fe3O4NP/CHI MB 0.36 pM 1 pM–1 μM Sun et al. (2011)
ErGO/PANI nanofibers [Ru(NH3)6]3+ 0.03 pM 0.1 pM–0.1 μM Du et al. (2012)
G/Co3O4 nanorod/CHI MB 0.43 pM 1 pM–1 μM Qi et al. (2012)
ErGO/AuNP MB 0.29 pM 1 pM–1 μM Sun et al. (2012)
Polylysine/rGO-COOH MB 0.17 pM 1 pM–1 μM Sun et al. (2012)
rGO/AuNP Adriamycin 40 fM 0.1 pM–10 nM Zhang and Jiang (2012)
rGO/Thi Daunomycin 0.13 pM 1 pM–0.1 μM Zhu et al. (2012)
Ac-G/PAMAM/AuNP MB 1.00 pM 1 pM–1 μM Jayakumar et al. (2012)
Thiolated CrGO/AuNP [Co(phen)2(Cl)(H2O)]+ 8.3 pM 25 pM–1.3 nM Niu et al. (2013)
ErGO/ZrO2/AuNP MB 12 fM 0.1 pM–0.1 μM Yang et al. (2013)
rGO/pyrenebutyric acid MB 0.4 fM 1 fM–5 pM Zhang et al. (2013)
GO/AE/Fe@AuNP BasicBlue 41 2 fM 10 fM–10 nM Yola et al. (2014)
CrGO/polydopamine/AgNP MB 3.2 fM 100 fM–10 nM Huang et al. (2014)
improvement achieved by an integration of conductive poly- sandwich genosensors is given in Table 11.3. A “supersand-
mers (Du et al. 2012) or metallic nanoparticles (Zhang and wich” arrangement was also tested relying on repetitive
Jiang 2012). Also, so-called graphene quantum dots, nanome- hybridization of captured target DNA and complementary
ter-sized graphene sheets deposited on a substrate electrode strands complexed with methylene blue. In this way, more
surface, were reported for efficient “signal on”-designed geno- molecules of the probe are detected on a substrate rGO-
sensor (Zhao et al. 2011). These biosensors could discriminate modified electrode (Wang et al. 2014). By a combination of
between complementary and single mismatched oligonucle- multi-G oligonucleotides and HRP amplification strategy on
otides, and exhibited a detection limit for complementary an rGO-modified electrode, an ultrasensitive DNA biosensor
strands in the range of pM, though detection in fM range was was fabricated with a limit of detection as low as aM level
reported with probe oligonucleotides attached on rGO surface (Dong et al. 2012). Using a similar approach, detection of
by π–π interactions (Zhang et al. 2013). This points out to the hepatocarcinoma-related microRNA was reported, though
importance of uninterrupted in-plane conductivity of a gra- with a higher detection limit of 60 fM (Yin et al. 2012). A
phene sheet, what was not the case for a genosensor based detection limit of the biosensor in the same range (10 fM)
on a chemically carboxylated (i.e., partially reoxidized) rGO was also published by Du et al. (2011), who developed a target
(Niu et al. 2013). As a proof of concept, Singh et al. tested DNA-templated polymerase chain reaction in order to multi-
their DNA biosensor based on GO/chitosan interface with an ply a number of probing dsDNA molecules (see Figure 11.6).
immobilized ssDNA probe for detection of Staphylococcus These probes were in a second step transferred to rGO/sil-
typhii antigen-coding gene in serum samples with a 100 fM ica/AuNP/ferrocene electrode interface with immobilized
detection limit reported, what is a promising step toward ssDNA capturing the prepared double strand probes in so-
commercial genosensors (Singh et al. 2013). called triplex DNA. This leads to a substantial decrease of
Very effective strategies to enhance the analytical signal recorded DPV peak of the incorporated ferrocene, the level
were developed based on a sandwich configuration (see Table of which depends on the amount of the PCR-prepared “prob-
11.3). The captured target oligonucleotides were hybridized ing” dsDNA, that is, primarily, on the amount of the tem-
with a tracer probe comprising an amperometric signal plating target molecules. Detection limit down to aM range
enabler/enhancer—for example, Au nanoparticles catalyz- has been also achieved with a tracer consisting of chemi-
ing a silver deposition (Lin et al. 2011), HRP catalyzing cally modified graphene sheets coated with a peroxidase-
detectable H2O2 reduction (Guo et al. 2013; Liu et al. 2013) mimicking porphyrine (Wang et al. 2013). This was shown
or multi-G nucleotides, where heterogeneous electrochemi- to be better than recently tested carboxylated graphene/
cal transformation of abundant guanine moieties resulted hemin probe employed in a sandwich-based genosensor with
in a detectable signal (so-called “biobarcode”) (Wang et al. LOD of 170 fM of microRNA (Zhou et al. 2014). Wang et al.
2012). The amplification itself relies on the fact that each employed so-called hairpin capture probes, folded oligo-
single tracer particle, attached to one molecule of hybrid- nucleotides which unfold upon hybridization with analyte.
ized target-capture dsDNA, consists of a nanoparticle coated This process exposed biotin molecules attached on the ends
with a number of electrochemically active molecules as can of capture probes and working as binder for streptavidin-
be seen in Figure 11.1b. An overview of graphene-based alkaline phosphatase (Wang et al. 2014).
TABLE 11.3
Characteristics of Different Genosensors Using Sandwich Arrangement or Biobarcode Assays
Electrode Modification
Tracer Probe Composition Electrochemical Detection LOD Linear Range Reference
Au electrode/rGO DPV of deposited silver 72 pM 200 pM–500 nM Lin et al. (2011)
AuNP with Ag staining
ErGO/AuNP-ssDNA DPV; mediated HRP catalysis 5 aM 10 aM–100 fM Dong et al. (2012)
CNS/streptavidina-HRP
GO-COOH/ssDNA SWV; Ru(bpy)32+-mediated guanine oxidation 1 pM 1 pM–1 nM Wang et al. (2012)
MagB/SA-BIO/poly[G]20
ErGO/AuNP/ssDNA CA; mediated HRP catalysis 3.4 fM 50 fM–5 pM Liu et al. (2013)
Reporter ssDNA/HRP
rGO/AuNP/ssDNA DPV; MB oxidation 350 aM 1 fM–100 nM Wang et al. (2014)
Reporter ssDNA/MBb
G/CHI/PANI/AuNP/CPc DPV; AP electrocatalysis 2.1 pM 10 pM–1 nM Wang et al. (2014)
SA/AP
AuNP/hairpin capture probe DPV; GO-hemin complex electrocatalysis 170 fM 0.5 pM–10 nM Zhou et al. (2014)
AuNP/ssDNA/GO-hemin
a Biotinylated target ssDNA needed.

b Multiplication of hybridization step, that is, increased amount of MB-labeled reporter ssDNA.
c CP, capture probe, that is, biotin-labeled hairpin probe.
11.2.2.2 Impedimetric Devices unpaired loop, can be employed instead of conventional oli-
Coupling of a target and a capture oligonucleotide on an elec- gonucleotides. Such folded strains constitute a very high RCT,
trode surface can result in hindrance of the heterogeneous elec- but they are very readily decoupled from the surface upon
trochemical transformation of a soluble electrochemical probe hybridization with complementary strands, as can be seen in
(Zhang et al. 2014). This can be measured by EIS, which was Figure 11.7. Applying hairpin capture probes, Bonanni et al.
employed, for example, in a DNA biosensor based on rGO/ (Bonanni and Pumera 2011) have reported detection of apo-
ssDNA interface. Hybridization of the target oligomer based lipoprotein E-related oligonucleotide with sub-pM detection
on methycilline-resistant Staphylococcus aureus gene was limit, while Chen et al. (2011) used hairpin probes attached
detected as the increase of surface charge transfer resistance to Au nanoparticles-modified graphene. By application of an
(RCT) with a detection limit of 100 fM. Hu et al. employed rGO additional exonuclease III-performed recycling amplification
grafted with Au nanoparticles with an even lower detection method, a detection limit as low as fM was obtained.
limit of 34 fM obtained (Hu et al. 2011) and a detection limit
of 4.2 fM was achieved employing rGO intercalated with con- 11.2.2.3 FET Devices
ducting polymer nanofibers deposited on the electrode surface The first report on an FET-based DNA device employing
and modified with the capture ssDNA (Yang et al. 2013). chemically modified GO is by Mohanty and Berry (2008). In
When the capture ssDNA is attached noncovalently their work, an increased conductivity of a GO-based channel
(using π–π interactions between bases and graphene), the is observed upon a covalent tethering of the ssDNA followed
hybridization with a complementary ssDNA releases the as- by another increase when hybridized with a target DNA. This
formed dsDNA from the surface, and thus, the measurable is explained by the presence of a partial negative charge of
RCT decreases. This was employed for a detection of damage DNA backbone inducing hole-type charge careers (negative
of DNA attached on a GO/hemin-modified electrode. It was gating) in p-type semiconductors (Mohanty and Berry 2008).
shown that exposure of such a modified electrode to a geno- Other FET-based DNA biosensors were then reported based
toxic agent decreased the overall RCT of an electrode surface on a pristine (Stine et al. 2010) or Pt nanoparticle-grafted rGO
(Ni et al. 2014). Employing a similar principle, genosensor (Yin et al. 2012) with ssDNA being covalently attached on a
was fabricated using PANI/rGO/ssDNA biorecognition inter- nanoparticle surface. Both biosensors exhibited the detection
face and exhibiting an impressive detection limit of 250 aM limit of about 2 nM, though in the latter case the detected
for a hosphinothricin acetyltransferase gene (Yang et al. negative shift of a minimum drain–source current was found
2013). Unmodified GO was also used in a similar manner, but out and explained by a so-called n-doping, that is, electron
probably because of poor electrochemical properties of GO supply from DNA to rGO channel. This was also observed
compared with rGO, the revealed limit of detection was only by Dong et al., who assembled a CVD-grown graphene with
110 fM (Hu et al. 2012). a physisorbed ssDNA into a biosensor with an LOD of 10 pM.
DNA “hairpin” probes, having one end of the strand hybrid- An integration of Au nanoparticles did not lower this value,
ized with the other end while the rest of the strand forms an but extended the dynamic range of the biosensor to higher
1 CTAB concentrations. Interestingly, even if the probe DNA strands

2 TEOS
(a) 3 Hydrazine AuNPs were attached covalently on the nanoparticles instead of the
4 APTES
physisorption on graphene, the detection pattern did not change
i ii
(Dong et al. 2010). To make it complete, FET biosensors were
(b) (A) reported based on few-layered CVD-grown graphene, where
(B) a negative gating and not n-doping was revealed to be a detec-
Fc-PEI
(A)
GSGHs
(B)
SH-ODN tion principle (Lin et al. 2010), with a limit of detection of
3 nM (Guo et al. 2011). In a work of Chen et al., a high-purity,
one-layer CVD-grown graphene was applied in a similar way.
Their findings suggest that its homogeneity together with a
Amplification detection DNA based on DNA polymerization:
(c) controlled purity of polymer substrate employed for a growth
Target DNA
of graphene sheets substantially lowered the detection limit
Recycled
i. Extension-based amplification blocker
of the biosensor down to 1 pM (Chen et al. 2013). Finally, the
LOD of 0.1 pM was observed on an rGO-based FET geno-
dNTPs and Klenow sensor with a peptide nucleic acid capture probe instead of
Hairpin template fragment polymerase
Duplex
conventional DNA oligonucleotides (Cai et al. 2014).
cumulation FET-based biosensors are sensitive to the pH of a sample,
which makes it practically impossible to repeat precise mea-
Primer
Extension product surements under slightly different conditions. Stine et al.
developed a two-channel FET genosensor which overcomes
ii. Triplex-based label-free signal amplification Ag+
this obstacle, with one channel being reference, unmodified,
and the other modified with a catching probe. The output
Triplex
Acceptor value is evaluated from the values read from both channels
and is free from nonspecific interactions and different pH
interferences (Stine et al. 2010).
A modification of FET called a high electron mobility
transistor was performed by a junction of FET with a “2D
electron gas,” a sheet of material with an enhanced electron
conductivity serving as an amplifier for the drain–source
FIGURE 11.6 (a) A scheme of fabrication of AuNP-modified CrGO.
current changes (see Figure 11.8). Zhang et al. prepared on
(b) Fabrication of DNA sensor. (c) A scheme of amplification mecha-
nism involving target-templated PCR and DNA triplex formation. this platform a DNA biosensor with rGO/ssDNA complex as
(Reprinted from Biomaterials, 32, Du, Y. et al., An integrated sensing a gate material. A very high amount of probe attached, plus
system for detection of DNA using new parallel-motif DNA triplex the aforementioned electron gas amplification strategy, were
system and graphene–mesoporous silica–gold nanoparticle hybrids, behind an excellent detection limit of 70 aM (Zhang et al.
8584–8592, Copyright 2011, with permission from Elsevier.) 2013).
Probe
12,000
Graphene
Wild
10,000
8000
–Zi (Ω)
6000
Mutant
4000
2000
0 NC
0 2000 4000 6000 8000 10,000 12,000 14,000 16,000
Zr (Ω)
FIGURE 11.7 Schematic depiction of impedance genosensors employing hairpin-DNA capture probes. (Reprinted with permission from
Bonanni, A., and M. Pumera. 2011. Graphene platform for hairpin-DNA-based impedimetric genosensing. ACS Nano 5 (3):2356–2361.
Copyright 2011 American Chemical Society.)
Ni/AuGe/Ni/Au AuNPs/antibody nanocomposite exhibiting an impressive

AlGaAs SiO2 isolation layer GaAs- detection limit of 50 ag mL −1 (0.25 aM) for a carcinoembry-
spacer doped onic antigen (Han et al. 2013). [Fe(CN)6]3− as the redox probe
Gate
Si cap
supplied directly into a measured solution exhibited redox
layer
AlGaAs-doped Si transformation at a potential not very far from 0 mV on most
2 DEG
nanoparticle-modified electrodes. Using this probe, Eissa
GaAs et al. assembled an inexpensive graphene screen-printed elec-
trode with covalently attached antibodies on its surface with
SI-GaAs substrate
a detection limit down to pg mL −1 range for β-lactoglobulin
(Eissa et al. 2012) and ovalbumin (Eissa et al. 2013). Yang
FIGURE 11.8 Scheme of a FET with 2D electron gas. Genosensor et al. developed a system of magnetic nanoparticles-modified
was prepared by modification of gate of this FET with rGO/ssDNA.
GO sheets possessing capture antibody. These modified GO
(Zhang, X.: Reduced graphene oxide-functionalized high electron
mobility transistors for novel recognition pattern label-free DNA nanoparticles were incubated with a sample solution and con-
sensors. Small. 2013. Copyright Wiley-VCH Verlag GmbH & Co. sequently attached to a Pt electrode by applying an external
KGaA. Reproduced with permission.) magnetic field. The detection step relied on change in ampero-
metric response of ferricyanide, but the system enabled high
11.2.3 Immunosensors reproducibility and regeneration of electrode surface (Yang
et al. 2014). Besides ferricyanide, hydroquinone freely dis-
11.2.3.1 Amperometric Devices solved in measuring solution was also tested as the redox
For biosensors with antibodies as biorecognition elements, probe (Li et al. 2011), while other works reported detection
amperometric methods were developed based on a simple with redox probes immobilized on the electrode surface (Han
selective “fouling” of a surface with respective antigen mol- et al. 2011, 2013; Mao et al. 2012).
ecules hindering the detection of a redox probe. In a typical Wu et al. took advantage of redox properties of Ag
arrangement, an electrode surface is covered with rGO in nanoparticles as the redox probe combined with additional
order to improve sensitivity to redox probe and to increase the Au nanolayer increasing the antibody loading with a detection
amount of the immobilized antibody selectively “catching up” limit as low as 3 pg mL −1 for determination of α-1-fetoprotein
desired antigens from a sample solution. Antigens detected by (Wu et al. 2013). A similar synergistic effect was reported
such biosensors are listed in Table 11.4. Basically, a level of also for bimetallic, for example, CoPt (Feng et al. 2013), or
GO reduction and its modification by metallic nanoparticles, combined Fe3O4 and Au nanoparticles grafted on an rGO sur-
the used redox probe and the manner of its application (i.e., face. Furthermore, magnetic properties of iron oxides allowed
in a solution or fixed at the electrode interface) are the main a flow-through mode of assays. Enzymatic catalysis was also
parameters affecting the biosensor’s performances. employed in this type of assay with HRP/antibody clusters
In most experiments, rGO was employed, though Han deposited on graphene/thionine/AuNPs electrode, where the
et al. reported the electrode modified with GO/thionine/ analyzed α-fetoprotein hindered the HRP-mediated H2O2
TABLE 11.4
Summary of Analytes Detected by Graphene-Based Amperometric Immunosensors
Modification of Electrochemical
Electrode Technique; Probe Detected Antigen LOD (mL−1) Linear Range Reference
rGO-NH3 DPV; [Fe(CN)6] 3+ β-LG 850 fg 1 pg–100 mg mL −1 Eissa et al. (2012)
rGO DPV; [Fe(CN)6]3+ CA 15-3 12 mU 0.1–20 U mL−1 Li et al. (2013)
GO/Thi/AuNP SWV; Thi CEA 50 ag 0.1 fg–1 μg mL−1 Han et al. (2013)
rGO/nanoporous Au CV; HQ Gonadotropin 34 pg 0.5–40 ng mL−1 Li et al. (2011)
rGO/MB/CHI CV; MB PSA 13 pg 50 pg–5 ng mL−1 Mao et al. (2012)
CrGO-COOH CV; [Fe(CN)6]3− Aflatoxin B1 120 pg 0.13–1 ng mL−1 Srivastava et al. (2013)
rGO SWV; [Fe(CN)6]3− Okadaic acid 19 ng up to 5 ng mL−1 Eissa and Zourob (2012)
G/AuNP/Naf-Cys DPV; [Fe(CN)6]3+ Hepatitis Ba 100 pg 0.8 pg–0.5 ng mL−1 Huang et al. (2012)
rGO-COOH DPV; [Fe(CN)6]3− Ovalbumin 830 fg 1 pg–0.5 μg mL−1 Eissa et al. (2013)
GO/Fe3O4NP/AuNP/HRP HRP catalysis α-fetoprotein 1 pg 10 pg–200 ng mL−1 Zhang et al. (2011)
rGO-CHI/CeO2-AuNP DPV; [Fe(CN)6]3− T. gondiib 0.44 U 0.75–24 U mL−1 Jiang et al. (2012)
CrGO-Thi/CoPtNP-CHI CV; Thi Zearalenone 13 pg 50 pg–5 ng mL−1 Feng et al. (2013)
Note: For each analyte an immunosensor with the best performance was selected.
a Hepatitis B virus surface antigen.
b Toxoplasma gondii-specific IgM.

reduction (Chen et al. 2011). While decrease of the detected is enhanced by HRP (Cai et al. 2011) or quantum dots (Yang
current upon a higher concentration of the analyte was et al. 2011) with rGO being usually an amplifier of a redox
observed in all cases, Shrivastava et al. (2013) were first to probe signal on the main biosensor interface. Predictably,
report an increase of the anodic peak of the redox probe (fer- integration of rGO with metallic (Su et al. 2011) or carbon
ricyanide) with growing aflatoxin B1 concentration. This was nanoparticles (Liu et al. 2012; Zhou et al. 2012) was reported
explained by a chemical composition of aflatoxin B1, which to increase either a redox probe reaction rate or a loading of
is electrochemically active, and in fact it undergoes a ferri- a primary antibody. The whole sandwich-based assembly
cyanide-mediated oxidation on the rGO-modified indium-tin can be simplified by an incubation of the antigen/primary
oxide (ITO) electrode. Simultaneous detection of two anti- antibody-loaded immunosensor surface only with secondary
gens has been reported by Kong et al. by employing a screen- antibodies conjugated with HRP (Liu et al. 2012) or ferro-
printed electrode with two separate parts, each one coated cene (Wang et al. 2013). Nevertheless, such biosensors did not
with rGO and a different mediator, thus, two different redox exhibit a high performance of detection even though graphene
peaks were visible on differential pulse voltammograms enhanced the surface redox probe rate reactions.
(DPV) taken. Consequently, two different antibodies were A secondary antibody labeled with ssDNA was also used as
immobilized on each part of the electrode, thus, respective tracer with a consequent hybridization with a complementary
antigens were captured and a decrease in the respective DPV strand and, finally, a detectable redox probe was intercalated
peak could be observed (Kong et al. 2013). into a formed dsDNA. The whole assay was performed on a
Like DNA biosensors, immunosensors can take advan- GO-modified electrode and allowed a detection of as low as
tage of a sandwich-based amplification. Antigens detected by 5.2 human ovarian cancer cells on average per mL (Xia et al.
sandwich-assay immunosensors are listed in Table 11.5. The 2012). In a study by Feng et al. (2013), nanoporous Pt–Co alloy
prepared tracers are typically metallic or silica nanoparticles particles were used as the redox tracers instead of HRP. The
integrated with a secondary antibody allowing a selective final detection of H2O2 reduction rate was performed at a sim-
attachment of tracers to the electrode surface, and a signal ilar potential as in the case of HRP-based tracers, but lower
TABLE 11.5
Summary of Analytes Detected by Graphene-Based Amperometric Immunosensors Using Sandwich Assay
Electrode Modification Detected LOD
Tracer Probe Composition Electrochemical Detection Antigen (mL−1) Linear Range Reference
rGO-PVP/thionine CV; mediated HRP catalysis BrCA1 4.9 pg 10 pg–15 ng mL −1 Cai et al. (2011)
MS-NH3 /HRP/Ab2/IL
ErGO/CHI LSV; deposited Ag CEA 120 fg 0.5 pg–0.5 fg mL−1 Lin et al. (2012)
PSCA beads/AuNP/Ab2
rGO-Thi/Au-AgNP DPV; mediated HRP catalysis AFP 500 fg 1 pg–200 ng mL−1 Su et al. (2011)
TiO2-AuNP/HRP/Ab2
(PDDA/CrGO/CNT)n CV; mediated HRP catalysis IgGA 200 pg 1–500 ng mL−1 Liu et al. (2012)
Ab2-HRP
GCE/pNiPAM SWASV; dissolved QDs Mesothelin 1 pg 1 pg–1 μg mL−1 Shiddiky et al. (2012)
GO-CdSe QDs/scVf Ab2
ITO-COOH/AB1-biotin SWASV; dissolved QDs EpCAM antigen 0.1/1 pga 6 orders of magnitude Shiddiky et al. (2012)
GO-CdSe QDs/Ab2/SA
Au electrode/AuNP Voltammetry; AgNP E. coli 10 CFU 50–106 CFU mL−1 Jiang et al. (2013)
GO-AgNP-Ab2
TrGO-AuNP DPV; mediated HRP catalysis CAE 350 fg 1 pg–10 ng mL−1 Sun et al. (2013)
AgNP-Thi/HRP-Ab2
CHI/G/Thi-Ab1 DPV; mediated HRP catalysis Phos p53 100 pg 0.2–10 ng mL−1 Xie et al. (2011)
Ab2/HRP(streptavidin-biotin)
PDDA-ErGO/AuNP DPV; mediated HRP catalysis human IgG 50 pg 0.1–200 ng mL−1 Liu et al. (2011)
PDDA-GO/AuNP/Ab2-HRP
N-doped G/AuNP DPV; mediated HRP catalysis MM 2 0.1 pg 0.5 pg–50 ng mL−1 Yang et al. (2013)
GO/PDA/HRP-Ab2
rGNR-Ab1 (Ab1′) SWV; AgNP and Cd2+ IL 6b 100 fg 1 pg−1 μg mL−1 Shi et al. (2014)
PS@PDA/AgNP/Ab2; MMP 9b 5 fg 10 fg−1 μg mL−1
PS@PDA/Cd2+/Ab2′
Note: For each analyte an immunosensor with best performance was selected.
a Value obtained in buffer/human serum.
b Simultaneous detection.
limit detection for zearalenone of 2.1 pg mL −1 was observed 11.2.3.2 FET Devices
(Feng et al. 2013), compared to an enzyme-employing biosen- Bioconjugation of antigens onto antibody-modified channel of
sor with a limit of detection of 4.9 pg mL −1 for a BRCa1 anti- an FET device was proved to induce a decrease of a p-type
gen (Cai et al. 2011). A detection limit as low as 120 fg mL −1 conductance (positive gating). This change is measurable and
was reported for a carcinoembryonic antigen biosensor with was employed for FET-based immunosensors fabrication with
gold nanoparticles coating polymeric microbeads applied as a pristine or metallic nanoparticle-modified rGO channel, pro-
a secondary antibody-modified tracer. These nanoparticles viding a detection limit in the sub-pM (ng mL−1) level (Mao
catalyzed deposition of silver, which was finally detected by et al. 2011). The fabrication of an FET channel from rGO sheets
a sensitive anodic stripping technique (Lin et al. 2012). This wrapping silica nanoparticles was tested in order to obtain
is, together with presence of rGO on the electrode surface of a higher surface for covalent antibody loading and detection
the biosensor, behind a low detection limit. Noteworthy is limits of 1 and 100 pM for human epidermal growth factor
a study of Jin et al. who used a solid CVD-grown electrode receptor 2, and epidermal growth factor receptor, respectively,
instead of common GO/rGO-modified ones in a sandwich- were obtained (Myung et al. 2011). By recording a potential
type immunosensor. Even though an obtained detection limit shift of a minimum conductivity point (i.e., the gate potential
of 5 ng mL −1 of CEA is quite high, it is most likely a result which sets a mobility of either holes or electrons in transistor
of an improperly chosen system for detection and caption of channel to minimum) instead of drain–source current, Kim
tracers (Jin et al. 2014). et al. (2013) were able to push a detection limit for prostate
In some studies, rGO (Zhong et al. 2010; Yang et al. 2011) cancer biomarker down to an fM range. In order to achieve
and GO sheets were tested as a template for a preparation this, only a simple FET with an rGO channel grafted with rel-
of nanosized tracers in combination with HRP (Chen et al. evant antibody was employed (Kim et al. 2013).
2013), metal nanoparticles (Jiang et al. 2013) or quantum dots FETs with CVD-grown graphene channels were also fab-
(Shiddiky et al. 2012). The latter assembly in the form of GO/ ricated, and their conjugation with surface area-enhancing
QD/Ab2 together with an antigen bound to a covalently fixed polymeric (Kwon et al. 2013), or metallic nanoparticles with
primary antibody on an ITO electrode showed an amazing a vertical arrangement of graphene sheets (Mao et al. 2013)
detection limit down to 100 fg mL −1 for EpCAM antigen with was tested with a limit of detection in a pM range. A sub-
a dynamic concentration range covering 6 orders of magni- stantial improvement was achieved with a channel prepared
tude due to the above mentioned sensitivity of anodic strip- by suspending a few layers of graphene sheets between drain
ping voltammetry (Shiddiky et al. 2012). Another approach and source electrodes (see Figure 11.9). Detection of a pros-
relied on GO-based tracers, which increased conductivity of a tate-specific antigen as a model analyte down to 400 ag mL −1
thiol-blocked electrode surface after the Ag was deposited on (11 aM) was reported, what was substantially lower compared
GO from a supplied solution (Qu et al. 2011). to the device using the same graphene channel deposited on
Besides the signal amplification, a sandwich-type assay the substrate (Zhang et al. 2012). An extremely low detection
brought an additional possibility for a simultaneous detec- limit was assigned to a strong signal noise suppressing abil-
tion. The rGO electrode with two primary antibodies was ity of a suspended configuration. A sandwich assay in FET
incubated with a sample followed by addition of two tracers immunosensor was reported by Zhang et al. (Zhang and Cui
containing the two secondary antibodies, HRP and media- 2011). They used HRP/secondary antibody labels attached
tors. Since each secondary antibody is combined with a par- to the antigen caught by surface-immobilized primary anti-
ticular mediator with distinguishable redox potential, two body. HRP-induced oxidation of supplied l-ascorbic acid led
signals can be detected and evaluated in one measurement to a local change of pH, consequently changing a channel
(Tang et al. 2011). Zhu et al. reported the simultaneous assay conductance. This approach offered a 100-fold decrease of a
of three analytes on a single electrode surface modified with detection limit (4 fg mL −1 i.e., 11 fM) for a prostate-specific
three different antibodies (Zhu et al. 2013). Similarly, poly- antigen compared to a performed label-free assay (Zhang and
meric tracers containing either Ag nanoparticles or Cd2+ ions Cui 2011).
detected at different potentials, were used for simultaneous Okamoto et al. reported an FET device with a physically
detection of interleukin-6 and matrix metallopeptidase-9 exfoliated graphene sheet channel, where intact antibod-
(Shi et al. 2014). ies were replaced by their binding-responsible fragments, so
A dual tracer assessment was also adapted for an enhance- the antigen recognition occurred much closer to the surface,
ment of sensitivity. Tracers with two different secondary anti- within an electrical double-layer thickness. A heat shock
bodies and quantum dots detectable at different potentials protein was detected from a 100 pM concentration level
were incubated with Hep3B-line cells being anchored to rGO/ (Okamoto et al. 2012), though the authors did not compare
chitosan electrode by a primary antibody. Since a Hep3B cell this value with the full antibody-based device.
possesses antigens for both secondary antibodies, by a sub-
sequent detection of both quantum dots, as few as 10 cells 11.2.3.3 Impedimetric Devices
per mL were detected (Wu et al. 2013). This holds a prom- EIS is another technique capable of recording the antigen–
ise for detection of other kinds of circulating tumor cells antibody coupling. Impedimetric immunosensor consisting
causing metastases and occurring in the blood at a very low of an rGO/conducting polymer/Au nanoparticles interface
concentration. onto which the antibody is anchored selectively binding an
accessible for a redox probe, and thus, with a higher concen-

tration of the target, a higher current response was detected
(Guo et al. 2013) (“signal on” arrangement). Another “signal
on” approach relied on a decoupling of passivating GO labels
from aptamer strands after biorecognition of a target protein
took place (Yuan et al. 2012). This triggers the effective elec-
trochemical transformation on the electrode surface, where
GOx, after glucose addition, generated H2O2 which was
consequently oxidized to water by a peroxidise-mimicking
hemin/azo-dye complex. The system of detection is shown in
Figure 11.10.
“Signal off” arrangements were reported, as well. For
example, redox probe-coated rGO sheets and the aptamer on
PSA capture antibody Electrodes SiO2 the surface of an electrode created a system, where a surface
electrochemistry of the redox probe used for generation of an
Self-assembled graphene PSA BSA Si analytical signal was substantially suppressed by a coupled
analyte (Sun et al. 2011; Yuan et al. 2011). Another version
FIGURE 11.9 Scheme of a FET immunosensor with suspended of this system is an electrochemical probe-modified aptamer
graphene channel modified with antibodies. (Reprinted from where current response decreased once it was complexed with
Biosens. Bioelectron, 31, Zhang, B. et al., Ultra-sensitive suspended analyte. Thus, the aptamer strands were folded in such a way
graphene nanocomposite cancer sensors with strong suppression of
that their labeled ends moved away from an electrode surface
electrical noise, 105–109, Copyright 2012, with permission from
Elsevier.)
(Wen et al. 2014). Similarly, the GO/azo-dye complex was
completely liberated from the surface upon a bioconjugation
with an analyte which also led to a “signal off” switch (Chen
analyte—aflatoxin B1 was reported (Linting et al. 2012). An et al. 2013). Du et al. used a redox probe complexed with a
increasing amount of the analyte is finally detected as an mixed aptameric nucleotide containing one part of selectively
increased RCT, which offers a lower detection limit (1 fg mL −1) binding thrombin and the other part selectively binding lyso-
(Linting et al. 2012) compared to the approach based on a zyme. If one of the analytes was present and conjugated to
CV reading (120 pg mL −1) (Srivastava et al. 2013). To scale the relevant part of the aptamer strand, a redox probe was
up biosensors, a free-standing “graphene paper” electrode partly released from a complex, resulting in a detected signal
was prepared by a simple self-assembly of rGO sheets. Anti- decrease. The same level of decrease in signal was observed
Escherichia coli O157:H7 antibodies were adsorbed on a when the other analyte was present, while a simultaneous
large surface electrode, and such a prepared flexible biosensor binding of both analytes caused the liberation of the whole
exhibited correlation between E. coli O157:H7 cell concen- aptamer–redox probe complex from the electrode surface
tration and an increase in a charge transfer resistance (Wang with the detection of a significantly lower signal. The device
et al. 2013). worked as a base for a Boolean NAND gate with a presence of
Sub-pg mL −1 limit of detection of carcinoembryonic anti- none, one, or two analytes being the output values (Du et al.
gen was obtained with an EIS-based biosensor employing a 2012). Besides this, GO was also used directly as the redox
sandwich-based signal amplification relying on a secondary probe, employing its ability to be easily electrochemically
antibody/HRP integrated with GO nanosheets as the molecu- reduced, however, this reaction occurs at a relatively unfa-
lar tracers. HRP-catalyzed precipitation of a supplied chloro- vorably low potential (Loo et al. 2013a). Also, DET of GOx
naphthtol, which was deposited on the surface and caused an loaded together with platelet-derived growth factor-binding
increase of RCT (Hou et al. 2013). aptamer on rGO-Au nanoclusters surface was employed as
a probe signal with decreasing enzyme’s FAD/FADH2 signal
with “fouling” of the electrode with detected growth factor
11.2.4 Aptamer-Based Sensors
(Deng et al. 2013).
Protein-recognizing nucleotides—aptamers—are another Either in “signal on” or “signal off” devices, rGO was
choice for the fabrication of selective biosensors, and prin- typically exploited as an enhancer of a surface area and of
cipally, such devices are in between immuno- and DNA- redox kinetics, but signal-amplifying tracer probes based on
biosensors. Therefore, the immobilization, signal reading, and rGO were also described adopting a HRP sandwich assay
amplification techniques are very similar to that described with either H2O2 directly supplied into the system (Wang
earlier, although aptamers as synthetically prepared mol- et al. 2011) or generated by coimmobilized GOx (Bai et al.
ecules provided a higher variability than natural antibodies. 2012). The latter system was suggested to improve H2O2 dis-
tribution due to its generation in proximity of the HRP and
11.2.4.1 Amperometric Devices the reported bienzymatic thrombin aptasensor-exhibited pM
Unlike antibodies, target-bound aptamers can be “decou- detection limit (Bai et al. 2012). A detection limit down to
pled” from the electrode surface making the electrode more 100 fM for thrombin was reached upon replacement of HPR
PTCA/hemin Au TBA GO-CTBA GOx

Gold electrode
No target
Potential
With target Thrombin
Potential
(ox)
Sensing interface
Gluconolactone H2O PTCA/hemin
GOx PTCA/hemin
Glucose H2O2 PTCA/hemin

(red)
FIGURE 11.10 Preparation and detection scheme of a “signal on” aptasensor using modified GO nanoparticles as passivating layer. (Yuan,
Y.: 3,4,9,10-Perylenetetracarboxylic acid/hemin nanocomposites act as redox probes and electrocatalysts for constructing a pseudobien-
zyme-channeling amplified electrochemical aptasensor. Chemistry – A European Journal. 2012. 18 (44):14186–14191. Copyright Wiley-
VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
and GOx with hemin/G-quadruplex, a DNAzyme complex force of a two-electrode system which enabled detection of as
capable of simultaneous NADH oxidation and H2O2 reduction low as 1 CFU mL −1 of cells. Interestingly, swapping of GO
(see Figure 11.11) (Yuan et al. 2013; Yi et al. 2014). Besides for rGO did not cause any significant change in operational
taking further an idea of in situ enzymatic H2O2 generation, characteristics, although covalent attachment of aptamer on
such small “biocatalytic” molecules can be loaded in large GO was more reproducible, while being slightly less effective
amounts, which further enhanced the device’s performance. (Hernández et al. 2014).
Nevertheless, rGO modified with azo-dye functioning as
a signal enhancer and a mediator for HRP-catalyzed H2O2 11.2.4.2 FET and Impedimetric Devices
reduction turned out to be the most effective arrangement Although not that extensively as in the case of DNA or anti-
with a detection limit down to 650 aM for thrombin (Peng bodies, aptamers were tested for a construction of impedimet-
et al. 2012), although a highly nanostructured probe substrate ric or FET-based biosensors. It was found out that, similarly to
formed by hollow Pt-Co nanochains wrapped with Pt nanopar- immuno- and DNA-based FET biosensors, a partial positive
ticles and thionine film concomitantly contribute to the excel- charge of the analyzed immunoglobulin E induced a detect-
lent biosensor performances as well (Peng et al. 2012). able decrease of p-type conductance, while a partially nega-
The sandwich configuration was employed also for tively charged aptamer caused the opposite effect (a negative
simultaneous detection on one interface with two different gating) (Ohno et al. 2010, 2011). An FET biosensor was
aptamer–analyte conjugates, each one binding specifically the reported using a GO channel coated with one subunit of a
respective tracer bearing a HRP/mediator. Diverse mediators thrombin-specific aptamer. Upon addition of thrombin and
secured detectable responses at distinguishable potentials, a second aptamer subunit, all three components fold into a
thus, thrombin and platelet-derived growth factor could be complex which was released from the channel surface affect-
assayed in one measurement (Bai et al. 2012). ing conductivity of the device. Detection limit of 25 nM was
Besides amperometry, a potentiometric measurement was observed (Sharon et al. 2013), what is quite high compared
also employed in aptasensors. Hernández et al. (2014) com- to a detection limit of 100 fM for antivascular endothelial
pared GO and rGO for modification of electrodes and immo- growth factor reported by Kwon et al. (2012). In their work,
bilization of S. aureus-detecting aptamers. Attachment of an N-doped, CVD-grown graphene channel was coated with
detected cells caused a measurable change in the electromotive an RNA aptamer, and the whole device was designed as a
(a) modified a pencil graphite electrode with GO/chitosan com-

posite and lysozyme-binding aptamer. Interestingly, upon an
PDDA attachment of lysozyme, decrease of RCT was observed which
Hydrazine ADH AuNPs allowed for biosensing of the protein with LOD of 28.5 nM
(Erdem et al. 2014).
GO rGO
11.2.5 Lectin–Glycan-Based Biosensors
Hemin/ BSA Lectins are proteins selectively binding glycan structures,
G-quadruplex
which are, among others, fundamental for in vivo recognition
at the molecular and cellular levels, and for an early diagnosis
(b) of some diseases. Highly selective lectin–glycan interaction
has already been employed in biosensor constructions, though
dep-Au
there are only very few reports on integration of graphene-
GCE TBA I
based materials in such devices. Alava et al. (2013) reported
BSA on a possible partial denaturation of lectin concanavalin A
upon adsorption directly on a graphene surface, and suggested
that the protein structure can be preserved when anchoring
onto a monolayer-modified graphene surface, which can be
TBA II Thrombin employed for future lectin-based biosensors development
Hemin
H2O2 NADH Alcohol
(Alava et al. 2013). The only graphene-based sensor was
(ox)
e– Hemin
H2O NAD+ Acetaldehyde
(red) O O–
O O–
PTCA/CCG
FIGURE 11.11 Scheme of pseudo triple-enzyme sandwich apta-
sensor. (a) Composition of rGO-based tracers. (b) Composition of
electrode with anchored primary aptamer. (Reprinted from Biosens.
Bioelectron, 54, Yi, H. et al., A pseudo triple-enzyme cascade
EDC/NHS
amplified aptasensor for thrombin detection based on hemin/G-qua- AS1411
druplex as signal label, 415–420, Copyright 2014, with permission
from Elsevier.)
flexible biosensor. A high-performance FET biosensor was

fabricated by immobilization of an aptamer coupling the
anthrax toxin, which caused a detectable shift of a minimum NH NH
NH NH
drain–source current voltage. Furthermore, a 10-fold lower O O O O
H 2O
detection limit of 1.2 aM was obtained upon application of a
sandwich arrangement with Au nanoparticles anchored via a Wash
secondary aptamer (Kim et al. 2013).
Several types of graphene-based electrodes, that is, modi-
fied with GO and rGOs, were compared for a fabrication of
cDNA K+
impedimetric aptasensors for thrombin. An increase in RCT
was measured after formation of an aptamer–thrombin conju-
gate, and it was found out that the GO-based biosensor offered
a dynamic range from 10 to 50 nM (Loo et al. 2012). The
same approach was used for detection of cancer cells, employ-
ing an rGO electrode grafted with the aptamer selectively
binding nucleolin overexpressed on the cancer cell surface
NH NH NH NH
(Feng et al. 2011). The device offered a calculated detection
O O Cell O O
limit of 794 HeLa cells per ml, and was designed as a reusable
biosensor by an application of complementary DNA strands, Capture
which released the anchored cells and prepared the surface for
another measurement (Figure 11.12). LOD of 3 CFU mL −1 of
FIGURE 11.12 Reusable impedimetric aptasensor for biosensing
Salmonella cells was obtained with impedimetric aptasensor of HeLa cells. (Reprinted from Biomaterials, 32, Feng, L. et al., A
prepared by modification of a glassy carbon electrode with graphene functionalized electrochemical aptasensor for selective
GO and Au nanoparticles (Ma et al. 2014). A miniaturized label-free detection of cancer cells, 2930–2937, Copyright 2011,
impedimetric aptasensor was introduced by Erdem et al., who with permission from Elsevier.)
reported by Chen et al. (Chen et al. 2011), when FET with a amperometric signal. On the basis of a similar principle,
CVD-grown graphene channel was modified by glycan struc- molecularly imprinted biosensor detecting as low as 230 pM
tures. The device selectively caught lectins from the sample of tetrabrombisphenol A was reported, also with an rGO
solution resulting in a detectable change in the channel con- substrate functioning as an electrochemical signal enhancer
ductance. Nevertheless, a higher sensitivity and specificity (Chen et al. 2013).
were seen when CNTs were employed instead of graphene Human serum albumin attached to a GO-modified elec-
sheets because of much easier change of conductance on a trode exhibited an effective binding for l- and d-tryptophan
1D CNT surface compared to a 2D graphene surface (Chen with possible amperometric detection of enantiomers at dif-
et al. 2011). These results together with observations made on ferent potentials. Thus, an amperometric biosensor enabling
devices using nanomaterials other than graphene will most the discriminative detection of both enantiomers in one mea-
likely be taken as a basis for development of inexpensive and surement could be developed (Zor et al. 2013).
high-performance graphene–lectin–glycan biosensors in the A glucose biosensor based on an enzymatically triggered
future. cascade reaction was described by Zhu et al. (Zhu et al. 2013).
A fixed amount of GOx, Fe2+, self-folded palindromatic DNA,
and different amounts of glucose triggering an enzymatic gen-
11.2.6 Other Graphene-Based Biosensors
eration of H2O2 were incubated. In the mixture, a Fe2+ cata-
Several other enzymatic, DNA or RNA, antibody, aptamer- lyzed H2O2 hydrolysis (a Fenton reaction) and the formed OHγ
based or lectin–glycan interactions have been used for con- radical cleaved the folded DNA. The whole mixture was con-
struction of graphene-based biosensors. These devices sequently incubated with an rGO-modified electrode, where
include, for example, a flexible FET-type bioelectronic nose, cleaved fragments adsorbed preferentially and their amount
where human olfactory receptors were immobilized on a sur- was quantified amperometrically and correlated to the ini-
face of a plasma-treated bilayer graphene channel and selec- tial glucose concentration with a detection limit of 5 μM for
tively detected amyl butyrate down to aM concentration (Park glucose (Zhu et al. 2013).
et al. 2012).
A specific hexapeptide structure was exploited as a cap-
11.3 BIOFUEL CELLS
ture probe for cyclin A (a regulatory protein overexpressed
in cancer cells) determination with a limit of detection of BFCs share with biosensors a need for an effective loading
320 fM in a buffer solution, and approximately around 1 pM of high amounts of biomolecules/bioelectrocatalysts on an
in cancer cells extracts. The device was designed as a simple electrode surface (see Figure 11.13). From this point of view,
rGO-based impedimetric biosensor with an anchored cap- nanomaterials, including graphene-based nanomaterials, play
ture probe using a hydrophobic interaction (Feng et al. 2012). basically the same role in their preparation as in the construc-
An interesting, impedimetric, enzyme-based biosensor tion of amperometric biosensors. This implies that there is a
was developed by Song et al. The authors used an electrode similarity in the way nanomaterials are processed and inte-
modified with rGO and lectin concanavalin A possessing a grated to become a part of a device. In this subchapter, not
certain surface electric charge. This charge is changed once only are the BFCs covered using graphene integrated with a
analytes, that is, glucose or urea are enzymatically processed biocatalyst, but also the devices with bioanodes and abiotic,
by added GOx and urease, resulting in a detected RCT value graphene-modified cathodes are included.
change. This could be correlated with enzymatic activity trig-
gered by added analytes (Song et al. 2014).
11.3.1 Microbial BFCs
Another option for biosensing is to capture analytes into
“molecularly imprinted” interfaces. These were fabricated by The majority of studies were focused on a preparation of a
a polymerization of the monomer with a strong affinity for graphene-based electrode interface possessing a better elec-
the detected molecules, which are present during the forma- tron exchange between the electrode and bacteria (see Table
tion of a polymer film as template molecules. After removal 11.6). The first work was published by Zheng et al., who
of the template, the patterned interface was formed with prepared a Teflon–graphene nanocomposite pressed into a
cavities capable of selective and reversible binding of tar- form of a compact disk electrode. This was placed into an
get molecules. By creating a molecularly imprinted film on anaerobic anodic chamber filled with a buffer containing
rGO/Pt nanoparticles surface, electrochemical biosensor for E. coli, glucose and a mediator and connected to an abiotic
17β-estradiol was developed with a limit of detection of 2 nM cathode with [Fe(CN)6]3− as a final electron acceptor. The
(Wen et al. 2012). The molecularly imprinted film served just device produced a maximum power output of 267 μW cm−2,
for selective capture of the analyte, which was consequently and using electron microscopy, it was found that E. coli cells
electrochemically transformed with the help of rGO/nanopar- adhere very densely on the graphene-based surface (Zhang,
ticles catalyst generating a detectable signal. Similarly, rGO/ Mo et al. 2011). Using a similar principle, vacuum-stripped
metallic nanoparticles signal-amplifying tracers were used graphene and chitosan were fabricated into a conductive,
by Wang et al. (Wang et al. 2013). They fabricated an albu- ordered, highly porous electrode interface of a microbial bio-
min-imprinted film for detection of BSA, and after BSA was fuel cell (MBFC) anode. A system was designed as mediator-
bound to the polymers a tracer was injected to give a readable less, yielding a maximum power of the anode of 153 μW cm−2
R 3.3 W m−3 was reported for a device using a crumpled gra-

e– e–
phene-modified biocathode and unmodified bioanode. Both
values were higher compared to a system employing a con-
ventional “flat” sheet graphene (Xiao et al. 2012). This was
SA SK
confirmed in another study reporting crumpled graphene-
modified anode in MBFC with power output of 240 μW cm−2
PA PK (Luo et al. 2011).
Integration of graphene nanoparticles with a conduc-
tive polymer (PANI) was also tested in combination with
Shewanella oneidensis cells, which have naturally developed
Anode Cathode surface cytochrome c system for transferring electrons to the
solid surfaces including electrodes (Yong et al. 2012). This
PA
heterogeneous electron transfer together with electron trans-
fer driven by self-secreted mediators was reported to be sub-
SA
stantially improved by the presence of graphene nanoribbons
PA
(Huang et al. 2011).
Synergistic effect of a conductive PANI nanofibers and
SA
ErGO deposited on a high surface area carbon cloth elec-
trode was revealed, with anaerobic sludge inoculum was
used in the anodic chamber. The device with such an anode
and an abiotic cathode exhibited a maximum power output of
Biocatalysts Nanomaterials 139 μW cm−2 (Hou et al. 2013). Guo et al. reported the employ-
ment of a different kind of polymer matrix for layer-by-layer
modification of carbon paper by negatively charged carbox-
FIGURE 11.13 A schematic depiction of a one compartment, ylated rGO, and positively charged poly(ethyleneimine).
membraneless enzymatic biofuel cell with biocatalysts integrated Such modification improved almost twice the power output
with nanomaterials on an anode and a cathode surface. While on an of BFC driven by a waste-water microbial consortium com-
anode, a biofuel/substrate (SA) is oxidized to the product (PA), and on pared with a device with a bare carbon paper anode (Guo
a cathode surface, oxygen (SK) is reduced to water (PK). et al. 2014).
An innovative approach was tested by Zhuang et al., who
(He et al. 2012). Xiao et al. (2012) explored an employment of exploited a surface-reducing capability of cathodic bacteria
crumpled graphene sheets dispersed in a polymer matrix for a (incubated on a carbon electrode from aerobic waste-water
modification of either microbial anode or microbial cathode. plant sludge). The authors added GO into a cathode compart-
A maximum power density of a two-chamber BFC using a ment where it underwent a bacterial reduction with simultane-
crumpled graphene-modified bioanode was 3.6 W m−3, while ous assembling into the electrode bacteria–rGO interface. In
TABLE 11.6
Characteristics of Graphene-Based Microbial Biofuel Cells Using Different Bacteria
Cathode Configuration; pMAX (μW cm−2)/ OCP
Anode Configuration; Biofuel Final Electron Acceptor Voltage of pMAX (mV) (mV) Reference
E. coli + HNQ
SSM/CrGO + PTFE; glucose CP; ferricyanide 267/530 NA Zhang et al. (2011)
CP/CrGO/PEDOT; glucose CP; ferricyanide 87/ ≈430 ≈700 Wang et al. (2013)
P. aeruginosa
VSG/CHI; glucose CC; ferricyanide 153/550 NA He et al. (2012)
S. oneidensis MR-1
Graphite felt/GO/PPy; lactate CF; ferricyanide 133/ ≈420 NA Lv et al. (2013)
CP/GNR/PANI; lactate CP; ferricyanide 86/ ≈175 NA Zhao et al. (2013)
Microbial Consortium
SSM/PU sponge-G; glucose CC-PtNP; dissolved O2 157/NA NA Xie et al. (2012)
CC/bacterially reduced GO; acetate CC-PtNP; air cathode 191/300 ≈600 Yuan et al. (2012)
CC/crumpled rGO; acetate Carbon brush; ferricyanide 240/520 ≈800 Luo et al. (2011)
CC/ErGO/PANI; acetate CF; ferricyanide 139/460 770 Hou et al. (2013)
connection with an anaerobic microbial anode, the obtained and mediator or CNTs. This interface was effective in reoxida-
two-compartment, mediatorless BFC provided twice the tion of reduced nicotinamid adenine dinucleotide, which was
maximum power density compared to a system without rGO proved by immobilization of glucose dehydrogenase, and an
(Zhuang et al. 2012). employment of such a bioanode into a BFC (Wang et al. 2011).
Ci et al. (2012) have compared N-doped rGO and Pt Liu et al. employed GOx and bilirubin oxidase coentrapped
nanoparticles-modified electrodes for their ability to reduce with respective mediators in rGO/silica matrix for fabrication
O2 in an MBFC, and revealed that almost 10% higher maxi- of a glucose-dioxygen BFC with a maximum power density of
mum power density was obtained when a graphene-based 24.3 μW cm−2 (Liu et al. 2010). A substantially higher power
cathode was used. Even 30% increase of power output of output of 78.3 μW cm−2 was achieved when a silica matrix was
MBFC was observed when porous nitrogen-doped carbon replaced by a PANI structure fabricated on the rGO/enzyme/
sheets on graphene was compared with a common Pt-based mediator electrodes by a simple electropolymerization (Chang
catalyst in an air-breathing cathode (Wen et al. 2014). This et al. 2011). Devadas et al. found out that GOx can be effec-
superiority of N-doped rGO over Pt nanoparticles was found tively adsorbed on the surface consisting of ErGO integrated
to be more obvious when the N-doping of rGO was accom- with CNTs. This bioanode was combined with an abiotic
panied with Fe doping. A twofold higher maximum power cathode forming a BFC with a power density of 46 μW cm−2
output was found for the MBFC with N-Fe-doped rGO cath- and operating with a mediator in an anodic chamber (Devadas
ode, although, interestingly, this superiority over a Pt-based et al. 2012). On the other side, power density of only tens of
electrode was not observed when each electrode was studied nW cm−2 was reported for a GOx-LAC BFC, where an anodic
separately (Li et al. 2012). enzyme/GO complex was electrochemically reduced on the
Similarly, a synergistic effect of rGO and MnO2 nanopar- electrode surface, and bilirubin oxidase was integrated with
ticles was observed; a MnO2/rGO cathode exhibited 18% CNTs (Palanisamy et al. 2012).
higher power output compared to a Pt-based device (Wen Graphite oxide, as a material sharing some features with
et al. 2012). Even though this value was obtained with 10 GO, was found to be a very promising template; upon its
times higher loading of MnO2/rGO catalyst on the elec- intercalation with Co(OH)2 and integration into a chitosan
trode surface compared to Pt nanoparticles, the result is still matrix, an electrode interface was obtained onto which GOx
promising, since expensive Pt nanoparticles can be quite or Lac was covalently attached. The bioanode and the bio-
effectively replaced by graphene-based nanocatalysts for cathode assembled a glucose-dioxygen BFC with a power
construction of (bio)cathodes of MBFCs. Graphene/MnO2 output of 517 μW cm−2 (Uk Lee et al. 2013). Prasad et al.
nanoparticles composite was also employed for fabrication employed CVD-deposited graphite foam as a free-standing,
of an abiotic air-breathing cathode. When connected to a highly porous electrode of glucose–oxygen BFC providing
microbial anode, MBFC with power output of 336 μW cm−2 an amazing power output of 2.7 mW cm−2 and an equally
was prepared. This was comparable to a device equipped amazing open circuit potential of 1.2 V (Prasad et al. 2014).
with a Pt/C catalyst-based cathode (Gnana Kumar et al. Nevertheless, these values were obtained only because of
2014). single walled CNTs deposited on the graphite foam surface
since GOx and Lac did not exhibit electrocatalytic activity
upon immobilization on the bare graphene foam surface. But
11.3.2 Enzymatic Biofuel Cells
even if the reported BFC was constructed as two compart-
The number of studies on using graphenic materials in the ments, with ABTS mediator dissolved in a cathodic compart-
construction of enzymatic biofuel cells (EBFC) is substan- ment, the study is valuable in terms of employment of 3D
tially lower than for enzymatic biosensors. EBFCs reported graphenic structures in EBFC.
so far are listed in Table 11.7. Besides the above described BFCs, some enzymatic elec-
A high surface area and a good conductivity of rGO sheets trochemical biosensors mentioned in a relevant subchapter
were exploited in a layer-by-layer assembling of rGO sheets exhibited enough sensitivity that they can be applied as BFC
TABLE 11.7
Characteristics of Graphene-Based Enzymatic Biofuel Cells
pMAX (μW cm−2)/Voltage
Anode Configuration Cathode Configuration of pMAX (mV) OCP (mV) Reference
Au/TMOS gel+CrGO+FM+GOx Au/TMOS gel+CrGO+ABTS+BOD 24.3/380 580 Liu et al. (2010)
GCE/GNP/GOx/Nafion GCE/GNP/Lac+BSA/Nafion 58.0/220 ≈550 Zheng et al. (2010)
Au/CrGO/FM+GOx/PPy Au/CrGO/ABTS+Lac/PPy 78.3/500 790 Chang et al. (2011)
GCE/ErGO-MWCNT/GOx/Nafion GCE/CrGO+PtNP+Nafion 46/≈80 400 Devadas et al. (2012)
GCE/ErGO+GOx GCE/MWCNT-ZnO/Lac 0.054/50 60 Palanisamy et al. (2012)
Au/GO+Co(OH)2+CHI/GOx Au/GO+Co(OH)2+CHI/Lac 517/460 600 Uk Lee et al. (2013)
3DG/SWCNT/GOx 3DG/SWCNT/Lac 2270/≈500 1200 Prasad et al. (2014)
anodes, thus, the graphene application in BFCs fabrication acknowledged. The research leading to these results has
can be substantially broadened. received funding from the European Research Council
under the European Union’s Seventh Framework Programme
(FP/2007-2013)/ERC Grant Agreement no. 311532, and this
11.4 SUMMARY, OUTLOOK, AND PERSPECTIVE
work has received funding from the European Union’s Seventh
For the construction of electrochemical biosensors, diverse Framework Programme for research, technological develop-
graphene-based materials have been extensively tested, and ment and demonstration under grant agreement no. 317420.
their functional integration with different kinds of biorecog- This work was funded by the National Priorities Research
nition molecules has been described and studied. In par- Program (Qatar National Research Fund), NPRP 6-381-1-078.
ticular, the following three main issues were preferentially
addressed: (1) a high surface area resulting from the geometry REFERENCES
of graphene-based nanoparticles, (2) an intrinsic electrocata- Akhavan, O., E. Ghaderi, and R. Rahighi. 2012. Toward single-DNA
lytic activity of nanoparticles enhancing surface reactions electrochemical biosensing by graphene nanowalls. ACS Nano
of redox probes, and (3) surface properties and adjustability 6 (4):2904–2916. doi: 10.1021/nn300261t.
of GO/rGO allowing an effective immobilization of desired Alava, T., J. A. Mann, C. Theodore, J. J. Benitez, W. R. Dichtel,
molecules. All these tools allowed a construction of electro- J. M. Parpia, and H. G. Craighead. 2013. Control of the gra-
chemical biosensors with performances matching or surpass- phene-protein interface is required to preserve adsorbed pro-
ing devices based on some more expensive nanomaterials. tein function. Anal. Chem. 85 (5):2754–2759. doi: 10.1021/
ac303268z.
Conclusively, different conceptions of graphene-based bio- Aravind, S. S. J., T. T. Baby, T. Arockiadoss, R. B. Rakhi, and S.
sensing were proved to be efficient enough to be applied even Ramaprabhu. 2011. A cholesterol biosensor based on gold
for commercially fabricated devices, which should be the nanoparticles decorated functionalized graphene nanoplate-
forthcoming step in graphene-based biosensor development. lets. Thin Solid Films 519 (16):5667–5672. doi: http://dx.doi.
Techniques developed for biosensors were consequently org/10.1016/j.tsf.2011.03.032.
applied in a BFC fabrication, although not to such an extent. Baby, T. T., S. S. J. Aravind, T. Arockiadoss, R. B. Rakhi, and S.
A combination of graphene and CNT was used for develop- Ramaprabhu. 2010. Metal decorated graphene nanosheets
as immobilization matrix for amperometric glucose biosen-
ment of enzymatic, glucose-oxygen BFC with power output sor. Sens. Actuat. B Chem. 145 (1):71–77. doi: http://dx.doi.
matching the most efficient EBFCs employing other nanoma- org/10.1016/j.snb.2009.11.022.
terials. Nevertheless, high-performance devices based only Bai, L., R. Yuan, Y. Chai, Y. Zhuo, Y. Yuan, and Y. Wang. 2012.
on graphene are still to come. One of the ways to fulfill this Simultaneous electrochemical detection of multiple analytes
task can be an application of oxidoreductases other than GOx. based on dual signal amplification of single-walled carbon
On the other side, graphenic materials were more broadly nanotubes and multi-labeled graphene sheets. Biomaterials
tested for integration with exoelectrogens, which allowed 33 (4):1090–1096. doi: http://dx.doi.org/10.1016/j.biomaterials.
2011.10.012.
for development of graphene-based MBFCs with operational Bo, Y., H. Yang, Y. Hu, T. Yao, and S. Huang. 2011. A novel electro-
characteristics similar to values obtained from most power- chemical DNA biosensor based on graphene and polyaniline
ful devices using other nanomaterials. Contrary to EBFCs, nanowires. Electrochim. Acta 56 (6):2676–2681. doi: http://
all types of common microbial biocatalysts were tested with dx.doi.org/10.1016/j.electacta.2010.12.034.
graphene. It should be noted that significant improvement of Bo, Y., W. Wang, J. Qi, and S. Huang. 2011. A DNA biosensor
the device characteristics can be achieved by optimization of, based on graphene paste electrode modified with Prussian
for example, electrode compartment volumes, pretreatment blue and chitosan. Analyst 136 (9):1946–1951. doi: 10.1039/
C1AN15084G.
of microbial biocatalysts and other features. By addressing Bonanni, A. and M. Pumera. 2011. Graphene platform for hairpin-
these issues in studies focused on graphene-based MBFCs, DNA-based impedimetric genosensing. ACS Nano 5 (3):2356–
much higher power outputs are expected. Another field which 2361. doi: 10.1021/nn200091p.
deserves to be studied more intensively is self-assembling of Cai, B., S. Wang, L. Huang, Y. Ning, Z. Zhang, and G.-J. Zhang.
graphenic nanoparticles and microbes. By this, fabrication 2014. Ultrasensitive label-free detection of PNA–DNA
will become even much cheaper and easier. Substantial effort hybridization by reduced graphene oxide field-effect tran-
has been made also to develop graphene-based inorganic cat- sistor biosensor. ACS Nano 8 (3):2632–2638. doi: 10.1021/
nn4063424.
alysts for cathodic reduction of oxygen which should replace Cai, Y., H. Li, B. Du, M. Yang, Y. Li, D. Wu, Y. Zhao, Y. Dai, and
commonly used, and quite expensive, Pt nanoparticles. There Q. Wei. 2011. Ultrasensitive electrochemical immunoassay for
are also studies focusing on graphene-based microbial bio- BRCA1 using BMIM · BF4-coated SBA-15 as labels and func-
cathodes, but an additional effort will be surely needed to bet- tionalized graphene as enhancer. Biomaterials 32 (8):2117–
ter understand this interesting field and to efficiently exploit 2123. doi: http://dx.doi.org/10.1016/j.biomaterials.2010.11.058.
revealed mechanisms thereafter. Casero, E., C. Alonso, M. D. Petit-Domínguez, L. Vázquez, A. M.
Parra-Alfambra, P. Merino, S. Álvarez-García, A. Andrés, E.
Suárez, F. Pariente, and E. Lorenzo. 2014. Lactate biosensor
ACKNOWLEDGMENTS based on a bionanocomposite composed of titanium oxide
nanoparticles, photocatalytically reduced graphene, and lac-
Financial support from the Slovak research and develop- tate oxidase. Microchim. Acta 181 (1–2):79–87. doi: 10.1007/
ment agency APVV 0282-11 and from VEGA 1/0229/12 is s00604-013-1070-z.
Castro Neto, A. H., F. Guinea, N. M. R. Peres, K. S. Novoselov, and Deng, K., Y. Xiang, L. Zhang, Q. Chen, and W. Fu. 2013. An
A. K. Geim. 2009. The electronic properties of graphene. Rev. aptamer-based biosensing platform for highly sensitive detec-
Mod. Phys. 81 (1):109–162. tion of platelet-derived growth factor via enzyme-mediated
Chang, L., C. Zhongfang, and L. Chen-Zhong. 2011. Surface engi- direct electrochemistry. Anal. Chim. Acta 759:61–65. doi:
neering of graphene-enzyme nanocomposites for miniatur- http://dx.doi.org/10.1016/j.aca.2012.11.018.
ized biofuel cell. IEEE Trans. Nanotechnol. 10 (1):59–62. doi: Devadas, B., V. Mani, and S. M. Chen. 2012. A glucose/O2 biofuel cell
10.1109/TNANO.2010.2050147. based on graphene and multiwalled carbon nanotube composite
Chen, H., B. Zhang, Y. Cui, B. Liu, G. Chen, and D. Tang. 2011. modified electrode. Int. J. Electrochem. Sci. 7 (9):8064–8075.
One-step electrochemical immunoassay of biomarker based Devi, R., S. Relhan, and C. S. Pundir. 2013. Construction of a chito-
on nanogold-functionalized graphene sensing platform. Anal. san/polyaniline/graphene oxide nanoparticles/polypyrrole/Au
Methods 3 (7):1615–1621. electrode for amperometric determination of urinary/plasma
Chen, H. F., Z. Q. Gao, Y. L. Cui, G. N. Chen, and D. P. Tang. oxalate. Sens. Actuat. B Chem. 186:17–26. doi: 10.1016/j.
2013. Nanogold-enhanced graphene nanosheets as multi- snb.2013.05.078.
enzyme assembly for sensitive detection of low-abundance Dong, H., Z. Zhu, H. Ju, and F. Yan. 2012. Triplex signal ampli-
proteins. Biosens. Bioelectron. 44:108–114. doi: 10.1016/j. fication for electrochemical DNA biosensing by coupling
bios.2012.12.054. probe-gold nanoparticles–graphene modified electrode with
Chen, H. J., Z. H. Zhang, R. Cai, X. Q. Kong, X. Chen, Y. N. Liu, enzyme functionalized carbon sphere as tracer. Biosens.
and S. Z. Yao. 2013. Molecularly imprinted electrochemi- Bioelectron. 33 (1):228–232. doi: http://dx.doi.org/10.1016/j.
cal sensor based on a reduced graphene modified carbon bios.2012.01.006.
electrode for tetrabromobisphenol A detection. Analyst 138 Dong, X., Y. Shi, W. Huang, P. Chen, and L.-J. Li. 2010. Electrical
(9):2769–2776. detection of DNA hybridization with single-base specificity
Chen, J. R., X. X. Jiao, H. Q. Luo, and N. B. Li. 2013. Probe-label- using transistors based on CVD-grown graphene sheets. Adv.
free electrochemical aptasensor based on methylene blue- Mater. 22 (14):1649–1653. doi: 10.1002/adma.200903645.
anchored graphene oxide amplification. J. Mater. Chem. B 1 Dreyer, D. R., R. S. Ruoff, and C. W. Bielawski. 2010. From con-
(6):861–864. doi: 10.1039/c2tb00267a. ception to realization: An historial account of graphene and
Chen, T.-Y., P. T. K. Loan, C.-L. Hsu, Y.-H. Lee, J. T.-W. Wang, some perspectives for its future. Angew. Chem. Int. Edit. 49
K.-H. Wei, C.-T. Lin, and L.-J. Li. 2013. Label-free detection (49):9336–9344. doi: 10.1002/anie.201003024.
of DNA hybridization using transistors based on CVD grown Du, M., T. Yang, C. Zhao, and K. Jiao. 2012. Electrochemical
graphene. Biosens. Bioelectron. 41:103–109. doi: http://dx.doi. logic aptasensor based on graphene. Sens. Actuat. B Chem.
org/10.1016/j.bios.2012.07.059. 169:255–260. doi: http://dx.doi.org/10.1016/j.snb.2012.04.078.
Chen, Y., B. Jiang, Y. Xiang, Y. Chai, and R. Yuan. 2011. Target Du, M., T. Yang, and K. Jiao. 2010. Immobilization-free direct
recycling amplification for sensitive and label-free impedi- electrochemical detection for DNA specific sequences based
metric genosensing based on hairpin DNA and graphene/Au on electrochemically converted gold nanoparticles/graphene
nanocomposites. Chem. Commun. 47 (48):12798–12800. doi: composite film. J. Mater. Chem. 20 (41):9253–9260. doi:
10.1039/C1CC14902D. 10.1039/C0JM01549K.
Chen, Y., H. Vedala, G. P. Kotchey, A. Audfray, S. Cecioni, A. Du, M., T. Yang, X. Li, and K. Jiao. 2012. Fabrication of DNA/gra-
Imberty, S. Vidal, and A. Star. 2011. Electronic detection of phene/polyaniline nanocomplex for label-free voltammetric
lectins using carbohydrate-functionalized nanostructures: detection of DNA hybridization. Talanta 88:439–444. doi:
Graphene versus carbon nanotubes. ACS Nano 6 (1):760–770. http://dx.doi.org/10.1016/j.talanta.2011.10.054.
doi: 10.1021/nn2042384. Du, Y., S. Guo, S. Dong, and E. Wang. 2011. An integrated sens-
Choi, B. G., J. Im, H. S. Kim, and H. S. Park. 2011. Flow-injection ing system for detection of DNA using new parallel-motif
amperometric glucose biosensors based on graphene/Nafion DNA triplex system and graphene–mesoporous silica–gold
hybrid electrodes. Electrochim. Acta 56 (27):9721–9726. doi: nanoparticle hybrids. Biomaterials 32 (33):8584–8592. doi:
http://dx.doi.org/10.1016/j.electacta.2011.07.118. http://dx.doi.org/10.1016/j.biomaterials.2011.07.091.
Ci, S. Q., Y. M. Wu, J. P. Zou, L. H. Tang, S. L. Luo, J. H. Li, and Z. Eissa, S., C. Tlili, L. L’Hocine, and M. Zourob. 2012. Electrochemical
H. Wen. 2012. Nitrogen-doped graphene nanosheets as high immunosensor for the milk allergen β-lactoglobulin based on
efficient catalysts for oxygen reduction reaction. Chinese Sci. electrografting of organic film on graphene modified screen-
Bull. 57 (23):3065–3070. doi: 10.1007/s11434-012-5253–5. printed carbon electrodes. Biosens. Bioelectron. 38 (1):308–
Clark, L. C. Jr. and C. Lyons. 1962. Electrode systems for continu- 313. doi: http://dx.doi.org/10.1016/j.bios.2012.06.008.
ous monitoring in cardiovascular surgery. Ann. NY Acad. Sci. Eissa, S., L. L’Hocine, M. Siaj, and M. Zourob. 2013. A graphene-
102:29–45. doi: 10.1111/j.1749-6632.1962.tb13623.x. based label-free voltammetric immunosensor for sensi-
Cui, L., Z. R. Chen, Z. Zhu, X. Y. Lin, X. Chen, and C. J. Yang. tive detection of the egg allergen ovalbumin. Analyst 138
2013. Stabilization of ssRNA on graphene oxide surface: (15):4378–84. doi: 10.1039/c3an36883a.
An effective way to design highly robust RNA probes. Anal. Eissa, S. and M. Zourob. 2012. A graphene-based electrochemi-
Chem. 85 (4):2269–2275. doi: 10.1021/ac303179z. cal competitive immunosensor for the sensitive detection of
Cui, M., B. Xu, C. Hu, H. B. Shao, and L. Qu. 2013. Direct elec- okadaic acid in shellfish. Nanoscale 4 (23):7593–7599. doi:
trochemistry and electrocatalysis of glucose oxidase on 10.1039/c2nr32146g.
three-dimensional interpenetrating, porous graphene modi- Erdem, A., E. Eksin, and M. Muti. 2014. Chitosan–graphene oxide
fied electrode. Electrochim. Acta 98:48–53. doi: http://dx.doi. based aptasensor for the impedimetric detection of lyso-
org/10.1016/j.electacta.2013.03.040. zyme. Colloid. Surface. B 115:205–211. doi: http://dx.doi.
Deng, K., J. Zhou, and X. Li. 2013. Noncovalent nanohybrid of fer- org/10.1016/j.colsurfb.2013.11.037.
rocene with chemically reduced graphene oxide and its appli- Fan, D.-H., D.-J. Niu, and K.-J. Huang. 2010. Simultaneous deter-
cation to dual biosensor for hydrogen peroxide and choline. mination of adenine and guanine in DNA based on carbox-
Electrochim. Acta 95:18–23. doi: http://dx.doi.org/10.1016/j. ylic acid functionalized graphene. Sens. Actuat. B Chem. 82
electacta.2013.02.009. (2):543–549. doi: 10.1016/j.snb.2010.03.023.
Fan, Y., K.-J. Huang, D.-J. Niu, C.-P. Yang, and Q.-S. Jing. 2011. Guo, Y., J. Li, and S. Dong. 2011. Hemin functionalized graphene
TiO2-graphene nanocomposite for electrochemical sensing of nanosheets-based dual biosensor platforms for hydrogen per-
adenine and guanine. Electrochim. Acta 56 (12):4685–4690. oxide and glucose. Sens. Actuat. B Chem. 160 (1):295–300.
doi: http://dx.doi.org/10.1016/j.electacta.2011.02.114. doi: http://dx.doi.org/10.1016/j.snb.2011.07.050.
Feng, L., L. Wu, J. Wang, J. Ren, D. Miyoshi, N. Sugimoto, and X. Guo, Y., J. H. Chen, and G. Chen. 2013. A label-free electro-
Qu. 2012. Detection of a prognostic indicator in early-stage chemical biosensor for detection of HIV related gene based
cancer using functionalized graphene-based peptide sensors. on interaction between DNA and protein. Sens. Actuat. B
Adv. Mater. 24 (1):125–131. doi: 10.1002/adma.201103205. Chem. 184:113–117. doi: http://dx.doi.org/10.1016/j.snb.2013.
Feng, L., Y. Chen, J. Ren, and X. Qu. 2011. A graphene functional- 04.046.
ized electrochemical aptasensor for selective label-free detec- Han, J., J. Ma, and Z. Ma. 2013. One-step synthesis of graphene
tion of cancer cells. Biomaterials 32 (11):2930–2937. doi: oxide-thionine-Au nanocomposites and its application for
http://dx.doi.org/10.1016/j.biomaterials.2011.01.002. electrochemical immunosensing. Biosens. Bioelectron.
Feng, Q., K. Liu, J. Fu, Y. Zhang, Z. Zheng, C. Wang, Y. Du, and 47:243–247.
W. Ye. 2012. Direct electrochemistry of hemoglobin based Han, J., Y. Zhuo, Y.-Q. Chai, L. Mao, Y.-L. Yuan, and R. Yuan. 2011.
on nano-composite film of gold nanopaticles and poly (dial- Highly conducting gold nanoparticles–graphene nanohybrid
lyldimethylammonium chloride) functionalized graphene. films for ultrasensitive detection of carcinoembryonic anti-
Electrochim. Acta 60:304–308. doi: http://dx.doi.org/10.1016/j. gen. Talanta 85 (1):130–135. doi: http://dx.doi.org/10.1016/j.
electacta.2011.11.048. talanta.2011.03.036.
Feng, R., Y. Zhang, H. Ma, D. Wu, H. Fan, H. Wang, H. Li, B. Du, He, Z. M., J. Liu, Y. Qiao, C. M. Li, and T. T. Y. Tan. 2012.
and Q. Wei. 2013. Ultrasensitive non-enzymatic and non- Architecture engineering of hierarchically porous chitosan/
mediator electrochemical biosensor using nitrogen-doped vacuum-stripped graphene scaffold as bioanode for high per-
graphene sheets for signal amplification and nanoporous alloy formance microbial fuel cell. Nano Lett. 12 (9):4738–4741.
as carrier. Electrochim. Acta 97:105–111. doi: http://dx.doi. doi: 10.1021/nl302175j.
org/10.1016/j.electacta.2013.02.093. Her, J.-L., T.-M. Pan, W.-Y. Lin, K.-S. Wang, and L.-J. Li. 2013.
Feng, R., Y. Zhang, H. Q. Yu, D. Wu, H. M. Ma, B. C. Zhu, Label-free detection of alanine aminotransferase using a gra-
C. X. Xu, H. Li, B. Du, and Q. Wei. 2013. Nanoporous PtCo- phene field-effect biosensor. Sens. Actuat. B Chem. 182:396–
based ultrasensitive enzyme-free immunosensor for zeranol 400. doi: http://dx.doi.org/10.1016/j.snb.2013.03.026.
detection. Biosens. Bioelectron. 42:367–372. doi: 10.1016/j. Hernández, R., C. Vallés, A. M. Benito, W. K. Maser, F. Xavier
bios.2012.10.031. Rius, and J. Riu. 2014. Graphene-based potentiometric bio-
Feng, W. and P. Ji. 2011. Enzymes immobilized on carbon nanosensor for the immediate detection of living bacteria. Biosens.
tubes. Biotechnol. Adv. 29 (6):889–895. doi: http://dx.doi. Bioelectron. 54:553–557. doi: http://dx.doi.org/10.1016/j.
org/10.1016/j.biotechadv.2011.07.007. bios.2013.11.053.
Feng, X., R. Li, C. Hu, and W. Hou. 2011. Direct electron transfer and Hossain, M. F. and J. Y. Park. 2014. Amperometric glucose biosensor
electrocatalysis of hemoglobin immobilized on graphene–Pt based on Pt-Pd nanoparticles supported by reduced graphene
nanocomposite. J. Electroanal. Chem. 657 (1–2):28–33. doi: oxide and integrated with glucose oxidase. Electroanalysis
http://dx.doi.org/10.1016/j.jelechem.2011.03.004. doi: 10.1002/elan.201400018.
Filip, J., J. Šefčovičová, P. Gemeiner, and J. Tkac. 2013. Hou, J. X., Z. L. Liu, and P. Y. Zhang. 2013. A new method for
Electrochemical features of bilirubin oxidase immobilized on fabrication of graphene/polyaniline nanocomplex modified
different carbon nanostructures. Key Eng. Mater. 543:13–17. microbial fuel cell anodes. J. Power Sources 224:139–144.
doi: 10.4028/www.scientific.net/KEM.543.13. doi: 10.1016/j.jpowsour.2012.09.091.
Gnana Kumar, G., Z. Awan, K. S. Nahm, and J. S. Xavier. 2014. Hou, L., Y. Cui, M. Xu, Z. Gao, J. Huang, and D. Tang. 2013.
Nanotubular MnO2/graphene oxide composites for the Graphene oxide-labeled sandwich-type impedimetric immu-
application of open air-breathing cathode microbial fuel noassay with sensitive enhancement based on enzymatic
cells. Biosens. Bioelectron. 53:528–534. doi: http://dx.doi. 4-chloro-1-naphthol oxidation. Biosens. Bioelectron. 47:
org/10.1016/j.bios.2013.10.012. 149–156.
Guo, C., H. Sun, and X. S. Zhao. 2012. Myoglobin within graphene Hu, Y., S. Hua, F. Li, Y. Jiang, X. Bai, D. Li, and L. Niu. 2011.
oxide sheets and Nafion composite films as highly sensitive Green-synthesized gold nanoparticles decorated graphene
biosensor. Sens. Actuat. B Chem. 164 (1):82–89. doi: http:// sheets for label-free electrochemical impedance DNA hybrid-
dx.doi.org/10.1016/j.snb.2012.01.057. ization biosensing. Biosens. Bioelectron. 26 (11):4355–4361.
Guo, K., K. Qian, S. Zhang, J. Kong, C. Yu, and B. Liu. 2011. Bio- doi: http://dx.doi.org/10.1016/j.bios.2011.04.037.
electrocatalysis of NADH and ethanol based on graphene Hu, Y., F. Li, D. Han, T. Wu, Q. Zhang, L. Niu, and Y. Bao. 2012.
sheets modified electrodes. Talanta 85 (2):1174–1179. doi: Simple and label-free electrochemical assay for signal-on
http://dx.doi.org/10.1016/j.talanta.2011.05.038. DNA hybridization directly at undecorated graphene oxide.
Guo, S.-R., J. Lin, M. Penchev, E. Yengel, M. Ghazinejad, C. S. Ozkan, Anal. Chim. Acta 753:82–89.
and M. Ozkan. 2011. Label free DNA detection using large area Huang, K. J., J. Li, Y. M. Liu, X. Cao, S. Yu, and M. Yu. 2012.
graphene based field effect transistor biosensors. J. Nanosci. Disposable immunoassay for hepatitis B surface antigen
Nanotechnol. 11 (6):5258–5263. doi: 10.1166/jnn.2011.3885. based on a graphene paste electrode functionalized with gold
Guo, W., Y. Cui, H. Song, and J. Sun. 2014. Layer-by-layer construc- nanoparticles and a Nafion-cysteine conjugate. Microchim.
tion of graphene-based microbial fuel cell for improved power Acta 177 (3–4):419–426.
generation and methyl orange removal. Bioproc. Biosyst. Eng. Huang, K.-J., D.-J. Niu, X. Liu, Z.-W. Wu, Y. Fan, Y.-F. Chang,
37 (9):1749–1758. doi: 10.1007/s00449-014-1148-y. and Y.-Y. Wu. 2011. Direct electrochemistry of catalase at
Guo, Y., Y. Han, Y. Guo, and C. Dong. 2013. Graphene-orange II amine-functionalized graphene/gold nanoparticles compos-
composite nanosheets with electroactive functions as label- ite film for hydrogen peroxide sensor. Electrochim. Acta
free aptasensing platform for signal-on detection of protein. 56 (7):2947–2953. doi: http://dx.doi.org/10.1016/j.elect-
Biosens. Bioelectron. 45 (1):95–101. acta.2010.12.094.
Huang, K.-J., Y.-J. Liu, H.-B. Wang, and Y.-Y. Wang. 2014. A sensi- Kong, F.-Y., S.-X. Gu, W.-W. Li, T.-T. Chen, Q. Xu, and W. Wang.
tive electrochemical DNA biosensor based on silver nanopar- 2014. A paper disk equipped with graphene/polyaniline/
ticles-polydopamine@graphene composite. Electrochim. Acta Au nanoparticles/glucose oxidase biocomposite modified
118:130–137. doi: http://dx.doi.org/10.1016/j.electacta.2013. screen-printed electrode: Toward whole blood glucose deter-
12.019. mination. Biosens. Bioelectron. 56:77–82. doi: http://dx.doi.
Huang, K.-J., L. Wang, J. Li, T. Gan, and Y.-M. Liu. 2013. Glassy org/10.1016/j.bios.2013.12.067.
carbon electrode modified with glucose oxidase–gra- Kong, F. Y., B. Y. Xu, Y. Du, J. J. Xu, and H. Y. Chen. 2013. A
phene–nano-copper composite film for glucose sensing. branched electrode based electrochemical platform: Towards
Measurement 46 (1):378–383. doi: http://dx.doi.org/10.1016/j. new label-free and reagentless simultaneous detection of two
measurement.2012.07.012. biomarkers. Chem. Commun. 49 (11):1052–1054. doi: 10.1039/
Huang, Y., X. Dong, Y. Shi, C. M. Li, L.-J. Li, and P. Chen. 2010. c2cc37675j.
Nanoelectronic biosensors based on CVD grown graphene. Kwak, Y. H., D. S. Choi, Y. N. Kim, H. Kim, D. H. Yoon, S.-S. Ahn,
Nanoscale 2 (8):1485–1488. doi: 10.1039/C0NR00142B. J.-W. Yang, W. S. Yang, and S. Seo. 2012. Flexible glucose
Huang, Y. X., X. W. Liu, J. F. Xie, G. P. Sheng, G. Y. Wang, Y. sensor using CVD-grown graphene-based field effect tran-
Y. Zhang, A. W. Xu, and H. Q. Yu. 2011. Graphene oxide sistor. Biosens. Bioelectron. 37 (1):82–87. doi: http://dx.doi.
nanoribbons greatly enhance extracellular electron trans- org/10.1016/j.bios.2012.04.042.
fer in bio-electrochemical systems. Chem. Commun. 47 Kwon, O. S., S. H. Lee, S. J. Park, J. H. An, H. S. Song, T. Kim, J. H.
(20):5795–5797. Oh, J. Bae, H. Yoon, T. H. Park, and J. Jang. 2013. Large-scale
Jang, H. D., S. K. Kim, H. Chang, K.-M. Roh, J.-W. Choi, and J. graphene micropattern nano-biohybrids: High-performance
Huang. 2012. A glucose biosensor based on TiO2–Graphene transducers for FET-Type flexible fluidic HIV immunoassays.
composite. Biosens. Bioelectron. 38 (1):184–188. doi: http:// Adv. Mater. 25 (30):4177–4185. doi: 10.1002/adma.201301523.
dx.doi.org/10.1016/j.bios.2012.05.033. Kwon, O. S., S. J. Park, J.-Y. Hong, A. Reum Han, J. S. Lee, J. S. Lee,
Jayakumar, K., R. Rajesh, V. Dharuman, R. Venkatasan, J. H. Hahn, J. H. Oh, and J. Jang. 2012. Flexible FET-type VEGF aptasensor
and S. Karutha Pandian. 2012. Gold nano particle decorated based on nitrogen-doped graphene converted from conducting
graphene core first generation PAMAM dendrimer for label polymer. ACS Nano 6 (2):1486–1493. doi: 10.1021/nn204395n.
free electrochemical DNA hybridization sensing. Biosens. Labroo, P. and Y. Cui. 2013. Flexible graphene bio-nanosensor for
Bioelectron. 31 (1):406–412. lactate. Biosens. Bioelectron. 41:852–856. doi: http://dx.doi.
Jiang, J., W. Fan, and X. Du. 2014. Nitrite electrochemical biosensing org/10.1016/j.bios.2012.08.024.
based on coupled graphene and gold nanoparticles. Biosens. Labroo, P. and Y. Cui. 2014. Graphene nano-ink biosensor arrays on
Bioelectron. 51:343–348. doi: http://dx.doi.org/10.1016/j. a microfluidic paper for multiplexed detection of metabolites.
bios.2013.08.007. Anal. Chim. Acta 813:90–96. doi: http://dx.doi.org/10.1016/j.
Jiang, S. T., Z. P. Chen, Q. Z. Diao, S. C. Sheng, G. M. Xie, M. J. aca.2014.01.024.
Zhang, and H. J. Xu. 2012. Disposable amperometric immu- Lei, H. Z., L. J. Mi, X. J. Zhou, J. J. Chen, J. Hu, S. W. Guo, and Y.
nosensor for detecting Toxoplasma gondii-specific IgM Zhang. 2011. Adsorption of double-stranded DNA to graphene
based on graphene and CeO2-Au. Acta Chimica Sinica 70 oxide preventing enzymatic digestion. Nanoscale 3 (9):3888–
(19):2085–2090. doi: 10.6023/a12050257. 3892. doi: 10.1039/c1nr10617a.
Jiang, X., K. Chen, J. Wang, K. Shao, T. Fu, F. Shao, D. Lu, J. Liang, Li, H., J. He, S. Li, and A. P. F. Turner. 2013. Electrochemical immu-
M. F. Foda, and H. Han. 2013. Solid-state voltammetry-based nosensor with N-doped graphene-modified electrode for label-
electrochemical immunosensor for Escherichia coli using free detection of the breast cancer biomarker CA 15-3. Biosens.
graphene oxide-Ag nanoparticle composites as labels. Analyst Bioelectron. 43 (1):25–29. doi: 10.1016/j.bios.2012.11.037.
138 (12):3388–3393. doi: 10.1039/C3AN00056G. Li, R., D. Wu, H. Li, C. Xu, H. Wang, Y. Zhao, Y. Cai, Q. Wei, and
Jin, B., P. Wang, H. Mao, B. Hu, H. Zhang, Z. Cheng, Z. Wu, X. Bian, B. Du. 2011. Label-free amperometric immunosensor for the
C. Jia, F. Jing, Q. Jin, and J. Zhao. 2014. Multi-nanomaterial detection of human serum chorionic gonadotropin based on
electrochemical biosensor based on label-free graphene for nanoporous gold and graphene. Anal. Biochem. 414 (2):196–
detecting cancer biomarkers. Biosens. Bioelectron. 55:464– 201. doi: http://dx.doi.org/10.1016/j.ab.2011.03.019.
469. doi: http://dx.doi.org/10.1016/j.bios.2013.12.025. Li, S., Y. Hu, Q. Xu, J. Sun, B. Hou, and Y. Zhang. 2012. Iron- and
Jin, L. L., K. Yang, K. Yao, S. Zhang, H. Q. Tao, S. T. Lee, Z. Liu, nitrogen-functionalized graphene as a non-precious metal
and R. Peng. 2012. Functionalized graphene oxide in enzyme catalyst for enhanced oxygen reduction in an air-cathode
engineering: A selective modulator for enzyme activity and microbial fuel cell. J. Power Sources 213:265–269. doi: http://
thermostability. ACS Nano 6 (6):4864–4875. doi: 10.1021/ dx.doi.org/10.1016/j.jpowsour.2012.04.002.
nn300217z. Li, Y., Y. Bai, G. Han, and M. Li. 2013. Porous-reduced graphene
Kang, X., J. Wang, H. Wu, I. A. Aksay, J. Liu, and Y. Lin. 2009. oxide for fabricating an amperometric acetylcholinesterase
Glucose oxidase–graphene–chitosan modified electrode biosensor. Sens. Actuat. B Chem. 185:706–712. doi: 10.1016/j.
for direct electrochemistry and glucose sensing. Biosens. snb.2013.05.061.
Bioelectron. 25 (4):901–905. doi: http://dx.doi.org/10.1016/j. Li, Z., Y. Huang, L. Chen, X. Qin, Z. Huang, Y. Zhou, Y. Meng,
bios.2009.09.004. J. Li, S. Huang, Y. Liu, W. Wang, Q. Xie, and S. Yao. 2013.
Kim, D.-J., H.-C. Park, I. Y. Sohn, J.-H. Jung, O. J. Yoon, J.-S. Amperometric biosensor for NADH and ethanol based on
Park, M.-Y. Yoon, and N.-E. Lee. 2013. Electrical graphene electroreduced graphene oxide–polythionine nanocomposite
aptasensor for ultra-sensitive detection of anthrax toxin with film. Sens. Actuat. B Chem. 181:280–287. doi: http://dx.doi.
amplified signal transduction. Small 9 (19):3352–3360. doi: org/10.1016/j.snb.2013.01.072.
10.1002/smll.201203245. Liang, B., L. Fang, G. Yang, Y. Hu, X. Guo, and X. Ye. 2013. Direct
Kim, D.-J., I. Y. Sohn, J.-H. Jung, O. J. Yoon, N. E. Lee, and J.-S. electron transfer glucose biosensor based on glucose oxidase
Park. 2013. Reduced graphene oxide field-effect transistor for self-assembled on electrochemically reduced carboxyl gra-
label-free femtomolar protein detection. Biosens. Bioelectron. phene. Biosens. Bioelectron. 43:131–136. doi: http://dx.doi.
41:621–626. doi: http://dx.doi.org/10.1016/j.bios.2012.09.040. org/10.1016/j.bios.2012.11.040.
Lin, D. J., J. Wu, M. Wang, F. Yan, and H. X. Ju. 2012. Triple sig- Liu, Y., G. Wang, C. Li, Q. Zhou, M. Wang, and L. Yang. 2014. A
nal amplification of graphene film, polybead carried gold novel acetylcholinesterase biosensor based on carboxylic gra-
nanoparticles as tracing tag and silver deposition for ultra- phene coated with silver nanoparticles for pesticide detection.
sensitive electrochemical immunosensing. Anal. Chem. 84 Mater. Sci. Eng. C 35:253–258. doi: http://dx.doi.org/10.1016/j.
(8):3662–3668. doi: 10.1021/ac3001435. msec.2013.10.036.
Lin, J., D. Teweldebrhan, K. Ashraf, G. Liu, X. Jing, Z. Yan, R. Loo, A. H., A. Bonanni, and M. Pumera. 2012. Impedimetric throm-
Li, M. Ozkan, R. K. Lake, A. A. Balandin, and C. S. Ozkan. bin aptasensor based on chemically modified graphenes.
2010. Gating of single-layer graphene with single-stranded Nanoscale 4 (1):143–147. doi: 10.1039/C1NR10966A.
deoxyribonucleic acids. Small 6 (10):1150–1155. doi: 10.1002/ Loo, A. H., A. Bonanni, and M. Pumera. 2013a. Thrombin apta-
smll.200902379. sensing with inherently electroactive graphene oxide nano-
Lin, L., Y. Liu, L. Tang, and J. Li. 2011. Electrochemical DNA sen- platelets as labels. Nanoscale 5 (11):4758–4762. doi: 10.1039/
sor by the assembly of graphene and DNA-conjugated gold c3nr00511a.
nanoparticles with silver enhancement strategy. Analyst 136 Loo, A. H., A. Bonanni, and M. Pumera. 2013b. An insight into the
(22):4732–4737. doi: 10.1039/C1AN15610A. hybridization mechanism of hairpin DNA physically immobi-
Linting, Z., L. Ruiyi, L. Zaijun, X. Qianfang, F. Yinjun, and L. lized on chemically modified graphenes. Analyst 138 (2):467–
Junkang. 2012. An immunosensor for ultrasensitive detection 471. doi: 10.1039/C2AN36199J.
of aflatoxin B1 with an enhanced electrochemical performance Lu, L., O. Zhang, J. Xu, Y. Wen, X. Duan, H. Yu, L. Wu, and T.
based on graphene/conducting polymer/gold nanoparticles/the Nie. 2013. A facile one-step redox route for the synthesis of
ionic liquid composite film on modified gold electrode with graphene/poly (3,4-ethylenedioxythiophene) nanocomposite
electrodeposition. Sens. Actuat. B Chem. 174:359–365. doi: and their applications in biosensing. Sens. Actuat. B Chem.
http://dx.doi.org/10.1016/j.snb.2012.06.051. 181:567–574. doi: http://dx.doi.org/10.1016/j.snb.2013.
Liu, A.-L., G.-X. Zhong, J.-Y. Chen, S.-H. Weng, H.-N. Huang, W. 02.024.
Chen, L.-Q. Lin et al. 2013. A sandwich-type DNA biosensor Luo, J., H. D. Jang, T. Sun, L. Xiao, Z. He, A. P. Katsoulidis, M. G.
based on electrochemical co-reduction synthesis of graphene- Kanatzidis, J. M. Gibson, and J. Huang. 2011. Compression
three dimensional nanostructure gold nanocomposite films. and aggregation-resistant particles of crumpled soft sheets.
Anal. Chim. Acta 767:50–58. doi: http://dx.doi.org/10.1016/j. ACS Nano 5 (11):8943–8949. doi: 10.1021/nn203115u.
aca.2012.12.049. Luo, Z., X. Ma, D. Yang, L. Yuwen, X. Zhu, L. Weng, and L. Wang.
Liu, C., S. Alwarappan, Z. Chen, X. Kong, and C.-Z. Li. 2010. 2013. Synthesis of highly dispersed titanium dioxide nano-
Membraneless enzymatic biofuel cells based on graphene clusters on reduced graphene oxide for increased glucose
nanosheets. Biosens. Bioelectron. 25 (7):1829–1833. doi: sensing. Carbon 57:470–476. doi: http://dx.doi.org/10.1016/j.
http://dx.doi.org/10.1016/j.bios.2009.12.012. carbon.2013.02.020.
Liu, F., Y. Piao, K. S. Choi, and T. S. Seo. 2012. Fabrication of free- Lv, Z., Y. Chen, H. Wei, F. Li, Y. Hu, C. Wei, and C. Feng. 2013.
standing graphene composite films as electrochemical biosen- One-step electrosynthesis of polypyrrole/graphene oxide com-
sors. Carbon 50 (1):123–133. doi: http://dx.doi.org/10.1016/j. posites for microbial fuel cell application. Electrochim. Acta
carbon.2011.07.061. 111:366–373.
Liu, J. Q., N. Kong, A. H. Li, X. Luo, L. Cui, R. Wang, and S. Y. Ma, X., Y. Jiang, F. Jia, Y. Yu, J. Chen, and Z. Wang. 2014. An
Feng. 2013. Graphene bridged enzyme electrodes for glu- aptamer-based electrochemical biosensor for the detection of
cose biosensing application. Analyst 138 (9):2567–2575. doi: Salmonella. J. Microbiol. Meth. 98:94–98. doi: http://dx.doi.
10.1039/c3an36929c. org/10.1016/j.mimet.2014.01.003.
Liu, K., J.-J. Zhang, C. Wang, and J.-J. Zhu. 2011. Graphene-assisted Mani, V., B. Devadas, and S.-M. Chen. 2013. Direct electrochemis-
dual amplification strategy for the fabrication of sensitive try of glucose oxidase at electrochemically reduced graphene
amperometric immunosensor. Biosens. Bioelectron. 26 oxide-multiwalled carbon nanotubes hybrid material modified
(8):3627–3632. doi: http://dx.doi.org/10.1016/j.bios.2011.02.018. electrode for glucose biosensor. Biosens. Bioelectron. 41:309–
Liu, K., J. Zhang, G. Yang, C. Wang, and J.-J. Zhu. 2010. Direct 315. doi: http://dx.doi.org/10.1016/j.bios.2012.08.045.
electrochemistry and electrocatalysis of hemoglobin based on Mao, K., D. Wu, Y. Li, H. Ma, Z. Ni, H. Yu, C. Luo, Q. Wei, and
poly(diallyldimethylammonium chloride) functionalized gra- B. Du. 2012. Label-free electrochemical immunosensor based
phene sheets/room temperature ionic liquid composite film. on graphene/methylene blue nanocomposite. Anal. Biochem.
Electrochem. Commun. 12 (3):402–405. doi: http://dx.doi. 422 (1):22–27. doi: http://dx.doi.org/10.1016/j.ab.2011.12.047.
org/10.1016/j.elecom.2010.01.004. Mao, S., K. H. Yu, J. B. Chang, D. A. Steeber, L. E. Ocola, and J. H.
Liu, T., H. Su, X. Qu, P. Ju, L. Cui, and S. Ai. 2011. Acetylcholinesterase Chen. 2013. Direct growth of vertically-oriented graphene for
biosensor based on 3-carboxyphenylboronic acid/reduced field-effect transistor biosensor. Sci. Rep. 3 :1–6. doi: 10.1038/
graphene oxide–gold nanocomposites modified electrode for srep01696.
amperometric detection of organophosphorus and carbamate Mao, S., K. Yu, G. Lu, and J. Chen. 2011. Highly sensitive protein
pesticides. Sensor. Actuat. B Chem. 160 (1):1255–1261. doi: sensor based on thermally-reduced graphene oxide field-
http://dx.doi.org/10.1016/j.snb.2011.09.059. effect transistor. Nano Res. 4 (10):921–930. doi: 10.1007/
Liu, X., J. Zhang, R. Yan, Q. Zhang, and X. Liu. 2014. Preparation s12274-011-0148-3.
of graphene nanoplatelet–titanate nanotube composite and Mohanty, N. and V. Berry. 2008. Graphene-based single-bacte-
its advantages over the two single components as biosensor rium resolution biodevice and DNA transistor: Interfacing
immobilization materials. Biosens. Bioelectron. 51:76–81. graphene derivatives with nanoscale and microscale bio-
doi: http://dx.doi.org/10.1016/j.bios.2013.07.029. components. Nano Lett. 8 (12):4469–4476. doi: 10.1021/
Liu, Y., Y. Liu, H. B. Feng, Y. M. Wu, L. Joshi, X. Q. Zeng, and nl802412n.
J. H. Li. 2012. Layer-by-layer assembly of chemical reduced Muti, M., S. Sharma, A. Erdem, and P. Papakonstantinou. 2011.
graphene and carbon nanotubes for sensitive electrochemi- Electrochemical monitoring of nucleic acid hybridization by
cal immunoassay. Biosens. Bioelectron. 35 (1):63–68. doi: single-use graphene oxide-based sensor. Electroanalysis 23
10.1016/j.bios.2012.02.007. (1):272–279. doi: 10.1002/elan.201000425.
Myung, S., A. Solanki, C. Kim, J. Park, K. S. Kim, and K.-B. Lee. Piao, Y., D. J. Han, and T. S. Seo. 2014. Highly conductive graph-
2011. Graphene-encapsulated nanoparticle-based biosensor ite nanoparticle based enzyme biosensor for electrochemical
for the selective detection of cancer biomarkers. Adv. Mater. glucose detection. Sens. Actuat. B Chem. 194:454–459. doi:
23 (19):2221–2225. doi: 10.1002/adma.201100014. http://dx.doi.org/10.1016/j.snb.2013.12.045.
Ni, Y., P. Wang, H. Song, X. Lin, and S. Kokot. 2014. Electrochemical Ping, J., Y. Wang, K. Fan, J. Wu, and Y. Ying. 2011. Direct electro-
detection of benzo(a)pyrene and related DNA damage using chemical reduction of graphene oxide on ionic liquid doped
DNA/hemin/nafion–graphene biosensor. Anal. Chim. Acta screen-printed electrode and its electrochemical biosensing
821:34–40. doi: http://dx.doi.org/10.1016/j.aca.2014.03.006. application. Biosens. Bioelectron. 28 (1):204–209. doi: http://
Niu, S., J. Sun, C. Nan, and J. Lin. 2013. Sensitive DNA biosen- dx.doi.org/10.1016/j.bios.2011.07.018.
sor improved by 1,10-phenanthroline cobalt complex as indi- Prasad, K. P., Y. Chen, and P. Chen. 2014. Three-dimensional
cator based on the electrode modified by gold nanoparticles graphene-carbon nanotube hybrid for high-performance
and graphene. Sensor. Actuat. B Chem. 176:58–63. doi: http:// enzymatic biofuel cells. ACS Appl. Mater. Interfaces 6
dx.doi.org/10.1016/j.snb.2012.08.085. (5):3387–3393. doi: 10.1021/am405432b.
Novoselov, K. S., A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. Qi, X., H. Gao, Y. Zhang, X. Wang, Y. Chen, and W. Sun. 2012.
V. Dubonos, I. V. Grigorieva, and A. A. Firsov. 2004. Electric Electrochemical DNA biosensor with chitosan-Co3O4 nanorod-
field effect in atomically thin carbon films. Science 306 graphene composite for the sensitive detection of Staphylococcus
(5696):666–669. doi: 10.1126/science.1102896. aureus nuc gene sequence. Bioelectrochemistry 88:42–47. doi:
Ohno, Y., K. Maehashi, K. Inoue, and K. Matsumoto. 2011. Label- http://dx.doi.org/10.1016/j.bioelechem.2012.05.007.
free aptamer-based immunoglobulin sensors using graphene Qiu, J.-D., J. Huang, and R.-P. Liang. 2011. Nanocomposite film
field-effect transistors. Jpn. J. Appl. Phys. 50 (7) :070120-1– based on graphene oxide for high performance flexible glu-
070120-4. doi: 10.1143/jjap.50.070120. cose biosensor. Sens. Actuat. B Chem. 160 (1):287–294. doi:
Ohno, Y., K. Maehashi, and K. Matsumoto. 2010. Label-free biosen- http://dx.doi.org/10.1016/j.snb.2011.07.049.
sors based on aptamer-modified graphene field-effect transis- Qu, F., H. Lu, M. Yang, and C. Deng. 2011. Electrochemical immu-
tors. J. Am. Chem. Soc. 132 (51):18012–18013. doi: 10.1021/ nosensor based on electron transfer mediated by graphene
ja108127r. oxide initiated silver enhancement. Biosens. Bioelectron. 26
Okamoto, S., Y. Ohno, K. Maehashi, K. Inoue, and K. Matsumoto. (12):4810–4814.
2012. Immunosensors based on graphene field-effect transis- Ruan, C., T. Li, Q. Niu, M. Lu, J. Lou, W. Gao, and W. Sun. 2012.
tors fabricated using antigen-binding fragment. Jpn. J. Appl. Electrochemical myoglobin biosensor based on graphene–
Phys. 51, article no. 06FD08. (6 PART 2). ionic liquid–chitosan bionanocomposites: Direct electro-
Oliveira, T. M. B. F., M. Fátima Barroso, S. Morais, M. Araújo, C. chemistry and electrocatalysis. Electrochim. Acta 64:183–189.
Freire, P. de Lima-Neto, A. N. Correia, M. B. P. P. Oliveira, doi: http://dx.doi.org/10.1016/j.electacta.2012.01.005.
and C. Delerue-Matos. 2013. Laccase–Prussian blue film–gra- Scott, C. L., G. Zhao, and M. Pumera. 2010. Stacked graphene nano-
phene doped carbon paste modified electrode for carbamate fibers doped polypyrrole nanocomposites for electrochemi-
pesticides quantification. Biosens. Bioelectron. 47:292–299. cal sensing. Electrochem. Commun. 12 (12):1788–1791. doi:
doi: http://dx.doi.org/10.1016/j.bios.2013.03.026. http://dx.doi.org/10.1016/j.elecom.2010.10.025.
Palanisamy, S., S. Cheemalapati, and S. M. Chen. 2012. An enzy- Shan, C., H. Yang, D. Han, Q. Zhang, A. Ivaska, and L. Niu. 2010.
matic biofuel cell based on electrochemically reduced graphene Graphene/AuNPs/chitosan nanocomposites film for glucose
oxide and multiwalled carbon nanotubes/zinc oxide modified biosensing. Biosens. Bioelectron. 25 (5):1070–1074. doi:
electrode. Int. J. Electrochem. Sci. 7 (11):11477–11487. http://dx.doi.org/10.1016/j.bios.2009.09.024.
Palanisamy, S., S. Cheemalapati, and S.-M. Chen. 2014. Sharon, E., X. Liu, R. Freeman, O. Yehezkeli, and I. Willner. 2013.
Amperometric glucose biosensor based on glucose oxidase Label-free analysis of thrombin or Hg2+ ions by nucleic
dispersed in multiwalled carbon nanotubes/graphene oxide acid-functionalized graphene oxide matrices assembled on
hybrid biocomposite. Mater. Sci. Eng. C 34:207–213. doi: field-effect transistors. Electroanalysis 25 (4):851–856. doi:
http://dx.doi.org/10.1016/j.msec.2013.09.011. 10.1002/elan.201200581.
Palanisamy, S., C. Karuppiah, and S.-M. Chen. 2014. Direct Shi, J.-J., T.-T. He, F. Jiang, E. S. Abdel-Halim, and J.-J. Zhu. 2014.
electrochemistry and electrocatalysis of glucose oxidase
Ultrasensitive multi-analyte electrochemical immunoas-
immobilized on reduced graphene oxide and silver nanopar- say based on GNR-modified heated screen-printed carbon
ticles nanocomposite modified electrode. Colloid Surf. B electrodes and PS@PDA-metal labels for rapid detection of
114:164–169. doi: http://dx.doi.org/10.1016/j.colsurfb.2013. MMP-9 and IL-6. Biosens. Bioelectron. 55:51–56. doi: http://
10.006. dx.doi.org/10.1016/j.bios.2013.11.056.
Paredes, J. I., S. Villar-Rodil, M. J. Fernandez-Merino, L. Guardia, A. Shiddiky, M. J. A., P. H. Kithva, D. Kozak, and M. Trau. 2012.
Martinez-Alonso, and J. M. D. Tascon. 2011. Environmentally An electrochemical immunosensor to minimize the nonspe-
friendly approaches toward the mass production of pro- cific adsorption and to improve sensitivity of protein assays
cessable graphene from graphite oxide. J. Mater. Chem. 21 in human serum. Biosens. Bioelectron. 38 (1):132–137.
(2):298–306. doi: 10.1039/C0JM01717E. doi:10.1016/j.bios.2012.05.014.
Park, S. J., O. S. Kwon, S. H. Lee, H. S. Song, T. H. Park, and J. Shiddiky, M. J. A., S. Rauf, P. H. Kithva, and M. Trau. 2012. Graphene/
Jang. 2012. Ultrasensitive flexible graphene based field-effect quantum dot bionanoconjugates as signal amplifiers in strip-
transistor (FET)-type bioelectronic nose. Nano Lett. 12 ping voltammetric detection of EpCAM biomarkers. Biosens.
(10):5082–5090. doi: 10.1021/nl301714x. Bioelectron. 35 (1):251–257. doi:10.1016/j.bios.2012.02.057.
Peng, K., H. Zhao, X. Wu, Y. Yuan, and R. Yuan. 2012. Ultrasensitive Singh, A., G. Sinsinbar, M. Choudhary, V. Kumar, R. Pasricha, H. N.
aptasensor based on graphene-3,4,9,10-perylenetetracarbox- Verma, S. P. Singh, and K. Arora. 2013. Graphene oxide-chi-
ylic dianhydride as platform and functionalized hollow PtCo tosan nanocomposite based electrochemical DNA biosensor
nanochains as enhancers. Sens. Actuat. B Chem. 169:88–95. for detection of typhoid. Sens. Actuat. B Chem. 185:675–684.
doi: http://dx.doi.org/10.1016/j.snb.2012.03.044. doi: 10.1016/j.snb.2013.05.014.
Singh, V., D. Joung, L. Zhai, S. Das, S. I. Khondaker, and S. Seal. Listeria monocytogenes with dendritic nanogold and elec-
2011. Graphene based materials: Past, present and future. Prog. trochemical reduced graphene modified carbon ionic liquid
Mater. Sci. 56 (8):1178–1271. doi: http://dx.doi.org/10.1016/j. electrode. Electrochim. Acta 85:145–151. doi: http://dx.doi.
pmatsci.2011.03.003. org/10.1016/j.electacta.2012.07.133.
Sohn, I.-Y., D.-J. Kim, J.-H. Jung, O. J. Yoon, T. N. Thanh, T. T. Sun, W., Y. Zhang, X. Ju, G. Li, H. Gao, and Z. Sun. 2012.
Quang, and N.-E. Lee. 2013. pH sensing characteristics and Electrochemical deoxyribonucleic acid biosensor based on
biosensing application of solution-gated reduced graphene carboxyl functionalized graphene oxide and poly-l-lysine
oxide field-effect transistors. Biosens. Bioelectron. 45:70–76. modified electrode for the detection of tlh gene sequence
doi: http://dx.doi.org/10.1016/j.bios.2013.01.051. related to vibrio parahaemolyticus. Anal. Chim. Acta 752:39–
Soldano, C., A. Mahmood, and E. Dujardin. 2010. Production, prop- 44. doi: http://dx.doi.org/10.1016/j.aca.2012.09.009.
erties and potential of graphene. Carbon 48 (8):2127–2150. Świetlikowska, A., M. Gniadek, and B. Pałys. 2013.
doi: http://dx.doi.org/10.1016/j.carbon.2010.01.058. Electrodeposited graphene nano-stacks for biosensor appli-
Song, H., Y. Ni, and S. Kokot. 2014. Investigations of an electro- cations. Surface groups as redox mediators for laccase.
chemical platform based on the layered MoS2–graphene and Electrochim. Acta 98:75–81. doi: http://dx.doi.org/10.1016/j.
horseradish peroxidase nanocomposite for direct electrochem- electacta.2013.03.055.
istry and electrocatalysis. Biosens. Bioelectron. 56:137–143. Tang, J., D. P. Tang, R. Niessner, G. N. Chen, and D. Knopp. 2011.
doi: http://dx.doi.org/10.1016/j.bios.2014.01.014. Magneto-controlled graphene immunosensing platform for
Song, Y., H. Liu, H. Tan, F. Xu, J. Jia, L. Zhang, Z. Li, and L. Wang. simultaneous multiplexed electrochemical immunoassay
2014. pH-switchable electrochemical sensing platform based using distinguishable signal tags. Anal. Chem. 83 (13):5407–
on chitosan-reduced graphene oxide/concanavalin A layer for 5414. doi: 10.1021/ac200969w.
assay of glucose and urea. Anal. Chem. 86 (4):1980–1987. doi: Uk Lee, H., H. Young Yoo, T. Lkhagvasuren, Y. Seok Song, C. Park,
10.1021/ac402742m. J. Kim, and S. Wook Kim. 2013. Enzymatic fuel cells based
Srivastava, S., V. Kumar, M. A. Ali, P. R. Solanki, A. Srivastava, on electrodeposited graphite oxide/cobalt hydroxide/chito-
G. Sumana, P. S. Saxena, A. G. Joshi, and B. D. Malhotra. san composite-enzyme electrode. Biosens. Bioelectron. 42
2013. Electrophoretically deposited reduced graphene oxide (1):342–348. doi:10.1016/j.bios.2012.10.020.
platform for food toxin detection. Nanoscale 5 (7):3043–3051. Wang, G., X. Gang, X. Zhou, G. Zhang, H. Huang, X. Zhang, and L.
doi: 10.1039/C3NR32242D. Wang. 2013. Electrochemical immunosensor with graphene/
Stine, R., J. T. Robinson, P. E. Sheehan, and C. R. Tamanaha. 2010. gold nanoparticles platform and ferrocene derivatives label.
Real-time DNA detection using reduced graphene oxide Talanta 103:75–80. doi:10.1016/j.talanta.2012.10.008.
field effect transistors. Adv. Mater. 22 (46):5297–5300. doi: Wang, G., X. Tan, Q. Zhou, Y. Liu, M. Wang, and L. Yang. 2014.
10.1002/adma.201002121. Synthesis of highly dispersed zinc oxide nanoparticles on
Su, B. L., D. P. Tang, Q. F. Li, J. Tang, and G. N. Chen. 2011. Gold- carboxylic graphene for development a sensitive acetylcholin-
silver-graphene hybrid nanosheets-based sensors for sensitive esterase biosensor. Sens. Actuat. B Chem. 190:730–736. doi:
amperometric immunoassay of alpha-fetoprotein using nano- http://dx.doi.org/10.1016/j.snb.2013.09.042.
gold-enclosed titania nanoparticles as labels. Anal. Chim. Wang, G.-X., Y. Qian, X.-X. Cao, and X.-H. Xia. 2012. Direct
Acta 692 (1–2):116–124. doi: 10.1016/j.aca.2011.02.061. electrochemistry of cytochrome c on a graphene/poly
Sun, G. Q., J. J. Lu, S. G. Ge, X. R. Song, J. H. Yu, M. Yan, and J. D. (3,4-ethylenedioxythiophene) nanocomposite modified elec-
Huang. 2013. Ultrasensitive electrochemical immunoassay for trode. Electrochem. Commun. 20:1–3. doi: http://dx.doi.
carcinoembryonic antigen based on three-dimensional mac- org/10.1016/j.elecom.2012.03.029.
roporous gold nanoparticles/graphene composite platform and Wang, J., A. Shi, X. Fang, X. Han, and Y. Zhang. 2014. Ultrasensitive
multienzyme functionalized nanoporous silver label. Anal. electrochemical supersandwich DNA biosensor using a glassy
Chim. Acta 775:85–92. doi: 10.1016/j.aca.2013.03.009. carbon electrode modified with gold particle-decorated sheets
Sun, T., L. Wang, N. Li, and X. Gan. 2011. Label-free electrochemi- of graphene oxide. Microchim. Acta 181 (9):935–940. doi:
cal aptasensor for thrombin detection based on the nafion@ 10.1007/s00604-014-1182-0.
graphene as platform. Bioproc. Biosyst. Eng. 34 (9):1081– Wang, J., Y. Zhao, F. X. Ma, K. Wang, F. B. Wang, and X. H. Xia.
1085. doi: 10.1007/s00449–011-0558-3. 2013. Synthesis of a hydrophilic poly-L-lysine/graphene
Sun, W., Y. Guo, X. Ju, Y. Zhang, X. Wang, and Z. Sun. 2013. Direct hybrid through multiple non-covalent interactions for bio-
electrochemistry of hemoglobin on graphene and titanium sensors. J. Mater. Chem. B 1 (10):1406–1413. doi: 10.1039/
dioxide nanorods composite modified electrode and itselectro- c2tb00454b.
catalysis. Biosens. Bioelectron. 42:207–213. doi: http://dx.doi. Wang, K., Q. Liu, Q.-M. Guan, J. Wu, H.-N. Li, and J.-J. Yan. 2011.
org/10.1016/j.bios.2012.10.034. Enhanced direct electrochemistry of glucose oxidase and bio-
Sun, W., Y. Guo, T. Li, X. Ju, J. Lou, and C. Ruan. 2012. sensing for glucose via synergy effect of graphene and CdS
Electrochemistry of horseradish peroxidase entrapped in gra- nanocrystals. Biosens. Bioelectron. 26 (5):2252–2257. doi:
phene and dsDNA composite modified carbon ionic liquid http://dx.doi.org/10.1016/j.bios.2010.09.043.
electrode. Electrochim. Acta 75:381–386. doi: http://dx.doi. Wang, L., E. Hua, M. Liang, C. Ma, Z. Liu, S. Sheng, M. Liu, G. Xie,
org/10.1016/j.electacta.2012.05.018. and W. Feng. 2014. Graphene sheets, polyaniline and AuNPs
Sun, W., X. Qi, Y. Chen, S. Liu, and H. Gao. 2011. Application of based DNA sensor for electrochemical determination of BCR/
chitosan/Fe3O4 microsphere–graphene composite modi- ABL fusion gene with functional hairpin probe. Biosens.
fied carbon ionic liquid electrode for the electrochemi- Bioelectron. 51:201–207. doi: http://dx.doi.org/10.1016/j.
cal detection of the PCR product of soybean Lectin gene bios.2013.07.049.
sequence. Talanta 87:106–112. doi: http://dx.doi.org/10.1016/j. Wang, Q., J. Lei, S. Deng, L. Zhang, and H. Ju. 2013. Graphene-
talanta.2011.09.047. supported ferric porphyrin as a peroxidase mimic for
Sun, W., X. Qi, Y. Zhang, H. Yang, H. Gao, Y. Chen, and Z. Sun. electrochemical DNA biosensing. Chem. Commun. 49
2012. Electrochemical DNA biosensor for the detection of (9):916–918. doi: 10.1039/C2CC37664D.
Wang, Q., J. Su, J. Xu, Y. Xiang, R. Yuan, and Y. Chai. 2012. Dual extracellular oxygen released from erythrocytes based on
amplified, sensitive electrochemical detection of pathogenic graphene film integrated with laccase and 2,2-Azino-bis(3-
sequences based on biobarcode labels and functional gra- ethylbenzothiazoline-6-sulfonic acid). Anal. Chem. 82
phene modified electrode. Sens. Actuat. B Chem. 163 (1):267– (9):3588–3596. doi: 10.1021/ac100621r.
271. doi: http://dx.doi.org/10.1016/j.snb.2012.01.050. Wu, Y., W. Xu, Y. Wang, Y. Yuan, and R. Yuan. 2013. Silver–graphene
Wang, X., J. Dong, H. Ming, and S. Ai. 2013. Sensing of glycopro- oxide nanocomposites as redox probes for electrochemical
tein via a biomimetic sensor based on molecularly imprinted determination of α-1-fetoprotein. Electrochim. Acta 88:135–
polymers and graphene-Au nanoparticles. Analyst 138 140. doi: http://dx.doi.org/10.1016/j.electacta.2012.10.081.
(4):1219–1225. doi: 10.1039/C2AN36297J. Wu, Y. F., P. Xue, Y. J. Kang, and K. M. Hui. 2013. Highly specific
Wang, X., J. F. Wang, H. J. Cheng, P. Yu, J. S. Ye, and L. Q. Mao. and ultrasensitive graphene-enhanced electrochemical detec-
2011. Graphene as a spacer to layer-by-layer assemble elec- tion of low-abundance tumor cells using silica nanoparticles
trochemically functionalized nanostructures for molecular coated with antibody-conjugated quantum dots. Anal. Chem.
bioelectronic devices. Langmuir 27 (17):11180–11186. doi: 85 (6):3166–3173. doi: 10.1021/ac303398b.
10.1021/la202018r. Xia, Y., P. Gao, Y. Bo, W. Wang, and S. Huang. 2012. Immunoassay
Wang, Y., J. Ping, Z. Ye, J. Wu, and Y. Ying. 2013. Impedimetric immu- for SKOV-3 human ovarian carcinoma cells using a graphene
nosensor based on gold nanoparticles modified graphene paper oxide-modified electrode. Microchim. Acta 179 (3–4):201–207.
for label-free detection of Escherichia coli O157: H7. Biosens. doi: 10.1007/s00604-012-0880-8.
Bioelectron. 49:492–498. doi:10.1016/j.bios.2013.05.061. Xiao, L., J. Damien, J. Luo, H. D. Jang, J. Huang, and Z. He. 2012.
Wang, Y., R. Yuan, Y. Chai, Y. Yuan, L. Bai, and Y. Liao. 2011. A Crumpled graphene particles for microbial fuel cell electrodes.
multi-amplification aptasensor for highly sensitive detection J. Power Sources 208:187–192. doi: http://dx.doi.org/10.1016/j.
of thrombin based on high-quality hollow CoPt nanoparticles jpowsour.2012.02.036.
decorated graphene. Biosens. Bioelectron. 30 (1):61–66. doi: Xie, X., G. Yu, N. Liu, Z. Bao, C. S. Criddle, and Y. Cui. 2012.
http://dx.doi.org/10.1016/j.bios.2011.08.027. Graphene-sponges as high-performance low-cost anodes for
Wang, Y., C. Zhao, D. Sun, J.-R. Zhang, and J.-J. Zhu. 2013. A gra- microbial fuel cells. Energ. Environ. Sci. 5 (5):6862–6866.
phene/poly(3,4-ethylenedioxythiophene) hybrid as an anode doi: 10.1039/C2EE03583A.
for high-performance microbial fuel cells. ChemPlusChem 78 Xie, Y., A. Chen, D. Du, and Y. Lin. 2011. Graphene-based immu-
(8):823–829. doi: 10.1002/cplu.201300102. nosensor for electrochemical quantification of phosphorylated
Wen, Q., S. Wang, J. Yan, L. Cong, Y. Chen, and H. Xi. 2014. Porous p53 (S15). Anal. Chim. Acta 699 (1):44–48. doi: http://dx.doi.
nitrogen-doped carbon nanosheet on graphene as metal-free org/10.1016/j.aca.2011.05.010.
catalyst for oxygen reduction reaction in air-cathode microbial Xu, H., H. Dai, and G. Chen. 2010. Direct electrochemistry and elec-
fuel cells. Bioelectrochemistry 95:23–28. doi: http://dx.doi. trocatalysis of hemoglobin protein entrapped in graphene and
org/10.1016/j.bioelechem.2013.10.007. chitosan composite film. Talanta 81 (1–2):334–338. doi: http://
Wen, Q., S. Wang, J. Yan, L. Cong, Z. Pan, Y. Ren, and Z. Fan. 2012. dx.doi.org/10.1016/j.talanta.2009.12.006.
MnO2–graphene hybrid as an alternative cathodic catalyst to Xu, Q., S.-X. Gu, L. Jin, Y. Zhou, Z. Yang, W. Wang, and X. Hu.
platinum in microbial fuel cells. J. Power Sources 216:187– 2014. Graphene/polyaniline/gold nanoparticles nanocom-
191. doi: http://dx.doi.org/10.1016/j.jpowsour.2012.05.023. posite for the direct electron transfer of glucose oxidase and
Wen, T., C. Xue, Y. Li, Y. Wang, R. Wang, J. Hong, X. Zhou, and H. glucose biosensing. Sens. Actuat. B Chem. 190:562–569. doi:
Jiang. 2012. Reduced graphene oxide-platinum nanoparticles http://dx.doi.org/10.1016/j.snb.2013.09.049.
composites based imprinting sensor for sensitively electrochem- Yang, G., L. Li, R. K. Rana, and J.-J. Zhu. 2013. Assembled gold
ical analysis of 17β-estradiol. J. Electroanal. Chem. 682:121– nanoparticles on nitrogen-doped graphene for ultrasensitive elec-
127. doi: http://dx.doi.org/10.1016/j.jelechem.2012.07.015. trochemical detection of matrix metalloproteinase-2. Carbon
Wen, W., T. Bao, J. Yang, M.-Z. Zhang, W. Chen, H.-Y. Xiong, X.-H. 61:357–366. doi: http://dx.doi.org/10.1016/j.carbon.2013.05.016.
Zhang, Y.-D. Zhao, and S.-F. Wang. 2014. A novel ampero- Yang, H.-W., C.-W. Lin, M.-Y. Hua, S.-S. Liao, Y.-T. Chen, H.-C.
metric adenosine triphosphate biosensor by immobilizing Chen, W.-H. Weng, C.-K. Chuang, S.-T. Pang, and C.-C. M.
graphene/dual-labeled aptamers complex onto poly(o-phen- Ma. 2014. Combined detection of cancer cells and a tumor
ylenediamine) modified electrode. Sens. Actuat. B Chem. biomarker using an immunomagnetic sensor for the improve-
191:695–702. doi: http://dx.doi.org/10.1016/j.snb.2013.10.061. ment of prostate-cancer diagnosis. Adv. Mater. 26 (22):3662–
Wen, W., W. Chen, Q.-Q. Ren, X.-Y. Hu, H.-Y. Xiong, X.-H. 3666. doi: 10.1002/adma.201305842.
Zhang, S.-F. Wang, and Y.-D. Zhao. 2012. A highly sensitive Yang, J., S. Deng, J. Lei, H. Ju, and S. Gunasekaran. 2011.
nitric oxide biosensor based on hemoglobin–chitosan/gra- Electrochemical synthesis of reduced graphene sheet–AuPd
phene–hexadecyltrimethylammonium bromide nanomatrix. alloy nanoparticle composites for enzymatic biosensing.
Sens. Actuat. B Chem. 166–167:444–450. doi: http://dx.doi. Biosens. Bioelectron. 29 (1):159–166. doi: http://dx.doi.
org/10.1016/j.snb.2012.02.086. org/10.1016/j.bios.2011.08.011.
Wu, H., J. Wang, X. Kang, C. Wang, D. Wang, J. Liu, I. A. Aksay, Yang, M., A. Javadi, and S. Gong. 2011. Sensitive electrochemi-
and Y. Lin. 2009. Glucose biosensor based on immobilization cal immunosensor for the detection of cancer biomarker
of glucose oxidase in platinum nanoparticles/graphene/chito- using quantum dot functionalized graphene sheets as labels.
san nanocomposite film. Talanta 80 (1):403–406. doi: http:// Sens. Actuat. B Chem. 155 (1):357–360. doi: http://dx.doi.
dx.doi.org/10.1016/j.talanta.2009.06.054. org/10.1016/j.snb.2010.11.055.
Wu, J., Y. Zou, X. Li, H. Liu, G. Shen, and R. Yu. 2005. A bio- Yang, T., Q. Guan, X. Guo, L. Meng, M. Du, and K. Jiao. 2012. Direct
sensor monitoring DNA hybridization based on polyaniline and freely switchable detection of target genes engineered by
intercalated graphite oxide nanocomposite. Sens. Actuat. reduced graphene oxide-poly(m-aminobenzenesulfonic acid)
B Chem. 104 (1):43–49. doi: http://dx.doi.org/10.1016/j. nanocomposite via synchronous pulse electrosynthesis. Anal.
snb.2004.04.097. Chem. 85 (3):1358–1366. doi: 10.1021/ac3030009.
Wu, X. M., Y. J. Hu, J. Jin, N. L. Zhou, P. Wu, H. Zhang, and Yang, T., X. Guo, Q. Kong, R. Yang, Q. Li, and K. Jiao. 2013.
C. X. Cai. 2010. Electrochemical approach for detection of Comparative studies on zirconia and graphene composites
obtained by one-step and stepwise electrodeposition for Zhang, B., Q. Li, and T. Cui. 2012. Ultra-sensitive suspended gra-
deoxyribonucleic acid sensing. Anal. Chim. Acta 786:29–33. phene nanocomposite cancer sensors with strong suppression
doi: 10.1016/j.aca.2013.05.023. of electrical noise. Biosens. Bioelectron. 31 (1):105–109. doi:
Yang, T., Q. Li, X. Li, X. Wang, M. Du, and K. Jiao. 2013. Freely switch- http://dx.doi.org/10.1016/j.bios.2011.09.046.
able impedimetric detection of target gene sequence based on Zhang, B., D. P. Tang, B. Q. Liu, H. F. Chen, Y. L. Cui, and G.
synergistic effect of ERGNO/PANI nanocomposites. Biosens. N. Chen. 2011. GoldMag nanocomposite-functionalized
Bioelectron. 42 (1):415–418. doi: 10.1016/j.bios.2012.11.007. graphene sensing platform for one-step electrochemical
Yang, T., Q. H. Li, L. Meng, X. H. Wang, W. W. Chen, and K. Jiao. immunoassay of alpha-fetoprotein. Biosens. Bioelectron. 28
2013. Synchronous electrosynthesis of poly(xanthurenic acid)- (1):174–180. doi: 10.1016/j.bios.2011.07.016.
reduced graphene oxide nanocomposite for highly sensitive Zhang, J., J. Lei, R. Pan, Y. Xue, and H. Ju. 2010. Highly sensitive
impedimetric detection of DNA. ACS Appl. Mater. Interfaces electrocatalytic biosensing of hypoxanthine based on func-
5 (9):3495–3499. doi: 10.1021/am400370s. tionalization of graphene sheets with water-soluble conduct-
Yi, H., W. Xu, Y. Yuan, L. Bai, Y. Wu, Y. Chai, and R. Yuan. 2014. ing graft copolymer. Biosens. Bioelectron. 26 (2):371–376.
A pseudo triple-enzyme cascade amplified aptasensor for doi: http://dx.doi.org/10.1016/j.bios.2010.07.127.
thrombin detection based on hemin/G-quadruplex as signal Zhang, L., H. Cheng, H.-M. Zhang, and L. Qu. 2012. Direct elec-
label. Biosens. Bioelectron. 54:415–420. doi: http://dx.doi. trochemistry and electrocatalysis of horseradish peroxidase
org/10.1016/j.bios.2013.11.036. immobilized in graphene oxide–Nafion nanocomposite film.
Yin, H., Y. Zhou, H. Zhang, X. Meng, and S. Ai. 2012. Electrochemical Electrochim. Acta 65:122–126. doi: http://dx.doi.org/10.1016/j.
determination of microRNA-21 based on graphene, LNA inte- electacta.2012.01.029.
grated molecular beacon, AuNPs and biotin multifunctional bio Zhang, Q., Y. Qiao, F. Hao, L. Zhang, S. Wu, Y. Li, J. Li, and X.-M.
bar codes and enzymatic assay system. Biosens. Bioelectron. Song. 2010. Fabrication of a biocompatible and conductive
33 (1):247–253. doi: http://dx.doi.org/10.1016/j.bios.2012.01.014. platform based on a single-stranded DNA/graphene nanocom-
Yin, Z., Q. He, X. Huang, J. Zhang, S. Wu, P. Chen, G. Lu, P. Chen, posite for direct electrochemistry and electrocatalysis. Chem.
Q. Zhang, Q. Yan, and H. Zhang. 2012. Real-time DNA detec- Eur. J. 16 (27):8133–8139. doi: 10.1002/chem.201000684.
tion using Pt nanoparticle-decorated reduced graphene oxide Zhang, S., S. Tang, J. Lei, H. Dong, and H. Ju. 2011. Functionalization
field-effect transistors. Nanoscale 4 (1):293–297. doi: 10.1039/ of graphene nanoribbons with porphyrin for electrocatalysis and
C1NR11149C. amperometric biosensing. J. Electroanal. Chem. 656 (1–2):285–
Yola, M. L., T. Eren, and N. Atar. 2014. A novel and sensitive elec- 288. doi: http://dx.doi.org/10.1016/j.jelechem.2010.10.005.
trochemical DNA biosensor based on Fe@Au nanoparticles Zhang, X., F. Gao, X. Cai, M. Zheng, F. Gao, S. Jiang, and Q. Wang.
decorated graphene oxide. Electrochim. Acta 125:38–47. doi: 2013. Application of graphene–pyrenebutyric acid nanocom-
http://dx.doi.org/10.1016/j.electacta.2014.01.074. posite as probe oligonucleotide immobilization platform in
Yong, Y. C., X. C. Dong, M. B. Chan-Park, H. Song, and P. Chen. a DNA biosensor. Mater. Sci. Eng. 33 (7):3851–3857. doi:
2012. Macroporous and monolithic anode based on polyani- 10.1016/j.msec.2013.05.022.
line hybridized three-dimensional graphene for high-perfor- Zhang, X., Q. Liao, M. Chu, S. Liu, and Y. Zhang. 2014. Structure
mance microbial fuel cells. ACS Nano 6 (3):2394–2400. doi: effect on graphene-modified enzyme electrode glucose sen-
10.1021/nn204656d. sors. Biosens. Bioelectron. 52:281–287. doi: http://dx.doi.
Yu, Y., Z. Chen, S. He, B. Zhang, X. Li, and M. Yao. 2014. Direct org/10.1016/j.bios.2013.07.022.
electron transfer of glucose oxidase and biosensing for glu- Zhang, X., Y. Zhang, Q. Liao, Y. Song, and S. Ma. 2013. Reduced
cose based on PDDA-capped gold nanoparticle modified graphene oxide-functionalized high electron mobility tran-
graphene/multi-walled carbon nanotubes electrode. Biosens. sistors for novel recognition pattern label-free DNA sensors.
Bioelectron. 52:147–152. doi: http://dx.doi.org/10.1016/j. Small 9 (23):4045–4050. doi: 10.1002/smll.201300793.
bios.2013.08.043. Zhang, Y., G. Mo, X. Li, W. Zhang, J. Zhang, J. Ye, X. Huang, and C.
Yuan, Y., X. Gou, R. Yuan, Y. Chai, Y. Zhuo, X. Ye, and X. Gan. Yu. 2011. A graphene modified anode to improve the perfor-
2011. Graphene-promoted 3,4,9,10-perylenetetracarboxylic mance of microbial fuel cells. J. Power Sources 196 (13):5402–
acid nanocomposite as redox probe in label-free electrochem- 5407. doi: http://dx.doi.org/10.1016/j.jpowsour.2011.02.067.
ical aptasensor. Biosens. Bioelectron. 30 (1):123–127. doi: Zhang, Y., J. Zhang, H. Wu, S. Guo, and J. Zhang. 2012. Glass car-
http://dx.doi.org/10.1016/j.bios.2011.08.041. bon electrode modified with horseradish peroxidase immo-
Yuan, Y., R. Yuan, Y. Chai, Y. Zhuo, X. Gan, and L. Bai. 2012. bilized on partially reduced graphene oxide for detecting
3,4,9,10-Perylenetetracarboxylic acid/hemin nanocompos- phenolic compounds. J. Electroanal. Chem. 681:49–55. doi:
ites act as redox probes and electrocatalysts for constructing http://dx.doi.org/10.1016/j.jelechem.2012.06.004.
a pseudobienzyme-channeling amplified electrochemical Zhang, Y.-Q., Y.-J. Fan, L. Cheng, L.-L. Fan, Z.-Y. Wang, J.-P.
aptasensor. Chem. Eur. J. 18 (44):14186–14191. doi: 10.1002/ Zhong, L.-N. Wu, X.-C. Shen, and Z.-J. Shi. 2013. A novel
chem.201103960. glucose biosensor based on the immobilization of glucose oxi-
Yuan, Y., S. Zhou, B. Zhao, L. Zhuang, and Y. Wang. 2012. dase on layer-by-layer assembly film of copper phthalocyanine
Microbially-reduced graphene scaffolds to facilitate extra- functionalized graphene. Electrochim. Acta 104:178–184. doi:
cellular electron transfer in microbial fuel cells. Bioresour. http://dx.doi.org/10.1016/j.electacta.2013.04.099.
Technol. 116:453–8. doi: 10.1016/j.biortech.2012.03.118. Zhang, Y. and W. Jiang. 2012. Decorating graphene sheets with gold
Yuan, Y. L., G. P. Liu, R. Yuan, Y. Q. Chai, X. X. Gan, and L. J. nanoparticles for the detection of sequence-specific DNA.
Bai. 2013. Dendrimer functionalized reduced graphene oxide Electrochim. Acta 71:239–245. doi: http://dx.doi.org/10.1016/j.
as nanocarrier for sensitive pseudobienzyme electrochemical electacta.2012.03.136.
aptasensor. Biosens. Bioelectron. 42:474–480. doi: 10.1016/j. Zhang, Z., S. Liu, M. Kang, G. Yang, Y. Li, F. Yan, L. He, X. Feng,
bios.2012.11.003. P. Wang, and S. Fang. 2014. Graphene nanostructures with
Zhang, B. and T. H. Cui. 2011. An ultrasensitive and low-cost gra- plasma-polymerized pyrrole as an adsorbent layer for bio-
phene sensor based on layer-by-layer nano self-assembly. Appl. sensors. Microchim. Acta 181 (9):1059–1067. doi: 10.1007/
Phys. Lett. 98 (7) :073116–073116-3. doi: 10.1063/1.3557504. s00604-014-1207-8.
Zhao, C., P. Gai, C. Liu, X. Wang, H. Xu, J. Zhang, and J. J. Zhu. 2013. Zhou, X.-H., L.-H. Liu, X. Bai, and H.-C. Shi. 2013. A reduced gra-
Polyaniline networks grown on graphene nanoribbons-coated phene oxide based biosensor for high-sensitive detection of
carbon paper with a synergistic effect for high-performance phenols in water samples. Sens. Actuat. B Chem. 181:661–667.
microbial fuel cells. J. Mater. Chem. A 1 (40):12587–12594. doi: http://dx.doi.org/10.1016/j.snb.2013.02.021.
doi: 10.1039/C3TA12947K. Zhou, Y., M. Wang, Z. Xu, C. Ni, H. Yin, and S. Ai. 2014.
Zhao, J., G. Chen, L. Zhu, and G. Li. 2011. Graphene quantum dots- Investigation of the effect of phytohormone on the expression
based platform for the fabrication of electrochemical biosen- of microRNA-159a in Arabidopsis thaliana seedlings based
sors. Electrochem. Commun. 13 (1):31–33. doi: http://dx.doi. on mimic enzyme catalysis systematic electrochemical bio-
org/10.1016/j.elecom.2010.11.005. sensor. Biosens. Bioelectron. 54:244–250. doi: http://dx.doi.
Zheng, W., H. Y. Zhao, J. X. Zhang, H. M. Zhou, X. X. Xu, Y. F. org/10.1016/j.bios.2013.11.026.
Zheng, Y. B. Wang, Y. Cheng, and B. Z. Jang. 2010. A glucose/ Zhou, Y., H. Yin, X. Meng, Z. Xu, Y. Fu, and S. Ai. 2012. Direct
O2 biofuel cell base on nanographene platelet-modified elec- electrochemistry of sarcosine oxidase on graphene, chito-
trodes. Electrochem. Commun. 12 (7):869–871. doi: 10.1016/j. san and silver nanoparticles modified glassy carbon elec-
elecom.2010.04.006. trode and its biosensing for hydrogen peroxide. Electrochim.
Zhiguo, G., Y. Shuping, L. Zaijun, S. Xiulan, W. Guangli, F. Yinjun, and Acta 71:294–301. doi: http://dx.doi.org/10.1016/j.electacta.
L. Junkang. 2011. An ultrasensitive electrochemical biosensor 2012.04.014.
for glucose using CdTe-CdS core–shell quantum dot as ultrafast Zhu, L., L. Luo, and Z. Wang. 2012. DNA electrochemical biosen-
electron transfer relay between graphene-gold nanocomposite sor based on thionine-graphene nanocomposite. Biosens.
and gold nanoparticle. Electrochim. Acta 56 (25):9162–9167. Bioelectron. 35 (1):507–511. doi: http://dx.doi.org/10.1016/j.
doi: http://dx.doi.org/10.1016/j.electacta.2011.07.117. bios.2012.03.026.
Zhong, Z., W. Wu, D. Wang, D. Wang, J. Shan, Y. Qing, and Z. Zhu, Q., Y. Chai, R. Yuan, Y. Zhuo, J. Han, Y. Li, and N. Liao.
Zhang. 2010. Nanogold-enwrapped graphene nanocomposites 2013. Amperometric immunosensor for simultaneous detec-
as trace labels for sensitivity enhancement of electrochemical tion of three analytes in one interface using dual functional-
immunosensors in clinical immunoassays: Carcinoembryonic ized graphene sheets integrated with redox-probes as tracer
antigen as a model. Biosens. Bioelectron. 25 (10):2379–2383. matrixes. Biosens. Bioelectron. 43 (1):440–445. doi: 10.1016/j.
doi: http://dx.doi.org/10.1016/j.bios.2010.03.009. bios.2012.12.030.
Zhou, J., J. Y. Zhuang, M. Miro, Z. Q. Gao, G. N. Chen, and D. P. Zhu, X., L. Sun, Y. Chen, Z. Ye, Z. Shen, and G. Li. 2013.
Tang. 2012. Carbon nanospheres-promoted electrochemical Combination of cascade chemical reactions with graphene–
immunoassay coupled with hollow platinum nanolabels for DNA interaction to develop new strategy for biosensor fab-
sensitivity enhancement. Biosens. Bioelectron. 35 (1):394– rication. Biosens. Bioelectron. 47:32–37. doi: http://dx.doi.
400. doi: 10.1016/j.bios.2012.03.025. org/10.1016/j.bios.2013.02.039.
Zhou, K., Y. Zhu, X. Yang, and C. Li. 2010. Electrocatalytic oxidation Zhuang, L., Y. Yuan, G. Q. Yang, and S. G. Zhou. 2012. In situ for-
of glucose by the glucose oxidase immobilized in graphene- mation of graphene/biofilm composites for enhanced oxygen
Au-Nafion biocomposite. Electroanalysis 22 (3):259–264. doi: reduction in biocathode microbial fuel cells. Electrochem.
10.1002/elan.200900321. Commun. 21:69–72. doi: 10.1016/j.elecom.2012.05.010.
Zhou, K., Y. Zhu, X. Yang, J. Luo, C. Li, and S. Luan. 2010. A novel Zor, E., I. H. Patir, H. Bingol, and M. Ersoz. 2013. An electro-
hydrogen peroxide biosensor based on Au–graphene–HRP– chemical biosensor based on human serum albumin/graphene
chitosan biocomposites. Electrochim. Acta 55 (9):3055–3060. oxide/3-aminopropyltriethoxysilane modified ITO electrode
doi: http://dx.doi.org/10.1016/j.electacta.2010.01.035. for the enantioselective discrimination of d- and l-trypto-
Zhou, M., Y. Zhai, and S. Dong. 2009. Electrochemical sensing and phan. Biosens. Bioelectron. 42:321–325. doi: http://dx.doi.
biosensing platform based on chemically reduced graphene org/10.1016/j.bios.2012.10.068.
oxide. Anal. Chem. 81 (14):5603–5613. doi: 10.1021/ac900136z.
12 Chemical Modification of Graphene
Xiao-Rong Li, Jing-Juan Xu, and Hong-Yuan Chen
CONTENTS
Abstract...................................................................................................................................................................................... 207
12.1 Covalent Modification on Graphene Carbon Network..................................................................................................... 207
12.1.1 Nucleophilic Addition........................................................................................................................................... 207
12.1.2 Electrophilic Substitution..................................................................................................................................... 208
12.1.2.1 Friedel–Crafts Acylation....................................................................................................................... 208
12.1.2.2 Hydrogen–Lithium Exchange................................................................................................................ 208
12.1.3 Cycloaddition........................................................................................................................................................ 209
12.1.3.1 [2 + 1] Cycloaddition...............................................................................................................................210
12.1.3.2 [2 + 2] Cycloaddition............................................................................................................................. 212
12.1.3.3 [3 + 2] Cycloaddition............................................................................................................................. 212
12.1.3.4 [4 + 2] Cycloaddition.............................................................................................................................. 213
12.1.4 Rearrangement...................................................................................................................................................... 213
12.2 Noncovalent Modification on Graphene Carbon Network................................................................................................214
12.2.1 van der Waals Force..............................................................................................................................................214
12.2.2 Electrostatic Interaction........................................................................................................................................ 215
12.2.3 Hydrogen Bonding.................................................................................................................................................216
12.2.4 Coordination Bonds...............................................................................................................................................216
12.2.5 π–π Stacking Interaction.......................................................................................................................................216
12.3 Chemical Doping Modifications on Graphene Carbon Network......................................................................................217
12.3.1 Interstital Doping...................................................................................................................................................217
12.3.2 Substitutional Doping............................................................................................................................................219
12.3.3 Selective Doping................................................................................................................................................... 220
12.4 Conclusions....................................................................................................................................................................... 221
References.................................................................................................................................................................................. 221
ABSTRACT modification of graphene via these methods offers good

arenas for versatile applications, such as intelligent sensors,
Chemical modification is a leading strategy with which green energy devices, and so on.
to intrinsically modify the properties of carbon materi-
als. Graphene, a two-dimensional, single-layer sheet of sp2
hybridized carbon atoms, has attracted tremendous attention 12.1 COVALENT MODIFICATION ON
and research interest, owing to its exceptional electronic con- GRAPHENE CARBON NETWORK
ductivity, thermal stability, and mechanical strength. In this
chapter, we focus mainly on the methodologies of chemi- Covalent modification of a graphene framework is associ-
cal modification of graphene, including covalent, noncova- ated with rehybridization of one or more sp2 carbon atoms of
lent, and chemical doping methods. Covalent modification the carbon network into the sp3 configuration accompanied
involves the nucleophilic addition, electrophilic substitution, by simultaneous loss of electronic conjugation. The covalent
cycloaddition, and rearrangement reactions; noncovalent modification of graphene can be achieved mainly in four
modification based on the van der Waals force, electrostatic different ways: nucleophilic addition, electrophilic substitu-
interaction, hydrogen bonding, coordination bonds, or the tion, cycloaddition, and rearrangement reactions (Table 12.1).
π–π interaction. Chemical doping modification contains inter- Here, we focus on the description of the general mechanisms
stitial doping, substitutional doping, and selective doping. of these reactions.
Many materials have been used for chemical modification of
graphene, namely, polymers, metal, semiconductor, organic
12.1.1 Nucleophilic Addition
materials, and so on, which leads to the desired solubility and
stability, tunable electric, thermal, and mechanical properties, The Bingel reaction utilizes a halide derivative of the diethyl
and enhanced catalytic and biological properties. Chemical malonate moiety in the presence of a base which originated
207
TABLE 12.1
Covalent Modification Techniques on Graphene Carbon Network
Modification Techniques Chemical Reaction Mechanism Chemical Reaction Agent References
Nucleophilic addition Bingel reaction Diethyl malonate and extended tetrathiafulvalene [2]
4,4′-[(1,3-Dioxo-1,3-propanediyl)bis(oxy)] [3]
bis[2,2,6,6-tetramethyl-1-piperidinyloxy]
Electrophilic substitution Friedel–Crafts acylation 4-Aminobenzoic acid [5]
Hydrogen–lithium exchange Triethylamine [6]
[2 + 1] cycloaddition Cyclopropane adduct Dichlorocarbene [10]
Aziridine adduct Azidotrimethylsilane [11]
Perfluorophenylazide [12]
Phenylalanine [14]
[2 + 2] cycloaddition Elimination–addition (benzyne adduct) 2-(Trimethylsilyl)aryl triate [15]
[3 + 2] cycloaddition 1,3-Dipolar cycloaddition Azomethine ylide [16]
Modified amino acid [17]
Porphyrin [18]
[4 + 2] cycloaddition Diels–Alder cycloaddition Tetracyanoethylene and maleic anhydride [20]
9-Methylanthracene and 2,3-dimethoxy-1,3-butadiene [20]
Rearrangement Claisen rearrangement Allylic alcohol [21]
from the cyclopropanation chemistry of fullerene. It has found analysis, X-ray photoelectron spectroscopy, Raman spectros-
usefulness in graphene chemistry given the ease of its reac- copy, and electron paramagnetic resonance measurements.
tion conditions. The halide-malonate moiety is typically gen-
erated in situ in a mixture of tetrahalomethane and a base.
12.1.2 Electrophilic Substitution
The base abstracts a proton from the halide-malonate to pro-
vide an enolate which subsequently nucleophilically attacks a The electron-rich nature of graphene allows the addition of
C=C bond on the graphene carbon framework.1 The resulting molecules onto the carbon backbone by electrophilic substi-
carbanion undergoes a subsequent nucleophilic substitution tution. This has resulted in the high reactivity of graphene
which displaces the halide atom to provide a cyclopropane toward strong electrophiles. Substitution reactions have been
adduct via intramolecular ring closure. successfully achieved via the Friedel–Crafts acylation and
On the basis of the Bingel cyclopropanation reac- hydrogen–lithium exchange methods.
tion, Tagmatarchis et al. introduced cyclopropane adducts
(diethyl malonate and extended tetrathiafulvalene moieties) 12.1.2.1 Friedel–Crafts Acylation
onto the exfoliated graphene sp2 carbon network as shown This reaction needs to introduce aryl ketone groups onto the
in Figure 12.1.2 The hybrid graphene materials were func- graphene platform. A typical Friedel–Crafts acylation reac-
tionalized up to ~23% and displayed good dispersibility in tion goes through an intermediate of an acyl anion catalyzed
organic solvents, such as dichloromethane, o-dichloroben- by the presence of a Lewis acid, which was adopted to initiate
zene, dimethylformamide, and toluene. The electroactive the Friedel–Crafts reaction, contributing to the improvement
extended tetrathiafulvalene–graphene materials exhibited of interfacial interaction.4
the formation of the radical ion pair, which encapsulated Modification of graphene via Friedel–Crafts reaction has
the one-electron reduction of graphene and a one-electron been successfully performed by Pumera and coworkers.5 The
oxidation of extended tetrathiafulvalene. Also, Coronado acylation was achieved with 4-aminobenzoic acid in the pres-
and his coworkers functionalized the exfoliated graphene ence of PPA/P2O5 (Figure 12.3). The hybrid graphene material
with 4,4′-[(1,3-dioxo-1,3-propanediyl)bis(oxy)]bis[2,2,6,6- presented a new peak corresponding to C=O in the C1s core
tetramethyl-1-piperidinyloxy] organic radicals by using the level spectrum. This was also further supported by Fourier
Bingel cyclopropanation reaction (Figure 12.2).3 Graphene transform infrared spectroscopy.
layers were produced by liquid-phase exfoliation of graphite
in o-dichlorobenzene (oDCB) or benzylamine with the aid 12.1.2.2 Hydrogen–Lithium Exchange
of ultrasounds. Then, the use of different solvents permits Compared with the simple aromatic moieties, metallated aro-
variation of the density of radicals anchored to the carbon matics (Ar–M) are usually more reactive. In the presence of
layers. This variable degree of functionalization will be bet- electrophiles, the Ar–M reacts readily to form a covalent bond
ter suited for the study of the effect of the radical concen- resulting in the extension of the aromatic ring.
tration over the graphene sheets on its electronic structures Yan et al., demonstrated the functionalization of a graphene
and conducting properties. In this reaction, covalent graft- carbon network based on the hydrogen–lithium exchange.6 In
ing is unambiguously demonstrated by thermo-gravimetric the effort to produce a solid basic catalyst with a triethylamine
Chemical Modification of Graphene 209
NH2 NH2
NH2
Benzylamine NH2 NH2 NH2

sonication
NH2 NH2 NH2
(1) or (2) CBr4, DBU,

MW irradiation
O O
O O
O O R O O
O O O
(1) R O O
O O
RO
S S
O O
O O S S
CH2CH3 G1
(2)
R:
S S
G2
S S
FIGURE 12.1 Functionalization of graphene with diethyl malonate (1) and extended tetrathiafulvalene (2) moieties based on the Bingel
reaction. (Adapted from Economopoulos, S. P. et al. 2010. ACS Nano 4:7499–7507.)
O. O.
N N
O. O.
O N N
O
O O O O
oDCB CBr4, DBU, oDCB O O
O
Ultrasounds
O
O O
O O
. N N .
O O
1 O O
. N N .
O O
FIGURE 12.2 Functionalization of graphene with paramagnetic organic radicals in a two-step procedure based on the Bingel reaction.
(Adapted from Bosch-Navarro, C. et al. 2013. J. Mater. Chem. C 1:4590–4598.)
moiety, an electrophilic substitution method was utilized. The cycloaddition. Cycloaddition reactions involve simultaneous
group exploited the possibility of a carbometalation reaction bond cleaving and bond formation processes. The types of
(Figure 12.4). The hybrid graphene showed extremely high reactions are termed as pericyclic. While cycloaddition reac-
catalytic activity and hydrolysis rate. tions have been accomplished on fullerene and carbon nano-
tubes, these reactions have provided beneficial effects for
graphene platforms in terms of improvement in its band gap
12.1.3 Cycloaddition
and solubility. Here, four major types of cycloaddition intro-
The most frequent synthetic route to covalently connect bio- duced in the functionalization of graphene will be highlighted
active molecules to sp2 graphene carbon network is through in the following section.
H2N
H2N
O
NH2 O
O
HO
PPA/P2O5
O
O
NH2
GR GR-FC NH2
FIGURE 12.3 Functionalization of graphene via Friedel–Crafts acylation with PPA/P2O5. (Adapted from Chua, C. K. and Pumera, M.
2012. Chem. Asian J. 7:1009–1012.)
Li
Br-CH2CH2-NEt2
nBuLi N
N
Deprotonation
Li
nBuLi
Br-CH2CH2-NEt2 N
Bu
N
Carbometalation
FIGURE 12.4 Amino grafting on graphene based on the hydrogen–lithium exchange. (Adapted from Yuan, C., Chen, W. and Yan, L. 2012.
J. Mater. Chem. 22:7456–7460.)
12.1.3.1 [2 + 1] Cycloaddition

There are generally two main approaches to functionalize the
graphene sp2 carbon network in [2 + 1] cycloaddition, namely,
the cycloaddition toward the formation of cyclopropane or
CHCl3
aziridine adducts. The cyclopropane adduct is achieved via NaOH
dichlorocarbene insertion reaction, whereas aziridine adduct PTC
is achieved via nitrene moiety as an intermediate. Cl Cl
Cyclopropane adduct. Carbene in the form of dichlorocar-
bene has been introduced on the graphene sp2 carbon network.
Cl Cl
The formation of carbene was achieved by the α-elimination
method in a mixture of chloroform and strong base. Carbene FIGURE 12.5 Functionalization of graphene via dichlorocarbene
generated by this method usually exists as singlet carbene cyclopropanation reaction. (Adapted from Chua, C. K., Ambrosi, A.
since the starting central carbon atom of the chloroform pre- and Pumera, M. 2012. Chem. Commun. 48:5376–5378.)
cursor inherited two paired electrons from the broken C–H
σ-bond, which readily react in their initial spin state, mean- exhibited a strong presence of C–Cl bond formation at 285.6
ing they did not switch to a triplet state. In this case, the sin- and 292 eV. The occurrence of Cl2p and Cl2s peaks also
glet carbene behaves as an electrophile reacts spontaneously highlighted the success of the functionalization. In addition,
with the graphene sp2 carbon atoms through a concerted the presence of C–C stretching bands at 1155 and 1219 cm–1
manner.7 The addition of dichlorocarbene occurred based on as well as the C–Cl bond stretch at 634 cm–1 were observed
two phases, where first, the empty π orbital that is the lowest under Fourier transform infrared spectroscopy analyses. This
unoccupied molecular orbital (LUMO) of the carbene inter- was also further supported by the functionalization of gra-
acts with the π bond, namely, the highest occupied molecular phene via dichlorocarbene cyclopropanation reaction.
orbital (HOMO) of the C=C, and second, the electron pair Aziridine adduct. Aziridine adduct onto the graphene sp2
of carbene (HOMO) interacts with the π* antibonding orbital carbon network is usually achieved via a nitrene intermedi-
(LUMO) of the C=C.8,9 ate, which is structurally related to a carbene molecule. This
Grafting of dichlorocarbene on graphene, by utilizing a nitrene intermediate is generally obtained through thermal or
mixture of chloroform with sodium hydroxide and triethyl- photo-decomposition of an azide group, and is subsequently
benzylammonium chloride as a phase transfer catalyst, was conjugated to the sp2 carbon atoms of graphene via cycloaddi-
demonstrated by Pumera and coworkers (Figure 12.5).10 They tion. The decomposition is driven by the release of a gaseous
established the formation of the hybrid graphene material by nitrogen molecule, which is the most stable leaving group.
X-ray photoelectron spectroscopy, whereby the C1s spectrum This results in the generation of a highly reactive singlet
(a) (b) (c) (d) (f )
–N2 Si ISC Si
Si Si
N N– Δ :N : N. ·
:
N+ N
N– N⊕
Si
:N
(e)
FIGURE 12.6 (a, b) Resonance modes of azidotrimethylsilane; (c, d) thermally generated nitrene radicals; (e) nitrene adsorbed graphene;
(f) covalently bound stretched graphene model to describe interface layer of epitaxial graphene (black circles: carbon atoms at interface
layer, dark gray circles: Si atoms, light gray circles: C atoms). (Adapted from Choi, J. et al. 2009. J. Phys. Chem. C 113:9433–9435.)
nitrene which subsequently conjugates to the sp2 carbon atoms and perfluoroalkyl chains were grafted on exfoliated gra-
of graphene via a cycloaddition mechanism. phene to improve the solubility of the graphene hybrid mate-
Using this reaction, the addition of aziridine adducts onto rials in aqueous and organic solvents. The resulting graphene
graphene sp2 carbon network was first illustrated by Kim and hybrid materials modified with alkyl, perfluoroalkyl chains,
coworkers (Figure 12.6).11 They reported the addition of azi- and ethylene oxide showed high solubility–dispersibility in
dotrimethylsilane onto an epitaxial graphene with a reactive o-dichlorobenzene or aqueous portion.
nitrene radical intermediate generated by thermal treatment. In addition, Borron and Kostarelos also exhibited the
The extent of functionalization had a profound impact on cycloaddition mechanism of an amino acid side chain, phe-
varying the band gaps of the graphene hybrid materials. nylalanine, via a [2 + 1] cycloaddition (Figure 12.8).13,14 The
The addition of aziridine adducts through thermal pho- graphene sheets were reacted with Boc-protected azidophenyl-
tochemical activation techniques were also highlighted by alanine. More importantly, the phenylalanine chains provided
Yan and coworkers (Figure 12.7).12 Derivatives of perfluoro- additional anchors for further functionalizations with the –
phenylazide moieties containing long alkyl, ethylene oxide, COOH and –NH2 arms, suggesting that the functionalizations
Graphene + PFPA 1
Δ or hv DCB
F F
F F O
O N
N N
N F F H
H F F
F F O
N3 N PFPA 1
F H
F
F
F O
O O O PFPA 2
N3 O OH
F
F
F
F O
F FF F F F F F
O F PFPA 3
N3
F F FF F F F F F F F
F
FIGURE 12.7 Functionalization of graphene with perfluorophenylazide moieties via nitrene addition. Reaction mixture of PFPA1, PFPA2,
and PFPA3 in DCB. (Adapted from Liu, L.-H., Lerner, M. M. and Yan, M. 2010. Nano Lett. 10:3754–3756.)
R 12.1.3.3 [3 + 2] Cycloaddition

H O N The [3 + 2] cycloaddition, known as the 1,3-dipolar cycload-
Boc N OH + N R dition, has been widely applied to functionalize fullerenes,
Δ, ODCB, and this chemical functionalization of graphene toward a five-
N3 4d N
R atom ring occurs through a six-electron cycloaddition between
Exfoliated R Phe-N-μG (1) a 1,3-dipole and a dipolarophile (graphene). The 1,3-dipole
μG
applied in many functionalization studies on graphene con-
FIGURE 12.8 Functionalization of graphene with amino acids
sists of unstable azomethine ylides obtained from a decarbox-
(phenylalanine moieties) via nitrene addition. (Adapted from Strom, ylation reaction of N-methylgylcine (sarcosine) and a carbonyl
T. A. et al. 2010. Chem. Commun. 46:4097–4099.) species. The decarboxylation method is very convenient in
terms of reaction conditions. The [3 + 2] cycloaddition prod-
uct is ultimately obtained via the reaction between azomethine
occurred not only at the edge planes, but on the basal planes ylide and graphene sp2 carbon network, which was first demon-
as well. strated by Trapalis and coworkers via a 1,3-dipolar cycloaddi-
tion reaction (Figure 12.10).16 The thermo-gravimetric analysis
revealed an extent of functionalization at one functional group
The formation of a four-electron cycloaddition on graphene per 40 carbon atoms. The graphene hybrid material was highly
sp2 carbon network goes through an aryne or benzyne inter- dispersed in ethanol solution without aggregation.
mediate with the elimination–addition mechanism.1 The for- Prato and coworkers highlighted that the 1,3-dipolar cyclo-
mation of arynes typically requires harsh conditions such as addition not only occurred at the edges of graphene, but also
strong base and metals, but a fluoride-induced decomposition on the basal planes.17 The group illustrated that the attached
methodology on an o-trimethylsilyl-phenyl triflate provides a long alkyl chains were deprotected to expose free amino
mild condition. The presence of a fluoride ion induced a desi- moieties (Figure 12.11). The thermo-gravimetric analysis
lylation step (driven by the formation of a strong F–Si bond) highlighted an extent of one functional group per 128 car-
to provide a carbanion with the sp2 orbital in the plane of the bon atoms. Further gold nanorods were incorporated onto the
ring. This proceeded by an elimination of the triflate group to functionalized graphene using the 1, 3-dipolar cycloaddition
give benzyne, which is electrophilic. Subsequent nucleophilic of azomethine ylides. The inserted amino groups selectively
attack by the C=C of graphene sp2 carbon network resulted in bound to gold nanorods, allowing to highlight the active sites
the [2 + 2] cycloaddition. created onto the graphene sheets, thus constructing the ver-
The modification of graphene based on the benzyne satile platforms and achieving a variety of biological tasks
cycloaddition strategy using 2-(trimethylsilyl)aryl triate as a as drug carriers, photothermal agents, imaging agents, and
precursor was demonstrated by Ma and coworkers (Figure radio-sensitizers.
12.9).15 Three derivatives of the benzyne adduct, mainly
with hydrogen (H-hybrid), fluoride (F-hybrid), or methyl Graphene
(Me-hybrid) groups, were covalently attached to the graphene
carbon network. The reaction resulted in the formation of a CHO
four-membered ring that connected the aromatic arene rings
to the graphene surface. The thermo-gravimetric analysis CH3NH2CH2COOH
provided solid evidence on the degree of functionalization OH DMF, Reflux 96 h
that was estimated to be one functional group per 17 carbon OH
atoms. The graphene hybrid materials showed stable disper-
sions in solvents, such as chloroform, 1,2-dichlorobenzene,
and water. Additionally, the aryl rings can be substituted by
several organic groups (i.e., methyl and fluoro), thereby giving HO
CH3 OH CH3 OH
a variety of interesting derivatives.
N OH
CH3 N
OH
N OH
R R′ R R′ R R′
H-AG; R=H, R′=H

TfO TMS CsF, CH3CN
F-AG; R=F, R′=F
45°C, 24h Me-AG; R=Me, R′=H
Graphene
R R′ Graphene-f-OH
FIGURE 12.9 Functionalization of graphene via benzyne cyclo- FIGURE 12.10 Functionalization of graphene via 1,3-dipolar
addition. (Adapted from Zhong, X. et al. 2010. Chem. Commun. cycloaddition of azomethine ylide. (Adapted from Georgakilas, V.
46:7340–7342.) et al. 2010. Chem. Commun. 46:1766–1768.)
NHBoc NH3+Cl–
NHBoc NH3+Cl–
O O
O O
O O
O O
N N
N N
HCl
FIGURE 12.11 Deprotection of amine groups to expose free amino moieties. (Adapted from Quintana, M. et al. 2010. ACS Nano
4:3527–3533.)
n
yri
r ph
N Po
ODCB, 160°C Graphene

Graphene + CH3NHCH2CO2H + Porphyrin CHO
Porphyrin N N
CHO =
N Pd N
NH HN or
N N
CHO CHO
TPP-CHO PdTPP-CHO
FIGURE 12.12 Functionalization of graphene with TPP-CHO and PdTPP-CHO. (Adapted from Zhang, X. et al. 2011. Chem. Eur. J.
17:8957–8964.)
The 1,3-dipolar cycloaddition method was further exploited On the basis of the Diels–Alder reaction, the versatility
in the incorporation of porphyrin moieties (tetraphenylporphy- and dual nature of graphene was demonstrated by Haddon
rin (TPP) and palladium tetraphenylporphyrin (PdTPP)) onto and coworkers, whereby graphene acted as both a diene and
graphene sheets by Feringa and coworkers (Figure 12.12).18 dienophile (Figure 12.13).19,20 For graphene as diene, tetra-
The thermo-gravimetric analysis estimated a degree of func- cyanoethylene and maleic anhydride have been introduced as
tionalization of one TPP/PdTTP group per 240 carbon atoms. dienophile. The results indicated the effects of graphene type
The graphene hybrid materials displayed quenching effects and temperature on the success and reversibility of the cyclo-
based on fluorescence spectroscopy analyses, which indicated addition. Subsequent application of graphene as dienophile
possible energy- or electron-transfer processes between the was also demonstrated by cycloaddition with 9-methylanthra-
graphene and the covalently attached porphyrin molecules. cene and 2,3-dimethoxy-1,3-butadiene.

12.1.4 Rearrangement
The [4 + 2] cycloaddition, known as the Diels–Alder cyclo-
addition, is the most powerful and elegant of reactions in The Claisen rearrangement is known as sigmatropic rear-
organic chemistry. The reaction proceeds in a single step via rangement, involving a rearrangement of allylic aryl ethers
heat treatment to give a six-membered ring, involving the to o-allylphenols upon tautomerization. The rearrangement
interaction between a conjugated diene (electron rich) and a undergoes a concerted pericyclic [3,3]-sigmatropic rear-
dienophile (electron deficient). The reaction proceeds only rangement mechanism, similar to that of cycloaddition. The
when the diene is locked in an s-cis conformation. As for rearrangement proceeds spontaneously under heat treatment.
the dienophile, the reaction would proceed favorably in the Several variations of the Claisen rearrangement, such as
presence of conjugation with an electron withdrawing group Ireland–Claisen and Eschenmoser–Claisen, involve different
(phenyl group, chlorine atom, etc.).1 Because of the good sym- precursors in the same mechanism.
metry and overlap between the HOMO of diene and LUMO of Swager and coworkers achieved the modification of the gra-
dienophile, the reaction proceeds to furnish ring formations. phene derivative via the Eschenmoser–Claisen rearrangement
NC CN Dual nature of graphene:

NC CN diene and dienophile H2
NC
NC NC , 0°C C OMe
NC RT MeO , 12 C OMe
10 MeO H2
0°C
°C
170
Me
O °C
O , 120
O , 13
O 160 0°C Me
°C Diene Graphene Dienophile °C
150
O resonance hybrid
O
FIGURE 12.13 Graphene as diene-dienophile and their respective reactivities. (Adapted from Sarkar, S., Bekyarova, E. and Haddon, R. C.
2012. Acc. Chem. Res. 45:673–682.)
R R R
HO O O O
Vinyl ether Claisen Graphene

Graphite oxide
formation rearrangement derivative
FIGURE 12.14 Functionalization toward a graphene derivative via Eschenmoser–Claisen rearrangement on graphite oxide. (Adapted
from Collins, W. R. et al. 2011. Angew. Chem. Int. Ed. 50:8848–8852.)
on graphite oxide (Figure 12.14).21 In this reaction, allylic graphitic chemical structure and thus secures the physical
alcohol on graphite oxide was converted to γ,δ-unsaturated and chemical properties of graphitic materials. Furthermore,
N,N-dimethylamides with a vinyl transfer reagent N,N- they are easy to achieve over the entire graphene surface and
dimethylacetamide dimethyl acetal. The evolution of the C=O reversible in some cases. When the graphene’s electrical con-
stretching mode at 1635 cm–1 in the Fourier transform infra- ductivity and large surface are required, noncovalent modifi-
red spectroscopy suggested amide incorporation, indicating cation methods are typically preferred.
a successful rearrangement process. The thermo-gravimetric
analysis supported the functionalization whereby a new ther-
mal decomposition at higher temperature (290–310°C) sug- 12.2.1 van der Waals Force
gested the occurrence of a new covalent functionalization that Assembling molecules on the graphene basal plane through
occurs in this reaction but not a simply physical adsorption van der Waals interactions represents an operationally simple
process. Moreover, the X-ray photoelectron spectroscopy method to introduce arbitrary functionality to the graphene
highlighted the presence of an N1s up to 3.1%. Further saponi- surface and to influence its charge carrier density. On the
fication treatment on the graphene hybrid material converted basis of this, it has become an increasingly desirable approach
the amide moieties to carboxylate moieties, showing a dra- to interface functional molecules or polymers to graphene
matical increase in the solubility of the graphene derivative (Figure 12.15).22 For example, graphene–polymer were pre-
in aqueous environments without the need for cosolvents or pared based on simple physical mixing, where the graphene
additives, and further expanding the scope and application sheets worked as conductive fillers to afford nanocomposites
of the transformation for Claisen rearrangements on the gra- with very low percolation concentrations.23,24 Percolation
phene derivatives. concentration refers to an onset concentration at which the
conductive filler forms an infinite network of connected paths
12.2 NONCOVALENT MODIFICATION ON through the insulating matrix and electrical conductivity of
composite materials. Solvent processing, in situ polymeriza-
GRAPHENE CARBON NETWORK
tion, and melt processing are the generally adopted methods
Noncovalent modification of graphene is based on van der to achieve physical mixed graphene composites as summa-
Waals forces, electrostatic interaction, hydrogen bonding, rized by Verdejo and coworkers.25 The solvent processing
coordination bonds, or the π–π interactions (Table 12.2). method is better for producing quite homogeneous graphene–
Noncovalent modification minimizes the damage to the polymer composites.26,27 In situ polymerization method can
TABLE 12.2
Noncovalent Nodification Techniques on Graphene Carbon Network
Modification Techniques Chemical Reaction Mechanism Chemical Reaction Agent References
van der Waals force Intermolecular forces Polymer [22,23,24]
Ionic liquids [30,31]
Trimethyl-indium, trimethylgallium [32]
Electrostatic interaction Self-assembly (electric charge) Lactoferrin [33]
Chitosan [33]
Peptide [35]
Hydrogen bonding C–OH •••P Poly(vinyl alcohol) [36]
Polymers (epoxy, poly(acrylonitrile) and polyaniline) [37,38,39]
C–OH •••DNA DNA [40]
Coordination bonds C‒M Metals [41]
π–π stacking interaction Aromatic rings interaction Pyrene terminated poly(N-isopropylacrylamide) [44,48]
Oligoethylene glycol acrylate and diethylene glycol ethyl ether acrylate [49]
Pyrene-terminated poly(2-N,N′-(dimethyl amino ethyl acrylate) and [50]
polyacrylic acid
Poly (acrylic acid) [51]
visible under a fluorescence microscope by quenching the

emission from a dye coating.29 The dye can be conveniently
removed afterwards by rinsing without disrupting the sheets.
Traditional imaging techniques for graphene-based sheets rely
on the use of special substrates. By contrast, the fluorescence
quenching mechanism is no longer limited by the type of sub-
van der Waals strate, therefore offering unprecedented imaging flexibility for
interactions graphene-based materials. Hu and coworkers demonstrated
that collective van der Waals forces between ionic liquids and
graphene are able to describe both the short-ranged cation−π
interaction and the long-ranged dispersion interaction, and this
microscopic interaction drives two graphene plates trapped
in their metastable state, while two graphene plates easily
self-assemble into graphite in water.30 Inspired by the suc-
cessful dispersion of graphene in ionic liquids, Ahmad et al.,
reported that the gels formed by ionic liquids and graphene
FIGURE 12.15 Noncovalent functionalization of graphene using have been studied as potential electrolytes for dye-sensitized
polymers or molecule adsorbates via van der Waals force. (Adapted solar cells.31 Li and coworkers have prepared high-density
from Jason, A. M. and William, R. D. 2013. J. Phys. Chem. Lett. vertical InAs nanowire arrays and self-organized core−shell
4:2649–2657.) heterostructured InAs−InGaAs NWs through a seed-free van
der Waals Epitaxy approach on graphene films. Such hetero-
be a good alternative for generating homogeneous composites geneous nanostructures can be exploited for the fabrication of
with much stronger interaction present between graphene and a novel class of functional hybrid materials for use in flexible
the polymer matrix. Direct chemical modification of gra- optoelectronic device applications.32
phene by photopolymerization was also performed to afford
polystyrene brushes on graphene. As the polymerization took
12.2.2 Electrostatic Interaction
place at the graphene defects, its conjugated structure was not
damaged; therefore, its electronic properties were maximally Electrostatic interaction can be used as a convenient tool for
protected.28 Melt processing requires milder conditions and the preparation of graphene hybrid materials via self-assem-
can be applied to a more extensive range of polymers. bly. Multifunctional graphene composites were prepared
Stankovich and coworkers have been pioneers in the prepa- through electrostatic interaction between the biocompatible
ration and modification of graphene.23 They have successfully ingredients.33 Different functionalities were added to gra-
prepared graphene nanosheets using the Hummers method phene by some anchoring materials such as lactoferrin (NLf),
and modified them with polystyrene with a solvent processing chitosan (Ch), and combinations thereof. Anchoring of Ch
method to achieve a percolation threshold of 0.1 volume percent. and NLf enhances the antibacterial property of graphene.
Van der Waals force interaction was also employed to The addition of Ch to graphene not only helped in forming
prepare graphene-based sheets that can be made highly stable dispersions but also helped in fabricating large-area
films. Functionalities, such as photoluminescence were added between graphene and metal can be obviously enhanced via
to Ch–graphene composites by anchoring Au@NLf on it. the modification of graphene–metal interface using the oxy-
The composites showed stable luminescence in the presence gen intercalation. When the metal ruthenium surface beneath
of various metal ions in the solid state. Another study that graphene could be selectively oxidized, the metal–carbon cou-
employs electrostatic interactions was conducted by Choi et pling was boosted, and the characteristic Dirac cones of iso-
al. where the residual carboxylic acid groups on graphene and lated single-layer graphene were correspondingly restored.42
amino groups on polystyrene chains were used to form a gra-
phene composite material.34
12.2.5 π–π Stacking Interaction
Electrostatic interactions were also used to prepare a novel
hybrid material of negative-charged sulfonated graphene and π−π stacking interactions are of special interest due to
positive-charged poly(3,4-ethyldioxythiophene)-based gra- the extended π orbitals of graphene, which usually occur
phene, which showed excellent transparency, electrical con- between two relatively large nonpolar aromatic rings with
ductivity, and good flexibility, together with high thermal the overlapping π orbitals.43 These are comparable to cova-
stability and ease of processing in both water and organic lent attachment in strength, and hence provide more stable
solvents. Kim and coworkers have also used electrostatic alternatives to the hydrogen bonding, electrostatic bond-
attraction to prepare a graphene–peptide hybrid by assembly ing and coordination bonding interactions discussed above.
into core–shell nanowires with a peptide core and a graphene Furthermore, π–π stacking modification does not disrupt the
shell, which exhibited the electroconductivity through their conjugation of the graphene sheets, thus it preserves the elec-
continuous graphene shell. The thermal calcinations of pep- tronic properties of graphene. Hence, such a modification
tide cores led to the formation of highly entangled networks strategy is envisaged to have enormous potential. Pyrene
with the hollow graphene shell, demonstrating its remarkable moiety is a π-orbital-rich group that easily forms strong π–π
performance as a supercapacitor electrode.35 stacking interactions with other polyaromatic materials.44
Pyrene-tethered precursors have been successfully prepared
and attached onto aromatic macromolecules such as carbon
12.2.3 Hydrogen Bonding
nanotubes, graphene oxide, and fullerene.45–47 A number of
Hydrogen bonding is ubiquitous in the biological and small π-orbital-rich aromatic molecules such as pentacene,
chemical world. An individual hydrogen bond is very weak benzopyrene, 1-pyrenebutyrate, doxorubicin, and dopamine
(2–8 kcal mol–1); however, multiple hydrogen bonds will have also been attached previously onto graphene sheets
afford strong interactions such as DNA hybridization. Liang using the π–π stacking approach.
and coworkers prepared the poly(vinyl alcohol) (PVA) nano- In recent studies, Liu and coworkers synthesized ther-
composites with graphene by dispersing graphene into PVA moresponsive graphene composites. They first synthesized
at molecular level.36 The hydrogen-bonding interaction a well-defined thermoresponsive pyrene-terminated poly(N-
between the residual oxygen-containing groups of graphene isopropylacrylamide) (PNIPAAm) using reversible addition
and hydroxyl groups of the PVA endow the composite with fragmentation chain transfer (RAFT) polymerization, fol-
increased tensile strength and Young’s modulus. Graphene– lowed by attachment onto the basal plane of graphene sheets
polymer composites with other hydrophilic polymers, epoxy, via π–π stacking interactions (Figure 12.16).48 The lower
poly(acrylonitrile), and polyaniline exhibited an extraordi- critical solution temperature (LCST) of pyrene-terminated
nary high increase in modulus or glass transition tempera- PNIPAAm was measured to be 33°C. However, when the
ture, attributed to hydrogen bonding interaction.37–39 On the pyrene-functional polymer was attached to graphene, the
other hand, nanomaterials for DNA adsorption are useful for resultant graphene composites were also thermoresponsive in
sequence-specific DNA detection. Current materials for DNA aqueous solutions, but with a lower LCST of 24°C.
adsorption employ electrostatic attraction, hydrophobic inter- Furthermore, thermal-responsive random copolymers of
action, or π−π stacking, none of which can achieve sequence oligoethylene glycol acrylate and diethylene glycol ethyl ether
specificity. Moreover, most materials reported so far have acrylate with controllable LCST from 31°C to 82°C were also
rather poor specificity for DNA sequences. Specificity might synthesized using the RAFT method.48 These thermorespon-
be improved by involving hydrogen bonding. Recently, Liu sive copolymers were further utilized to modify graphene via
and coworkers designed such materials that could be used for π–π stacking interaction to afford graphene–polymer com-
sequence-specific DNA detection based on hydrogen-bonding posites with controllable but reduced LCSTs from 22°C to
interaction.40 Therefore, functionalized graphene and DNA 72°C compared to the random polymers (Figure 12.17).49
via hydrogen bonding will probably be a good starting point Using π–π stacking interactions, Liu and coworkers also
with the enormous potential for DNA analysis. prepared pH-sensitive graphene composites by functionaliza-
tion of graphene with a pyrene-terminated positive-charged
polymer, poly(2-N,N′-(dimethyl amino ethyl acrylate)
12.2.4 Coordination Bonds
(PDMAEA), and a negatively charged polymer, polyacrylic
Coordination bonds interaction, particularly with the tran- acid (PAA).50 The pyrene-terminated PDMAEA and PAA
sition metals coordination, is a well-studied chemical pro- were synthesized using a pyrene-terminated RAFT agent.
cess.41 Sutter and coworkers demonstrated that the interaction These charged graphene–polymer composites were found to
S be an efficient drug carrier for pH-controlled release of drug

S OH DCC, DMAP
+ HO S in the neutral intestine environment. In addition, the RAFT
S S COOH THF
[1] OH mechanism has also been utilized to synthesize π-orbital-rich
O polymers for modification of graphene to afford the water-
O
HO S S S
S O DCC, DMAP soluble graphene composites via π–π stacking interactions.
[2] S
O
O S S S NIPAAm
S O 12.3 CHEMICAL DOPING MODIFICATIONS
O S AIBN/70°C
[3] ON GRAPHENE CARBON NETWORK
O
O S S S
Chemical doping is a common approach to control semicon-
S O n ducting properties in the conventional semiconductor commu-
O O S
[4]
HN
nity. In case of graphene, theoretic studies show that doping
can effectively modulate its electrical properties, leading to
a metal–semiconductor transition. Recently, increasing atten-
tion has been paid here due to its fascinating properties and
Graphene widespread potential applications such as super-conduction,
[4] ferromagnetism, etc. Here, we focus on three main means to
dope graphene, which are interstitial doping, substitutional
doping and selective doping (Table 12.3).
12.3.1 Interstital Doping
FIGURE 12.16 Synthesis of pyrene-terminated PNIPAAm using In the case of interstitial doping, dopant atoms or molecules
a pyrene-functional RAFT agent, and the subsequent attachment of are introduced as adatoms or admolecules on the surface of
the polymer to graphene. (Adapted from Liu, J. et al. 2010. J. Polym. graphene. In 2004, Geim et al., demonstrated the possibility
Sci., Part A: Polym. Chem. 48:425–433.) of doping graphene by absorbed water vapor, NO2 or NH3
molecules.52 Since then, many efforts have demonstrated the
demonstrate phase transfer behavior between aqueous and interstitial doping of graphene by gas,53 metal,54–58 organic
organic media at different pH values. These two oppositely molecules,59,60 and so on. This is a simple route for doping,
charged graphene–polymer composites can, via self-assem- which just needs to deposit the dopants on graphene or expose
bly, give layer-by-layer structures (Figure 12.18). Furthermore, graphene in the dopant vapor. Electron donors such as NH3,
a pH-sensitive, mechanically strong, and thermally stable CO, ethanol, and alkali metals transfer electrons to graphene,
graphene/poly (acrylic acid) (graphene/PAA) hydrogel was and shift up the Femi level, resulting in n-type behavior of
prepared via RAFT polymerizations (Figure 12.19).51 This graphene, while electron acceptors like NO2, H2O, and Au
pH-sensitive porous graphene/PAA hydrogel was shown to receive electrons from graphene and shift down the Femi
O
O S S S OEG-A/DEG-A
S O
O S AIBN/70°C
O Polymer chain
O S S S
S O m n
O
O OO O S
Pyrene
O O
8 2
Graphene
sheet
FIGURE 12.17 Synthesis of graphene–polymer nanocomposites via π–π stacking interactions. (Adapted from Liu, J. Q. 2011. Aust. J.
Chem. 64:1414–1414.)
O
DMAEA or AA
O S S S
S O
AIBN/65–70°C
O S
O
S S
O S S O n
O S
R
Graphene sheet
+ – – + – +
+ – –
+ +
– Self-assembly
+
+ or – –
+ – – –+
+ – + + +
+
R= N O O R = HO O
FIGURE 12.18 Synthesis of pH-sensitive pyrene–polymer composites via π–π stacking interactions for the self-assembly of functional-
ized graphene into layered structures. (Adapted from Liu, J. et al. 2010. Langmuir 26:10068–10075.)
COOH COOH
HOOC N2BF4 RAFT-OH
RT, 7h
COOH
COOH COOH
Graphene
Graphene-COOH
S CN O O CN S
S O O O
S O S S
O O
S CN O
O O CN S
S S O O
O O S S
O
Graphene-RAFT
O
OH O
O
O O
Monomer Cross-linking agent
Graphene/PAA hydrogels
FIGURE 12.19 Synthesis of pH-sensitive graphene/PAA hydrogels via π–π stacking interactions. (Adapted from Liu, J. et al. 2014. Eur.
Polym. J. 50:9–17.)
TABLE 12.3
Doping Modification Techniques on Graphene Carbon Network
Modification Techniques Chemical Doping Method Chemical Doping Species References
Interstital doping Adsorbate-induced doping H2O2, NO2, N2O4, NH3 [52,53]
Wet-chemical method Potassium [54,55]
Chemical vapor deposition Potassium, rubidium [56,57]
Microwave-assisted method Platinum [58]
π–π interactions Aromatic molecules [59,60]
(1,5-naphthalenediamine,
9,10-dimethylanthracene,
9,10-dibromoanthracene
Tetrasodium 1,3,6,8-pyrenetetrasulfonic acid
9-Cyanophenanthrene)
Electron beam evaporator Bismuth, antimony [61]
Annealing method Gold [61]
Substitutional doping Chemical vapor deposition NH3 [64,65]
Solvothermal method Lithium nitride [66]
Annealing method NH3, polyaniline, or polypyrrole [67]
Hydrothermal treating NH3 [68]
N plasma treatment method Nitrogen plasma [69]
Annealing method B2O3 [71]
Thermal treating method H2S [73]
Annealing method Benzyl disulfide [74]
Annealing method Triphenylphosphine [75]
Two-step solution process Borane tetrahydrofuran and hydrazine monohydrate [78]
Annealing method NH3 and H3BO3 [79]
Selective doping Patterned doping Polymethylmethacrylate [80]
Depositing method Potassium [81]
level, resulting in p-type behavior of graphene. These doping Owing to the atomic thickness, almost all conducting
effects can be clearly observed by angle-resolved photoemis- channels are exposed to surface adsorbates, thus, the inter-
sion spectroscopy (Figure 12.20),61 and have possibility of stitial doping effect is highly sensitive, which can be used in
doping graphene by electrons or holes in high concentration inorganic anions or molecule detection. Xu and coworkers
without deterioration of its mobility. used potassium-doped graphene as an advanced electrode
material for the detection of nitrite, exhibiting an extremely
low detection limit and high sensitivity. On the basis of this, it
can be further applied in the determination of nitrite released
0.2
0 Bi atoms/u.c. 0.28 Bi atoms/u.c. from living cells.54 In addition, Robinson and coworkers used
0.0 reduced graphene oxide for high-performance molecule sen-
Initial state energy E (eV)
sors, which showed high sensitivity as low as parts-per-billion

–0.2
ED
level with significantly reduced noise levels over carbon nano-
–0.4 tube-based sensors, while the response and recovery charac-
–0.6 teristics of the conductance could be tailored by adjusting the
reduction process.62 Moreover, graphene is an exceptionally
–0.8
low-noise material electronically, which makes it capable of
–1.0 detecting the absorption of individual molecules.63
–0.2 –0.1 0.0 0.1 0.2 –0.2 –0.1 0.0 0.1 0.2 12.3.2 Substitutional Doping
Wave vector k (1/Å) Wave vector k (1/Å)
In the case of substitutional doping, the dopant atoms replace
the lattice carbon atoms, and form sp2 bonds in the lattice
FIGURE 12.20 Experimental band structure (angle-resolved pho-
toemission spectroscopy) of graphene before and after interstitial of graphene, thus, it should be more stable and reliable for
doping with bismuth atoms. After doping, the Dirac point clearly applications compared with the interstitial doping. N-doped
shifts toward the Fermi level. (Adapted from Gierz, I. et al. 2008. graphene has been the most intensively studied doping model.
Nano Lett. 8:4603–4607.) N is next to C in the periodic table of chemical elements. As
Pyrridinic N
Pyridinic N N
N N Anili nia
ne
mmo N
N N N A N
0°C
55
N-RG-O 550°C
N
PANI/RG-O
Polymerization; Am
carbonization ole mo
P yrr 85
nia Pyrridinic /
Pyrrolic N 0° graphite N
C
N
Ammonia
N
1000°C
N
N N
N N
N
N N-RG-O 850°C
Ppy/RG-O
Polymerization;
carbonization
Graphitic/
Pyridinic N
N
N N N-RG-O 1000°C
N N
FIGURE 12.21 Schematic diagram for the preparation of N-doped graphene via substitutional doping with different N states. (Adapted
from Lai, L. et al. 2012. Energy Environ. Sci. 5:7936–7942.)
such, the introduction of N into graphene sheets could modify were also developed,70−79 and the resultant products would be
its local electronic structures. The N heteroatom could be useful for various practical applications of the substitutional
introduced directly during the graphene growth, by chemical graphene.
vapor deposition method using NH3 as the nitrogen source
and CH4 as the carbon source.64 It has been well established 12.3.3 Selective Doping
that the incorporation of N atoms into the graphene matrix
can lead to three main types of N formats, including graphitic In the case of selective doping, doping can be performed on
N with direct substitution structure, and pyridinic N and pyr- both a selective area and selective layers of graphene. By elec-
rolic N structures.65 Graphitic N means the doping N atom tron beam lithography, Lohmann and coworkers realized the
is combined into a hexagonal ring. Pyridinic and pyrrolic N patterned doping of graphene (Figure 12.22).80 They patterned
donate one and two p electrons to the p system, forming sp2
and sp3 hybridized bonds, respectively.66 Moreover, Ruoff
and coworkers obtained N-doped graphene with different N
S
doping formats via annealing GO with different N sources,67
and the schematic diagram is shown in Figure 12.21. Through
annealing of GO with ammonia, polyaniline, or polypyr- D
1
role, the content of graphitic N species, pyridinic N centers,
and pyrrolic N moieties could be tuned, making this process
1
D
2
2
3
more suitable for practical applications. Importantly, electri-

4
5
cal measurement showed an n-type behavior with mobilities

S
in 200–450 cm2 V−1 s−1 after nitrogen doping, thus, the sub- 4 μm
6
3
stitutional doping is an effective means to modulate the elec-
trical properties of graphene. In general, N-doped graphene
has shown many superior features in metal-free catalysis
FIGURE 12.22 A contacted single-layer graphene with a pat-
including high efficiency, high stability, recyclability, and terned layer of PMMA on top. The white dashed line highlights the
biocompatibility,68,69 making it highly attractive for further edges of the graphene flake. Two windows in the PMMA allow us
investigations. Furthermore, with the development of doping to selectively dope the flake underneath in these regions, resulting
technology, many other metal-free catalysts, such as B-, S-, in the three independent p–n junctions. (Adapted from Lohmann,
P-, Si-doped graphene, as well as B and N co-doped graphene, T., Klitzing, K. V. and Smet, J. H. 2009. Nano Lett. 9:1973–1979.)
polymethylmethacrylate (PMMA) films on graphene by elec- 10. Chua, C. K., Ambrosi, A. and Pumera, M. 2012. Introducing
tron beam lithography, which acted as the doping mask, thus dichlorocarbene in graphene. Chem. Commun. 48:5376–5378.
protecting the parts of the graphene sample, while the uncov- 11. Choi, J., Kim, K. -J., Kim, B., Lee, H. and Kim, S. 2009.
Covalent functionalization of epitaxial graphene by azidotri-
ered parts were exposed to dopants, resulting in the p–n junc-
methylsilane. J. Phys. Chem. C 113:9433–9435.
tions of graphene. Ohta and coworkers realized the selective 12. Liu, L.-H., Lerner, M. M. and Yan, M. 2010. Derivitization of
doping of the surface layer of graphene by depositing potas- pristine graphene with well-defined chemical functionalities.
sium atoms onto the surface of the bilayer graphene grown Nano Lett. 10:3754–3756.
on a SiC substrate.81 The potassium atoms donated their lone 13. Strom, T. A., Dillon, E. P., Hamilton, C. E. and Barron, A.
valence electrons to the surface layer, forming a dipole. This R. 2010. Nitrene addition to exfoliated graphene: A one-step
dipole field together with interface dipole field acted as the route to highly functionalized graphene. Chem. Commun.
46:4097–4099.
symmetry-breaking factor, which controlled the presence or
14. Servant, A., Bianco, A., Prato, M. and Kostarelos, K. 2014.
absence of the band gap. Therefore, by selectively doping the Graphene for multi-functional synthetic biology: The last
surface layer, the magnitude of the band gap of the bilayer “zeitgeist” in nanomedicine. Bioorg. Med. Chem. Lett.
graphene could be manipulated close to the Dirac crossing. 24:1638–1649.
15. Zhong, X., Jin, J., Li, S. W., Niu, Z. Y., Hu, W. Q., Li, R. and
Ma, J. T. 2010. Aryne cycloaddition: Highly efficient chemical
12.4 CONCLUSIONS modification of graphene. Chem. Commun. 46:7340–7342.
16. Georgakilas, V., Bourlinos, A. B., Zboril, R., Steriotis, T. A.,
In summary, chemical modifications of graphene can be
Dallas, P., Stubos, A. K. and Trapalis, C. 2010. Organic func-
achieved via covalent, noncovalent, or chemical doping meth- tionalisation of graphenes. Chem. Commun. 46:1766–1768.
odologies. Among them, the covalent modification can be 17. Quintana, M., Spyrou, K., Grzelczak, M., Browne, W. R.,
used to form graphene composites with strong affinity; the Rudolf, P. and Prato, M. 2010. Functionalization of graphene
noncovalent modification will maximally protect graphene’s via 1,3-dipolar cycloaddition. ACS Nano 4:3527–3533.
intrinsic properties; chemical doping modification is a lead- 18. Zhang, X., Hou, L., Cnossen, A., Coleman, A. C., Ivashenko,
ing strategy to tailor the electronic properties of graphene O., Rudolf, P., van Wees, B. J., Browne, W. R. and Feringa,
B. L. 2011. One-pot functionalization of graphene with
composites in microelectronics. More importantly, chemical
porphyrin through cycloaddition reactions. Chem. Eur. J.
modifications of graphene would help tackle the problems 17:8957–8964.
regarding its solubility and stability, tunable electric, thermal, 19. Sarkar, S., Bekyarova, E., Niyogi, S. and Haddon, R. C. 2011.
and mechanical properties, and enhanced catalytic and bio- Diels-Alder chemistry of graphite and graphene: Graphene as
logical properties. These properties make them desired mate- diene and dienophile. J. Am. Chem. Soc. 133:3324–3327.
rials for the fabrication of intelligent sensors, green energy 20. Sarkar, S., Bekyarova, E. and Haddon, R. C. 2012. Chemistry
devices, etc. Modifications of graphene via these methods at the Dirac point: Diels-Alder reactivity of graphene. Acc.
Chem. Res. 45:673–682.
offer extremely excellent arenas for versatile applications.
21. Collins, W. R., Lewandowski, W., Schmois, E., Walish, J.
and Swager, T. M. 2011. Claisen rearrangement of graphite
REFERENCES oxide: A route to covalently functionalized graphenes. Angew.
Chem. Int. Ed. 50:8848–8852.
1. Chua, C. K. and Pumera, M. 2013. Covalent chemistry on gra- 22. Jason, A. M. and William, R. D. 2013. Noncovalent function-
phene. Chem. Soc. Rev. 42:3222–3233. alization of graphene by molecular and polymeric adsorbates.
2. Economopoulos, S. P., Rotas, G., Miyata, Y., Shinohara, H. J. Phys. Chem. Lett. 4:2649–2657.
and Tagmatarchis, N. 2010. Exfoliation and chemical modifi- 23. Stankovich, S., Dikin, D. A., Dommett, G. H. B., Kohlhaas,
cation using microwave irradiation affording highly function- K. M., Zimney, E. J., Stach, E. A., Piner, R. D., Nguyen, S. T.
alized graphene. ACS Nano 4:7499–7507. and Ruoff, R. S. 2006. Graphene-based composite materials.
3. Bosch-Navarro, C., Busolo, F., Coronado, E., Duan, Y., Martí- Nature 442:282–286.
Gastaldo, C. and Prima-Garciaa, H. 2013. Influence of the 24. Noél, A., Faucheu, J., Rieu, M. Viricelle, J. and Shin, C. M.
covalent grafting of organic radicals to graphene on its mag- 2014. Tunable architecture for flexible and highly conduc-
netoresistance. J. Mater. Chem. C 1:4590–4598. tive grapheme-polymer composites. Combust. Sci. Technol.
4. Ran, S. Y., Guo Z. H, Han, L. G. and Fang Z. P. 2014. Effect 95:82–88.
of a Lewis acid catalyst on the performance of HDPE/BFR/ 25. Verdejo, R., Bernal, M. M., Romasanta, L. J. and Lopez-
GNPs composites. Ind. Eng. Chem. Res. 53:4711–4717. Manchado, M. A. 2011. Graphene filled polymer nanocom-
5. Chua, C. K. and Pumera, M. 2012. Friedel-Crafts acylation on posites. J. Mater. Chem. 21:3301–3310.
graphene. Chem. Asian J. 7:1009–1012. 26. Kim, H., Miura, Y. and Macosko, C. W. 2010. Graphene/poly-
6. Yuan, C., Chen, W. and Yan, L. 2012. Amino-grafted gra- urethane nanocomposites for improved gas barrier and electri-
phene as a stable and metal-free solid basic catalyst. J. Mater. cal conductivity. Chem. Mater. 22:3441–3450.
Chem. 22:7456–7460. 27. Kim, H., Kobayashi, S., AbdurRahim, M. A., Zhang, M. L. J.,
7. Clayden, J., Greeves, N., Warren, S. and Wothers, P. 2001. Khusainova, A., Hillmyer, M. A., Abdala, A. A. and Macosko,
Organic Chemistry. Oxford: Oxford University Press. C. W. 2011. Graphene/polyethylene nanocomposites: Effect
8. Pérez, D., Peña, D. and Guitián, E. 2013. Aryne cycloaddition of polyethylene functionalization and blending methods.
reactions in the synthesis of large polycyclic aromatic com- Polymer 52:1837–1846.
pounds. Eur. J. Org. Chem. 2013:5981–6013. 28. Steenackers, M., Gigler, A. M., Zhang, N., Deubel, F., Seifert,
9. Bettinger, H. F. 2006. Addition of carbenes to the sidewalls of M., Hess, L. H., Lim, C. H. Y. X. et al. 2011. Polymer brushes
single-walled carbon nanotubes. Chem. Eur. J. 12:4372–4379. on graphene. J. Am. Chem. Soc. 133:10490–10498.
29. Kim, J., Cote, L. J., Kim, F. and Huang, J. 2009. Visualizing reversible phase-transition behavior, and tribological prop-
graphene based sheets by fluorescence quenching microscopy. erties as lubricant additives. J. Polym. Sci., Part A: Polym.
J. Am. Chem. Soc. 132:260–267. Chem. 46:7225–7237.
30. Zhao, Y. and Hu, Z. 2013. Graphene in ionic liquids: Collective 47. Yuan, W. Z., Yuan, J. Y., Zhou, M. and Pan, C. Y. 2008.
van der waals interaction and hindrance of self-assembly path- Synthesis, characterization, and fluorescence of pyrene-con-
way. J. Phys. Chem. B 117:10540–10547. taining eight-arm star-shaped dendrimer-like copolymer with
31. Ahmad, I., Khan, U. and Gun’ko, Y. K. 2011. Graphene, car- pentaerythritol core. J. Polym. Sci., Part A: Polym. Chem.
bon nanotube and ionic liquid mixtures: Towards new quasi- 46:2788–2798.
solid state electrolytes for dye sensitised solar cells. J. Mater. 48. Liu, J., Yang, W., Tao, L., Li, D., Boyer, C. and Davis, T. P.
Chem. 21:16990–16996. 2010. Thermosensitive graphene nanocomposites formed
32. Mohesni, P. K., Behnam, A., Wood, J. D., English, C. D., using pyrene-terminal polymers made by RAFT polymeriza-
Lyding, J. W., Pop, E. and Li, X. 2013. InxGa1−xAs nanowire tion. J. Polym. Sci., Part A: Polym. Chem. 48:425–433.
growth on graphene: Van der Waals epitaxy induced phase 49. Liu, J. Q. 2011. Imparting polymeric properties to graphene
segregation. Nano Lett. 13:1153–1161. nanosheets by surface modification via π–π stacking. Aust. J.
33. Sreeprasad, T. S., Maliyekkal, M. S., Deepti, K., Chaudhari, Chem. 64:1414–1414.
K., Xavier, P. L. and Pradeep, T. 2011. Transparent, lumines- 50. Liu, J., Tao, L., Yang, W., Li, D., Boyer, C., Wuhrer, R., Braet,
cent, antibacterial and patternable film forming composites F. and Davis, T. P. 2010. Synthesis, characterization, and mul-
of graphene oxide/reduced graphene oxide. ACS Appl. Mater. tilayer assembly of pH sensitive graphene-polymer nanocom-
Interface 3:2643–2654. posites. Langmuir 26:10068–10075.
34. Choi, E.-Y., Han, T. H., Hong, J., Kim, J. E., Lee, S. H., Kim, 51. Liu, J., Cui, L., Kong, N., Barrow, C. J. and Yang, W. 2014.
H. W. and Kim, S. O. 2010. Noncovalent functionalization RAFT controlled synthesis of graphene/polymer hydrogel
of graphene with end-functional polymers. J. Mater. Chem. with enhanced mechanical property for pH-controlled drug
20:1907–1912. release. Eur. Polym. J. 50:9–17.
35. Han, T. H., Lee, W. J., Lee, D. H., Kim, J. E., Choi, E.-Y. and 52. Novoselov, K. S., Geim, A. K., Morozov, S. V., Jiang, D.,
Kim, S. O. 2010. Peptide/graphene hybrid assembly into core/ Zhang, Y., Dubonos, S. V., Grigorieva, I. V. and Firsov, A.
shell nanowires. Adv. Mater. 22:2060–2064. A. 2004. Electric field effect in atomically thin carbon films.
36. Liang, J., Huang, Y., Zhang, L., Wang, Y., Ma, Y., Guo, T. and Science 306:666–669.
Chen, Y. 2009. Molecular-level dispersion of graphene into 53. Wehling, T. O., Novoselov, K. S., Morozov, S. V., Vdovin,
poly(vinyl alcohol) and effective reinforcement of their nano- E. E., Katsnelson, M. I., Geim, A. K. and Lichtenstein, A. I.
composites. Adv. Funct. Mater. 19:2297–2302. 2008. Molecular doping of graphene. Nano Lett. 8:173–177.
37. Rafiee, M. A., Rafiee, J., Wang, Z., Song, H., Yu, Z.-Z. and 54. Li, X. R., Kong, F. Y., Liu, J., Xu, J. J. and Chen, H. Y.
Koratkar, N. 2009. Enhanced mechanical properties of nano- 2012. Synthesis of potassium-modified graphene and its
composites at low graphene content. ACS Nano 3:3884–3890. application in nitrite-selective sensing. Adv. Funct. Mater.
38. Ramanathan, T., Abdala, A. A., Stankovich, S., Dikin, D. A., 22:1981–1988.
Herrera-Alonso, M., Piner, R. D., Adamson, D. H. et al. 2008. 55. Xue, M., Chen, G., Yang, H., Zhu, Y., Wang, D., He, J. and
Functionalized graphene sheets for polymer nanocomposites. Cao, T. 2012. Superconductivity in potassium-doped few-
Nat. Nanotechnol. 3:327–331. Layer graphene. J. Am. Chem. Soc. 134:6536–6539.
39. Wang, H., Hao, Q., Yang, X., Lu, L. and Wang, X. 2011. Effect 56. Bianchi, M., Rienks, E. D. L., Lizzit, S., Baraldi, A., Balog,
of graphene oxide on the properties of its composite with poly- R., Hornekaer, L. and Hofmann, P. 2010. Electron-phonon
aniline. ACS Appl. Mater. Interfaces 2:821–828. coupling in potassium-doped graphene: Angle-resolved pho-
40. Wang, F., Liu, B., Huang, P. J. and Liu J. 2013. Rationally toemission spectroscopy. Phys. Rev. B 81:041403-1–041403-4.
designed nucleobase and nucleotide coordinated nanoparti- 57. Jung, N., Kim, B., Crowther, A. C., Kim, N., Nuckolls, C.
cles for selective DNA adsorption and detection. Anal. Chem. and Brus, L. 2011. Optical reflectivity and Raman scattering
85:12144–12151. in few-layer-thick graphene highly doped by K and Rb. ACS
41. Liu, J. Q., Yang, W. R., Zareie, H. M., Gooding, J. J. and Nano 5:5708−5716.
Davis, T. P. 2009. pH-detachable polymer brushes formed 58. Ullah, K., Ye, S., Zhu, L., Jo, S. B., Jang, W. K., Cho, K. and
using titanium-diol coordination chemistry and living radical Oh, W. 2014. Noble metal doped graphene nanocomposites
polymerization (RAFT). Macromolecules 42:2931–2939. and its study of photocatalytic hydrogen evolution. Solid State
42. Sutter, P., Sadowski, J. T. and Sutter, E. A. 2010. Chemistry Sci. 31:91–98.
under cover: Tuning metal-graphene interaction by reactive 59. Dong, X. C., Fu, D. L., Fang, W. J., Shi, Y. M., Chen, P. and
intercalation. J. Am. Chem. Soc. 132:8175–8179. Li, L. J. 2009. Doping single-later graphenee with aromatic
43. Zhang, X. and Shao, X. 2014. π–π binding ability of differ- molecules. Small 5:1422–1426.
ent carbon nano-materials with aromaticphthalocyanine mol- 60. Dong, X. C., Shi, Y. M., Chen, P., Ling, Q. D. and Huang,
ecules: Comparison between graphene, grapheneoxide and W. 2010. Aromatic molecules doping in single-layer gra-
carbon nanotubes. J. Photoch. Photobio. C 278:69–74. phene probed by Raman spectroscopy and electrostatic force
44. Cao, M., Fu, A., Wang, Z., Liu, J., Kong, N., Zong, X., Liu, H. microscopy. Jpn. J. Appl. Phys. 49:01AH04-1–01AH04-3.
and Gooding, J. J. 2014. Electrochemical and theoretical study 61. Gierz, I., Riedl, C., Starke, U., Ast, C. R. and Kern, K. 2008.
of π–π stacking interactions between graphitic surfaces and Atomic hole doping of graphene. Nano Lett. 8:4603–4607.
pyrene derivatives. J. Phys. Chem. C 118:2650–2659. 62. Robinson, J. T., Perkins, F. K., Snow, E. S., Wei, Z. Q. and
45. Gacal, B. N., Koz, B., Gacal, B., Kiskan, B., Erdogan, M. Sheehan, P. E. 2008. Reduced graphene oxide molecular sen-
and Yagci, Y. 2009. Pyrene functional poly(vinyl alcohol) sors. Nano Lett. 8:3137–3140.
by “click” chemistry. J. Polym. Sci., Part A: Polym. Chem. 63. Schedin, F., Geim, A. K., Morozov, S. V., Hill, E. W., Blake,
47:1317–1326. P., Katsnelson, M. I. and Novoselov, K. S. 2007. Detection of
46. Pei, X. W., Xia, Y. Q., Liu, W. M., Yu, B. and Hao, J. C. individual gas molecules adsorbed on graphene. Nat. Mater.
2008. Polyelectrolyte-grafted carbon nanotubes: Synthesis, 6:652–655.
64. Wei, D. C., Liu, Y. Q., Wang, Y., Zhang, H. L., Huang, L. P. and graphene based on ultrathin graphene oxide-porous silica
Yu, G. 2009. Synthesis of N-doped graphene by chemical vapor sheets for oxygen reduction reactions. Adv. Funct. Mater.
deposition and its electrical properties. Nano Lett. 9:1752–1758. 22:3634–3640.
65. Sheng, Z. H., Shao, L., Chen, J. J., Bao, W. J., Wang, F. B. 74. Yang, Z., Yao, Z., Li, G., Fang, G., Nie, H., Liu, Z., Zhou,
and Xia, X. H. 2011. Catalyst-free synthesis of nitrogen-doped X., Chen, X. and Huang, S. 2012. Sulfur-doped graphene as
graphene via thermal annealing graphite oxide with melamine an efficient metal-free cathode catalyst for oxygen reduction.
and its excellent electrocatalysis. ACS Nano 5:4350–4358. ACS Nano 6:205–211.
66. Bai, J., Zhu, Q., Lv, Z., Dong, H., Yu, J. and Dong, L. 2013. 75. Zhang, C., Mahmood, N., Yin, H., Liu, F. and Hou, Y. 2013.
Nitrogen-doped graphene as catalysts and catalyst supports Synthesis of phosphorus-doped graphene and its multifunc-
for oxygen reduction in both acidic and alkaline solutions. Int. tional applications for oxygen reduction reaction and lithium
J. Hydrogen Energy 38:1413–1418. ion batteries. Adv. Mater. 25:4932–4937.
67. Lai, L., Potts, J. R., Zhan, D., Wang, L., Poh, C. K., Tang, C., 76. Wang, H., Wang, H., Chen, Y., Liu, Y. Zhao, J. Cai, Q. and
Gong, H., Shen, Z., Lin, J. and Ruoff, R. S. 2012. Exploration of Wang, X. 2013. Phosphorus-doped graphene and (8, 0) car-
the active center structure of nitrogen-doped graphene-based bon nanotube: Structural, electronic, magnetic properties, and
catalysts for oxygen reduction reaction. Energy Environ. Sci. chemical reactivity. Appl. Surf. Sci. 273:302–309.
5:7936–7942. 77. Chen, Y., Liu, Y., Wang, H., Zhao, J., Cai, Q., Wang, X. and
68. Kong, X., Sun, Z., Chen, M., Chen, C. and Chen, Q. 2013. Ding, Y. 2013. Silicon-doped graphene: An effective and
Metal-free catalytic reduction of 4-nitrophenol to 4-aminophe- metal-free catalyst for NO reduction to N2O? ACS Appl.
nol by N-doped graphene. Energy Environ. Sci. 6:3260–3266. Mater. Interfaces 5:5994–6000.
69. Wang, Y., Shao, Y., Matson, D. W., Li, J. and Lin, Y. 2010. 78. Han, J., Cheon, J. Y., Joo, S. H. and Park, S. 2014. Synthesis of
Nitrogen-doped graphene and its application in electrochemi- boron and nitrogen co-doped graphene nano-platelets using a
cal biosensing. ACS Nano 4:1790–1798. two-step solution process and catalytic properties for oxygen
70. Kong, X., Chen, C. and Chen, Q. 2014. Doped graphene for reduction reaction. Solid State Sci. 33:1–5.
metal-free catalysis. Chem. Soc. Rev. 43:2841–2857. 79. Zheng, Y., Jiao, Y., Ge, L., Jaroniec, M. and Qiao, S. Z. 2013.
71. Sheng, Z. H., Gao, H. L., Bao, W. J., Wang, F. B. and Xia, Two-step boron and nitrogen doping in graphene for enhanced
X. H. 2012. Synthesis of boron doped graphene for oxygen synergistic catalysis. Angew. Chem. Int. Ed. 52:3110–3116.
reduction reaction in fuel cells. J. Mater. Chem. 22:390–395. 80. Lohmann, T., Klitzing, K. V. and Smet, J. H. 2009. Four-
72. Nachimuthu, S., Lai, P. and Jiang, J. 2014. Efficient hydro- terminal magneto-transport in graphene p–n junctions created
gen storage in boron doped graphene decorated by transition by spatially selective doping. Nano Lett. 9:1973–1979.
metals—A first-principles study. Carbon 73:132–134. 81. Ohta, T., Bostwick, A., Seyller, T., Horn, K. and Rotenberg, E.
73. Yang, S., Zhi, L., Tang, K., Feng, X., Maier, J. and Müllen, 2006. Controlling the electronic structure of bilayer graphene.
K. 2012. Efficient synthesis of heteroatom (N or S)-doped Science 313:951–954.
13 Graphene Synthesis by Chemical
Vapor Deposition on Copper
Kemal Celebi, Ning Yang, Matthew T. Cole,
Kenneth B. K. Teo, and Hyung Gyu Park
CONTENTS
Abstract...................................................................................................................................................................................... 225
13.1 Introduction...................................................................................................................................................................... 225
13.2 CVD by Carbon Precipitation on Transition Metals........................................................................................................ 227
13.3 Surface-Mediated Graphene CVD on Cu......................................................................................................................... 227
13.3.1 Substrate Preparation............................................................................................................................................ 227
13.3.2 Crystal Nucleation................................................................................................................................................ 229
13.3.3 Growth Kinetics................................................................................................................................................... 229
13.3.4 Reaction Energetics.............................................................................................................................................. 230
13.3.5 Crystal Morphology.............................................................................................................................................. 232
13.3.6 Secondary Crystal................................................................................................................................................ 233
13.4 Transfer and Direct Growth on Insulators........................................................................................................................ 234
13.4.1 Graphene Transfer................................................................................................................................................ 235
13.4.2 Transfer Cleanliness............................................................................................................................................. 235
13.4.3 Direct Growth on Dielectrics............................................................................................................................... 237
13.5 Electronic Properties of CVD-Synthesized Graphene..................................................................................................... 238
13.6 Scaling-Up Schemes......................................................................................................................................................... 239
13.7 Summary.......................................................................................................................................................................... 240
References.................................................................................................................................................................................. 240
ABSTRACT charge mobilities up to 1,000,000 cm2 V−1 s−1,2 broadband

optical transparency, linear dispersion resulting in massless
Progress in the synthesis of high-quality monolayer graphene Dirac fermions,3 a near-zero intrinsic bandgap resulting in sat-
has taken a giant leap forward with the introduction of copper urable optical absorption,4 tunable bandgap,5 unprecedented
as the growth catalyst. This progress has also drawn interest mechanical strength (1 TPa tensile modulus),6 and one of the
from the industry because of its promise to yield large-area highest thermal conductivities ever reported (5300 W m−1 K−1)
graphene at low cost. In this chapter, we provide a detailed second only to superfluid He.7
description of the chemical vapor deposition of graphene Most investigations on the fundamental properties of
on copper, the fundamental mechanisms that govern it, the graphene have been carried out using mechanically exfoli-
effects of the growth conditions on the morphology, and the ated graphene from highly oriented pyrolytic graphite or
electrical properties. We also present graphene transfer meth- kish sources.1 The high graphitic quality and cleanliness of
ods and cleanliness of the transfer processes. The chapter this method have provided a crystallographically ideal mate-
concludes with a review of the prospects on the scaling-up rial perfectly suited for a test bed for fundamental explora-
schemes for the industrial production of graphene. tion. Indeed, graphene obtained through this method is what
allowed the measurement of many, if not all, of the above
properties in practice. However, exfoliation techniques can
13.1 INTRODUCTION
only provide micron- and submicron-sized graphene crystals
The dimensionality paradigm in materials science has under- at random and uncontrolled locations. Despite a wide variety
gone a phenomenal shift with the introduction of graphene. of chemical and mechanical exfoliation techniques available
Unlike other low-dimensional materials, such as the fuller- and compatible with an arbitrary range of substrates, such
enes and nanotubes, graphene has provided unprecedented approaches are unsuitable for technological purposes due to
applicability along with intrinsic processability. Since its first issues with scaling and reproducibility.
isolation,1 graphene has been widely studied and has been Given the unquestionably large catalog of excellent mechan-
shown to possess unique properties such as extremely large ical, optical, and electronic properties demonstrated, graphene
225
is evidently a very promising material for a lot of technologi- approach has been evidenced, the use of this method of syn-
cal applications such as flexible transparent conductors,8 inte- thesis remains limited because of poor areal coverage and
grated circuit elements,9 high-frequency (GHz) transistors,10 graphitic quality. More efforts are ongoing to overcome these
ultracapacitors,11 battery electrodes,12 porous membranes for issues.
filtration13 and desalination,14 passivation layers,15 catalytic Perhaps the most industrially viable approach is chemical
support,16 plasmonic sensor,17 and composite materials.18 All vapor deposition (CVD) on catalytic metal substrates. CVD
of these applications require large-area graphene produced of graphene is emerging as the preferred method of synthe-
via scalable, economical means. Thus, alternative methods sizing graphene for almost all applications, attributed, in
of isolation to mechanical cleavage are in critical need, with part, to a wide process window, large-area compatibility, and
particular interest in direct chemical synthesis (Table 13.1). technical familiarity with the semiconductor industry. Early
One example of this alternative approach is to use other top- CVD processes employed transition metal substrates such as
down methods based on chemical exfoliation19 that employ Ir,24 Ru,25 Pt,26 Pd,27 and Ni28 with exploiting the high car-
agitation of intercalated graphite or chemical reduction of bon solubility of these metals to precipitate carbon from the
graphene oxide in order to produce large quantities of crystals supersaturated solid-solution reservoir in their bulk lattices to
that can coat extremely large surfaces8a,20 if coupled with vari- form graphene layers on their surfaces. Among them, Ni has
ous mechanical deposition techniques such as casting, spray recently proven to be favorable. Nevertheless, the cost of some
casting, Meyer rod, Langmuir Blodgett, screen printing, and of these high-purity catalysts, especially Ir, coupled with dif-
electrophoretic deposition. However, the resulting layer is an ficulties in controlling the number of precipitated layers has
ensemble of graphitic material that is far from having the high urged researchers to search for metals with lower carbon solu-
quality (and subsequent properties therein) of the graphene bility. More recently, Cu has proven to be the most widely
monolayer, ascribed to the out-of-plane interflake connectiv- adopted substrate for producing polycrystalline graphene
ity, tunnel barrier dominance, and altered chemical structure. monolayers.29 State-of-the-art Cu-based CVD graphene has
As such, atomistic (or bottom-up) synthesis methods are the been evolved to the extent that it is now extremely graphitic
sole option to obtain a high-quality continuous monolayer of in nature, and approaching a point where it can yield almost
graphene. as good electronic quality as an individual graphene flake
Early work on graphene synthesis focused on epitaxy of exfoliated mechanically30 while capably scaled to hundreds
SiC by removal of the Si atoms from the crystalline surface. of meters by emerging roll-to-roll processes.31
This method became popular among researchers because of This chapter reviews graphene synthesis by CVD, with par-
the high quality of graphene obtained21 and has been used to ticular reference to Cu-catalyzed CVD throughout. An insight
demonstrate some of the fastest GHz circuits ever produced.22 on the precipitation-based CVD synthesis is first given, fol-
Unfortunately, the SiC substrate is particularly expensive, lowed by the underlying surface catalysis. The nucleation and
confining the commercial feasibility to specialized applica- growth mechanisms, and the effects of growth conditions on
tions such as electronics of very high performance. Moreover, crystal morphology are then reviewed in detail. The important
the Si sublimation process requires quite high temperatures aspects of graphene transfer and cleanliness are subsequently
(ca. 1500°C) making the epitaxial growth incompatible with explained, and the efforts toward transfer-free direct growth
many substrates as well as further increasing the per-area cost on insulators are summarized. Next, the quality of CVD gra-
of the synthesized material. Direct chemical assembly of ben- phene and the film quality implications of the growth condi-
zene subunits, as well as material properties thereof, has also tions are discussed. Lastly, the recent scale-up schemes for the
been investigated.23 Although the feasibility of this promising mass production of large-area materials are presented.
TABLE 13.1
Comparison of Major Methods to Obtain Graphene
Method Mechanical Exfoliation Chemical Exfoliation SiC Epitaxy CVD on Ni CVD on Cu
T(°C) Ambient Ambient ~1500 900–1000 ~1000
Mobility (cm2 V−1 s−1) Up to 106 100a 3000–11,000 3700 Up to 65,000
Thickness Single Few Single Single-few Single
Crystallite size (µm) >1000 ≤0.1 50 20 ≤5000
Scalability Low Medium High High High
Transferability High – Low High High
Sample size (mm) >1 Infinitea 100 10 762 (30 inch)
Cost Very high Low High Low Low
References 1,2 19 21 28,36 30,31
a Overlapping graphitic flakes.

Graphene Synthesis by Chemical Vapor Deposition on Copper 227
13.2 CVD BY CARBON PRECIPITATION 13.3 SURFACE-MEDIATED

ON TRANSITION METALS GRAPHENE CVD ON CU
The history of the growth of graphene on metals can be The practically zero carbon solubility of Cu44 (10−3%–10−2%
traced back to more than half a century, with the first reports for Cu at 1000°C) makes it an ideal catalyst to synthesize gra-
of monolayer and few-layer graphite on Pt32 and Ni.33 At that phene. Surface catalysis and the lack of bulk carbon cause
time, the interpretation of the observed monolayers to be of the reaction to be self-terminating under particular growth
a graphitic nature was not always immediate.34 The lack of conditions, yielding monolayer graphene.29a Surface-based
interest in graphene has limited further research for a few growth was experimentally evidenced by Li et al. using iso-
decades. It was not until 2004 when the mechanical exfo- tope labeling of the carbon precursor during growth.29b They
liation method by Novoselov et al.1 nurtured the research showed that the graphene grows by attachment of the surface-
on graphene and its production, much to the surprise of adsorbed carbon to the edge, resulting in a periodic, nested
fervent Mermin followers who have believed that the isola- alternating isotope pattern (Figure 13.1a), which is distinct
tion of monolayer materials at finite temperatures was non- from the carbon precipitation from the bulk, which would
physical because of inherent thermodynamic instabilities of have produced a random isotope spatial distribution (Figure
their lattices.35 Thus, traditional methods based on direct 13.1b).29b Cu is also preferable for graphene transfer because
growth were revived. Initial works in this period started with it can be etched selectively and at low cost by chemicals, for
Ni-based growths36 and continued with the help of demon- example, ammonium persulfate. Though it was extensively
strations on other metals. So far, graphene has been synthe- argued that the financial burden associated with the etching
sized on polycrystalline Ni,28 Fe,37 Au,38 stainless steel,39 and of the catalyst material dramatically increased the manufac-
single crystals of Ru(0001),25 Ir(111),24 Rh(111),40 Pt(111),26 turing cost, this issue has recently been circumnavigated with
and Pd(111).27 Since these metals can dissolve a large amount the advent of frame-assisted bubble transfer,45 the reuse and
of carbon in bulk, the growth mechanism roots mainly in pre- purification of low-cost etchants, such as ferric chloride, and
cipitation with additional contributions from decomposition the recovery of Cu from the solution.46
of the precursors on the metal surface. During cooling down, A typical growth on Cu starts with conditioning of the
the carbon from the bulk precipitates on the surface and forms substrate to reduce any native oxide (>500°C or with plasma
the graphitic layers. Nevertheless, many of these catalysts, in reducing gas, such as H2), high-temperature annealing
including materials processing cost, are quite expensive, as to crystallize the Cu (i.e., 700–1000°C), and then, finally,
most of them are single crystalline metals of extremely high exposure to the hydrocarbon precursor (e.g., C2H2,47 C2H4,48
purity. Furthermore, controlling the number of layers during CH4,29a C6H6,49 etc.), solid50 or liquid50b at similar or lower
the growth has also proved to be challenging. Nevertheless, temperatures. Graphene crystals then nucleate and enlarge
conventional metallurgical approaches were adopted to after consuming the reactants formed by the catalysis of the
achieve monolayer growth, including rapid quenching and carbon precursor on Cu. The available parametric window for
more exotic means of direct engineering of the catalyst, such continuous monolayer synthesis is wide.51 For CH4, growth
as developing binary alloys.41 Meanwhile, lattice mismatch temperatures can lie between 800°C and 1050°C, at chamber
between graphene and the catalyst metals (with the exception pressures of 0.01 mTorr to 760 Torr. However, the variation
of Ni, with a mismatch of less than 1%) became an important of the morphology by the growth parameters causes impor-
hurdle. Lattice mismatch stimulates corrugation and wrinkle tant variations in the physical properties of the deposited
formation, altering the intrinsic properties of graphene, such materials. In this section, we describe the growth process in
as the strain-dependent work function, and producing Moire further detail and explain the fundamental mechanisms and
patterns,42 often undesired for most purposes. Substrate inter- kinetics that govern it. Flake morphology and secondary lay-
actions, in particular with Ru(0001), can also cause modifica- ers are also discussed.
tions in the electronic band structure.25b So far, in this carbon
precipitation method, only polycrystalline Ni has been dem-
13.3.1 Substrate Preparation
onstrated to yield graphene of reasonable quality (i.e., a
charge mobility of 3700 cm2 V−1 s−1) at low cost, scalable Substrate preparation is a crucial step prior to growth of gra-
size, and low lattice mismatch.28a However, it is extremely phene (Table 13.2). If the surface of the substrate is rough,
difficult to control the number of layers on polycrystalline an initial chemical mechanical polishing step is often neces-
Ni, which has different carbon saturation depending on the sary.52 Scratches and defects on the substrate could function as
metallic grains, causing spatial variation of the graphene nucleation sites during the synthesis (Figure 13.1c). Smoother
thickness. It is actually easier to grow monolayer graphene substrates have less nucleation sites, thereby reducing the
on Cu and transfer it three times to a triple-layer graphene nucleation density and enlarging the attainable grain size.53
stack than to precipitate exactly three layers of graphene on However, excessive polishing may disturb the spatial unifor-
polycrystalline Ni. Therefore, even Ni has been surpassed by mity of the sizes of graphene grains. Although the nucleation
the controlled growths enabled by Cu. At present, Cu-based sites are less dense compared with an unpolished surface, tiny
CVD is, by far, the most popular production method for the defects on the polished surface can act as nucleation sites. It
production of graphene on a large scale.43 is these sparsely and inhomogeneously distributed defects
Graphene grown on Cu by Multilayer graphene grown

surface adsorption on Ni by precipitation
(a) C12 1591 (b) C12 1563
C13 C13
Rel. Rel.
1/cm 1/cm
1523 1544
(c) Scratching Pre-heating Growing
FIGURE 13.1 Surface growth mechanism. (a, b) Micro-Raman characterization of isotope-labeled graphene monolayers grown on Cu
(a) and Ni (b) foils. The integrated intensity of the G peak is shown, indicating spatially alternating regions of C12 and C13 isotopes. (c)
The schematics of the graphene nucleation, indicating the higher nucleation density at the scratch sites. (Reprinted with permission from
Li, X. et al., Evolution of graphene growth on Cu and Ni studied by carbon isotope labeling. Nano Letters, 9, 4268–4272. Copyright 2009
American Chemical Society; Reprinted with permission from Han, G. H. et al., Influence of copper morphology in forming nucleation seeds
for graphene growth. Nano Letters, 11 (10), 4144–4148. Copyright 2011 American Chemical Society.)
that cause an inhomogeneous nucleation density and uneven small particles are difficult to remove and can remain attached
distribution of grain sizes. Consequently, the resulting mono- to graphene even after transferring to other substrates.
layer film could have poor grain connectivity, high sheet Besides the composition, Cu crystallinity57 and grain
resistances, and perturbed and undulating surfaces owing to conformation have also been shown to be important for the
polishing.54 catalytic process. Single or polycrystalline Cu can be used
The next step of preparation is the chemical reduction for graphene growth.58 However, owing to their low cost and
often involving acetic or hydrochloric acid pretreatments.55 wide availability, the most popular forms of Cu so far have
A high-temperature annealing under pressures of 0.01 mTorr been commercial foils and polycrystalline physical vapor-
to 760 Torr at 900–1000°C is the main furnace process to pre- deposited thin films. Larger grains are favorable, and the
pare the surface right before the growth step. Here, higher grain sizes of evaporation-deposited Cu films can be enlarged
annealing pressures are known to smoothen the Cu surface.56 from tens of nanometers to a few micrometers by anneal-
This process of preparation can take up to 30 min (for coming at 900–1000°C.59 This increase in the grain size can be
mercial 25-μm-thick foils). What is critical in this preparation more dramatic in Cu foils, with sizes increasing up to hun-
step is the metallurgical composition of the Cu foil. Additional dreds of micrometers.29a Crystalline orientation is also critical
coatings or solvent impurities from the rolling manufacturing for graphene growth.57 Thermal annealing can recrystallize
processes may cause nanometer-sized particles to form on the the catalysis-sensitive surface, yielding mostly monocrys-
foils during the growth of graphene, visible under a scanning talline orientation over the entire surface.48 Electron back-
electron microscope (secondary electron detection).48,54 These scattering diffraction (Figure 13.2a) shows that the annealing
TABLE 13.2
Cu Substrate Preparation Procedure
Step Method Condition Rational
1 Chemical mechanical 3.0–6.0 V 60s in electropolishing solution Reduce defects sites thereby eliminate
polishing nucleation density52,53
2 Reduction Acetic or hydrochloric acid pretreatments Remove surface oxide55
3 Annealing 900–1000°C 15–30 min for commercial Yield larger Cu grains up to hundreds
25 µm thick Cu foil of micrometers56
(a) (b)
2.5 I2D/IG Cu(632)
Intensity ratio (unitless)

ID/IG
2.0 Cu(533)
r2 = 0.76
1.5 Cu(111)
Cu(100) Cu(511)
1.0
111
0.5 r2 = 0.56
0.0 0.2 0.4 0.6 0.8 1.0

1 mm
001 101 Percentage of Cu(111) Surface
FIGURE 13.2 Influence of the copper crystalline orientation. (a) EBSD microscan of the Cu foil surface after high-temperature annealing
at 900°C for 30 min. Inset: Inverse pole figure for the surface normal direction, showing the dominant copper crystallographic orientation
is (100). (b) Intensity ratios IG′/IG and ID/IG for different Cu surface crystallographic orientations (the G′ peak shown as 2D in the figure).
Cu(111) has monolayer graphene, while others have multilayer coverage. Defect density appears to be independent of the surface orientation.
(Reprinted with permission from Celebi, K. et al., Evolutionary kinetics of graphene formation on copper. Nano Letters, 13 (3), 967–974.
Copyright 2013 American Chemical Society; Reprinted with permission from Wood, J. D. et al., Effects of polycrystalline Cu substrate on
graphene growth by chemical vapor deposition. Nano Letters, 11 (11), 4547–4554. Copyright 2011 American Chemical Society.)
pretreatment renders the dominant surface orientation of Cu also cause spatial inhomogeneity in the nucleation distribu-
foils to become (100); while for Cu films, annealing crys- tion (Figure 13.3a(ii)) which is often critically important in
tallizes the otherwise amorphous film to (111) surfaces.59 horizontal-flow tube furnaces with short precursor residence
Low-index Cu facets tend to yield higher areal densities of time that constitute perhaps the most common type of CVD
monolayer graphene with less defects (Figure 13.2b).57 On the systems. The nucleation density is proportional to decreasing
other hand, Tao et al. has reported that Cu films can produce temperature and increasing hydrocarbon partial pressure as
higher-quality graphene compared to foils,60 where it is pro- well as flow rate (largely for atmospheric pressure CVD).30a
posed that the higher H2 content of the films may replace the Temperature also determines the degree of supersaturation
need for the gas-phase hydrogen and reduce the defect density. because the carbon capture and reactant desorption rates are
highly dependent on temperature.65 Figure 13.3c shows that
at lower CVD temperatures, the nucleation is limited by the
13.3.2 Crystal Nucleation
capture events of the reactants with an activation energy (EA)
Most carbon precursors, such as CH4, C2H4, C2H2 and simi- of ca. 1 eV, while at higher temperatures, desorption becomes
larly reactive hydrocarbons, readily chemisorb onto a faceted the dominant physicochemical mechanism with a higher EA
Cu surface and face only a comparatively small energy barrier of ca. 3 eV.54
on the order of 0.1 eV.61 As the adsorbed carbon species, the
form of which is still presently unknown, populate the sur-
13.3.3 Growth Kinetics
face and subsurface, supersaturation is reached. The surface
defects, facets, and lattice impurities then trigger heteroge- The time dependence of the growth rate is critical to under-
neous nucleation. Direct experimental evidence of the growth standing the fundamental mechanisms that govern the CVD
nucleation at lattice defects, a scratch in this instance, was of graphene. The information about the growth rate can be
observed53 and is shown in Figure 13.1c. Upon initial nucle- extracted by observing the variation in graphene flake size
ation, these graphene seeds begin to deplete the nearby solid over the period of growth. Although real-time observation
solution of carbon reactants. The form of these reactants is of an Ru-based graphene growth using low-energy electron
disputed, with indications ranging from carbon adatoms54 microscopy (LEEM) has been reported,25a,66 it is challenging
to carbon clusters62 or even hydrocarbons.61,63 As the nucle- to employ this technique for Cu due to low carbon density
ation is governed by the initial carbon supersaturation of the and substrate sublimation.25a,67 As such, the literature on the
catalyst and the availability of suitable lattice defects, the Cu-based CVD kinetics is, at present, rather limited. Recently,
hydrocarbon partial pressure and the catalyst temperature CH4-based graphene CVD investigations by Chhowalla and
determine, in part, the spatial number density of the nucle- colleagues have proposed a supersaturation-driven growth,
ation events. Figure 13.3a shows a reduction in the nucleation a claim that attributes the entire growth to the depletion of
density caused by depletion of the available CH4 gas (Figure the initial population of carbon adatoms from the surface
13.3b), which reduces the number of available carbon reac- (Figure 13.4a).54 An exponential-like kinetics curve on Figure
tants at the moment of the seeding event.64 Any disturbances 13.4b was fitted by the Johnson–Mehl–Avrami–Kolmogorov
in the gas flow related to local facets or turbulence may (JMAK) crystallization model,68 with the assumptions of
(a) (b)
(i) Gas flow
Gas flow Cgas
CH4 Mass transport

Csurface
Cgas
1 cm Cu catalyst
(ii) (iii) (iv) (c) Temperature (°C)

1000 950 900 850 800 750
ELT = 1 eV
Nucleation density (μm–2)

10
EHT = 3 eV Desorption
Nucleation
1
Capture
Cu
20 μm 20 μm 20 μm 7.5 × 10–4 8.0 × 10–4 8.5 × 10–4 9.0 × 10–4 9.5 × 10–4 1.0 × 10–3
1/T (K–1)
FIGURE 13.3 Nucleation density. (a) (i–iv) Optical photograph and micrographs of the Cu foil after 1 min graphene growth at 1000°C in
a horizontal tube furnace. (b) Schematic of the carbon source distribution near the catalyst surface for the horizontal flow. (Reprinted by
permission from Macmillan Publishers Ltd. Scientific Reports, Jia, C. et al., Direct optical characterization of graphene growth and domains
on growth substrates. 2, 00707, copyright 2012.) (c) Arrhenius plot of the nucleation density on a Cu foil. Higher slope curve indicates the
desorption-limited nucleation regime with an EA of 3 eV. Higher slope indicates the desorption-limited nucleation regime with an EA of
3 eV, while the lower-slope represents capture-controlled regime with an EA of 1 eV. (Reprinted with permission from Celebi, K. et al.,
Evolutionary kinetics of graphene formation on copper. Nano Letters, 13 (3), 967–974. Copyright 2013 American Chemical Society.)
instant nucleation and carbon attachment-limited growth, reactions compares the energetic weight of the corresponding
where they incorrectly neglected any contribution of the step in the overall reaction potential scheme. A brief sum-
additional hydrocarbon input during the growth. Their fitting mary of the mechanisms that govern the growth is provided in
implied that the growth saturates and a continuous graphene Figure 13.5a. When the hydrocarbon feedstock comes in the
layer can never be achieved unless there is sufficient source hot CVD reactor, gas-phase reactions could thermally rear-
of carbon available at the initial growth stage. On the basis of range it prior to adsorption.70 However, unlike the rich litera-
similar assumptions, the continuity of the graphene layer can ture on the gas-phase reactions involved in carbon nanotube
be modeled as a function of the ratio of the CH4 partial pres- growth by often employing in situ mass spectroscopy,71 analo-
sure to the square of the hydrogen partial pressure: PCH4 /PH2 2. gous study for probing the growth atmosphere during the gra-
Growth continuity is achieved for partial pressure ratios of phene growth is not abundant. Moreover, since the adsorption
>0.01 for growth temperatures greater than 900°C.69 Other energy barrier of most hydrocarbons on Cu is negligibly small
carbon precursors such as C2H4 can undergo a more complex (~0.1 eV),61 any transiently derived CxHy species formed in
reaction cascade due to the wide number of thermally dissoci- the gas phase may rapidly adsorb, disabling direct measure-
ated intermediaries, as well as a range of possible dehydroge- ment and leading to the complexity of the underlying catalytic
nation processes on the Cu surface. Using an C2H4-based CVD reactions. Moreover, thermal variations within CVD reactors
system, Park and colleagues have proposed a Gompertzian only support the formation of such transients in hot zones,
sigmoidal kinetics for graphene growth (Figure 13.4c), indi- particularly, which further complicates direct measurement.
cating that the growth is not only influenced by supersatu- Currently, very little is known about these hydrocarbon reac-
ration but also by continual hydrocarbon supply as well as tion cascades on Cu. Consequently, it is very difficult to make
associated surface adsorption/desorption dynamics.48 any assertions on the effects of the gas-phase reactions on the
overall CVD process.
Carbon reactant diffusion follows hydrocarbon adsorp-
13.3.4 Reaction Energetics
tion.72 The energy barrier for the diffusion is also low, typi-
Determination of the reaction energetics is critical in order cally less than 1 eV on Cu.61,72b Thus, under equilibrium
to determine the rate-limiting processes during the CVD conditions, diffusion is rather unlikely to be a rate-limit-
growth. The associated energy barrier for each of the possible ing step. Diffusion may become dominant under certain
(a) (b) 0.9

0.8
900 °C
0.7
Crystal area coverage (%)

CH4
0.6
1. Chemisorption
2. Desorption 0.5
0.4 720 °C
3. Diffusion 4. Cluster 0.3

formation 5. Attachment/detachment
0.2
Cu
0.1
0.0
0 400 800 1200 1600
iii
Time (sec)
i ii Asat < 1
(c) 0.30
iv 0.3 8 Gompertz fit
Flake area (μm2)

6
Cover (%)
0.25 0.2
Asat = 1 4
0.1
2
Crystal area (μm2)

0.20
Ccu
0 0
CH4 flow 0 1 2 3 4
ON Cnuc 0.15 Time (min)
Ceq
0 0.10
Time Max. flake area at 770°C
0.05 770°C
800°C
Inflection points
0
0 1 2 3 4
Time (min)
FIGURE 13.4 Growth kinetics. (a) Schematic for the graphene growth controlled by adatom attachment mechanism (top) and the carbon
adatom density variation on the Cu surface (bottom). (b, c) Time dependence of the graphene covering on the Cu surface for CH4-based
growth, fitted by the JMAK model (b), and for C2H4-based growth, fitted by the Gompertz model (c). ((a, b) Reprinted with permission from
Kim, H. et al., Activation energy paths for graphene nucleation and growth on Cu. ACS Nano, 6 (4), 3614–3623. Copyright 2012 American
Chemical Society; (c) Reprinted with permission from Celebi, K. et al., Evolutionary kinetics of graphene formation on copper. Nano Letters,
13 (3), 967–974. Copyright 2013 American Chemical Society.)
conditions, however, in particular when the hydrocarbon on the carbon precursor, and (2) attachment of carbon in the
reactions are simpler and diffusion occurs at the subsurface form of adatom, chain/cluster, or hydrocarbon at the crystal
level. Nie et al. have concluded that diffusion is the plau- edges of growing graphene. The exact nature of these reac-
sible rate-limiting mechanism when pure carbon is used as tions is elusive. A simplistic diagram that displays the approx-
the precursor on Cu(111) surfaces under ultrahigh vacuum imate energy levels of the adsorption and dehydrogenation is
conditions.67 Besides surface diffusion, gas diffusion can shown in Figure 13.5b.49 The dehydrogenation energy barrier
particularly be important under atmospheric-pressure CVD indicates that the overall reaction is unfavorable and must be
(APCVD).73 During APCVD, a boundary layer forms above supported by excessive hydrocarbon feed. This finding also
the substrate surface, and gas diffusion through this bound- suggests that reactive precursors can be used for the growth
ary layer could become rate limiting. Care must be taken of graphene at lower temperatures because once these species
when extracting the activation energies from the resultant are adsorbed, desorption would become more difficult.
Arrhenius curves. In order to determine correct energy barri- Which reaction step of the entire graphene growth process
ers responsible for chemical processes, these kinds of mass- should be regarded as rate limiting is currently in dispute.
transport-limited regimes must be ruled out. Accurate measurement of EA is required for the determina-
When gaseous transport of hydrocarbon is not the limit- tion of the reaction that determines the speed of growth. For a
ing factor, catalytic reactions on the Cu surface determine CH4 precursor, Kim et al. have found out an EA of 2.6 eV54 and
the rate of graphene crystal enlargement. Two main reaction concluded that the carbon adatom attachment (ca. 2 eV energy
types occur on the surface: (1) a dehydrogenation cascade, barrier)25a is the rate-limiting step, claiming that the dehydro-
which may also involve dissociation of the radicals depending genation energy barrier is lower (1.7–1.9 eV).74 The species that
(a)
Copper sublimation
Adsorption
Desorption
Dehydrogenation
Dissociation Diffusion
Copper catalyst Subsurface diffusion
(b)
1.5 CH4
C6H6 1.36
1
C18H14 0.84
0.5 Eb = 1.53 eV 0.55
Potential energy (eV)
Eb = 1.51 eV 0.21
0
–0.17
–0.5 –0.38
–0.67
–1
Eb = 1.55 eV –1.13
–1.5
–1.2 –1.93
–2.5 GS AS TS DS
Desorption Dehydrogenation
FIGURE 13.5 Mechanisms and energetics. (a) Illustration of the mechanisms that govern graphene growth on Cu. Atoms are rep-
resented by different shades and sizes: C (gray), H (dark gray), Cu (light gray). (Reprinted with permission from Celebi, K. et al.,
Evolutionary kinetics of graphene formation on copper. Nano Letters, 13 (3), 967–974. Copyright 2013 American Chemical Society.)
(b) Diagram for the energy levels associated with different carbon precursors on copper surface. GS, AS, TS, and DS represent the gas-
eous, adsorbed, transition, and dehydrogenated states, respectively. (Reprinted by permission from Macmillan Publishers Ltd. Scientific
Reports, Choi, J.-H. et al., Drastic reduction in the growth temperature of graphene on copper via enhanced London dispersion force. 3,
01925, copyright 2013.)
attach to the flake edges are also widely debated. While some the physical and chemical properties of the as-synthesized
reports focus on the adatoms,54,69,75 there are also reports that graphene, in particular, charge mobility. This subsection
compare different species to conclude that few-carbon-atom addresses competing mechanistic models that determine the
chains are also important reactant species.62 An increase in the crystal morphology. The morphologies reported so far can
CVD pressure has also been claimed to favor carbon chains be largely described as a mixture of dendritic or hexagonal
versus adatom formation.76 Alternatively, Celebi et al. have geometries, as well as compound structures. Substrate and
pointed out that the attachment is not the limiting step; instead, growth conditions are certainly responsible for the way in
a reaction cascade of the dissociative dehydrogenation on Cu which the crystal enlargement occurs.
determines the growth rate.48 The reported EA, peaking at Wofford et al. have reported in situ observations of four-
3.1 eV, was found to be time dependent, as a result of the dis- lobed flake enlargement on Cu(100) under ultrahigh vacuum
persive nature of this reaction cascade.77 However, the precur- using a pure carbon precursor.79 The growth was claimed to
sor used in this study (C2H4) may well support a more complex be attachment limited. The rate of flake enlargement is ori-
reaction scheme than the simpler hydrocarbon (CH4).78 For even entation dependent, and has a twofold symmetry as a result
more complex precursors, the understanding of the associated of the interaction between the sixfold and fourfold symme-
reactions is much weaker. More investigations are necessary to tries of the graphene and Cu, respectively. However, a similar
elucidate the reactants and mechanisms of the reactions. study by Nie et al. states that for Cu(111) surfaces, the flake
morphology changes from dendritic to hexagonal by simply
increasing the temperature (Figure 13.6a through d).67 They
13.3.5 Crystal Morphology
rationalized these findings via a diffusion-limited model.
Graphene grown by CVD is polycrystalline. The flake Indeed, unlike surface diffusion on Cu(100), carbon adatoms
crystallinity and edge morphology dramatically influence
diffuse through subsurface sites on Cu(111), which requires
(a) (c) (d)
Graphene
Cu
(b) (e)
Graphene
Cu
FIGURE 13.6 Graphene flake morphology. (a) LEEM image of dendritic growth of graphene on Cu(111) surface at 690°C. (b) Low-energy
electron diffraction (LEED) pattern showing that the dendritic graphene is polycrystalline. (c) LEEM image of a similar growth at 900°C.
(d, e) LEED patterns of the two different regions of graphene in (c), indicating monocrystallinity (d) and a 3° rotation by Cu step-edges (e).
(Reprinted with permission from Nie, S. et al., Origin of the mosaicity in graphene grown on Cu(111). Physical Review B, 84 (15), 155425.
Copyright 2011 by the American Physical Society.)
surfacing of the adatom, and thus manifests higher energies suppression of Cu sublimation,48,67 and (3) difference in the
compared to attachment-limited systems. However, this sur- carbon-to-crystal-edge attachment mechanism related to the
facing mechanism is yet to be explained in any great detail. formation of carbon chains on the Cu surface.62 Hexagonal
At higher temperatures on Cu(111), the graphene crystal also morphology is more favored compared with the dendritic
presents lobed geometries that are coupled to the sixfold morphology for better grain crystallinity and connectivity.
symmetry of the substrate due to the inhomogeneous den-
sity distribution of Cu steps. Regarding the crystallinity, the
13.3.6 Secondary Crystal
hexagonal crystal at higher temperatures has more uniform
crystalline orientation compared with the dendritic crystal CVD of graphene on Cu is, under the correct growth condi-
(Figure 13.6b,d). However, when such crystals enlarge and tions, self-terminating. That is, the growth terminates when
grow over the catalyst step edges, the crystalline orientation monolayer covering is complete. However, even for high-
may change by up to 3° (Figure 13.6e). quality monolayers, two or more additional layers often exist,
During the thermal dissociation of the hydrocarbon pre- covering the catalyst surface area up to a few percent of it. These
cursor, atomic hydrogen is liberated. Thus, the effect of con- layers result from secondary nucleations that occur at areas
comitant lattice etching may not be neglected during growth. that are covered with already-growing crystals. Conflicting
Vlassiouk et al. have reported etching of the graphene crystal reports exist on the nature of these nucleations. Robertson
by excessive hydrogen during the growth.80 The reported etch et al. have used atomic force microscopy data to claim that the
rate was low for zigzag edges, thereby favoring the hexago- secondary layers exist on top of the primary layer84; however,
nality of the crystal. The flake size initially benefits from the equivalent surface profiles can also occur when these second-
catalytic role of hydrogen, until etching becomes dominant ary nucleations are under the primary layer. Furthermore, it
at higher H2 partial pressure. However, the role of hydrogen is counterintuitive that secondary growth can be catalyzed on
on etching is disputed. A contrasting study by Choubak et al. top of the primary graphene layer. Other reports indicate the
claimed that it is the O2 impurities in the H2 line that might additional layers grow under the first, where the growth is still
cause the observed etching of graphene crystals.81 Indeed, catalyzed by the Cu and supported by diffusion of the carbon
negligible etching of graphene occurred when Celebi et al. reactants.56b,85 Figure 13.7a through c shows the LEEM data
grew equally graphitic graphene crystals with and without reported by Nie et al. supporting growth from below.56b The
a H2 ballast and under typical H2 partial pressures during intensity decrease in the secondary layer’s diffraction pattern
growth. Nonetheless, regardless of the exact etching species, (Figure 13.7c) clearly indicates that the second layer is below
perhaps the most important finding about the morphology of the first one. This claim has been recently supported by Kong
the graphene flake is that the degree of hexagonality is cor- and colleagues through carbon feedstock alteration between
related with total pressure of a reacting chamber.80,82 APCVD C12 and C13.86
growths report flakes of hexagonal shapes with good crys- Various strategies can be employed to minimize the forma-
tallinity,83 although the reason for this behavior is unclear. tion of the secondary layers. Han et al. have presented a pulsed
Associated explanations include various mechanisms that supply scheme of the CH4 precursor with incubation intervals
occur at higher pressures, namely (1) gas diffusion limita- designed to allow the excess reactants to be consumed by the
tion,73 (2) prevention of the Cu step advancement due to the first layer.87 Figure 13.7d through g describes this approach.
(a) (b) (c)
(d) (e)
CH4 Cooling
C Storage C Segregation
CxHy
Pulsed Growth
Species
(f ) (g)
Cooling
Cu
Cu Grain
C Depletion Boundaries Defect site
FIGURE 13.7 Secondary nucleations. (a) LEEM image of a graphene crystal on the Cu surface with a secondary nucleation. (b, c) LEED
patterns of the two highlighted regions (inside (b) and outside (c) of the dark hexagon) of (a), showing the secondary layer diffraction pat-
tern having less intensity, which indicates the secondary nucleation occurs below the first one. (From Nie, S. et al., New Journal of Physics
2012, 14 (9), 093028. With permission.) (d–g) Pulse growth schematic for the suppression of the secondary layer growth. (Reproduced from
Han, Z. et al., Suppression of multilayer graphene patches during CVD graphene growth on copper. arXiv preprint arXiv:1205.1337, 2012.
With permission.)
When the excess reactants exist, the probability of having with the APCVD results, where the crystals are also more
additional nucleations under the first layer is higher. Indeed, hexagonal with less lobes compared with the low-pressure
Wassei et al. have revealed that increasing the carbon con- CVD (LPCVD) growths.62 Such a similarity bolsters the sub-
tent in the precursor gas can promote secondary nucleations.88 limation-based reasoning for the determination of the flake
Alternatively, a carbon sink under a thin catalyst layer can morphology.
also be used to limit excessive carbon content on the surface.
However, this method has only been demonstrated for gra- 13.4 TRANSFER AND DIRECT
phene growth on Ni thin films with a Mo underlayer acting as
GROWTH ON INSULATORS
the carbon sink.89
Another interesting feature of secondary nucleations is Metal-catalyzed graphene may not be applied directly to the
that their hexagonal morphology is more pronounced. Lobe characterization of electronic, optical, or mechanical proper-
formation is restricted under the primary graphene cover, ties, or to the applications using these properties. For these
attributed to the suppression of Cu sublimation and the delay applications and characterizations, it is necessary to transfer
of the Cu step advancement.48 This behavior is analogous the graphene onto various dielectric substrates, such as SiO2
and HfO2, or to lay it freestanding. On the one hand, graphene by an air drying (ca. 30 min) and a vacuum (ca. 3 h) drying.
is chemically inert against most of the common solvents and Lastly, the PMMA layer is removed by annealing at 400°C
acids/bases even though it can be readily doped. On the other under Ar and H2 (e.g., 500 sccm: 500 sccm) flows for 2 h.
hand, it is possible to selectively remove the catalyst metal Other than the common wet-transfer technique as
and the transfer media (i.e., polymer) to which graphene is described above, numerous wet/dry methods have been
attached. This selective removability is, in part, one of the demonstrated.92b,99 For example, a direct transfer method
main reasons for the success of the metalcatalytic CVD. exploiting the adhesion force of surface tension has been illus-
While the polymer-mediated transfer process is easy, com- trated by sticking as-grown graphene to transmission electron
plications associated with wrinkle and crack formation as microscope (TEM) grids.100 Here, a drop of isopropanol is
well as incomplete cleanup of the residues from the transfer placed on a TEM grid, which stays on the Cu foil with gra-
medium could all degrade the physical properties of the trans- phene. During solvent evaporation, the surface tension draws
ferred graphene.90 Hence, there have been recent attempts the graphene from the surface of Cu to the amorphous carbon
focusing on direct graphene synthesis on insulators, removing grid. Because this method has low yields, the PMMA-based
the need for any transfer. This section reviews common meth- method is still preferred by many. Alternatively, a dry trans-
ods of graphene transfer and discusses the various cleaning fer technique based on polydimethyl-siloxane (PDMS) can be
procedures employed. We conclude by summarizing advance- used, albeit with less conformity.92b All these methods share
ments in direct growth on dielectric substrates. complete removal of polymer residues as the critical issue.
Large-area, mass-scale transfer techniques include roll-
to-roll processes comprising major steps of (1) adhering gra-
13.4.1 Graphene Transfer
phene to various polymers, which act as thermal, optical, or
Early methods for transfer of graphene from metal substrates chemical release layers; (2) etching of Cu and reattaching
have proven the possibility of relocating graphene by adher- to the destination substrate; and (3) removal of the release
ing it to a submicrometer-thick polymer layer.91 A common layers.94a More recently, renewable procedures to avoid the
wet transfer procedure begins with spin coating of a polymer Cu etch step have been introduced. Figure 13.8c shows the
layer, for example, poly methyl-methacrylate (PMMA), on the delaminating method by Yoon et al.101 Here, a flexible sub-
as-grown graphene/catalyst stack. The stack is then cured at strate with an additional epoxy layer is adhered to the gra-
120–180°C in the case of PMMA. Next, the catalyst metal phene. By precisely measuring the graphene–Cu adhesion
(e.g., Cu) is etched away with the use of aqueous etchant energy (0.72 J m−2), coupled with an appropriate choice of
solutions such as FeCl3,92 Fe(NO3)3,29a HCl,92a HNO3,93 and the epoxy/substrate pair, they removed graphene from its Cu
(NH4)2S2O894 at concentrations of 0.1–5 M. After etching Cu, catalyst without any etching, with the aim of recycling the
the floating polymer/graphene stack is rinsed in water and Cu for future growths. Another alternative method to retain
“picked up” with a destination substrate and dried. Finally, the Cu is a so-called bubble transfer method (Figure 13.8d).45–46,102
polymer layer is removed by a proper etchant such as acetone. Electrochemical reduction of water generates H2 bubbles and
Improvements on this wet transfer scheme have been achieved detaches the graphene layer from the Cu foil grounded in a
over the past few years. For PMMA, a secondary casting pro- NaOH aqueous solution (0.25 M), or polarized at −5 V in a
cess has been introduced to release the strain after the initial K2S2O8 solution (0.05 M). However, this method still necessi-
polymer cure.95 Groups have also recently abandoned pre-etch tates the use of a PMMA layer for mechanical integrity of the
curing of polymer,94b which appears to encourage the PMMA/ graphene layer. The procedures of several transfer methods
graphene to conform better to the destination dielectric sub- are summarized in Table 13.3.
strate, and so to reduce the detrimental wrinkling and cracks.
Alternative methods for removing the graphene–PMMA
13.4.2 Transfer Cleanliness
from the solution have also been demonstrated, such as pick-
ing up of the film from above, slow lowering of the liquid So far, no other wet or dry transfer method has surpassed the
level (Figure 13.8a), and reduced-angle fishing with a needle popularity of the PMMA-based transfer, for almost all emerg-
(Figure 13.8b).94b To retain the novel properties of the gra- ing methods have their own distinct drawbacks as described
phene, it is critical to ensure the cleanliness of the graphene above. The PMMA-based transfer method is the most com-
during the transfer process. To achieve this goal, PMMA mon for research purposes using centimeter-sized or larger-
removal has been widely investigated. Heated acetone,96 ace- area graphene, owing to its simplicity, high conformity,
tone vapor,97 and additional annealing98 have been reported. and compatibility with most of the destination substrates.
At the moment, acetone is being phased out completely, being However, PMMA and other polymer-based methods often
replaced by high-temperature gas annealing, which will be produce residues remaining on the graphene after transfer.
discussed in detail in the following section. To exemplify the While many organic solvents such as acetone efficiently dis-
current state-of-the-art transfer,94b optimal wet-transfer trend solve the bulk PMMA, a thin layer (1–2 nm) often remains
for centimeter-sized graphene starts by the spin-coating of adsorbed. These residues could induce weak p-doping of the
PMMA (2% in anisole) at ca. 4000 rpm, continued with Cu graphene.90 A complicated wet-cleaning process based on the
etching by the (NH4)2S2O8 solution (ca. 0.1–0.5 M) in two standard semiconductor cleaning recipes could improve resi-
steps (first step to remove the graphene on the rear), followed due removal.103 Nevertheless, presently, the most efficient way
PMMA Graphene
Cu (c)
Etchant
Target substrate
(a) (b)
PMMA Graphene Epoxy
Graphene Delamination
Cu
Substrate Spacer Water SiO2
Substrate Si
Before heating
Needle
(d) + –
After heating
Glassy carbon
PMMA/Graphene
Heat Heat
Bubble Copper foil
FIGURE 13.8 Graphene transfer methods. (a, b) Schematics for different removal methods to attach the graphene/PMMA layer onto the
destination substrate. Heating at 180°C for 30 min after drying facilitates the flattening of the PMMA layer on the destination substrate.
(Reprinted with permission from Suk, J. W. et al., Transfer of CVD-grown monolayer graphene onto arbitrary substrates. ACS Nano, 5 (9),
6916–6924. Copyright 2011 American Chemical Society.) (c) Schematics of the delamination method that peels of graphene without etch-
ing Cu catalyst. (Reprinted with permission from Yoon, T. et al., Direct measurement of adhesion energy of monolayer graphene As-grown
on copper and its application to renewable transfer process. Nano Letters, 12 (3), 1448–1452. Copyright 2012 American Chemical Society.)
(d) A bubbling method to remove graphene/PMMA layer from Cu. Electrochemical reactions cause gas generation on the copper surface,
which helps detach the graphene layer mechanically. (Reprinted with permission from de la Rosa, C. J. L. et al., Frame assisted H2O elec-
trolysis induced H2 bubbling transfer of large area graphene grown by chemical vapor deposition on Cu. Applied Physics Letters, 102 (2),
022101–022101-4. Copyright 2013, American Institute of Physics.)
TABLE 13.3
Summary of Graphene Transfer Methods
References Method Description Procedure
95–98 Wet transfer Polymer (PMMA) based, (1) Cu/graphene spin-coat polymer (2) cure in air at 120–180°C (3) etch Cu (4) pick up
high yield graphene/polymer layer with desired substrate (5) remove PMMA with heated acetone,
acetone vapor or annealing
92b Dry transfer PDMS based (1) Coat graphene/Cu with self-release layer (2) contact with elastomeric stamp (3) etch
Cu (4) contact with target substrate (5) remove self-release layer
100 Direct transfer Low yield (1) Place a drop of isopropanol on TEM grid which stays on the Cu foil with graphene,
on TEM grid (2) evaporate isopropanol. graphene draw from Cu foil towards amorphous TEM grid by
surface tension
101 Etch-free transfer Epoxy based, reusable Cu (1) Precisely measure the graphene-Cu adhesion energy (2) chose an epoxy/substrate pair
(3) remove graphene from its Cu catalyst without any etching
45–46,102 Bubble transfer Reusable Cu, minimized (1) Spin-coat PMMA (2) press Kapton tape on PMMA with a teflon roller (3) immerse
unintentional doping in DI water at 90°C (4) press the stack on to target substrate for 40 min at 140°C
(5) remove PMMA with acetone and RTA
99d Direct transfer Applicable to flexible (1) Place graphene/Cu in between target substrate and protective paper (2) put the stack in
via lamination substrates between two PET films (3) insert into lamination machine (4) remove PET films and
protective paper (5) etch Cu
(a) (b) (c)
50 nm 10 nm
(e) (f )
200 nm
(d)
Copper
PMMA-A
PMMA-G
Clean 50 nm 10 nm
Graphene area
FIGURE 13.9 Transfer cleanliness. (a–c) TEM images of graphene after PMMA-based wet transfer and two-step annealing under air
and H2/Ar flows at 200°C for 2 hours. (d) Schema representing Cu remaining, PMMA facing air (PMMMA-A), and graphene (PMMA-G).
(e, f) Same TEM images as (b, c) with shade coding showing the residues as indicated by the schematic in (d). (Reprinted with permis-
sion from Lin, Y.-C. et al., Graphene annealing: How clean can it be? Nano Letters, 12 (1), 414–419. Copyright 2011 American Chemical
Society.)
to clean graphene is high-temperature annealing. Lin et al. quartz substrate (Figure 13.10a).108 However, the continuity
have characterized the removal of these residues by high- and the quality of the resulting graphene were low with par-
temperature annealing.98 A 200°C anneal under H2 removed ticular regard to film wrinkling. Alternatively, thin-film Cu or
most of the adsorbed polymer facing air, while retaining poly- Ni on insulating substrates can be used as a medium through
mer (PMMA-G) and some Cu particles, which are in contact which to diffuse and catalyze carbon reactants for forming
with graphene (Figure 13.9). Researchers have found out a way graphene at the interface between the metal catalyst and the
of removing the remaining Cu remainder (or tiny precipitated substrate.107a,107b Su et al. have reported a moderate charge
particles) by replacing the common Cu etchant (FeCl3), which mobility of graphene (670 cm2 V−1 s−1) from this method
creates poorly water-soluble Cu(I) chloride, with (NH4)2S2O8 (Figure 13.10b).107a This method is promising but still requires
that produces water-soluble Cu(II) chloride.104 Unlike the removal of metal after the growth. To avoid this etch step, a
remaining Cu, the removal strategy of PMMA-G is not yet remote catalysis method has been demonstrated.70 CH4 is first
satisfactory.105 Use of an additional O2 annealing may improve flown over a Cu foil placed upstream in a tube furnace, which
the result through defect generation by partial oxidation of catalyzes CH4 dissociation at 1000°C. Next, the active carbon
graphene is a concern. Graphene annealing in H2 at higher species are desorbed from the Cu surface, in part, along with
temperatures such as 250°C does not yield better cleaning due the subliming Cu. These species then deposit onto the desti-
to the formation of covalent bonds between PMMA and gra- nation substrate while flowing downstream where graphene is
phene that already produce sp3 hybridization.98 Nevertheless, grown, yielding mobilities around 100–600 cm2 V−1 s−1.
there are reports about improved cleanliness at 400°C anneal- The most promising way of graphene synthesis on insula-
ing in H2.94b,106 While some processes yield transfers of CVD tors is a metal-free growth. Hwang et al. in 2010 demonstrated
graphene with almost similar quality as the mechanically C3H8-based CVD of graphene on sapphire (0001) at tempera-
exfoliated graphene, strong needs for better transfer and clean- tures above 1350°C.109b Later, a lower temperature growth was
ing methods are calling for immediate attention. achieved by employing alcohol as the carbon source, albeit
giving inferior electronic quality.109c More recently, a high
charge mobility of 3000 cm2 V−1 s−1 has been demonstrated
13.4.3 Direct Growth on Dielectrics
by Fanton et al. by using a CH4-based CVD on sapphire at
Direct synthesis of graphene on insulating substrates is the 1550°C (Figure 13.10c).109a High-growth temperatures could
ultimate solution to avoid transfer and its associated problems be a drawback, for they limit the available range of substrates.
described above. Several studies have demonstrated graphene- Molecular beam epitaxy from pure carbon filaments can be
like film deposition on SiO2,107 quartz,108 sapphire,109 boron used on biotite mica (001) surfaces heated to 500–1000°C.111
nitride,110 mica,111 and MgO.112 Early work by Ismach et al. Furthermore, MgO-based substrates can support the growth of
has showed that growth on thin-film Cu on quartz, followed graphene at even lower temperatures of 325°C using C2H2.112
by the sublimation of the Cu layer, left the graphene on the With limited choice of substrate and temperature, direct CVD
(a) Graphene (c) H2 CH4

Cu
phire
Sap H2 + CH4 Flow
H2
+
300
CH
1425 – 1200
– 17 sec
4
Fl
00°C
ow
Graphene
Cu
2 μm Cool down
(b)
GB
Cu grain
GB
SiO /Si
2
C-atoms diffused through GB Formation of graphene under Cu-film
FIGURE 13.10 Direct synthesis on insulators. (a) SEM image of graphene wrinkles near a Cu bead after growth. Graphene is in direct
contact with the substrate without a need for transfer. (b) Schematics for direct graphene synthesis on SiO2 by diffusion of the catalyzed
carbon precursor through Cu grain boundaries. (c) Graphene synthesis on sapphire by CVD at high temperatures. (Reprinted with permis-
sion from Ismach, A. et al., Direct chemical vapor deposition of graphene on dielectric surfaces. Nano Letters, 10 (5), 1542–1548. Copyright
2010 American Chemical Society; Reprinted with permission from Su, C.-Y. et al., Direct formation of wafer scale graphene thin layers on
insulating substrates by chemical vapor deposition. Nano Letters, 11 (9), 3612–3616. Copyright 2011 American Chemical Society; Reprinted
with permission from Fanton, M. A. et al., Characterization of graphene films and transistors grown on sapphire by metal-free chemical
vapor deposition. ACS Nano, 5 (10), 8062–8069. Copyright 2011 American Chemical Society.)
on insulators remains to be much improved before graphene of the resistance introduced by the grains (ca. 250 nm in
thus produced reaches the quality comparable with graphene diameter).92a However, intergrain resistivity is highly depen-
produced by Cu-based CVD methods. dent on the growth conditions. Figure 13.11a through c shows
the boundary conductivities for graphene grown in fast and
13.5 ELECTRONIC PROPERTIES OF slow growth regimes, yielding approximately 1 and 50 µm
in grain sizes. Contrary to the expectation, the smaller-grain
CVD-SYNTHESIZED GRAPHENE graphene presents better grain connectivity by an order of
Charge mobility and sheet resistance are two of the most magnitude. However, for large domains (more than 10 µm),
common indicators of the electronic quality of graphene. the larger-grain graphene showed a slightly more conductive
While the highest mobility values have been reported for film (Figure 13.11d). Furthermore, overlapping regions (i.e.,
mechanically exfoliated graphene, CVD-based graphene is at a thickness of 325 nm) for the larger-grain graphene can
becoming comparable, having room-temperature mobilities improve the boundary conductivity by around 45%.30b
exceeding several thousand cm2 V−1 s−1.30a,56a On the other While overlapping crystals may be beneficial in enhanc-
hand, in order to utilize graphene as transparent conduc- ing intergrain connectivity, wrinkles and ripples are unques-
tors, sheet resistances less than 10 Ω/ϒ are required.113 Thus tionably detrimental to the conductivity. Zhu et al. have
far, this requirement over large areas has been difficult to calculated that ripple heights of up to 6 nm may occur during
reach, with most reported values ranging between 100 and folding, resulting in boundary resistivity values of the order
1000 Ω/ϒ. This section introduces the growth-related factors of 200 Ω-µm regardless of the width of the fold.114 Here,
that define the mobility and sheet resistance values for CVD- diffusive transport and interlayer charge tunneling could
grown graphene. increase resistivity. Nanometer-scale ripples can also add
Grain boundaries and connectivity are critical in defining considerable resistances. A report by Ni et al. has claimed
the resistance associated with boundary scattering by dis- that increasingly perturbed Cu topographies can cause the
torted crystallinity. Huang et al. have reported an intergrain formation of quasiperiodic nanoripples.115 These ripples can
resistivity of 60 Ω-µm, corresponding to roughly one-third then act as sources for flexural phonon scattering, which
(c)
(a) Growth A
10
Growth A
σGB( 2e2 μm )
1
h
Growth C
Ld = 2.02 × 107/cm
R = 2.17 nm
0.1
1010 1011 1012
n (cm–2)
(d)
Growth C
(b) Growth C
1.4 D
(1 GB)
1.3
L
L
R/ρ
1.2
1.1
1.0 Growth A
0 2 4 6 8 10 12 14 16
L (μm)
FIGURE 13.11 Charge mobility and graphene domains. (a, b) False-color TEM images of graphene with different domain sizes (scale
bars: 1 µm). (c) Conductivity of the grain boundaries with respect to the charge carrier density, indicating the grain connectivity for small
domains is electronically better than the connectivity of larger domains. (d) Resistance with respect to channel length. For short channels,
small grains have better conductivity, while for distances above 10 times the smaller domain size, large-domain graphene becomes more
conductive due to less number of boundaries. (From Tsen, A. W. et al., Tailoring electrical transport across grain boundaries in polycrystal-
line graphene. Science 2012, 336 (6085), 1143–1146. Reprinted with permission of AAAS.)
limits the minimum monolayer sheet resistance to 80 Ω/ϒ Initial reports on large-scale graphene growth have been
for a carrier density of 1012 cm−2. To avoid these ripples, based on growths on Cu foils inside large-diameter tube fur-
strain engineering can be employed,116 or alternatively, dop- naces, with foils up to 20 cm having been demonstrated cou-
ing can increase the charge carrier density by enhancing pled with roll-to-roll transfer processes for coating plastic
the conductivity without the need of strain engineering.113 substrates.94a Later, a small roll-to-roll system has been fitted
In contrast to intentional graphene doping to minimize the into a 25-mm-diameter quartz tube to produce low-quality
sheet resistance, substrate-induced doping could be detri- graphene on a narrow but continuous Cu foil.121 Yamada et al.
mental. A careful selection or treatment of the substrate with have employed a more advanced rolling system in a microwave
self-assembled monolayers may be one possible bypassing plasma oven to obtain graphene, albeit with low crystallin-
route. So far, boron nitride117 and self-assembled monolayer ity (at a width of 294 mm) at temperatures below 400°C and
dielectrics have been demonstrated for effective passivation with rather high sheet resistances of 10 k–700 kΩ/ϒ, which
of SiO2 substrates.118 is perhaps well suited for flexible transparent conductor appli-
cations.122 An increasingly advanced roll-to-roll system was
reported by Kobayashi et al.31 Here, a 100-m-long sheet of
13.6 SCALING-UP SCHEMES graphene, with a charge mobility of 900 cm2 V−1 s−1, has been
Presently, meter-scale production of graphene can only be grown and transferred onto flexible plastic substrates by a series
possible by two methods: CVD growth and liquid-phase exfo- of roll-to-roll processes, as shown in Figure 13.12a through c.
liation. The latter can be ruled out for applications requiring Meanwhile, Tao et al. have demonstrated better-quality gra-
higher electronic conductivity/mobililty because of low flake phene, uniform over a 150-mm wafer, with Raman G′ peak
concentrations (on the order of 0.1 mg mL −1),119 very low widths of 25–35 cm−1 and a charge mobility of 4900 cm2 V−1 s−1
truly monolayer flake yields (on the order of 1%),120 and most (Figure 13.12d and e).60 They employed a Cu thin film on an
importantly, the inherent difficulties in achieving continuous SiO2/Si wafer, which restricted the rolling process but enhanced
monolayers. Hence, CVD, in particular, the Cu-based growth, the material quality.59 Work is still ongoing throughout the
is most probably the only promising and viable method for industry to improve the quality and scalability of roll-to-roll
scaling-up the graphene synthesis currently. and large-scale wafer-based CVD growth schemes.
(a) (b) (d)

CVD
Gas out Spray etching 40 T
35
30 L C R
Cu foil Water
Counts (%)
25
J CuCl2 20 B
Winder CH4
15
PET/epoxy/gra./Cu 10
5
Gra./Cu H2+CH4 PET/epoxy/gra. 0
20 25 30 35 40
Pl FWHM2D (cm–1)
(c) (e)
2D
UV irradiation
Intensity (a.u.)
G
T
R
L
Gra./Cu C
Bonding B
1200 1700 2200 2700
PET/epox/gra./Cu Raman shift (cm–1)
PET Gravure coating
FIGURE 13.12 Scaling-up schemes. (a–c) Roll-to-roll CVD synthesis on Cu foils and transfer onto flexible PET support layers. (Reprinted
with permission from Kobayashi, T. et al., Production of a 100-m-long high-quality graphene transparent conductive film by roll-to-roll
chemical vapor deposition and transfer process. Applied Physics Letters, 102 (2), 023112-4. Copyright 2013, American Institute of Physics.)
(d, e) Uniform growth of high-quality monolayer graphene on a 150-mm wafer coated with a Cu film. (Reprinted with permission from Tao,
L. et al., Uniform wafer-scale chemical vapor deposition of graphene on evaporated Cu (111) film with quality comparable to exfoliated
monolayer. The Journal of Physical Chemistry C, 116 (45), 24068–24074. Copyright 2012 American Chemical Society.)
13.7 SUMMARY REFERENCES
Chemical vapor deposition on Cu has become the most 1. Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.;
convenient and popular synthesis method to produce high- Zhang, Y.; Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A.,
quality graphene. Unlike other transition metals, Cu does not Electric field effect in atomically thin carbon films. Science
2004, 306 (5696), 666–669.
dissolve carbon in bulk and thus supports, under the correct 2. (a) Bolotin, K. I.; Sikes, K. J.; Jiang, Z.; Klima, M.; Fudenberg,
growth conditions, the self-limiting growth of a monolayer G.; Hone, J.; Kim, P.; Stormer, H. L., Ultrahigh electron mobil-
graphene with low wrinkling and rippling. Advances in the ity in suspended graphene. Solid State Communications 2008,
understanding of the fundamental growth mechanisms have 146 (9–10), 351–355; (b) Du, X.; Skachko, I.; Barker, A.; Andrei,
permitted the synthesis of high-quality polycrystalline gra- E. Y., Approaching ballistic transport in suspended graphene.
phene with larger and better-connected grains. Meanwhile, Nature Nanotechnology 2008, 3 (8), 491–495; (c) Morozov, S.;
optimization of the transfer and cleaning procedures has Novoselov, K.; Katsnelson, M.; Schedin, F.; Elias, D.; Jaszczak,
J.; Geim, A., Giant intrinsic carrier mobilities in graphene and
resulted in graphene with charge mobilities close to that of a its bilayer. Physical Review Letters 2008, 100 (1), 016602;
mechanically exfoliated counterpart, at scalable dimensions. (d) Wang, L.; Meric, I.; Huang, P. Y.; Gao, Q.; Gao, Y.; Tran,
Direct deposition of the monolayer graphene on insulating H.; Taniguchi, T. et al., One-dimensional electrical contact to a
substrates promises the ultimate progress in the cleanli- two-dimensional material. Science 2013, 342 (6158), 614–617.
ness and electronic exploitation of graphene, removing the 3. (a) Geim, A. K.; Novoselov, K. S., The rise of graphene. Nature
necessity of using sacrificial metals and transfer processes. Materials 2007, 6 (3), 183–191; (b) Neto, A. C.; Guinea, F.;
However, direct deposition is still in its infancy, and sig- Peres, N.; Novoselov, K.; Geim, A., The electronic proper-
ties of graphene. Reviews of Modern Physics 2009, 81 (1),
nificant progress is needed for it to outperform the current 109; (c) Wallace, P. R., The band theory of graphite. Physical
Cu-based synthesis protocols. In prospect, mass production Review 1947, 71 (9), 622–634.
of the monolayer graphene is likely to be realized in the near 4. Bao, Q.; Zhang, H.; Wang, Y.; Ni, Z.; Yan, Y.; Shen, Z. X.;
future by roll-to-roll processes and handling meters-long Cu Loh, K. P.; Tang, D. Y., Atomic-layer graphene as a saturable
foils as the growth medium and flexible plastics as the desti- absorber for ultrafast pulsed lasers. Advanced Functional
nation substrate. Materials 2009, 19 (19), 3077–3083.
5. Bonaccorso, F.; Sun, Z.; Hasan, T.; Ferrari, A., Graphene 18. Stankovich, S.; Dikin, D. A.; Dommett, G. H.; Kohlhaas, K.
photonics and optoelectronics. Nature Photonics 2010, 4 (9), M.; Zimney, E. J.; Stach, E. A.; Piner, R. D.; Nguyen, S. T.;
611–622. Ruoff, R. S., Graphene-based composite materials. Nature
6. Bunch, J. S.; van der Zande, A. M.; Verbridge, S. S.; Frank, 2006, 442 (7100), 282–286.
I. W.; Tanenbaum, D. M.; Parpia, J. M.; Craighead, H. G.; 19. Hernandez, Y.; Nicolosi, V.; Lotya, M.; Blighe, F. M.; Sun,
McEuen, P. L., Electromechanical resonators from graphene Z.; De, S.; McGovern, I. T. et al., High-yield production of
sheets. Science 2007, 315 (5811), 490–493. graphene by liquid-phase exfoliation of graphite. Nature
7. Balandin, A. A.; Ghosh, S.; Bao, W.; Calizo, I.; Teweldebrhan, Nanotechnology 2008, 3 (9), 563–568.
D.; Miao, F.; Lau, C. N., Superior thermal conductivity of sin- 20. Eda, G.; Fanchini, G.; Chhowalla, M., Large-area ultrathin
gle-layer graphene. Nano Letters 2008, 8 (3), 902–907. films of reduced graphene oxide as a transparent and flex-
8. (a) Becerril, H. A.; Mao, J.; Liu, Z.; Stoltenberg, R. M.; Bao, Z.; ible electronic material. Nature Nanotechnology 2008, 3 (5),
Chen, Y., Evaluation of solution-processed reduced graphene 270–274.
oxide films as transparent conductors. ACS Nano 2008, 2 (3), 21. (a) Berger, C.; Song, Z.; Li, X.; Wu, X.; Brown, N.; Naud,
463–470; (b) Wassei, J. K.; Kaner, R. B., Graphene, a promising C.; Mayou, D. et al., Electronic confinement and coherence
transparent conductor. Materials Today 2010, 13 (3), 52–59. in patterned epitaxial graphene. Science 2006, 312 (5777),
9. Areshkin, D. A.; White, C. T., Building blocks for integrated 1191–1196; (b) De Heer, W. A.; Berger, C.; Wu, X.; First, P. N.;
graphene circuits. Nano Letters 2007, 7 (11), 3253–3259. Conrad, E. H.; Li, X.; Li, T.; Sprinkle, M.; Hass, J.; Sadowski,
10. (a) Lin, Y.-M.; Dimitrakopoulos, C.; Jenkins, K. A.; Farmer, M. L., Epitaxial graphene. Solid State Communications 2007,
D. B.; Chiu, H.-Y.; Grill, A.; Avouris, P., 100-GHz transis- 143 (1), 92–100; (c) Emtsev, K. V.; Bostwick, A.; Horn, K.;
tors from wafer-scale epitaxial graphene. Science 2010, 327 Jobst, J.; Kellogg, G. L.; Ley, L.; McChesney, J. L. et al.,
(5966), 662; (b) Lin, Y.-M.; Jenkins, K. A.; Valdes-Garcia, Towards wafer-size graphene layers by atmospheric pressure
A.; Small, J. P.; Farmer, D. B.; Avouris, P., Operation of gra- graphitization of silicon carbide. Nature Materials 2009, 8 (3),
phene transistors at gigahertz frequencies. Nano Letters 2008, 203–207; (d) Pallecchi, E.; Lafont, F.; Cavaliere, V.; Schopfer,
9 (1), 422–426; (c) Moon, J.; Curtis, D.; Hu, M.; Wong, D.; F.; Mailly, D.; Poirier, W.; Ouerghi, A., High electron mobil-
McGuire, C.; Campbell, P.; Jernigan, G.; Tedesco, J.; VanMil, ity in epitaxial graphene on 4H-SiC(0001) via post-growth
B.; Myers-Ward, R., Epitaxial-graphene RF field-effect tran- annealing under hydrogen. Scientific Reports 2014, 4, 04558.
sistors on Si-face 6H-SiC substrates. Electron Device Letters, 22. Lin, Y.-M.; Valdes-Garcia, A.; Han, S.-J.; Farmer, D. B.;
IEEE 2009, 30 (6), 650–652. Meric, I.; Sun, Y.; Wu, Y.; Dimitrakopoulos, C.; Grill, A.;
11. Stoller, M. D.; Park, S.; Zhu, Y.; An, J.; Ruoff, R. S., Graphene- Avouris, P., Wafer-scale graphene integrated circuit. Science
based ultracapacitors. Nano Letters 2008, 8 (10), 3498–3502. 2011, 332 (6035), 1294–1297.
12. (a) Wang, C.; Li, D.; Too, C. O.; Wallace, G. G., Electrochemical 23. (a) Cai, J.; Ruffieux, P.; Jaafar, R.; Bieri, M.; Braun, T.;
properties of graphene paper electrodes used in lithium bat- Blankenburg, S.; Muoth, M.; Seitsonen, A. P.; Saleh, M.; Feng,
teries. Chemistry of Materials 2009, 21 (13), 2604–2606; X., Atomically precise bottom-up fabrication of graphene
(b) Xiao, J.; Mei, D.; Li, X.; Xu, W.; Wang, D.; Graff, G. nanoribbons. Nature 2010, 466 (7305), 470–473; (b) Ruffieux,
L.; Bennett, W. D. et al., Hierarchically porous graphene as P.; Cai, J. M.; Plumb, N. C.; Patthey, L.; Prezzi, D.; Ferretti, A.;
a lithium–air battery electrode. Nano Letters 2011, 11 (11), Molinari, E. et al., Electronic structure of atomically precise
5071–5078. graphene nanoribbons. ACS Nano 2012, 6 (8), 6930–6935.
13. (a) Koenig, S. P.; Wang, L.; Pellegrino, J.; Bunch, J. S., 24. (a) Coraux, J.; N’Diaye, T.; Engler, M.; Busse, C.; Wall, D.;
Selective molecular sieving through porous graphene. Nature Buckanie, N.; zu Heringdorf, F. J. M.; van Gaste, R.; Poelsema,
Nanotechnology 2012, 7 (11), 728–732; (b) O’Hern, S. C.; B.; Michely, T., Growth of graphene on Ir (111). New Journal
Boutilier, M. S. H.; Idrobo, J. C.; Song, Y.; Kong, J.; Laoui, T.; of Physics 2009, 11, 023006; (b) van Gastel, R.; N’Diaye, A. T.;
Atieh, M.; Karnik, R., Selective ionic transport through tun- Wall, D.; Coraux, J.; Busse, C.; Buckanie, N. M.; Heringdorf,
able subnanometer pores in single-layer graphene membranes. F. J. M. z.; von Hoegen, M. H.; Michely, T.; Poelsema, B.,
Nano Letters 2014, 14 (3), 1234–1241; (c) Celebi, K.; Buchheim, Selecting a single orientation for millimeter sized graphene
J.; Wyss, R. M.; Droudian, A.; Gasser, P.; Shorubalko, I.; Kye, sheets. Applied Physics Letters 2009, 95 (12), 121901–121903.
J. I.; Lee, C.; Park, H. G., Ultimate permeation across atomi- 25. (a) Loginova, E.; Bartelt, N. C.; Feibelman, P. J.; McCarty,
cally thin porous graphene. Science 2014, 344 (6181), 289–292. K. F., Evidence for graphene growth by C cluster attachment.
14. Suk, M. E.; Aluru, N. R., Water transport through ultrathin New Journal of Physics 2008, 10 (9), 093026; (b) Sutter, P. W.;
graphene. The Journal of Physical Chemistry Letters 2010, Flege, J.-I.; Sutter, E. A., Epitaxial graphene on ruthenium.
1 (10), 1590–1594. Nature Materials 2008, 7 (5), 406–411.
15. (a) Chen, S.; Brown, L.; Levendorf, M.; Cai, W.; Ju, S.-Y.; 26. (a) Fujita, T.; Kobayashi, W.; Oshima, C., Novel structures
Edgeworth, J.; Li, X.; Magnuson, C. W.; Velamakanni, A.; of carbon layers on a Pt(111) surface. Surface and Interface
Piner, R. D., Oxidation resistance of graphene-coated Cu and Analysis 2005, 37 (2), 120–123; (b) Land, T. A.; Michely, T.;
Cu/Ni alloy. ACS Nano 2011, 5 (2), 1321–1327; (b) Sutter, E.; Behm, R. J.; Hemminger, J. C.; Comsa, G., STM investiga-
Albrecht, P.; Camino, F. E.; Sutter, P., Monolayer graphene tion of single layer graphite structures produced on Pt(111) by
as ultimate chemical passivation layer for arbitrarily shaped hydrocarbon decomposition. Surface Science 1992, 264 (3),
metal surfaces. Carbon 2010, 48 (15), 4414–4420. 261–270.
16. Kamat, P. V., Graphene-based nanoassemblies for energy 27. Kwon, S.-Y.; Ciobanu, C. V.; Petrova, V.; Shenoy, V. B.;
conversion. The Journal of Physical Chemistry Letters 2011, Bareño, J.; Gambin, V.; Petrov, I.; Kodambaka, S., Growth of
2 (3), 242–251. semiconducting graphene on palladium. Nano Letters 2009, 9
17. Ling, X.; Xie, L. M.; Fang, Y.; Xu, H.; Zhang, H. L.; Kong, (12), 3985–3990.
J.; Dresselhaus, M. S.; Zhang, J.; Liu, Z. F., Can graphene be 28. (a) Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim,
used as a substrate for Raman enhancement? Nano Letters K. S.; Ahn, J.-H.; Kim, P.; Choi, J.-Y.; Hong, B. H., Large-scale
2010, 10 (2), 553–561. pattern growth of graphene films for stretchable transparent
electrodes. Nature 2009, 457 (7230), 706–710; (b) Reina, A.; 40. Wang, B.; Caffio, M.; Bromley, C.; Früchtl, H.; Schaub, R.,
Jia, X.; Ho, J.; Nezich, D.; Son, H.; Bulovic, V.; Dresselhaus, Coupling epitaxy, chemical bonding, and work function at the
M.; Kong, J., Large area, few-layer graphene films on arbitrary local scale in transition metal-supported graphene. ACS Nano
substrates by chemical vapor deposition. Nano Letters 2009, 9 2010, 4 (10), 5773–5782.
(1), 30–35. 41. (a) Weatherup, R. S.; Bayer, B. C.; Blume, R.; Ducati, C.;
29. (a) Li, X.; Cai, W.; An, J.; Kim, S.; Nah, J.; Yang, D.; Piner, R. Baehtz, C.; Schlögl, R.; Hofmann, S., In situ characterization
et al., Large-area synthesis of high-quality and uniform gra- of alloy catalysts for low-temperature graphene growth. Nano
phene films on copper foils. Science 2009, 324 (5932), 1312– Letters 2011, 11 (10), 4154–4160; (b) Lin, T.; Huang, F. Q.;
1314; (b) Li, X.; Cai, W.; Colombo, L.; Ruoff, R., Evolution Wan, D.; Bi, H.; Xie, X.; Jiang, M., Self-regulating homog-
of graphene growth on Cu and Ni studied by carbon isotope enous growth of high-quality graphene on Co-Cu compos-
labeling. Nano Letters 2009, 9, 4268–4272. ite substrate for layer control. Nanoscale 2013; (c) Liu, N.;
30. (a) Li, X.; Magnuson, C. W.; Venugopal, A.; An, J.; Suk, J. Fu, L.; Dai, B.; Yan, K.; Liu, X.; Zhao, R.; Zhang, Y.; Liu,
W.; Han, B.; Borysiak, M. et al., Graphene films with large Z., Universal segregation growth approach to wafer-size
domain size by a two-step chemical vapor deposition pro- graphene from non-noble metals. Nano Letters 2010, 11 (1),
cess. Nano Letters 2010, 10 (11), 4328–4334; (b) Tsen, A. 297–303.
W.; Brown, L.; Levendorf, M. P.; Ghahari, F.; Huang, P. Y.; 42. (a) Liao, Q.; Zhang, H.; Wu, K.; Li, H.; Bao, S.; He, P.,
Havener, R. W.; Ruiz-Vargas, C. S.; Muller, D. A.; Kim, P.; Nucleation and growth of monodispersed cobalt nanoclusters
Park, J., Tailoring electrical transport across grain boundar- on graphene moire on Ru (0001). Nanotechnology 2011, 22
ies in polycrystalline graphene. Science 2012, 336 (6085), (12), 125303; (b) N’Diaye, A. T.; Bleikamp, S.; Feibelman, P.
1143–1146; (c) Hao, Y. F.; Bharathi, M. S.; Wang, L.; Liu, Y. J.; Michely, T., Two-dimensional Ir cluster lattice on a gra-
Y.; Chen, H.; Nie, S.; Wang, X. H. et al., The role of surface phene moire on Ir (111). Physical Review Letters 2006, 97
oxygen in the growth of large single-crystal graphene on cop- (21), 215501.
per. Science 2013, 342 (6159), 720–723; (d) Lee, J. H.; Lee, 43. Mattevi, C.; Kim, H.; Chhowalla, M., A review of chemical
E. K.; Joo, W. J.; Jang, Y.; Kim, B. S.; Lim, J. Y.; Choi, S. H. vapour deposition of graphene on copper. Journal of Materials
et al., Wafer-scale growth of single-crystal monolayer gra- Chemistry 2011, 21 (10), 3324–3334.
phene on reusable hydrogen-terminated germanium. Science 44. Hansen, M.; Shunk, F. A.; Elliott, R. P.; Anderko, K.,
2014, 344 (6181), 286–289; (e) Zhou, H. L.; Yu, W. J.; Liu, Constitution of Binary Alloys. McGraw-Hill, New York, 1965.
L. X.; Cheng, R.; Chen, Y.; Huang, X. Q.; Liu, Y.; Wang, Y.; 45. de la Rosa, C. J. L.; Sun, J.; Lindvall, N.; Cole, M. T.; Nam,
Huang, Y.; Duan, X. F., Chemical vapour deposition growth Y.; Loffler, M.; Olsson, E.; Teo, K. B.; Yurgens, A., Frame
of large single crystals of monolayer and bilayer graphene. assisted H2O electrolysis induced H2 bubbling transfer of large
Nature Communications 2013, 4, 2096. area graphene grown by chemical vapor deposition on Cu.
31. Kobayashi, T.; Bando, M.; Kimura, N.; Shimizu, K.; Kadono, Applied Physics Letters 2013, 102 (2), 022101–022101-4.
K.; Umezu, N.; Miyahara, K. et al., Production of a 100-m-long 46. Wang, Y.; Zheng, Y.; Xu, X.; Dubuisson, E.; Bao, Q.; Lu, J.;
high-quality graphene transparent conductive film by roll-to- Loh, K. P., Electrochemical delamination of CVD-grown gra-
roll chemical vapor deposition and transfer process. Applied phene film: Toward the recyclable use of copper catalyst. ACS
Physics Letters 2013, 102 (2), 023112-4. Nano 2011, 5 (12), 9927–9933.
32. Morgan, A. E.; Somorjai, G. A., Low energy electron diffrac- 47. Qi, M.; Ren, Z.; Jiao, Y.; Zhou, Y.; Xu, X.; Li, W.; Li, J.;
tion studies of gas adsorption on the platinum (100) single Zheng, X.; Bai, J., Hydrogen kinetics on scalable graphene
crystal surface. Surface Science 1968, 12 (3), 405–425. growth by atmospheric pressure chemical vapor deposition
33. Blakely, J.; Kim, J.; Potter, H., Segregation of carbon to the with acetylene. The Journal of Physical Chemistry C 2013,
(100) surface of nickel. Journal of Applied Physics 1970, 41 117 (27), 14348–14353.
(6), 2693–2697. 48. Celebi, K.; Cole, M. T.; Choi, J. W.; Wyczisk, F.; Legagneux,
34. May, J. W., Platinum surface LEED rings. Surface Science P.; Rupesinghe, N.; Robertson, J.; Teo, K. B. K.; Park, H. G.,
1969, 17 (1), 267–270. Evolutionary kinetics of graphene formation on copper. Nano
35. (a) Landau, L., Zur Theorie der phasenumwandlungen II. Letters 2013, 13 (3), 967–974.
Physikalische Zeitschrift der Sowjetunion 1937, 11, 26–35; (b) 49. Choi, J.-H.; Li, Z.; Cui, P.; Fan, X.; Zhang, H.; Zeng, C.;
Mermin, N. D., Crystalline order in two dimensions. Physical Zhang, Z., Drastic reduction in the growth temperature of
Review 1968, 176 (1), 250; (c) Peierls, R. E., Quelques propri- graphene on copper via enhanced London dispersion force.
etes typiques des corpses solides. Annales de l’Institut Henri Scientific Reports 2013, 3, 01925.
Poincaré 1935, 5, 177–222. 50. (a) Sun, Z.; Yan, Z.; Yao, J.; Beitler, E.; Zhu, Y.; Tour, J. M.,
36. Somani, P. R.; Somani, S. P.; Umeno, M., Planer nano-gra- Growth of graphene from solid carbon sources. Nature 2010,
phenes from camphor by CVD. Chemical Physics Letters 468 (7323), 549–552; (b) Li, Z.; Wu, P.; Wang, C.; Fan, X.;
2006, 430 (1–3), 56–59. Zhang, W.; Zhai, X.; Zeng, C.; Li, Z.; Yang, J.; Hou, J., Low-
37. Kondo, D.; Sato, S.; Yagi, K.; Harada, N.; Sato, M.; Nihei, temperature growth of graphene by chemical vapor deposition
M.; Yokoyama, N., Low-temperature synthesis of graphene using solid and liquid carbon sources. ACS Nano 2011, 5 (4),
and fabrication of top-gated field effect transistors without 3385–3390.
using transfer processes. Applied Physics Express 2010, 3 51. Kidambi, P. R.; Ducati, C.; Dlubak, B.; Gardiner, D.;
(2), 5102. Weatherup, R. S.; Martin, M.-B.; Seneor, P.; Coles, H.;
38. Oznuluer, T.; Pince, E.; Polat, E. O.; Balci, O.; Salihoglu, O.; Hofmann, S., The parameter space of graphene chemi-
Kocabas, C., Synthesis of graphene on gold. Applied Physics cal vapor deposition on polycrystalline Cu. The Journal of
Letters 2011, 98 (18), 183101–183103. Physical Chemistry C 2012, 116 (42), 22492–22501.
39. John, R.; Ashokreddy, A.; Vijayan, C.; Pradeep, T., Single- 52. Zhang, B.; Lee, W. H.; Piner, R.; Kholmanov, I.; Wu, Y.; Li,
and few-layer graphene growth on stainless steel substrates H.; Ji, H.; Ruoff, R. S., Low-temperature chemical vapor
by direct thermal chemical vapor deposition. Nanotechnology deposition growth of graphene from toluene on electropol-
2011, 22 (16), 165701. ished copper foils. ACS Nano 2012, 6 (3), 2471–2476.
53. Han, G. H.; Güneş, F.; Bae, J. J.; Kim, E. S.; Chae, S. J.; Shin, 71. (a) Kuwana, K.; Li, T.; Saito, K., Gas-phase reactions during
H.-J.; Choi, J.-Y.; Pribat, D.; Lee, Y. H., Influence of copper CVD synthesis of carbon nanotubes: Insights via numerical
morphology in forming nucleation seeds for graphene growth. experiments. Chemical Engineering Science 2006, 61 (20),
Nano Letters 2011, 11 (10), 4144–4148. 6718–6726; (b) Robertson, J.; Zhong, G.; Esconjauregui, S.;
54. Kim, H.; Mattevi, C.; Calvo, M. R.; Oberg, J. C.; Artiglia, Zhang, C.; Fouquet, M.; Hofmann, S., Chemical vapor deposi-
L.; Agnoli, S.; Hirjibehedin, C. F.; Chhowalla, M.; Saiz, E., tion of carbon nanotube forests. Physica Status Solidi (b) 2012,
Activation energy paths for graphene nucleation and growth 249 (12), 2315–2322; (c) Youn, S. K.; Frouzakis, C. E.; Gopi,
on Cu. ACS Nano 2012, 6 (4), 3614–3623. B. P.; Robertson, J.; Teo, K. B. K.; Park, H. G., Temperature
55. Levendorf, M. P.; Ruiz-Vargas, C. S.; Garg, S.; Park, J., gradient chemical vapor deposition of vertically aligned car-
Transfer-free batch fabrication of single layer graphene tran- bon nanotubes. Carbon 2013, 54 (0), 343–352.
sistors. Nano Letters 2009, 9 (12), 4479–4483. 72. (a) Hu, B.; Ago, H.; Ito, Y.; Kawahara, K.; Tsuji, M.; Magome,
56. (a) Yan, Z.; Lin, J.; Peng, Z.; Sun, Z.; Zhu, Y.; Li, L.; Xiang, C.; E.; Sumitani, K.; Mizuta, N.; Ikeda, K.-i.; Mizuno, S., Epitaxial
Samuel, E. L.; Kittrell, C.; Tour, J. M., Toward the synthesis growth of large-area single-layer graphene over Cu(111)/sap-
of wafer-scale single-crystal graphene on copper foils. ACS phire by atmospheric pressure CVD. Carbon 2012, 50 (1),
Nano 2012, 6 (10), 9110–9117; (b) Nie, S.; Wu, W.; Xing, S.; 57–65; (b) Riikonen, S.; Krasheninnikov, A. V.; Halonen,
Yu, Q.; Bao, J.; Pei, S.-s.; McCarty, K. F., Growth from below: L.; Nieminen, R. M., The role of stable and mobile carbon
Bilayer graphene on copper by chemical vapor deposition. adspecies in copper-promoted graphene growth. Journal of
New Journal of Physics 2012, 14 (9), 093028. Physical Chemistry C 2012, 116 (9), 5802–5809.
57. Wood, J. D.; Schmucker, S. W.; Lyons, A. S.; Pop, E.; Lyding, 73. Bhaviripudi, S.; Jia, X.; Dresselhaus, M. S.; Kong, J., Role of
J. W., Effects of polycrystalline Cu substrate on graphene kinetic factors in chemical vapor deposition synthesis of uni-
growth by chemical vapor deposition. Nano Letters 2011, 11 form large area graphene using copper catalyst. Nano Letters
(11), 4547–4554. 2010, 10 (10), 4128–4133.
58. Gao, L.; Guest, J. R.; Guisinger, N. P., Epitaxial graphene on 74. (a) Au, C.-T.; Ng, C.-F.; Liao, M.-S., Methane dissociation and
Cu(111). Nano Letters 2010, 10 (9), 3512–3516. syngas formation on Ru, Os, Rh, Ir, Pd, Pt, Cu, Ag, and Au:
59. Tao, L.; Lee, J.; Chou, H.; Holt, M.; Ruoff, R. S.; Akinwande, A theoretical study. Journal of Catalysis 1999, 185 (1), 12–22;
D., Synthesis of high quality monolayer graphene at reduced (b) Gajewski, G.; Pao, C.-W., Ab initio calculations of the reac-
temperature on hydrogen-enriched evaporated copper (111) tion pathways for methane decomposition over the Cu (111) sur-
films. ACS Nano 2012, 6 (3), 2319–2325. face. The Journal of Chemical Physics 2011, 135 (6), 064707–9.
60. Tao, L.; Lee, J.; Holt, M.; Chou, H.; McDonnell, S. J.; Ferrer, 75. Liu, L.; Zhou, H.; Cheng, R.; Chen, Y.; Lin, Y.-C.; Qu, Y.; Bai,
D. A.; Babenco, M. G. et al., Uniform wafer-scale chemical J. et al., A systematic study of atmospheric pressure chemical
vapor deposition of graphene on evaporated Cu (111) film with vapor deposition growth of large-area monolayer graphene.
quality comparable to exfoliated monolayer. The Journal of Journal of Materials Chemistry 2012, 22 (4), 1498–1503.
Physical Chemistry C 2012, 116 (45), 24068–24074. 76. Mehdipour, H.; Ostrikov, K., Kinetics of low-pressure, low-
61. Zhang, W.; Wu, P.; Li, Z.; Yang, J., First-principles thermo- temperature graphene growth: Toward single-layer, single-
dynamics of graphene growth on Cu surfaces. Journal of crystalline structure. ACS Nano 2012, 6 (11), 10276–10286.
Physical Chemistry C 2011, 115 (36), 17782–17787. 77. (a) Plonka, A., 4 Dispersive kinetics. Annual Reports Section
62. Luo, Z.; Kim, S.; Kawamoto, N.; Rappe, A. M.; Johnson, A. “C” (Physical Chemistry) 2001, 97, 91–147; (b) Siebrand, W.;
T. C., Growth mechanism of hexagonal-shape graphene flakes Wildman, T. A., Dispersive kinetics: A structural approach to
with zigzag edges. ACS Nano 2011, 5 (11), 9154–9160. nonexponential processes in disordered media. Accounts of
63. Treier, M.; Pignedoli, C. A.; Laino, T.; Rieger, R.; Müllen, K.; Chemical Research 1986, 19 (8), 238–243; (c) Skrdla, P. J.,
Passerone, D.; Fasel, R., Surface-assisted cyclodehydrogena- Roles of nucleation, denucleation, coarsening, and aggrega-
tion provides a synthetic route towards easily processable and tion kinetics in nanoparticle preparations and neurological
chemically tailored nanographenes. Nature Chemistry 2011, 3 disease. Langmuir 2012, 28 (10), 4842–4857.
(1), 61–67. 78. Chen, Y.; Vlachos, D. G., Hydrogenation of ethylene and
64. Jia, C.; Jiang, J.; Gan, L.; Guo, X., Direct optical characteriza- dehydrogenation and hydrogenolysis of ethane on Pt(111) and
tion of graphene growth and domains on growth substrates. Pt(211): A density functional theory study. Journal of Physical
Scientific Reports 2012, 2, 00707. Chemistry C 2010, 114 (11), 4973–4982.
65. Robinson, V. N. E.; Robins, J. L., Nucleation kinetics of gold 79. Wofford, J. M.; Nie, S.; McCarty, K. F.; Bartelt, N. C.; Dubon, O.
deposited onto UHV cleaved surfaces of NaCl and KBr. Thin D., Graphene Islands on Cu foils: The interplay between shape,
Solid Films 1974, 20 (1), 155–175. orientation, and defects. Nano Letters 2010, 10 (12), 4890–4896.
66. McCarty, K. F.; Feibelman, P. J.; Loginova, E.; Bartelt, N. C., 80. Vlassiouk, I.; Regmi, M.; Fulvio, P.; Dai, S.; Datskos, P.; Eres,
Kinetics and thermodynamics of carbon segregation and gra- G.; Smirnov, S., Role of hydrogen in chemical vapor deposi-
phene growth on Ru (0001). Carbon 2009, 47 (7), 1806–1813. tion growth of large single-crystal graphene. ACS Nano 2011,
67. Nie, S.; Wofford, J. M.; Bartelt, N. C.; Dubon, O. D.; McCarty, 5 (7), 6069–6076.
K. F., Origin of the mosaicity in graphene grown on Cu (111). 81. Choubak, S.; Biron, M.; Levesque, P. L.; Martel, R.;
Physical Review B 2011, 84 (15), 155425. Desjardins, P., No graphene etching in purified hydrogen. The
68. Avrami, M., Kinetics of phase change. I general theory. Journal of Physical Chemistry Letters 2013, 4 (7), 1100–1103.
Journal of Chemical Physics 1939, 7 (12), 1103–1112. 82. Jacobberger, R. M.; Arnold, M. S., Graphene growth dynam-
69. Kim, H.; Saiz, E.; Chhowalla, M.; Mattevi, C., Modeling of ics on epitaxial copper thin films. Chemistry of Materials
the self-limited growth in catalytic chemical vapor deposition 2013, 25 (6), 871–877.
of graphene. New Journal of Physics 2013, 15 (5), 053012. 83. Yu, Q.; Jauregui, L. A.; Wu, W.; Colby, R.; Tian, J.; Su, Z.;
70. Teng, P.-Y.; Lu, C.-C.; Akiyama-Hasegawa, K.; Lin, Y.-C.; Cao, H.; Liu, Z.; Pandey, D.; Wei, D., Control and character-
Yeh, C.-H.; Suenaga, K.; Chiu, P.-W., Remote catalyzation for ization of individual grains and grain boundaries in graphene
direct formation of graphene layers on oxides. Nano Letters grown by chemical vapour deposition. Nature Materials 2011,
2012, 12 (3), 1379–1384. 10 (6), 443–449.
84. Robertson, A. W.; Warner, J. H., Hexagonal single crystal and conductive interconnected graphene networks grown by
domains of few-layer graphene on copper foils. Nano Letters chemical vapour deposition. Nature Materials 2011, 10 (6),
2011, 11 (3), 1182–1189. 424–428.
85. Nie, S.; Walter, A. L.; Bartelt, N. C.; Starodub, E.; Bostwick, A.; 97. Jiao, L.; Zhang, L.; Wang, X.; Diankov, G.; Dai, H., Narrow
Rotenberg, E.; McCarty, K. F., Growth from below: Graphene graphene nanoribbons from carbon nanotubes. Nature 2009,
bilayers on Ir(111). ACS Nano 2011, 5 (3), 2298–2306. 458 (7240), 877–880.
86. Fang, W.; Hsu, A. L.; Caudillo, R.; Song, Y.; Birdwell, A. G.; 98. Lin, Y.-C.; Lu, C.-C.; Yeh, C.-H.; Jin, C.; Suenaga, K.; Chiu,
Zakar, E.; Kalbac, M.; Dubey, M.; Palacios, T.; Dresselhaus, P.-W., Graphene annealing: How clean can it be? Nano Letters
M. S., Rapid identification of stacking orientation in isoto- 2011, 12 (1), 414–419.
pically labeled chemical-vapor grown bilayer graphene by 99. (a) Chen, X.-D.; Liu, Z.-B.; Zheng, C.-Y.; Xing, F.; Yan, X.-Q.;
Raman spectroscopy. Nano Letters 2013, 13 (4), 1541–1548. Chen, Y.; Tian, J.-G., High-quality and efficient transfer of
87. Han, Z.; Kimouche, A.; Allain, A.; Arjmandi-Tash, H.; large-area graphene films onto different substrates. Carbon
Reserbat-Plantey, A.; Pairis, S.; Reita, V.; Bendiab, N.; 2013, 56 (0), 271–278; (b) Choi, J.-Y., Graphene transfer: A
Coraux, J.; Bouchiat, V., Suppression of multilayer graphene stamp for all substrates. Nature Nanotechnology 2013, 8 (5),
patches during CVD graphene growth on copper. arXiv pre- 311–312; (c) Song, J.; Kam, F.-Y.; Png, R.-Q.; Seah, W.-L.;
print arXiv:1205.1337 2012. Zhuo, J.-M.; Lim, G.-K.; Ho, P. K.; Chua, L.-L., A general
88. Wassei, J. K.; Mecklenburg, M.; Torres, J. A.; Fowler, J. D.; method for transferring graphene onto soft surfaces. Nature
Regan, B. C.; Kaner, R. B.; Weiller, B. H., Chemical vapor Nanotechnology 2013, 8 (5), 356–362; (d) Martins, L. G. P.;
deposition of graphene on copper from methane, ethane and Song, Y.; Zeng, T. Y.; Dresselhaus, M. S.; Kong, J.; Araujo,
propane: Evidence for bilayer selectivity. Small 2012, 8 (9), P. T., Direct transfer of graphene onto flexible substrates.
1415–1422. Proceedings of the National Academy of Sciences of the
89. Dai, B.; Fu, L.; Zou, Z.; Wang, M.; Xu, H.; Wang, S.; Liu, Z., United States of America 2013, 110 (44), 17762–17767.
Rational design of a binary metal alloy for chemical vapour 100. Regan, W.; Alem, N.; Alemán, B.; Geng, B.; Girit, C.;
deposition growth of uniform single-layer graphene. Nature Maserati, L.; Wang, F.; Crommie, M.; Zettl, A., A direct trans-
Communications 2011, 2, 522. fer of layer-area graphene. Applied Physics Letters 2010, 96
90. Pirkle, A.; Chan, J.; Venugopal, A.; Hinojos, D.; Magnuson, (11), 113102-113102-3.
C.; McDonnell, S.; Colombo, L.; Vogel, E.; Ruoff, R.; 101. Yoon, T.; Shin, W. C.; Kim, T. Y.; Mun, J. H.; Kim, T.-S.; Cho,
Wallace, R., The effect of chemical residues on the physical B. J., Direct measurement of adhesion energy of monolayer
and electrical properties of chemical vapor deposited gra- graphene As-grown on copper and its application to renewable
phene transferred to SiO2. Applied Physics Letters 2011, 99 transfer process. Nano Letters 2012, 12 (3), 1448–1452.
(12), 122108–122108-3. 102. Gao, L.; Ren, W.; Xu, H.; Jin, L.; Wang, Z.; Ma, T.; Ma, L.-P.;
91. Reina, A.; Thiele, S.; Jia, X.; Bhaviripudi, S.; Dresselhaus, Zhang, Z.; Fu, Q.; Peng, L.-M., Repeated growth and bub-
M.; Schaefer, J.; Kong, J., Growth of large-area single- and bling transfer of graphene with millimetre-size single-crystal
Bi-layer graphene by controlled carbon precipitation on poly- grains using platinum. Nature Communications 2012, 3, 699.
crystalline Ni surfaces. Nano Research 2009, 2 (6), 509–516. 103. Liang, X.; Sperling, B. A.; Calizo, I.; Cheng, G.; Hacker, C.
92. (a) Huang, P. Y.; Ruiz-Vargas, C. S.; van der Zande, A. M.; A.; Zhang, Q.; Obeng, Y. et al., Toward clean and crackless
Whitney, W. S.; Levendorf, M. P.; Kevek, J. W.; Garg, S.; transfer of graphene. ACS Nano 2011, 5 (11), 9144–9153.
Alden, J. S.; Hustedt, C. J.; Zhu, Y., Grains and grain bound- 104. O’Hern, S. C.; Stewart, C. A.; Boutilier, M. S.; Idrobo, J.-C.;
aries in single-layer graphene atomic patchwork quilts. Nature Bhaviripudi, S.; Das, S. K.; Kong, J.; Laoui, T.; Atieh, M.;
2011, 469 (7330), 389–392; (b) Lee, Y.; Bae, S.; Jang, H.; Jang, Karnik, R., Selective molecular transport through intrinsic
S.; Zhu, S.-E.; Sim, S. H.; Song, Y. I.; Hong, B. H.; Ahn, J.-H., defects in a single layer of CVD graphene. ACS Nano 2012, 6
Wafer-scale synthesis and transfer of graphene films. Nano (11), 10130–10138.
Letters 2010, 10 (2), 490–493. 105. Gonçalves, G.; Marques, P. A.; Barros-Timmons, A.; Bdkin,
93. Srivastava, A.; Galande, C.; Ci, L.; Song, L.; Rai, C.; Jariwala, I.; Singh, M. K.; Emami, N.; Grácio, J., Graphene oxide modi-
D.; Kelly, K. F.; Ajayan, P. M., Novel liquid precursor-based fied with PMMA via ATRP as a reinforcement filler. Journal
facile synthesis of large-area continuous, single, and few- of Materials Chemistry 2010, 20 (44), 9927–9934.
layer graphene films. Chemistry of Materials 2010, 22 (11), 106. Ishigami, M.; Chen, J. H.; Cullen, W. G.; Fuhrer, M. S.;
3457–3461. Williams, E. D., Atomic structure of graphene on SiO2. Nano
94. (a) Bae, S.; Kim, H.; Lee, Y.; Xu, X.; Park, J.-S.; Zheng, Y.; Letters 2007, 7 (6), 1643–1648.
Balakrishnan, J. et al., Roll-to-roll production of 30-inch gra- 107. (a) Su, C.-Y.; Lu, A.-Y.; Wu, C.-Y.; Li, Y.-T.; Liu, K.-K.;
phene films for transparent electrodes. Nature Nanotechnology Zhang, W.; Lin, S.-Y. et al., Direct formation of wafer scale
2010, 5 (8), 574–578; (b) Suk, J. W.; Kitt, A.; Magnuson, C. graphene thin layers on insulating substrates by chemical
W.; Hao, Y.; Ahmed, S.; An, J.; Swan, A. K.; Goldberg, B. vapor deposition. Nano Letters 2011, 11 (9), 3612–3616;
B.; Ruoff, R. S., Transfer of CVD-grown monolayer graphene (b) Peng, Z.; Yan, Z.; Sun, Z.; Tour, J. M., Direct growth
onto arbitrary substrates. ACS Nano 2011, 5 (9), 6916–6924. of bilayer graphene on SiO2 substrates by carbon diffusion
95. Li, X.; Zhu, Y.; Cai, W.; Borysiak, M.; Han, B.; Chen, D.; through nickel. ACS Nano 2011, 5 (10), 8241–8247; (c) Bi,
Piner, R. D.; Colombo, L.; Ruoff, R. S., Transfer of large-area H.; Sun, S.; Huang, F.; Xie, X.; Jiang, M., Direct growth of
graphene films for high-performance transparent conductive few-layer graphene films on SiO2 substrates and their photo-
electrodes. Nano Letters 2009, 9 (12), 4359–4363. voltaic applications. Journal of Materials Chemistry 2012,
96. (a) Stoberl, U.; Wurstbauer, U.; Wegscheider, W.; Weiss, D.; 22 (2), 411–416.
Eroms, J., Morphology and flexibility of graphene and few- 108. Ismach, A.; Druzgalski, C.; Penwell, S.; Schwartzberg, A.;
layer graphene on various substrates. Applied Physics Letters Zheng, M.; Javey, A.; Bokor, J.; Zhang, Y., Direct chemical
2008, 93 (5), 051906–051906-3; (b) Chen, Z.; Ren, W.; Gao, vapor deposition of graphene on dielectric surfaces. Nano
L.; Liu, B.; Pei, S.; Cheng, H.-M., Three-dimensional flexible Letters 2010, 10 (5), 1542–1548.
109. (a) Fanton, M. A.; Robinson, J. A.; Puls, C.; Liu, Y.; Hollander, 115. Ni, G.-X.; Zheng, Y.; Bae, S.; Kim, H. R.; Pachoud, A.; Kim,
M. J.; Weiland, B. E.; LaBella, M. et al., Characterization Y. S.; Tan, C.-L. et al., Quasi-periodic nanoripples in graphene
of graphene films and transistors grown on sapphire by grown by chemical vapor deposition and its impact on charge
metal-free chemical vapor deposition. ACS Nano 2011, 5 transport. ACS Nano 2012, 6 (2), 1158–1164.
(10), 8062–8069; (b) Hwang, J.; Shields, V. B.; Thomas, C. 116. Bao, W.; Miao, F.; Chen, Z.; Zhang, H.; Jang, W.; Dames, C.;
I.; Shivaraman, S.; Hao, D.; Kim, M.; Woll, A. R.; Tompa, Lau, C. N., Controlled ripple texturing of suspended graphene
G. S.; Spencer, M. G., Epitaxial growth of graphitic carbon and ultrathin graphite membranes. Nature Nanotechnology
on C-face SiC and sapphire by chemical vapor deposition 2009, 4 (9), 562–566.
(CVD). Journal of Crystal Growth 2010, 312 (21), 3219– 117. Dean, C.; Young, A.; Meric, I.; Lee, C.; Wang, L.; Sorgenfrei,
3224; (c) Miyasaka, Y.; Nakamura, A.; Temmyo, J., Graphite S.; Watanabe, K.; Taniguchi, T.; Kim, P.; Shepard, K., Boron
thin films consisting of nanograins of multilayer graphene on nitride substrates for high-quality graphene electronics.
sapphire substrates directly grown by alcohol chemical vapor Nature Nanotechnology 2010, 5 (10), 722–726.
deposition. Japanese Journal of Applied Physics 2011, 50 (4), 118. Mattevi, C.; Colléaux, F.; Kim, H.; Lin, Y.-H.; Park, K. T.;
04DH12. Chhowalla, M.; Anthopoulos, T. D., Solution-processable
110. Liu, Z.; Song, L.; Zhao, S.; Huang, J.; Ma, L.; Zhang, J.; Lou, organic dielectrics for graphene electronics. Nanotechnology
J.; Ajayan, P. M., Direct growth of graphene/hexagonal boron 2012, 23 (34), 344017.
nitride stacked layers. Nano Letters 2011, 11 (5), 2032–2037. 119. Ou, E.; Xie, Y.; Peng, C.; Song, Y.; Peng, H.; Xiong, Y.; Xu,
111. Lippert, G.; Dabrowski, J.; Lemme, M.; Marcus, C.; Seifarth, W., High concentration and stable few-layer graphene dis-
O.; Lupina, G., Direct graphene growth on insulator. Physica persions prepared by the exfoliation of graphite in different
Status Solidi (b) 2011, 248 (11), 2619–2622. organic solvents. RSC Advances 2013, 3 (24), 9490–9499.
112. Rummeli, M. H.; Bachmatiuk, A.; Scott, A.; Borrnert, F.; 120. Du, W.; Lu, J.; Sun, P.; Zhu, Y.; Jiang, X., Organic salt-assisted
Warner, J. H.; Hoffman, V.; Lin, J.-H.; Cuniberti, G.; Buchner, liquid-phase exfoliation of graphite to produce high-quality
B., Direct low-temperature nanographene CVD synthesis over graphene. Chemical Physics Letters 2013, 568–569, 198–201.
a dielectric insulator. ACS Nano 2010, 4 (7), 4206–4210. 121. Hesjedal, T., Continuous roll-to-roll growth of graphene films
113. De, S.; Coleman, J. N., Are there fundamental limitations on by chemical vapor deposition. Applied Physics Letters 2011,
the sheet resistance and transmittance of thin graphene films? 98 (13), 133106-3.
ACS Nano 2010, 4 (5), 2713–2720. 122. Yamada, T.; Ishihara, M.; Kim, J.; Hasegawa, M.; Iijima, S., A
114. Zhu, W.; Low, T.; Perebeinos, V.; Bol, A. A.; Zhu, Y.; Yan, H.; roll-to-roll microwave plasma chemical vapor deposition pro-
Tersoff, J.; Avouris, P., Structure and electronic transport in cess for the production of 294 mm width graphene films at low
graphene wrinkles. Nano Letters 2012, 12 (7), 3431–3436. temperature. Carbon 2012, 50 (7), 2615–2619.
14 Chemically Modified Graphene and Its
Applications in Electrochemical Sensing
Raghu G. Kempegowda and Pandurangappa Malingappa
CONTENTS
Abstract...................................................................................................................................................................................... 247
14.1 Introduction...................................................................................................................................................................... 247
14.2 Modification Routes.......................................................................................................................................................... 248
14.2.1 Covalent Modification........................................................................................................................................ 249
14.2.2 Diazonium Salt Reduction................................................................................................................................. 249
14.2.3 Modification through Oxygen-Containing Functional Groups.......................................................................... 249
14.2.4 Covalent Polymerization.................................................................................................................................... 251
14.2.5 Microwave Irradiation........................................................................................................................................ 252
14.2.6 Ball Milling........................................................................................................................................................ 253
14.2.7 Photochemical Irradiation.................................................................................................................................. 254
14.2.8 Noncovalent Modification.................................................................................................................................. 255
14.2.9 Modification by Metal/Metal-Oxide Nanoparticles........................................................................................... 256
14.2.10 Chemical Reduction Method.............................................................................................................................. 257
14.2.11 Electrochemical Deposition............................................................................................................................... 258
14.2.12 Hydrothermal and Solvothermal Techniques..................................................................................................... 260
14.2.13 Microwave-Assisted Modification with Metal Nanoparticles............................................................................ 260
14.2.14 Sonochemical Modification with Metal Nanoparticles...................................................................................... 261
14.3 Applications...................................................................................................................................................................... 261
14.3.1 Metal Ion Sensing............................................................................................................................................... 261
14.3.2 Hydrogen Peroxide (H2O2) Sensing................................................................................................................... 262
14.4 Conclusions....................................................................................................................................................................... 265
References.................................................................................................................................................................................. 265
ABSTRACT zero-dimensional fullerenes, one-dimensional nanotubes and

three-dimensional graphite materials [1]. The term graphene
Chemically modified graphene electrodes constitute a new class was initially used in the year 1987 to describe monolayer atomic
of materials that possess desired properties and are tuned toward sheets of graphite. It has been recognized as one of the novel
specific analyte quantitative studies. As a result, in this chap- types of material in the carbon family since its discovery in
ter, we have mainly focussed on modification strategies such as 2004 by Andre Geim and Konstantine Novoselov at Manchester
covalent, noncovalent and metal nanoparticle-based routes. The Metropolitan University, UK [2]. They were awarded the Nobel
covalent modification includes diazonium salt reduction, amide Prize in 2010 for this discovery [3]. It can be prepared by several
bond formation, polymerization, microwave irradiation and sol- routes, however, the basic mechanism involved in the synthesis
vent-free methods like ball milling and photochemical methods. of graphene is from graphene oxide and its reduction either by
Metal nanoparticles modification of graphene includes chemi- a chemical or electrochemical approach [4–6]. It is the purest
cal reduction, electrochemical deposition, hydrothermal, solvo- form of carbon due to the absence of metal impurities which
thermal, microwave and sonochemical routes. The applicability are normally incorporated during their catalytically assisted
of chemically modified graphene toward the electrochemical synthetic procedures [7]. Graphene has been characterized
quantification of selective metal ions and hydrogen peroxide by its basal or edge planes (Figure 14.1) and these planes are
sensing at trace level have been discussed. known to participate in heterogenous electron transfer kinetics
[8]. Its high electrical and thermal conductivities originate from
extended long-range π conjugation, hence, graphene became a
14.1 INTRODUCTION
very promising and useful material in many potential applica-
Graphene is a two-dimensional carbon nanostructure which tions in the fields of electronics, composites, sensors, energy
serves as a structural motiff for all forms of carbon, mainly storage and conversion devices [9].
247
Basal plane material obtained by its surface modification with modifier

molecules through specific interactions like covalent and non-
covalent linkages between them [16]. The modifier molecule
imparts specific properties to the graphene after its attachment,
and hence, the modified substrates exhibit a new set of proper-
ties which are not available in their unmodified counterparts.
Hence, these properties can be tuned toward achieving speci-
ficity and sensitivity for a particular target analyte. Chemically
modified graphene (CMG) mainly of consists two components,
Edge plane
that is, substrate material and the modifier molecule. The choice
of these two components depends mainly on the nature of the
FIGURE 14.1 Conceptual model depicting the structure of gra- analytical work proposed in the quantitative study [17].
phene and its basal as well as edge plane sites. (Adapted from D. The main objectives of the modification of graphene can
A. C. Brownson and C. E. Banks. Physical Chemistry Chemical
be attributed to the following:
Physics 14, 2012, 8264.)
1. To minimize the irreversible aggregation of gra-
Graphene has been shown to exhibit a unique array of physi- phene layers into platelets.
cal and chemical properties due to its large surface area, high 2. To anchor selective functional moieties which are
electrical and thermal conductivities, high mechanical strength highly specific toward target analyte.
and good biocompatibility [4]. Graphene-modified electrodes 3. To enhance the electron transfer reactivity at the sub-
have received considerable attention in recent years mainly in strate surface through the electron mediators.
the area of electroanalytical chemistry [10], but these electrodes 4. To improve the electrode kinetics and reduction of
possess several limitations in terms of selectivity, sensitivity, overpotential required in specific redox processes.
high overpotential and sluggish electrode kinetics due to the
absence of specific and selective interactions of electrode inter- Various strategies have been reported for the modification of
face with the target analyte. Hence, efforts have been directed graphene which have been discussed thoroughly in this chapter.
toward minimizing these limitations through chemical modi- All the strategies used for the covalent and noncovalent modi-
fication procedures. This can be achieved by altering the sur- fication of graphene with metal nanoparticles and its potential
face structure of the graphene by modifying its surface with applications toward the sensing of different metal ions and
selective molecules since all electrochemical reactions depend hydrogen peroxide (H2O2) quantification were discussed.
on the nature of interfacial structure. This modification results
in the formation of a new surface called chemically modified
graphene (CMG) [11–14]. The concept of a chemically modi-
14.2 MODIFICATION ROUTES
fied electrode (CME) was first introduced by electrochemist, Graphene can be modified in several ways with diverse func-
and these electrodes were favored by the scientific community tional groups. In this chapter, modification routes were mainly
in recent years due to their selectivity, sensitivity, as well as focused on covalent, noncovalent and metal nanoparticles-based
the possibility of having direct control over the surface struc- methodologies. The schematic representations of all these func-
ture of the electrode [15]. CMG forms a new type of electrode tionalization strategies discussed in this chapter are shown below.
Graphene
Functionalization
Covalent Noncovalent Metal nanoparticles
1. Diazoniun salt reduction 1. Electrostatic interaction 1. Chemical reduction

2. Oxygen containing functionalities 2. π–π interaction 2. Electrochemical deposition
3. Covalent polymerization 3. Hydrothermal and solvothermal
4. Microwave irradiation methods
5. Solvent free ball milling 4. Microwave irradiation
6. Photochemical method 5. Sonochemical method
Applications
(Metal ions and H2O2)
Different functionalization routes
Chemically Modified Graphene and Its Applications in Electrochemical Sensing 249
14.2.1 Covalent Modification COOH OH COOH

HOOC OH
Covalent modification of graphene is a process of formation O O
of a covalent bond between graphene substrate surface and COOH
HO
modifier molecules. It involves activation and coupling of
HOOC COOH
oxygen containing functional groups like –COOH, –OH on
carbon surface with amines, direct reaction between graphene HOOC
HO
and diazonium salts, and also through cycloaddition reactions
[18–20]. The covalent modification of graphene results in
the destruction of its conjugated structure at the sites where NH2 HCl + NaNO2 N2+ 0°C
modifier molecules are anchored covalently. Even though all 0°C 12 h
these problems are associated with covalent modification of
graphene, extensive research is going on in this area due to
its unique property gained after the modification of graphene OH COOH
COOH
which is more essential when other properties of graphene are HOOC OH
desirable except its inherent properties [21]. O O
COOH
HO
14.2.2 Diazonium Salt Reduction HOOC
Diazonium salts have been extensively used as modifiers in the HO HOOC

covalent modification of carbon-based substrate materials due
its availability with diverse functional groups which can be used
in many industrial and analytical applications [22]. In this direc- SCHEME 14.2 Schematic representation of derivatization of graphene
tion James M. Tour and coworkers have functionalized the ther- oxide with anthracene using corresponding diazonium ion. (Adapted
from Y. Lu et al. Journal of Material Chemistry 22, 2012, 2929.)
mally converted graphene with aryl diazonium salts containing
various functional groups such as –Cl, –NH2, –Br, –I, –COOH. 14.2.3 Modification through Oxygen-
The modification has been carried out by simple mechano-
Containing Functional Groups
chemical grinding of graphene, diazonium salt and K2CO3 in
presence of an ionic liquid 1-octyl-3-methylimidazolium tetra- The most popular methodology for the synthesis of graphene
fluroborate using pestle and mortar (Scheme 14.1) [19]. from natural graphite is the chemical reduction of graphene
The direct use of chemically converted graphene suf- oxide in presence of reducing agents like ascorbic acid,
fers from poor dispersion ability in most of the solvents. This hydrazine, sodium borohydride, ethylene glycol, and so on.
problem has been resolved by using surfactant, that is, sodium [26]. It is well known that the complete removal of oxygen-
dodecyl benzene sulfonate in the modification process [23]. containing functional groups of graphene oxide during the
Another approach in this strategy is in situ synthesis and sub- reduction process is not possible. As a result, very few of
sequent attachment of the diazonium salt. It involves the use them are still retained on the graphene layers even after its
of amine as a starting material from which a diazonium salt reduction, and these groups are commonly called as residual
can be prepared in situ through usual diazotization reaction. oxygen-containing functional groups [27]. These functional
It is more facile and advantageous because it reduces the time groups are amenable for further modification with desired
required for the synthesis and characterization of diazonium modifier molecules. On the basis of the existence of resid-
salt. Yizhang et al., have used this strategy and modified the ual oxygen-containing functional moieties, many research
graphene oxide nanosheets with 2-amino anthracene as a modi- groups have explored the possibility of modifying the gra-
fier molecule through diazonium chemistry (Scheme 14.2) [24]. phene and graphene oxide covalently with suitable indica-
Gao et al., have used a similar strategy to modify the tor species to tune the modified graphene toward specific
graphene oxide with thiol-containing groups, and the modi- goals. One of the most common approach is the activation
fied substrate was used in the water purification process of residual oxygen-containing functional groups like car-
(Scheme 14.3) [25]. boxylic groups using suitable activators like thionyl chloride
R R
–
– BF4
R IL: N N
C8H17
+
Thermally converted K2CO3
graphene (TCG) N2+BF4–
RT, Grinding with agate
R = Cl, Br, I, COOH mortar and pestle, 45 min R
SCHEME 14.1 Exfoliation and functionalization of graphene sheets by grinding TCG and diazonium salt in presence of ionic liquid using
a mortar and pestle. (Adapted from Z. Jin et al. Chemistry of Materials 21, 2009, 3045.)
N⊕ R R
2 R
O OH
OH O OH
OH OH O O
O OH O OH O O
OH O COOH
O COOH HO
HO R = –SH O O
OH
O O OH OH
OH
OH OH R
R
Graphite oxide (GO) Functionalized GO
SCHEME 14.3 Schematic representation of functionalization of graphene oxide with thiol-containing molecules. (Adapted from W. Gao
et al. Applied Materials and Interfaces 3, 2011, 1821.)
(1) SOCI2, 24 h
Graphene oxide TPP-NHCO-SPF graphene
(2) TPP-NH2, EtN3,
DMF, 130°C, 72 h
NH N
TPP-NH2 = NH2
N HN
N HN N
Excitation HN
NH N NH
N
HN
HN
C O O C
N Donor-Acceptor
N HH N HO
N
O
N C
H OH
O
O
ET
OC C O
NH
NH
PET
N N
NH HN NH HN Fluorescence
N N
SCHEME 14.4 Schematic representation of the structure of the covalent TPP-NHCO-SPF graphene. (SPF, solution processable function-
alized graphene). (Adapted from Y. Xu et al. Advanced Materials 21, 2009, 1275.)
(SOCl2), N,N′-dicyclohexylcarbodiimide and ethyl-3-(3- acid groups at the edge sites were activated into acyl chlo-
dimethylaminopropyl) carbodiimide (EDC), and then cou- ride using (SOCl2) as an activating agent. Then the resulting
pling with amine-terminated modifier species [18]. acyl chloride was coupled with 5-(4-aminophenyl)-10,15,20-
Chen and coworkers have covalently functionalized the gra- triphenyl porphyrin selectively at the amine position through
phene sheets with porphyrin. In this strategy, the carboxylic the formation of an amide bond (Scheme 14.4) [28].
HOOC COOH
O OH O OH O OH O O
HOOC COOH (1) CICH2COONa, NaOH HOOC COOH
(2) HCI
OH OH O O
HOOC COOH
GO GO-COOH
(1) SOCI2
(2) NH2
O NH O NH
N
N N
Ru O O OH O O NH
NH2 =
N N O
N NH2 HN O
N
H O O
HN O O NH
Ru-GO
SCHEME 14.5 Schematic pathway of insertion of Ru complex into graphene oxide (GO) sheets. (Adapted from Y. Yu et al. Carbon 50,
2012, 2539.)
Krishna et al., have modified graphene oxide with 3-amino the polymers wherein the polymerization takes place through
phenyl boronic acid through the formation of an amide bond existing chemical reactions, whereas the other one is the
in presence of 1-ethyl-3-(3-dimethylaminopropyl)-carbodi- “grafting to” method in which the polymer can be anchored
imide) (EDC) as a coupling agent [29]. to the graphene [32]. Pramoda et al. have developed a simple
Yu et al., have modified graphene oxide using a ruthenium and efficient route for the synthesis of a polymer graphene
complex. The modification proceeds through the attachment composite based on the in situ polymerization process [34].
of a ruthenium complex containing a long-chain alkyl amino The covalent modification has been carried out by the reac-
group into the carboxylic group of the graphene oxide via the tion between the oxygen-containing functional groups of
formation of an amide bond (Scheme 14.5) [30]. the graphene with an octadecylamine molecule. The amine-
Murali Krishna et al., have explored a novel way to cova- functionalized graphene further reacts with methacryloyl
lently anchor the metal-containing Cu, Zn, Sn, and metal- chloride to introduce a polymerizable carbon double bond
free porphyrin onto the graphene oxide sheets. The covalent which, in turn, facilitates the in situ polymerization process.
modification involves the conversion of carboxylic groups of Salavagione et al. have reported the covalent modification of
graphene oxide into acyl chloride in presence of SOCl2 in graphene with poly (vinyl alcohol) in two ways. The one way
dimethyl formamide (DMF) solvent under nitrogen atmo- involves the esterification of carboxylic groups of graphene,
sphere and coupling with known amounts of porphyrin [31]. and the other proceeds through the formation of an acyl chlo-
ride intermediate [35]. Jingchao et al., have reported the prepa-
ration of polymethylmethacrylate on graphene sheets by the
14.2.4 Covalent Polymerization
in situ suspension of polymerization as well as reduction of
Covalent grafting of polymers to the graphene represents a new graphene oxide using hydrazine hydrate and ammonia. Chen
era in the development of novel composite materials for a wide et al., have proposed another strategy to anchor poly(tert-butyl
variety of applications. In recent years, polymers have been methacrylate) (PtBMA) polymer chains on the surface of gra-
considered as the most exciting class of materials due to their phene through surface-initiated, single-electron transfer, radi-
low cost, easy preparation and availability in various chemical cal polymerization. Initially, the method proceeds through
forms [32]. The combination of polymer and graphene produces the covalent attachment of the Br group as an initiating group
a synergetic effect which can be essentially tuned in sensors and through diazonium chemistry, then the polymerization of
actuators based on the modulation of functional properties [33]. PtBMA chains through surface-initiated, single-electron
The derivatization of graphene with the polymers can be transfer, living, radical, polymerization [36]. Mohamadi et al.,
achieved in many ways. Among them, only two are most have prepared graphene nanoplatelets covalently functional-
exciting; one is the “grafting from” method, in which method ized with polymetha-acrylic anhydride chains by the intro-
the graphene itself can act as a site to attach the initiator for duction of vinyl groups onto the surface of graphene via
PPA/P2O5
Graphite 4-aminobenzoic acid (ABA) ABF-G
(NH4)2S2O8
1M HCl
Aniline
H H H O O H H H
N N N N N C C N N N N N
w y
H H H O O H H H
N N N N N C C N N N N N
x z
Polyaniline/graphene composites (PAGCs)
SCHEME 14.6 Schematic pathway of preparation of polyaniline/graphene composite. (Adapted from C. H. Chang et al. Carbon 50, 2012, 5044.)
simple esterification reaction between its hydroxyl groups community, as evidenced by the number of papers and reviews
and methacrylic anhydride followed by its polymerization appearing in the literature. The chemistry under microwave
[37]. Zhen et al., have prepared nylon-6-(PA6-)-modified gra- irradiation is known to be faster and more efficient than con-
phene by the in situ polymerization of caprolactam ring at its ventional methods. Microwave-assisted synthesis has been
carboxylic groups under thermal activation [38]. Tung et al., exploited in various fields of organic synthesis including het-
have reported the polyaniline-based water-soluble graphene erocyclic, organometallic, combinatorial chemistry, synthetic
composite by the simultaneous reduction of graphene oxide transformations, radical reactions, organocatalysis, cyclo-
and poly (sodium 4-styrenesulfonate) in presence of hydra- addition, and metathesis metal-mediated transformations.
zine as a chemical reducing agent [39]. It can be achieved by Microwave-assisted reactions have gained more importance
the in situ polymerization of aniline monomer with sodium as it reduces time, volume of solvents, and amount of reac-
4-styrenesulfonate through oxidative mechanism in aque- tants required for the completion of reaction, and results in
ous acidic solution. Rodney S Ruoff and his coworkers have good selectivity, high purity, and good yield. The basic prin-
introduced a new methodology for the covalent anchor- ciple lying in this method is that when the molecules with
ing of polymer brushes onto the surface of graphene oxide permanent dipole moment are subjected to the electric field,
using surface-initiated atom transfer radical polymerization the molecules get aligned with respect to the field. If the field
(ATRP). The hydroxyl groups present on the surface of gra- oscillates, then the molecule also oscillates with respect to the
phene oxide were first functionalized with bromoisobutyryl field, and the molecules align and realign continuously giving
bromide through ATRP, and then polymers of styrene, butyl rise to an intense internal heat energy which, in turn, agitates
acrylate, or methyl methacrylate were grown directly through the molecules which leads to the formation of products [42,43].
surface-initiated polymerization [40]. Xu et al., have reported a simple and scalable methodology
Chang et al., have modified graphene with polyaniline for the covalent modification of reduced graphene oxide with
followed by the modification with 4-aminobenzoyl groups. hydroxypropyl-β-cyclodextrin under microwave irradiation
Initially, the graphene was modified with aminobenzoyl through the formation of an ester bond between the carboxyl
groups in polyphosphoric acid (PPA) and phosphorous pent- groups of reduced graphene oxide and the hydroxyl groups of
oxide medium followed by the polymerization of aniline cyclodextrin [44]. In this modification, the graphene oxide and
monomers through oxidative mechanism (Scheme 14.6) [41]. cyclodextrin were uniformly dispersed in water using sonica-
tion and then subsequently irradiated by the microwaves for
different time intervals. Hydrazine hydrate and ammonia
14.2.5 Microwave Irradiation
have been added to the dispersion as reducing agents, and the
Microwave technology is a relatively new field which irradiation continued for another 30 min. Finally, the mix-
has received considerable attention among the scientific ture was centrifuged and washed repeatedly with anhydrous
OH
OH O
OH OH
OH
O
O O Hydrazine hydrate
HO
O O
O Ammonia
OH
HO O
OH
O OH O
C C
O X O
HO
O O
O O
X
HO HO e
OH OH
X = NO2/NH2/Cl/OH/CH3
SCHEME 14.7 Schematic representation of covalent modification of reduced graphene oxide with hydroxypropyl-β-cyclodextrin under
microwave irradiation. (Adapted from C. Xu et al. Journal of Material Chemistry 21, 2011, 10463.)
OH COOH OH COOH
HOOC O OH O
98% H2SO4, NaNO3 Sonication OH
O OH
KMnO4 DMA HOOC OH
COOH COOH
HOOC OH
Graphene oxide nanosheets (GONS)
Graphite
O
OH
O NH C
O NH
C
CS Hydrazine hydrate C
NH
DMA, microwave DMA, microwave
Graphene nanosheets (GNS)
covalently modified by chitosan
: Chitosan (CS) chain H3C
C O
OH NH
CH 2 HO
Chemical structure of chitosan: O O O
O O
HO CH2OH n
NH2
SCHEME 14.8 Schematic representation of modification of graphene nanosheets with chitosan. (Adapted from H. Hu et al. Applied
Surface Science 257, 2011, 2637.)
ethanol in order to remove the unreacted cyclodextrin, and Recently Hu et al., have covalently modified the graphene
then dried under vacuum at 70°C to yield functionalized gra- with a natural biopolymer, that is, chitosan (CS) under micro-
phene (Scheme 14.7). wave assistance in presence of N,N-dimethyl formamide
Economopoulos et al., have developed a new type of (DMF) as a solvent. The mechanism of the strategy is the
microwave methodology for covalent modification of gra- exfoliation of graphene oxide and its interaction with the
phene layers suspended in benzyl amine with malonate deriv- amino functional group of the chitosan (CS) under continuous
ative through Bingel reaction. In this method, the authors have microwave irradiation (Scheme 14.8) [46].
selected an efficient solvent capable of creating stable gra-
phene suspensions from graphite. This allowed the use of the
14.2.6 Ball Milling
exfoliated graphene dispersion to covalently graft malonate
moieties along the graphitic skeleton following the Bingel The ball-milling approach comes under solvent-free derivatiza-
reaction, followed by microwave irradiation [45]. tion protocols since it minimizes the use of organic solvents
H OH On the basis of this approach, various types of carbon sub-

OH O
strates like glassy carbon spheres and carbon nanotubes have
HO OH
KOH been covalently anchored with different species of interest under
Ball milling HO OH ball-milling conditions [48,49]. Yong Liu and coworkers have
O
HO H OH developed a strategy for the covalent bulk modification of car-
HO OH bon nanotubes with hydroxyl groups by taking nanotubes and
Graphene sheets Modified graphene sheets potassium hydroxide in an agate capsule and vigorously shaking
for 8 h at a rate of 200 rpm in air at room temperature using a
SCHEME 14.9 Schematic representation of modification of graphene
planetary ball mill [50]. Similar methodology has been used for
sheets with hydroxyl groups through ball milling approach. (Adapted
from L. Yan et al. Journal of Materials Chemistry 22, 2012, 8367.) the modification of graphene as a substrate material for covalent
modification (Scheme 14.9) [51].
during synthesis. It is a well established method for the genera-
tion of active centers on the surface of carbon-based substrate
14.2.7 Photochemical Irradiation
materials during the functionalization process. Simple grinding
facilitates the activation of substrate material and modifies the This approach forms a new platform for the formation of a
substrates with suitable indicator species often used in inorganic covalent bond between carbon and the modifier molecule.
and material chemistry for the preparation and modification of Liu et al., have covalently modified the graphene with ben-
solids both at the laboratory as well as at the industrial-scale lev- zoyl peroxide through the formation of a radical intermediate
els. The basic principle involved in this process is the generation under laser irradiation. The laser irradiation introduces the
of frictional forces caused by the collision between the surface spatially localized defects on the basal planes of the graphene.
of the reactants which increases the internal energy, temperature During the process of irradiation, graphene undergoes excita-
and pressure. In chemical synthesis, the ball-milling approach tion, and results in the transformation of electrons, and this
provides the required conditions very efficiently, thereby, the electron transfer results in the formation of benzoyl peroxide
chemical reactions take place in small time intervals [47]. The radical anion, which irreversibly decomposes to produce phe-
mechanism of modification takes place either by changing the nyl radicals, and subsequently gets anchored to the substrate
reactivity of the reagents or by breaking the molecular bonds. (Table 14.1, Scheme 14.10).
TABLE 14.1
Different Functionalization Strategies
Substrate Indicator Molecule Modification Strategy References
Graphene Chloro benzene Diazonium salt reduction [19]
2-amino anthracene Diazonium salt reduction [24]
Thiol Diazonium salt reduction [25]
Triphenyl porphyrin Amidation [28]
3-amino phenyl boronic acid Amidation [29]
Ruthenium complex Amidation [30]
Polyvinyl alcohol Covalent polymerization [35]
Polymethyl methacrylate Covalent polymerization [36]
Polymethacrylic anhydride Covalent polymerization [37]
Poly(sodium 4-styrenesulfonate) Covalent polymerization [39]
4-aminobenzoyl groups Covalent polymerization [41]
Hydroxypropyl-β-cyclodextrin Microwave irradiation [44]
Malonate Microwave irradiation [45]
Chitosan Microwave irradiation [46]
Hydroxyl groups Solvent free ball milling [50]
Benzoyl peroxide Photochemical [52]
O O
hv
O O +
–CO2
Radical Graphene Modified graphene
SCHEME 14.10 Schematic pathway of photochemical modification of graphene. (Adapted from H. Liu et al. Journal of American
Chemical Society 131, 2009, 17099.)
14.2.8 Noncovalent Modification large number of monolayer and double layer graphene sheets
can be easily fabricated on the substrates of interest.
Noncovalent approaches for the modification of graphene with Yang et al., have developed graphene–biopolymer com-
desired molecules have received considerable interest in the posites based on noncovalent interactions. They have modi-
development of graphene-based nanocomposites for a wide fied the graphene nanosheets with environmentally friendly
variety of analytical applications like sensors, biosensors, bat- and biologically compatible polymers like sodium lignosul-
teries, fuel cells, and electrocatalysts for oxygen reduction reac- fonate, sodium carboxymethyl cellulose, and pyrene-con-
tion. The noncovalent interactions between substrate and the taining hydroxypropyl cellulose in order to obtain highly
modifier molecule modify the graphene by retaining its intrin- stable graphene aqueous suspensions [58]. Mann et al., have
sic properties without any damage. Recently, the noncovalent reported a novel type of noncovalent modification proce-
functionalization of GO has shown exciting potential in terms dure using a tripodal-binding motiff Co(II) bis-terpyridyl
of loading efficiency and manipulation. This type of modifi- complex (Scheme 14.12) [59]. This complex comprises three
cation involves supramolecular interactions like π–π stacking, pyridine rings which can strongly adsorb onto the basal
electrostatic interaction, hydrogen bonding, and the formation planes of the graphene. From electrochemical kinetic and
of coordinate bonds. Graphene or graphene oxide undergoes thermodynamic study, it has been found that in the modified
this kind of interactions with the modifier molecules because graphene the tripod units with a variety of functional groups
of the presence of negative charge density at its surface due to are projected away from the surface plane of the graphene
the presence of ionizable oxygen-containing functional groups, sheets which has been confirmed by measuring the differ-
and also the presence of extended π conjugation [21,53]. ence in rate of electron transfer reaction between tripodal
The fundamental unit of the graphene network is hexago- and monovalent units.
nal rings of sp2 hybridized carbon atoms. Each carbon atom is Kim and coworkers have proposed a novel method to dis-
covalently bonded with each other through sigma bonds due perse the graphene sheets in various organic solvents [60]. The
to the pairing of 1 electron in its 2p orbital. The remaining dispersion can be obtained by the noncovalent functionaliza-
electron in 2p orbital is delocalized across the carbon atoms to tion of graphene with selective functional polymers. In this
form π bonds and, hence, it can be considered as a conjugated method, the oxygen-containing functional groups like cabox-
macromolecule. Thus, the molecules having a conjugated ylate groups acted as anchoring sites for the protonated amine
structure can form a strong interaction with the basal plane of groups of the functional polymers. This modification process
graphene through π–π interactions [54]. facilitates the phase transfer of graphene sheets from water
Xu et al., have developed a strategy for the noncovalent phase to an organics phase through simple sonication. The
functionalization of graphene with pyrenebutyric acid. The nature of organic solvents as well as the functional group of
functionalization process proceeds through the strong π–π the polymer and its concentration level played a very impor-
interactions between the pyrenyl rings and basal plane sites tant role in controlling as well as determining the dispersing
of graphene [55]. Similarly, Bai et al., have established the ability of the modified graphene.
same kind of interaction between sulfonated polyaniline and Methyl green (MG) is a water-soluble cationic dye, and
graphene sheets to get sulfonated polyaniline graphene com- hence, it was expected to be strongly adsorbed on the sur-
posite [56]. This modified graphene has been nicely dispersed face of graphene through π–π stacking, and subsequently
in water and has also shown enhanced electrocatalytic activ- reduce the Van der Waals force of interaction between the
ity toward electrochemical sensing of ascorbic acid. Su and graphene layers. The adsorption of MG onto graphene greatly
coworkers have tuned the electronic properties of graphene by improves the dispersibility of graphene sheets in water [61].
modifying its surface with the molecules capable of donating Zhang et al., have functionalized the graphene nanoribbons
and accepting the electrons. They have used a large aromatic with water-soluble iron (III) meso-tetrakis(N-methylpyridi-
donor, as well as acceptor molecules, through noncova- num-4-yl) porphyrin via π–π interactions [62]. The result-
lent interactions in order to retain their intrinsic properties ing composite film showed excellent electrocatalytic activity
[57]. This approach results in the formation of a novel class toward the reduction of oxygen, and also it has been applied
of graphene composite material with enhanced electronic as glucose biosensor. Fang et al., have functionalized the gra-
properties over its unmodified counterparts. To serve this phene with chitosan (CS) by chemical reduction of graphene
purpose, two organic molecules like Pyrene-1-sulfonic acid oxide. The modification proceeds through the interaction of
sodium salt which is an electronic donor, and disodium salt graphene with the amino and hydroxyl groups of CS through
of 3,4,9,10-perylenetetracarboxylic diimide bisbenzenesul- ionic interaction and hydrogen bonding. The graphene sus-
fonic acid which is an electronic acceptor have been selected pension is highly biocompatible which, in turn, makes the
as modifier molecules. Since these molecules have large system highly desirable for potential biological and medicinal
planar aromatic structures, they have strong affinity toward applications [63]. Zhang et al., have prepared the graphene
the surface of graphene sheets and, hence, these molecules oxide phthalocyanine hybrid material through π–π stacking.
have been strongly anchored onto its surface via π–π interac- The phthalocyanine used here is 2,11,20,29-tetra-tert-butyl-
tions without altering any of its original electronic properties 2,3-naphthalocyanine, and the intermolecular force of inter-
(Scheme 14.11). Therefore, the stability of the aqueous colloi- actions between this and the GO takes place immediately
dal dispersion of graphene sheets was greatly enhanced and a after the mixing of two components (Scheme 14.13) [64].
(a) (b) (c)
O O
O3S N N SO3
H2O
O O
PDI
SO3
or
Deposit
PyS
COO COO COO
OOC
OOC COO
OOC COO
ReG
SCHEME 14.11 Schematic illustration and images of aqueous dispersions of graphene sheets and composites on the surface: (a) reduced
graphene aqueous dispersion, black precipitate after reduction; (b) reduced graphene—PDI aqueous dispersion, without precipitate after
centrifugation; (c) reduced graphene-PyS aqueous dispersion, without precipitate after centrifugation. (Adapted from Q. Su et al. Advanced
Materials 21, 2009, 3191.)
1. NaH/DMF/rt N
N N N N • 2PF6
N N OTs
N 4 N Co N
N N
1. Pd(dppf )Cl2/Cul
3 6
THF/Et3N/rt
2.
2. BBr3/CH2Cl2/–78°C
Br N
N N
+ 48% Co
(two steps) CH3CN Cl Cl
HO THF/MeOH/rt O
OH O
HO 3. NH4PF6/H2O/rt O
H3CO 5
OCH3
H3CO 38%
4 1 · 2PF6
(three steps)
SCHEME 14.12 Synthesis of tripodal graphene. (Adapted from J. A. Mann et al. Journal of American Chemical Society 133, 2011, 17614.)
Rao and coworkers have demonstrated a simple and effi- 14.2.9 Modification by Metal/Metal-Oxide
cient methodology for the preparation of stable aqueous Nanoparticles
solutions of single and few layers of graphene by exploiting
noncovalent interactions with a coronene carboxylate accep- Incorporation of catalyst particles onto an individual graphene
tor molecule [65]. The functionalization of graphene with dif- sheet can provide greater versatility in carrying out selective
ferent polymers is summarized in Table 14.2. catalytic or sensing processes [72]. The metal nanoparticles
COOH COOH
R R
O
HO N
OH N NH N
= COOH N NH N
O = N
HOOC CH3
R R
R=C CH3
COOH COOH CH3
SCHEME 14.13 Schematic representation of formation of GO-NPC hybrid. (Adapted from X. Zhang et al. Carbon 48, 2010, 211.)
14.2.10 Chemical Reduction Method

TABLE 14.2
Modification Strategies of Graphene with Different It is one of the most widely used methods in the fabrication
of graphene–metal nanoparticle-based composites, and the
Polymers
name itself implies that the reduction process is accompanied
Substrate Polymer Type of Modification References by a chemical reagent. Various reducing agents have been
Graphene PEG Esterification [66] used in the reduction of bulk metal salts into metal nanopar-
PVA Esterification [35]
ticles [77].
P3HT Esterification [67]
TPAPAM Amidation [68]
Sodium borohydride which is a well known reducing agent
PMMA Radical grafting [69] which can reduce both graphene oxide into graphene and bulk
PMA Radical coupling [70] metal salt into nanosized metal particles [78]. In this direction,
Nylon Condensation [38] reduced graphene oxide (RGO)–silver (Ag) nanocomposite has
PEK, PP Electrophilic, Ziegler–Natta [71] been prepared by the reduction of AgNO3 with NaBH4 in a GO
suspension [79]. Dey et al., have designed a graphene–platinum
Note: PEG, polyethylene glycol; PVA, polyvinyl alcohol; PMMA, poly-
methyl methacrylate; P3HT, poly 3-hexylthiophene; PP, polyprope-
composite by the simultaneous and sequential reduction of gra-
lene; TPAPAM, triphenylamine–polyazomethine; PMA, polymethyl phene oxide and platinum precursor for cholesterol biosensor
acrylate; PEK, poly ether ketone. [80]. Zhuo et al. have reported a novel type of methodology
which involves the room temperature reduction of graphene
oxide into graphene in presence of metal nanoparticles. These
are of special interest in developing specific interface due nanoparticles (NPs) act as catalysts in the hydrolysis of NaBH4
to their unique catalytic and electronic properties [73]. The reagent under ambient conditions which can be recycled after
catalytic properties of these particles are entirely different several times under ambient conditions [26] (Scheme 14.14).
from their bulk counterpart. For instance, bulk Au is known Ethylene glycol (EG) is another important eco-friendly,
to be inactive toward many reactions, whereas the nanosized mild, reducing agent which is well known to have great control
Au particles have excellent catalytic activity [74]. The cata-
lytic properties of the nanoparticles are associated with the
O
quantum-scale dimensions, large surface-to-volume ratio, and OH
special binding sites available on the surface of the particles. OH

The particle size, morphology and its distribution over the O OH O
substrate material mainly depends on the type of synthesis Metal salt
O OH OH
as well as its corresponding reaction conditions. The size- NaBH4
HO
dependent properties have given rise to enormous applications O
OH
that have emerged in areas of data storage, sensing devices,
Metal salt: Cu, Fe, Co, Ni
catalysts, and biological interfaces in nanoelectromechanical GO GO-MNP
devices [75,76]. The most commonly used routes are chemical
reduction method, electro deposition, electroless deposition, SCHEME 14.14 Schematic illustration of formation of GO-MNP.
sol–gel method, hydrothermal method, thermal evaporation, (Adapted from M. C. Daniel and D. Astruc. Chemical Reviews 104,
photo-assisted reduction, and microwave-assisted deposition. 2004, 293.)
over the particle size and its dispersion over the surface of the Rodney S. Ruoff and his coworkers have developed a
substrate. It is due to the fact of rapid in situ generation of the method for one-pot synthesis of platinum nanoparticles embed-
species responsible for the reduction process [81]. ded on the surface of reduced graphene oxide and subsequently
Chao et al., have successfully developed a general method- applied the synthesized nanomaterial in the study of oxygen
ology for the preparation of graphene–metal particle nanocom- reduction and methanol oxidation [81]. Similarly, Li et al., have
posite using ethylene glycol (EG) as a reducing agent. Using adopted the same methodology for the modification of graphene
this methodology, they have modified graphene with platinum, with platinum nanoparticles of size 1–5 nm as per the follow-
palladium, and gold nanoparticles. These particles are initially ing scheme, and the modified graphene was used in methanol
adsorbed onto the surface of graphene oxide sheets, and then electro-oxidation reactions (Scheme 14.16) [83].
subsequently play a vital role in the catalytic reduction of GO A composite of graphene copper oxide was prepared by the
with EG (Scheme 14.15) [82]. adsorption of copper acetate salt on to the surface of graphene
oxide since it provides oxygen-containing functional groups
wherein these groups act as binding sites to adsorb acetate
(1) (2) salts. The reduction has been carried out by the influence
of EG which results in the formation of a graphene–copper
oxide composite (Scheme 14.17) [84].
(3)
14.2.11 Electrochemical Deposition
Electrochemical methodology is a powerful tool in the modi-
(4) fication of carbon-based substrate materials with metal nano-
particles as well as with metal oxide. This approach is very
simple, fast, and green toward environmental concerns, which
results in the formation of clean products. The simplicity of
this method involves the preparation of graphene oxide or gra-
SCHEME 14.15 Schematic representation of general mechanism phene-modified glassy carbon electrode and its reduction in
of anchoring of metal nanoparticles onto exfoliated graphite oxide: presence of metal salt precursors using a cyclic voltammetric
(1) Oxidation of graphite to graphite oxide. (2) Exfoliation of graph-
technique, or by the application of constant potential at a par-
ite oxide by sonication in water. (3) Attachment of metal particles
on the graphene oxide sheets. (4) Formation of graphene-supported ticular time period. It is superior due to the possibility of hav-
metal particle composites by reduction of the graphene oxide ing good control over the degree of nucleation and growth by
sheets. (Adapted from B. K. Jena and C. R. Raj. Journal of Physical controlling nucleation potential, deposition potential, as well
Chemistry C 112, 2008, 3496.) as the concentration of the precursor metal salts [5,85,86].
HO OH HO COOH
O
HOOC O
OH
O
O OH
OH COOH
HO OH HO COOH
HO O
Oxidation HOOC O
OH
(1) O
O OH
OH COOH
HO COOH
HO O OH HO
HOOC O
OH
O
O OH
Graphite OH COOH
HO )
tCl 6 x (2 Graphite oxide
H 2P eflu
ol r
e glyc
ylen
O Eth
HOOC OH
O
HOOC OH
O
HOOC OH
Pt/CCG
SCHEME 14.16 Formation strategy to anchor platinum (Pt) nanoparticles onto chemically converted graphene (CCG) nanosheets. (1) Oxidation
of pure graphite powder to graphite oxide. (2) Formation of Pt/CCG hybrids. (Adapted from B. K. Jena and C. R. Raj. Langmuir 23, 2007, 4064.)
Ele
ct
(1) (3) co- roche
red mic
uct
Oxidation Adsorption ion al
Formation of
(2) Reduction (4) graphene and AuCl4–
particles e H2 H2 e
Graphite oxide Au-nanoparticle Graphene

SCHEME 14.17 Schematic illustration of graphene and graphene–
Cu2O composites. (1) Oxidation of graphite to graphite oxide with SCHEME 14.18 Schematic representation of the fabrication of
greater interlayer distance. (2) Direct reduction of graphite oxide DNA biosensor based on modified graphene gold nanocomposite.
with reductant. (3) Adsorption of copper acetate on the graphene (Adapted from L. Liu et al. Analytica Chimica Acta 767, 2013, 50.)
oxide sheets. (4) Formation of graphene-supported Cu2O particles
by the reduction of graphene oxide sheets and the in situ crystal
growth of Cu2O. (Adapted from C. Tan, X. Huang and H. Zhang. graphene using mesoporous silica templates via electrochem-
Materials Today 16, 2013, 29.) istry. This approach enables direct contact between the CdSe
quantum dots and the graphene film without intervening layers
Xia et al., have proposed three routes to modify the graphene which, in turn, enhance the electrical and mechanical properties
with nanoparticles. In the first route, they have reduced the of the resulting material (Scheme 14.19) [90].
graphene oxide-modified electrode electrochemically prior to Zhu et al., have demonstrated a facile and effective meth-
the electrodeposition of platinum NPs. In the second route, odology for large-scale synthesis of homogenously dispersed
platinum nanoparticles were first deposited on graphene oxide MnO2 nanoarrays on the graphene thin film coated on the sur-
and then reduction performed. In the third case, they have face of the glassy carbon electrode. The method involves the
reduced both graphene oxide as well as metal precursors cyclic voltammetric scanning of the GO-modified electrode
simultaneously to obtain a graphene–platinum nanocompos- in the potential window ranging from 0 to −1.5 V with a scan
ite [87]. Liu et al., have prepared gold nanoparticles-modified rate of 50 mV/s for about 10 cycles in a solution of 0.1 mol/L
graphene sheets by the simultaneous reduction of graphene Na2SO4 containing 0.05 mol/L of Mn(OAc)2 [91].
oxide, as well as gold precursor, by using a cyclic voltammetric Zhang et al., have reported the growth of Pt nanoflowers
technique in the potential range of 0.0 to −1.5 V by immersing through electrodeposition protocol. The obtained Pt nano-
the graphene oxide-coated electrode into a solution contain- flowers have shown superior electrochemical response toward
ing 2.8 mM of HAuCl4 and 0.1 M H2SO4 (Scheme 14.18) [88]. the oxidation of methanol due to the exposure of different fac-
Jiao et al., have modified the graphene with gold nanopar- ets of the platinum [92]. They have modified the carbon cloth
ticles of average size 10 nm by the electrochemical coreduction with graphene via electrochemical reduction of GO at −0.9 V
of graphene oxide and Au salts in presence of an ionic liquid at in sodium phosphate buffer solution. Then, the prepared RGO/
a constant potential of −2.0 V [89]. Yong et al., have proposed CCE was used as working electrode for the electrodeposition
a novel methodology for the production of homogenous thin of Pt at −0.2 V in a solution consisting of 3.0 mM H2PtCl6 and
films of CdSe quantum dot arrays on the basal plane sites of the 0.5 M H2SO4 (Scheme 14.20) [93].
(a) (b) (c) CdSe quantum dots in

Graphene sheet Mesoporous silica template mesoporous silica template
Ni layer
SiO2/Si wafer
(d)
CdSe quantum dots
SCHEME 14.19 Schematic of CdSe quantum dot array on the basal plane of a graphene sheet using a mesoporous silica thin film as a
template: (a) formation of a mesoporous silica film on the graphene surface by spin-casting, aging, and calcination, (b) electrochemical
deposition of CdSe onto the graphene surface through the pores of the mesoporous silica film template, (c) removal of the mesoporous silica
template, and (d) removal of the Ni layer underneath the graphene sheet. (Adapted from Y. T. Kim et al. Advanced Materials 22, 2010, 515.)
(a)
CdS
Electrochemical
Pt deposition
Cd2+, DMSO
reduction GO
180°C, 12 h
Graphene oxide Graphene-CdS

RGO Pt (b)
Carbon cloth
Carbon cloth RGO/Carbon cloth Pt/RGO/Carbon cloth
DMSO
SCHEME 14.20 Schematic illustration of the formation of Pt 180°C, 12 h
nanostructure/RGO/GCE. (Adapted from Z. Yao et al. Journal of
Material Chemistry 22, 2012, 13707.)
Graphene oxide Graphene
14.2.12 Hydrothermal and Solvothermal Techniques (c)
Graphene can be modified with different types of nano-
structures through hydrothermal as well as solvothermal Cd2+, DMSO Wash Resuspend
approaches. These methods have been frequently used in the 180°C, 12 h
designing of nanocomposites, and are generally performed (I) (II) (III) (IV)
in an autoclave at high temperature and pressure. The most
striking feature of this method is the ability to form nanopar-
ticles with high crystallinity without any synthetic annealing SCHEME 14.21 Schematic representation of solvothermal reduc-
or calcination treatments [53,94]. tion of (a) graphene oxide to graphene-CdS composite in presence
of cadmium ion solution and (b) graphene oxide to graphene in
Cao et al., have prepared graphene cadmium sulfide nano-
DMSO. (c) Bottle I, graphene oxide suspension in DMSO, bottle II,
composite (G-CdS) through one-pot solvothermal reaction. The completion of reaction as graphene–cadmium sulfide settles down,
reaction has been carried out by dispersing the GO and cad- bottle III, graphene–cadmium sulfide after washing with acetone
mium acetate in dimethyl sulphoxide (DMSO) in a Teflon-lined and ethanol, bottle IV, resuspension of graphene–cadmium sulfide.
stainless-steel autoclave at 180°C for 12 h (Scheme 14.21) [95]. (Adapted from A. Cao et al. Advanced Materials 22, 2010, 103.)
Shen et al., have also prepared graphene–silver composite
by one-pot hydrothermal synthesis. In this method, ascorbic
acid was used as a green reducing agent, and the reducing Marquardt et al., have proposed a very simple and repro-
ability of ascorbic acid has been greatly enhanced under ducible pathway for the synthesis of very small metal nanopar-
hydrothermal conditions (Scheme 14.22). The same group has ticles from their binary metal carbonyl complexes in presence
applied a similar strategy for the modification of graphene of ionic liquids (Scheme 14.24). They have described a simple,
with TiO2 solvothermally [96,97]. rapid, and efficient strategy to deposit metal nanoparticles of
ruthenium and rhodium on the surface of chemically derived
graphene by the decomposition of their metal carbonyls under
14.2.13 Microwave-Assisted Modification
microwave irradiation conditions in presence of an ionic l iquid,
with Metal Nanoparticles
that is, 1-butyl-3-methylimidazolium tetrafluroborate [101].
Microwave-assisted synthesis of metal nanoparticle graphene Nethravathi et al., have synthesized graphene vanadate
composite is a fascinating approach since it allows the simul- nanotube hybrid material under hydrothermal conditions. The
taneous reduction of graphene oxide and metal ions with min- hydrothermal reaction was carried out by taking the mixture
imum time [98]. On the basis of the use of microwaves, Wang of colloidal dispersion of graphene oxide and ammonium van-
et al., have developed a novel one-pot synthesis strategy to adate in Teflon-coated autoclave, and subjected to heat treat-
prepare both SnO2/graphene nanocomposites using a micro- ment at 180°C for 48 h. The reduction of GO accompanied the
wave-assisted polyol process. In this approach, it has been formation of VO2 nanotubes leading to vanadium nanotube
realized that the reduction of both graphene oxide to graphene graphene nanohybrid material (Scheme 14.25) [102].
and SnO2 nanoparticles from SnCl2 precursors takes place Liu et al., have prepared silver nanoparticles-decorated
simultaneously [99]. Zhang et al., have designed a magnetite/ graphene under microwave irradiation in a single step. The
graphene composite by depositing Fe3+ in the interspaces of reaction has been carried out by taking suspension of GO
graphene sheets through microwave irradiation. This method in a DMF solvent which can act as both solvent as well as
involves the reduction of graphene oxide using ascorbic acid, reducing agent followed by the addition of AgNO3 and sub-
and the deposition of Fe3+ ions by urea with continuous nucle- jected to microwave irradiation for about 2 min. The pro-
ation as well as growth rates by a microwave irradiation process results in the formation of Gr–Ag composite (Scheme
cess (Scheme 14.23) [100]. 14.26) [103].
Ionic liquid GO
Stir Sonication Ascorbic acid
AgNO3
NH4OH
– HO OH
– – OH
O O
– – Hydrothermal O OHC
OH
OH
Ag+ method
– – O
Ag+ Ag+ HOOC
– – GO
OHC
– – Ag RGO HO
HO
O O
HO OH
SCHEME 14.22 Schematic pathway of preparation of silver-reduced graphene oxide. (Adapted from J. Shen et al. Journal of Material
Chemistry 21, 2011, 7795.)
H H H
H
O O O O O O
OH OH O O O O
OH
H HO H
O O
O O
O O OH
H H O
O O Fe3+, Urea, AC Ar, 873 K
O O Microwave 8h
OH
H H
O O
O
O OH
OH
O O
H
O OH OH OH O OH
O O O O O O H
H H H
Fe3+ FexOvHz Fe3O4
SCHEME 14.23 A synthetic pathway of magnetite particles dispersed on graphene sheets by a microwave irradiation method. (Adapted
from M. Zhang et al. Journal of Material Chemistry 20, 2010, 5538.)
14.2.14 Sonochemical Modification concentration levels has been considered as a serious issue

with Metal Nanoparticles relating to environmental pollution [106]. As a result of its
toxicity and longer stability in the environment, exposure
In recent years, sonochemistry has been used to prepare gra- of these metal ions to living systems may cause irreversible
phene by the reduction of graphene oxide and also to mod- adverse health effects. Hence, the importance in controlling
ify it with nanoparticles. The basis for this approach is the the levels of these pollutants in natural waterways and potable
cavitation effects generated during sonolysis process which, water has generated increasing interest in the development
in turn, results in the formation of oxidative and reductive of new sensors for the detection of these metal ions at trace
radicals which can be further tuned to initiate specific chemi- levels [107]. CMEs have become a major area of research in
cal reactions [104]. Kamath and coworkers have developed electroanalysis due to their tailor-made properties which meet
a novel strategy to produce and disperse gold nanoparticles specific requirements like specificity, sensitivity, and selectiv-
on the surface of reduced graphene oxide as a support. Both ity toward a particular analyte, as it reduces the overpotential
GO as well as gold salt have been simultaneously reduced compared to unmodified electrodes [108]. Several carbon-
under sonication conditions for about 5 h at a frequency of based substrate materials have been widely used as substrate
200 KHz which is sufficient enough to produce reactive radi- materials because of availability in various forms, and are
cals at higher concentration much faster than the conventional amenable to further modification with suitable modifier mol-
chemical reduction processes (Scheme 14.27) [105]. ecules [109]. Graphene and graphene derivatives have been
used in the development of metal ion sensors. Willemse et al.
14.3 APPLICATIONS have prepared mercury film-modified nafion-coated graphene
electrodes toward the sensitive and selective electrochemical
14.3.1 Metal Ion Sensing
determination of Zn2+, Cd2+, Pb2+ and Cu2+ ions in an acetate
The presence of heavy metal ions like lead, cadmium, arse- buffer solution of pH 4.6 using square wave anodic stripping
nic, mercury, chromium, etc. in the environment even at trace voltammetry (Figure 14.2) [110–112].
in the environmental samples. This selectivity has been attrib-

Graphite uted to the synergistic effect of graphene nanosheets as well as
the ion exchange capacity of nafion films (Figure 14.3).
NaNO3, KMnO4
The sharper and higher peak currents toward the target
H2SO4
metal ions with respect to the nafion–graphene modified elec-
trode are mainly due to the following two factors.
Graphite
oxide (GO) 1. The change of the morphology and the structure.
2. The interfacial electron-transfer properties due to
increased surface area.
N2, Δ, 30 s
Chemically Ramesha et al., have designed a reduced graphene-oxide
derived
graphene (CDG)
lead-dioxide composite by the one-pot electrochemical reduc-
tion of exfoliated graphene oxide on the surface of glassy
Mx(CO)y/IL (BMImBF4) carbon electrode (GCE) and lead ions in 1 M hydrochloric
microwave irradiation
6 min, 20 W (Ar)
acid solution [114]. This composite showed excellent elec-
trocatalytic activity toward the determination of arsenic at
nanomolar concentration level. Different types of oxygen-
containing functional groups present on the exfoliated gra-
phene oxide are known to be in ionizable form in a wide pH
range of 2–11 which, in turn, is responsible for the favorable
M-NP/CDG interaction between metal ions in solution. The detection limit
M = Ru, Rh of this sensor for arsenic has been found to be 10 nM, and
hence, it opens up the possibility of using the type of graphene
= M-NP composite for sensing applications and possibly simultaneous
detection of arsenic and lead.
Gong et al., have prepared Au nanoparticles-decorated gra-
SCHEME 14.24 Schematic representation of synthesis of graphene phene as an enhanced sensing platform for ultrasensitive strip-
oxide from graphite and subsequent modification with nanoparticles ping voltammetric detection of mercury (II) [115]. The platform
using metal carbonyls. (Adapted from D. Marquardt et al. Carbon
was constructed by the homogenous dispersion of graphene in
49, 2011, 1326.)
a solution containing CS. This nanoparticle-based graphene
The mechanism of the above electrode reaction proceeds composite showed good electroactivity toward m ercury due to
as follows: the synergetic effect produced by the presence of gold nanopar-
ticles as well as graphene nanosheets (Figure 14.4).
Mn+ + ne + Hg → M(Hg): Deposition step
M(Hg) → Mn+ + ne + Hg: Stripping step. 14.3.2 Hydrogen Peroxide (H2O2) Sensing

Recently, Jing Li et al., have modified graphene nanosheets H2O2 is a powerful oxidizing agent generally used as an anti-
with nafion followed with bismuth film. This modified electrode microbial agent in food samples, and also as a sterilizing agent
has been successfully applied in the simultaneous determina- in various fruit juices as well as milk products. However, per-
tion of lead and cadmium ions using differential pulse anodic oxide of concentration more than 3% may result in vomiting,
stripping voltammetry [113]. The fabricated modified graphene mild irritation to mucosa, and burns in the mouth, throat and
electrode showed improved sensitivity toward the target ana- stomach. Therefore, the rapid and reliable quantification of
lytes with least interference from commonly encountered ions H2O2 in food samples became a very important aspect of food
NH4VO3 formic acid
Hydrothermal 180°C
Aqueous colloidal dispersion VO2 (B) nanotubes

of graphene oxide graphene hybrid
SCHEME 14.25 Schematic illustration of the synthesis of VO2 (B) nanotubes–graphene hybrid. (Adapted from C. Nethravathi et al.
Carbon 50, 2012, 4839.)
COOH COO–
OH
HOOC –OOC
O
OH
O COOH COO–
OH OH
DMF, AgNO3 –OOC
HOOC
O HO
OH O
–OOC
HOOC COOH COO–
DMF: N,N-dimethylformamide Microwave Ag nanoparticles
SCHEME 14.26 Schematic representation of AgNPs/GN formation from GO and AgNO3 using DMF as a reducing agent and solvent.
(Adapted from S. Liu et al. Journal of Nanoparticle Resources 13, 2011, 4539.)
COOH
OH
20
O Nafion-G
O 16 Nafion
Current (μA)
12
•H –•CH2O
RGO-Au 8
Sonolysis 4
SCHEME 14.27 Dispersion of graphene sheets by anchoring 0

metal nanoparticles. (Adapted from K. Vinodgopal et al. Journal of –1.0 –0.8 –0.6 –0.4 –0.2
Physical Chemistry 1, 2010, 1987.) Potential/V (vs. Ag/AgCl)
FIGURE 14.3 Overlaid differential pulse anodic stripping voltam-

3.2 × 10–5 mograms of 20 mg L −1 of each Cd2+ and Pb2+ ions at in situ plated
Graphene-HgFE Nafion–bismuth film electrode, Nafion-graphene–bismuth film elec-
Nafion-G-HgFE
2.8 × 10–5 Nafion-HgFE trode. (Adapted from D. Li et al. Talanta 83, 2010, 332.)
a HgFE
2.4 × 10–5 GCE
0.8
2.0 × 10–5 (a) 0.6 60
(b)
Current (μA)
b 0.5 50
i/mA
0.7 40
1.6 × 10–5 0.4
30
i/mA
0.3 20
0.6
1.2 × 10–5 c 0.2 0.00 0.02 0.04 0.06
0.1 (Hg2+)/ppb
0.5 0.0
8.0 × 10–6 e
0 10 20 30 40 50 60
i/mA
(Hg2+)/ppb
0.4
4.0 × 10–6
d 0.3
0.0
–1.4 –1.2 –1.0 –0.8 –0.6 –0.4 –0.2 0.2
Potential (V)
0.1
–0.4 –0.2 0.0 0.2 0.4 0.6 0.8
FIGURE 14.2 Overlaid square wave anodic stripping voltam-
mograms of 30 μg L −1 of Zn2+, Cd2+ and Pb2+ ions recorded at (a) E/V vs (SCE)
Nafion-graphene mercury film electrode, (b) graphene mercury film
electrode, (c) Nafion mercury film electrode, (d) mercury film elec- FIGURE 14.4 Overlaid stripping voltammograms recorded with
trode, and (e) glassy carbon electrode in 0.1 M acetate buffer solu- the increased concentration of mercury. Inset plots (a) and (b) of dif-
tion of pH 4.6. (Adapted from C. M. Willemse et al. Sensors 11, ferent concentration range. (Adapted from J. Gong et al. Sensors and
2011, 3970.) Actuators B: Chemical 150, 2010, 491.)
0 H2O2 H2O
0.1 mM Catalase
1 mM
–10 N
Thionine H 2N
.
S NH2 Catalase Compound I
400
Current (μA)
350 TCA
–20 300
Current (μA)
250 Compound II Thioninered

200
150
–30 Graphene oxide
100
50 Thionineox
0
Catalase
0 20 40 60 80 100 Stirring for
–40 Concentration (mM) AuCl4–
2 h at 90°C
100 200 300 400 500 600
Time (s)
FIGURE 14.5 Typical steady-state response obtained at sil-

ver nanoparticle-reduced graphene oxide-modified glassy carbon
electrode (AgNP/rGO/GCE) with the successive addition of H2O2. SCHEME 14.28 Schematic illustration of catalytic-sensing pro-
Inset is the calibration plot. (Adapted from S. Liu et al. Journal of cess of hydrogen peroxide.
Nanoparticle Resources 13, 2011, 4539.)
technology [116]. In order to quantify H2O2, Liu and group of H2O2 at bio composite. A typical cyclic voltammetric pro-
members have used graphene–silver nanocomposite for the file with increased concentrations of H2O2 is shown.
reduction of (H2O2) ampherometrically [103] (Figure 14.5). It showed high sensitivity and fast response of 3 S in pH
The above figure represents the typical current–time plot 6.5 solution, with a potential of −0.3 V. The composite showed
for H2O2 analyte at AgNP/rGO/GCE with the successive linear range from 5 × 10−6 M to 5.13 × 10−3 M with a detection
additions of H2O2 concentrations under optimized conditions. limit of 1.7 × 10−6 M (S/N = 3) (Figure 14.6).
The composite showed fast amperometric response within Huang et al., have used amino functionalized graphene
2 s with a linear range from 100 µM to 100 mM. Zhou et al., gold nanoparticles in the development of H2O2 biosensors
have prepared gold graphene horseradish peroxidase (HRP) [118]. The newly developed biosensor showed good perfor-
CS biocomposite as a novel H2O2 biosensing platform [117]. mance such as high sensitivity, selectivity, improved stability
The platform was constructed by immobilizing graphene and, mainly, short response time and wide linear range owing
and HRP into CS matrix which was coated on the surface of to the synergic effects posed by both amine-functionalized
glassy carbon electrode followed by electrodeposition of Au graphene and gold nanoparticles. Zhang et al., have devel-
nanoparticles on the surface to fabricate Au/graphene/HRP/ oped a biosensor for the quantification of H2O2 present in food
CS/GCE. Cyclic voltammetry has been used to decipher the samples using well dispersed gold nanoparticles-modified
direct electron transfer of HRP and electrocatalytic reduction graphene oxide (Scheme 14.28) [119]. The sensor was initially
designed by using ultrafine gold nanoparticles-dispersed gra-
phene oxide assembled onto the surface of glassy carbon elec-
25
trode. The sensor exhibited good electrocatalytic efficiency
0.25 mM toward the reduction of H2O2 in acetate buffer solution of
20
pH 5.8. At the optimal conditions, the dynamic responses of
the biosensor toward H2O2 were achieved in the range from
Current (μA)
15
0 0.1 to 2.3 mM with a detection limit of 0.01 mM. In order to
10 establish the wide range applicability of the sensor, it has been
successfully applied in the sensing of H2O2 present in natu-
5 rally contaminated samples like sterilized milk, apple juices,
watermelon juice, coconut milk, and mango juice.
0 Ye et al., have introduced a nonenzymatic H2O2 sensor
based on graphene ferroferric oxide nanocomposite [120].
–5 The composite was fabricated by the coprecipitation method.
0.4 0.2 0.0 –0.2 –0.4 –0.6
Potential/V It showed enhanced catalytic activity toward the reduction of
peroxide. The developed sensor showed a linear range from
FIGURE 14.6 Overlaid cyclic voltammograms of Au/GS/HRP/ 0.1 to 6 mM with less than 5 s response time and the detection
CS/GCE with successive addition of 0.05 mM H2O2 in 0.1 M PBS limit was found to be 3.2 µM (S/N = 3).
at pH 6.5 at a scan rate of 0.1 V s−1. (Adapted from K. Zhou et al. Zhou et al., have utilized the direct electrochemistry of
Electrochemica Acta 55, 2010, 3055.) sacrosine oxidase (SOX)-immobilized graphene modified
FAD
Graphene 2e–/2H+
FADH2
GCE
FAD H2O
H2O2
2e–/2H+
0.1 M PBS
FADH2 H2O2
Graphene Silver nanoparticle Sarcosine oxidase (SOX)
SCHEME 14.29 Schematic illustration of biosensor fabrication and detection strategy of H2O2. (Adapted from Y. Zhou et al. Electrochimica
Acta 71, 2012, 294.)
TABLE 14.3
Applications of Graphene–Metal Nanoparticle Composite
Substrate Metal Nanoparticle Analyte Technique Detection Limits (3σ) References
Graphene Ag H2O2 Amperometric 2.8 µM [103]
Au H2O2 Amperometric 0.44 µM [117]
Fe3O4 H2O2 Amperometric 0.6 µM [120]
CdS H2O2 Amperometric 1.7 µM [122]
PbO As3+ SWASV 10 nM [114]
Pt As3+ DPASV 1 nM [123]
Au As3+ DPASV 1 nM [115]
Note: SWASV, square wave anodic stripping voltammetry; DPASV, differential pulse anodic stripping voltammetry.
with CS and silver nanoparticles (Scheme 14.29) [121]. The discussion recently because of its unique structure proper-
immobilized SOX exhibits surface confined to two protons ties. In this chapter, various strategies of graphene modifi-
and two electron transfer kinetics. It showed excellent sensi- cation, their methodologies and some of its applications in
tivity toward H2O2 in 0.1 M PBS (pH 7.0) with a wide linear electrochemical sensing have been presented.
range from 1.0 to 177 µM with a detection limit of 1.0 µM.
The excellent performance of the biosensor is attributed to
a large specific surface area, and high conductivity of gra- ACKNOWLEDGMENTS
phene and AgNPs, and good biocompatibility of CS which The authors acknowledge the financial support and award of
improve the enzyme absorption as well as in the promotion of Senior Research Fellowship to Raghu GK from the Council
direct electron transfer between SOX and the electrode sur- of Scientific and Industrial Research (CSIR), New Delhi,
face (Table 14.3). India.
14.4 CONCLUSIONS REFERENCES
Chemical modification provides an easy and useful strat- 1. C. N. R. Rao, A. K. Sood and K. S. Subhramanyam. Journal
egy to impart selective, specific properties on the surface of of Physical Chemistry C Letters 1, 2010, 572.
2. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y.
carbon substrate materials. Hence, CMEs have become sig-
Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov.
nificant in the area of electroanalytical chemistry in recent Science 306, 2004, 666.
years due to their high sensitivity and selectivity toward the 3. A. Du and S. C. Smith. Journal of Physical Chemistry C
target analyte. Chemical modification of graphene and its Letters 2, 2011, 73.
application in environmental sensing has been the point of 4. S. Park and R. S. Ruoff. Nature Nanotechnology 4, 2009, 217.
5. Y. Shao, J. Wang, M. Engelhard, C. Wang and Y. Lin. Journal 35. H. J. Salavagione, M. A. Gomez and G. Martinez.
of Materials Chemistry 20, 2010, 743. Macromolecules 42, 2009, 6331.
6. H. L. Guo, X. F. Wang, Q. Y. Qian, F. B. Wang and X. H. Xia. 36. X. Chen, L. Yuan, P. Yang, J. Hu and Dong Yang. Journal of
ACS Nano 3, 2009, 2653. Polymer Science Part A: Polymer Chemistry 49, 2011, 4977.
7. M. Pumera. The Chemical Record 9, 2009, 211. 37. S. Mohamadi, N. S. Sanjani and H. Mahdavi. Journal
8. D. A. C. Brownson and C. E. Banks. Physical Chemistry of Macromolecular Science, Part A: Pure and Applied
Chemical Physics 14, 2012, 8264. Chemistry 48, 2011, 577.
9. D. A. C. Brownson and C. E. Banks. Analyst 135, 2010, 2768. 38. Z. Xu and C. Gao. Macromolecules 43, 2010, 6716.
10. M. Pumera. Chemical Society Reviews 39, 2010, 4146. 39. N. T. Tung, T. V. Khai, M. Jeon, Y. J. Lee, H. Chung, J. H.
11. I. Zaman, H. C. Kuan, Q. Meng, A. Michelmore, N. Bang and D. Sohn. Macromolecular Research 19, 2011, 203.
Kawashima, T. Pitt, L. Zhang, S. Gouda, L. Luong and J. Ma. 40. S. H. Lee, D. R. Dreyer, J. An, A. Velamakanni, R. D. Piner,
Advanced Functional Materials 22, 2012, 2735. S. Park, Y. Zhu, S. O. Kim, C. W. Bielawski and R. S. Ruoff.
12. G. G. Wallace, J. Chen, D. Li, S. E. Moulton and J. M. Razal. Macromolecules. Rapid Communications 31, 2010, 281.
Journal of Materials Chemistry 20, 2010, 3553. 41. C. H. Chang, T. C. Huang, C. W. Peng, T. C. Yeh, H. I. Lu,
13. D. Pan, S. Wang, B. Zhao, M. Wu, H. Zhang, Y. Wang and Z. W. I. Hung, C. J. Weng, T. I. Yang and J. M. Yeh. Carbon 50,
Jiao. Chemistry of Materials 21, 2009, 3136. 2012, 5044.
14. X. Kang, J. Wang, H. Wu, J. Liu, I. A. Aksay and Y. Lin. 42. I. J. Cutress, F. Marken and R. G. Compton. Electroanalysis
Talanta 81, 2010, 754. 21, 2009, 113.
15. R. W. Murray, A. G. Ewlng and R. A. Dust. Analytical 43. S. Caddick and R. Fitzmaurice. Tetrahedron 65, 2009, 3325.
Chemistry 59, 1987, 379. 44. C. Xu, J. Wang, L. Wan, J. Lin and X. Wang. Journal of
16. J. Oni, N. Diab and S. Reiter. Sensors and Actuators B: Material Chemistry 21, 2011, 10463.
Chemical 105, 2005, 208. 45. S. P. Economopoulos, G. Rotas, Y. Miyata, H. Shinohara and
17. D. W. M. Arrigan. Analyst 119, 1994, 1953. N. Tagmatarchis. ACS Nano 4, 2010, 7499.
18. Z. B. Liu, Y. F. Xu, X. Y. Zhang, X. L. Zhang, Y. S. Chen and 46. H. Hu, X. Wang, J. Wang, F. Liu, M. Zhang and C. Xu. Applied
J. G. Tian. Journal of Physical Chemistry B 113, 2009, 9681. Surface Science 257, 2011, 2637.
19. Z. Jin, J. R. Lomeda, B. K. Price, W. Lu, Y. Zhu and J. M. 47. B. Rodrguez, A. Bruckmann, T. Rantanen and C. Bolm.
Tour. Chemistry of Materials 21, 2009, 3045. Advanced Synthetic Catalysis 314, 2007, 2213.
20. X. Zhong, J. Jin, S. Li, Z. Niu, W. Hu, R. Li and J. Ma. 48. M. Pandurangappa, T. Ramakrishnappa and R. G. Compton.
Chemical Communication 46, 2010, 7340. Carbon 47, 2009, 2186.
21. J. Liu, J. Tang and J. J. Gooding. Journal of Materials 49. Y. Lin, K. Kent, A. Watson, S. Ghose, J. G. Smith, T. V.
Chemistry 22, 2012, 12435. Williams, R. E. Crooks, W. Cao and J. Connell. Journal of
22. J. Pinson and F. Podvorica. Chemical Society Reviews 34, Physical Chemistry C 113, 2009, 14858.
2005, 429. 50. H. Pan, L. Liu, Z. X. Guo, L. Dai, F. Zhang, D. Zhu, R. Czerw
23. J. R. Lomeda, C. D. Doyle, D. V. Kosynkin, W. F. Hwang and and D. L. Carroll. Nano Letters 3, 2003, 29.
J. M. Tour. Journal of American Chemical Society 130, 2008, 51. L. Yan, M. Lin, C. Zeng, Z. Chen, S. Zhang, X. Zhao, A. Wu,
16201. Y. Wang, L. Dai, J. Qu, M. Guod and Yong Liu. Journal of
24. Y. Lu, A. Y. Jiang, A. W. Wei, A. H. Wu, A. M. Liu, A. L. Materials Chemistry 22, 2012, 8367.
Niua and W. Chen. Journal of Material Chemistry 22, 2012, 52. H. Liu, S. Ryu, Z. Chen, M. L. Steigerwald, C. Nuckolls and
2929. L. E. Brus. Journal of American Chemical Society 131, 2009,
25. W. Gao, M. Majumder, L. B. Alemany, T. N. Narayanan, M. 17099.
A. Ibarra, B. K. Pradhan and P. M. Ajayan. Appllied Materials 53. S. Bai and X. Shen. RSC Advances 2, 2012, 64.
and Interfaces 3, 2011, 1821. 54. M. Song and D. Cai. RSC Nano Science and Nanotechnology,
26. Q. Zhuo, J. Gao, M. Peng, L. Bai, J. Deng, Y. Xia, Y. Ma, J. Polymer-Graphene Nanocomposites, NO 26, Royal Society
Zhong and X. Sun. Carbon 52, 2013, 559. of Chemistry, Cambridge, UK, ISBN: 978-1-84973-567-4.
27. M. C. Hsiao, S. H. Liao, M. Y. Yen, P. I. Liu, N. W. Pu, C. A. 55. Y. Xu, H. Bai, G. Lu, C. Li and G. Shi. Journal of American
Wang and C. C. M. Ma. Applied Materials and Interfaces 2, Chemical Society 130, 2008, 5856.
2010, 3092. 56. H. Bai, Y. Xu, L. Zhao, C. Li and G. Shi. Chemical
28. Y. Xu, Z. Liu, X. Zhang, Y. Wang, J. Tian, Y. Huang, Y. Ma, Communications 2009, 1667.
X. Zhang and Y. Chen. Advanced Materials 21, 2009, 1275. 57. Q. Su, S. Pang, V. Alijani. C. Li, X. Feng and K. Mullen.
29. S. R. Krishna, V. Navakantha, B. Amurutur, K. Chakrapani Advanced Materials 21, 2009, 3191.
and S. Sampath. Institute of Electrical and Electronics 58. Q. Yang, X. Pan, F. Huang and K. Li. Journal of Physical
Engineers 2011, 1–4. Chemistry C 114, 2010, 3811.
30. Y. Yu, M. Zhou, W. Shen, H. Zhang, Q. Cao and H. Cui. 59. J. A. Mann, J. R. Lopez, H. D. Abruna and W. R. Dichtel.
Carbon 50, 2012, 2539. Journal of American Chemical Society 133, 2011, 17614.
31. M. B. Murali Krishna, N. Venkatramaiah, R. Venkatesan and 60. E. Y. Choi, T. H. Han, J. Hong, J. E. Kim, S. H. Lee, H. W.
D. N. Rao. Journal of Materials Chemistry 22, 2012, 3059. Kima and S. O. Kim. Journal of Material Chemistry 20, 2010,
32. H. J. Salavagione, G. Martínez and G. E. Macromol. Rapid 1907.
Communications 32, 2011, 1771. 61. H. Liu, J. Gao, M. Xue, N. Zhu, M. Zhang and T. Cao.
33. T. Kuilla, S. Bhadra, D. Yaoa, N. H. Kimc, S. Bosed and J. H. Langmuir 25, 2009, 12006.
Leea. Progress in Polymer Science 35, 2010, 1350. 62. W. Tu, J. Lei, S. Zhang and H. Ju. Chemistry European
34. K. P. Pramoda, H. Hussain, H. M. Koh, H. R. Tan and C. B. Journal 16, 2010, 10771.
He. Journal of Polymer Science Part A: Polymer Chemistry 63. Y. Fang, S. Guo, C. Zhu, Y. Zhai and E. Wang. Langmuir 26,
48, 2010, 4262. 2010, 11277.
64. X. Zhang, Y. Feng, S. Tang and W. Feng. Carbon 48, 2010, 94. X. Huang, X. Qi, F. Boey and H. Zhang. Chemical Society
211. Reviews 41, 2012, 666.
65. A. Ghosh, K. V. Rao, S. J. George and C. N. R. Rao. Chemistry 95. A. Cao, Z. Liu, S. Chu, M. Wu, Z. Ye, Z. Cai, Y. Chang, S.
European Journal 16, 2010, 2700. Wang, Q. Gong and Y. Liu. Advanced Materials 22, 2010, 103.
66. Z. Liu, J. T. Robinson, X. Sun and H. Dai. Journal of American 96. J. Shen, M. Shi, B. Yan, H. Ma, N. Li and M. Ye. Journal of
Chemical Society 30, 2008, 10876. Material Chemistry 21, 2011, 7795.
67. D. Yu, Y. Yang, M. Durstock, J. B. Baek and L. Dai. ACS 97. J. Shen, M. Shi, B. Yan, H. Ma, N. Li and M. Ye. Nano
Nano 4, 2010, 5633. Resources 4, 2011, 795.
68. X. D. Zhuang, Y. Chen, G. Liu, P. P. Li, C. X. Zhu, E. T. Kang, 98. H. M. A. Hassan, V. Abdelsayed, A. E. R. S. Khder, K. M.
K. G. Noeh, B. Zhang, J. H. Zhu and Y. X. Li. Advanced AbouZeid, J. Terner, M. S. ElShall, S. I. A. Resayes, A. A. E.
Materials 22, 2010, 1731. Azhary. Journal of Material Chemistry 19, 2009, 3832.
69. D. Valuga, J. M. Thmson, I. Molenberg, I. Huynen, B. Gilbert, 99. S. Wang, S. P. Jiang and X. Wang. Electrochimica Acta 56,
C. Jerome, M. Alexandree and C. Detrembleur. Chemical 2011, 3338.
Communications 47, 2011, 2544. 100. M. Zhang, D. Lei, X. Yin, L. Chen, Q. Li, Y. Wang and T.
70. L. Kan, Z. Xu and C. Gao. Macromolecules 44, 2011, 444. Wang. Journal of Material Chemistry 20, 2010, 5538.
71. Y. Huang, Y. Qin, Y. Zhou, H. Niu, Z. Z. Yu and J. Y. Dong. 101. D. Marquardt, C. Vollmer, R. Thomann, P. Steurer, R.
Chemistry of Materials 22, 2010, 4096. Mulhaupt, E. Redel, C. Janiak. Carbon 49, 2011, 1326.
72. P. V. Kamat. Journal of Physical Chemistry C Letters 1, 102. C. Nethravathi, B. Viswanath, J. Michael and M. Rajamath.
2010, 520. Carbon 50, 2012, 4839.
73. M. C. Daniel and D. Astruc. Chemical Reviews 104, 2004, 293. 103. S. Liu, J. Tian, L. Wang and X. Sun. Journal of Nanoparticle
74. P. C. Biswas. Nodasaka, M. Enyo and M. Haruta. Journal of Resources 13, 2011, 4539.
Electroanalytical Chemistry 381, 1995, 167. 104. G. Cravotto and P. Cintas. Chemistry European Journal 16,
75. B. K. Jena and C. R. Raj. Journal of Physical Chemistry C 2010, 5246.
112, 2008, 3496. 105. K. Vinodgopal, B. Neppolian, I. V. Lightcap, F. Grieser, M.
76. B. K. Jena and C. R. Raj. Langmuir 23, 2007, 4064. Ashokkumar and P. V. Kamat. Journal of Physical Chemistry
77. C. Tan, X. Huang and H. Zhang. Materials Today 16, 2013, 29. 1, 2010, 1987.
78. M. Chatenet, F. Micoud, I. Roche and E. Chainet. 106. M. Patra, N. Bhowmik, B. Bandopadhyay and A. Sharma.
Electrochimica Acta 51, 2006, 5459. Environmental and Experimental Botany 52, 2004, 199.
79. M. R. Das, R. K. Sarma, R. Saikia, V. S. Kale, M. V. Shelke 107. R. G. Kempegowda and M. Pandurangappa. Analytica
and P. Sengupta. Colloids and Surfaces B: Biointerfaces 83, Chimica Acta 728, 2012, 9.
2011, 16. 108. P. Ramesha and S. Sampath. Analyst 126, 2001, 1872.
80. R. S. Dey and C. R. Raj. Journal of Physical Chemistry. C 109. R. L. McCreery. Chemical Reviews 108, 2008, 2646.
114, 2010, 21427. 110. D. A. C. Brownson and C. E. Banks. Electrochemistry
81. H. W. Ha, I. Y. Kim, S. J. Hwang and R. S. Ruoff. Communications 13, 2011, 111.
Electrochemical and Solid State Letters 14, 2011, B70. 111. C. M. Willemse, K. Tlhomelang, N. Jahed, P. G. Baker and E.
82. C. Xu, X. Wang, J. Zh. Journal of Physical Chemistry C 112, I. Iwuoha. Sensors 11, 2011, 3970.
2008, 19841. 112. J. Li, S. Guo, Y. Zhai and E. Wang. Analytica Chimica Acta
83. Y. Li, W. Gao, L. Ci, C. Wang and P. M. Ajayan. Carbon 48, 649, 2009, 196.
2010, 1124. 113. D. Li, J. Jia and J. Wang. Talanta 83, 2010, 332.
84. C. Xu, X. Wang, L. Yang and Y. Wu. Journal of Solid State 114. G. K. Ramesha and S. Sampath. Sensors and Actuators B:
Chemistry 182, 2009, 2486. Chemical 160, 2011, 306.
85. R. S. Sundaram, C. G. Navarro, K. Balasubramanian, M. 115. J. Gong, T. Zhou, D. Song and L. Zhang. Sensors and
Burghard and K. Kern. Advanced Materials 20, 2008, 3050. Actuators B: Chemical 150, 2010, 491.
86. R. Muszynski, B. Seger and P. V. Kamat. Journal of Physical 116. Y. Liu, D. Wang, L. Xu, H. Hou and T. You. Biosensors and
Chemistry C 112, 2008, 5265. Bioelectronics 26, 2011, 4585.
87. Y. G. Zhou, J. J. Chen, F. B. Wang, Z. H. Sheng and X. H. Xia. 117. K. Zhou, Y. Zhu, X. Yang, J. Luo, C. Li and S. Luan.
Chemical Communications 46, 2010, 5951. Electrochemica Acta 55, 2010, 3055.
88. A. L. Liu, G. X. Zhong, J. Y. Chen, S. H. Weng, H. N. Huang, 118. K. J. Huang, D. J. Niu, X. Liu, Z. W. Wu, Y. Fan, Y. F. Chang
W. Chen, L. Q. Lin et al. Analytica Chimica Acta 767, 2013, 50. and Y. Y. Wu. Electrochimica Acta 56, 2011, 2947.
89. X. X. Jiao, H. Q. Luo and N. B. Li. Journal of Electroanalytical 119. B. Zhang, Y. Cui, H. Chen, B. Liu, G. Chen, and D. Tang.
Chemistry 691, 2013, 83. Electroanalysis 23, 2011, 1821.
90. Y. T. Kim, J. H. Han, B. H. Hong and Y. U. Kwon. Advanced 120. Y. Ye, T. Kong, X. Yu, Y. Wu, K. Zhang and X. Wang. Talanta
Materials 22, 2010, 515. 89, 2012, 417.
91. C. Zhu, S. Guo, Y. Fang, L. Han, E. Wang and S. Dong. Nano 121. Y. Zhou, H. Yin, X. Meng, Z. Xu, Y. Fu and S. Ai.
Research 4, 2011, 648. Electrochimica Acta 71, 2012, 294.
92. H. Zhang, W. Zhou, Y. Du, P. Yang and C. Wang. 122. A. Salimi, R. Rahmatpanah, R. Hallaj and M. Roushani.
Electrochemistry Communications 12, 2010, 882. Electrochimica Acta 95, 2013, 60.
93. Z. Yao, M. Zhu, F. Jiang, Y. Du, C. Wang and P. Yang. Journal 123. G. K. Raghu, D. Antony and M. Pandurangappa. International
of Material Chemistry 22, 2012, 13707. Journal of Smart and Nano Materials 5, 2014, 17.
15 Electrochemical Exfoliation and Applications
Graphene
Quang Duc Truong and Itaru Honma
CONTENTS
Abstract...................................................................................................................................................................................... 269
15.1 Introduction...................................................................................................................................................................... 269
15.2 An Overview of Graphene Synthesis Method.................................................................................................................. 270
15.3 Electrochemical Exfoliation............................................................................................................................................. 273
15.3.1 Exfoliation in IL................................................................................................................................................... 274
15.3.2 Exfoliation in Common Electrolyte...................................................................................................................... 275
15.3.3 Exfoliation in Alkali Metal Salt........................................................................................................................... 277
15.3.4 Exfoliation in Acid................................................................................................................................................ 279
15.3.5 Exfoliation in Surfactants..................................................................................................................................... 280
15.4 Applications...................................................................................................................................................................... 283
15.4.1 Precursor for the Synthesis of N-Doped Graphene.............................................................................................. 283
15.4.2 Conductive Paper.................................................................................................................................................. 284
15.4.3 Electrochemical Sensors....................................................................................................................................... 285
15.4.4 Water Purification................................................................................................................................................. 286
15.4.5 Energy Storage and Conversion Devices.............................................................................................................. 287
15.5 Concluding Remarks........................................................................................................................................................ 290
References.................................................................................................................................................................................. 291
ABSTRACT studies have proved its outstanding mechanics, electrical,

chemical properties, enhanced thermal stability, large specific
Graphene, with hexagonally, sp2-hybridized and one-atom- surface areas, high mobility of charge carriers, and thermal
thick layer structures, possesses high conductivity, superior conductivity due to its unique structure with all the carbon
electron mobility, and extremely high specific surface area, atoms linked together by strong and flexible bonds [4]. The
and can be produced on a large scale for various industrial outstanding properties of graphene reported so far include
applications. Thus, many techniques have been proposed for its high values of Young’s modulus (~1100 GPa), fracture
the synthesis of graphene including chemical vapor deposi- strength (125 GPa) [5], thermal conductivity (~5000 W/m K)
tion, exfoliation of graphite oxide, electrochemical synthe- [6], mobility of charge carriers (200,000 cm2/V s) [7], chemi-
sis, etc. Especially, electrochemical exfoliation routes have cal stability, and optical transmittance, as well as fascinating
attracted extensive attention because of their fast and simple transport phenomena such as quantum Hall effect at room
implementation and potential for mass production. This chap- temperature [8], an ambipolar electric field effect along with
ter summarizes the recent progress in the design of the syn- ballistic conduction of charge carriers [9], tunable bandgap,
thetic method of graphene via electrolysis in various kinds and so on. Therefore, graphene holds great promise for many
of electrolytes including ionic liquids, surfactants, propylene technological applications such as in nanoelectronics [10],
carbonate, etc. Furthermore, the applications of the resulting sensors [11], and composite materials [12]. Furthermore,
graphene are described in relation to their physical and chem- graphene could be soluble in conventional solvents or poly-
ical properties. The perspectives on the challenges and new mers without reducing its ideal intrinsic properties by chemi-
directions in this emerging area of research are also discussed. cal variations [13,14]. Hence, a myriad of applications have
been demonstrated in photocatalysis and energy conversion
such as in transparent conductive electrodes or active materi-
15.1 INTRODUCTION
als in solar cells [15,16], counter electrodes in dye-sensitized
Graphene, the new emerging two-dimensional (2D) sheet of solar cells [17,18], electrocatalysts for oxygen reduction in
carbon atoms, has attracted extraordinary research interest fuel cells [19,20], photocatalysts in water splitting [21,22],
since Geim and Novoselov, who were awarded the Nobel Prize high-performance electrodes in supercapacitors [23], and in
in Physics 2010, explored for the first time its properties and lithium-ion batteries (LIBs) [24,25]. Owing to its outstand-
its potential applications [1–3]. Theoretical and experimental ing properties and versatile applications, the study on the
269
synthesis, modification, and applications of graphene is cur- These approaches also enable to provide processable graphene
rently attracting extraordinary research interest. that can be easily coated onto various structures or integrate
graphene with other materials to form nanocomposites [33].
15.2 AN OVERVIEW OF GRAPHENE Traditionally, graphene has been synthesized by micro-
mechanical cleavage [1]. The method, the so-called “scotch
SYNTHESIS METHOD
tape,” involves exfoliation of graphite in which an adhesive
Efforts have been made on the preparation of graphene via a tape is used to repeatedly slide down the graphene layers on
number of physical and chemical methods. A number of syn- the substrate, causing the layers to cleave apart [1]. This tech-
thetic methods could provide high-quality graphene, opening nique is capable of producing monolayer, bilayer, and few-
up new possible routes to address the challenges in the prepara- layer graphene which are identified by optical microscopy
tion and molecular engineering of high-quality graphene. The over specially prepared SiO2/Si substrates. The mechanical
synthetic methods can be largely divided into two different cleaving using atomic force microscopy (AFM) cantilever has
types of approaches: top-down and bottom-up (Table 15.1). The also been developed, enabling to produce few-layer graphene.
most commonly used bottom-up methods, to make the high- These processes are of advantage in terms of high-quality
quality graphene films, include selective epitaxial growth of graphene sheets which is suitable for the study of the funda-
graphene on metal or nonmetal substrates using chemical vapor mental properties of graphene.
deposition (CVD) [26,27] or by thermal decomposition of SiC The top-down chemical routes are intensively developed
[28,29]. The graphene nanosheets have been produced by sub- as a promising approach for large-scale production. Currently,
strate-free CVD such as microwave-enhanced CVD of ethanol there are three popular chemical approaches to obtain gra-
or thermal decomposition of sodium ethoxide in ethanol [30– phene: exfoliation and reduction of graphite oxide, exfoliation
32]. For bottom-up methods, high levels of graphitization must of graphite/graphene oxide/graphite intercalation compounds
be promoted, so these methods generally require high tempera- (GICs), and direct exfoliation without chemical modification
tures, performed in ultrahigh vacuum conditions, or in inert gas in suitable organic solvents or surfactants [7]. Traditionally,
atmospheres. Therefore, the top-down methods, which involve exfoliation and reduction of graphite oxide has received the
breaking apart the stacked layers of graphite to yield single- most attention in which graphene oxide can be synthesized
layer graphene sheets, have been developed to tackle the task by oxidation of graphite using concentrated acids and strong
of graphene synthesis at low cost and large scale. Generally, oxidants following the reduction [34,40]. Graphene oxide
the top-down approaches are cost-efficient, using inexpensive with oxygen functional groups was able to disperse in solu-
graphite, and versatile, as they can be combined with other tion; however, it is electrically insulating due to the inevitable
chemical treatments, for example, chemical functionalization, disruption of long-range conjugation which compromises
to produce a variety of graphene and graphene-based materials. its unique properties precluding its direct application as an
TABLE 15.1
A Brief Summary of the Synthesis Method for Producing Graphene
Synthesis Method Description References
Bottom-Up
Pyrolysis Flowing an aerosol consisting of liquid ethanol droplets and argon gas directly into microwave-generated argon [30,31]
plasma at atmospheric pressure. Ethanol droplets evaporated and dissociated in the plasma-producing graphene
Epitaxial growth Conversion of SiC substrate to graphene via sublimation of silicon atoms on the surface at high temperature [28,29]
(~1300°C) and ultrahigh vacuum condition
CVD thermal Most promising, inexpensive method for production of single CVD plasma layer or multilayer graphene using [26,27]
transition metal (Ni, Cu, etc.) substrates or thin films as catalyst. Flowing carbon source (CH4) and reactant
gases (H2) at high temperature (~1000°C) for the growth of graphene. The method can be scaled up for
large-area graphene production for practical applications such as transparent electrode applications. Thermal
CVD was carried out in heating furnace and plasma CVD consists of plasma-assisted growth
Top–Down
Mechanical The method use Scotch tape, involves the exfoliation of graphite where adhesive tape was used to repeatedly slide [1]
exfoliation down the graphene layers on the substrate, leading to cleavage the layers apart. This technique was capable of
producing monolayer, bilayer, and few-layer graphene which are identified by optical microscopy over specially
prepared SiO2/Si substrates
Chemical exfoliation Exfoliation and reduction of graphite oxide [34,35]
Exfoliation of graphite/graphene oxide/graphite intercalation compounds [36,37]
Direct exfoliation without chemical modification in suitable organic solvents or surfactants [38,39]
Chemical synthesis Low temperature process involves the synthesis of graphite oxide, dispersed in solution and reducing to graphene
with hydrazine
Graphene 271
electronic material. The properties of graphene oxide are of monolayer graphene, while cyclopentanone is the solvent
also different from those of graphene due to the high level of giving the highest absolute concentration (monolayer and few-
defects created in the harsh oxidation process. layer graphene) [39]. Pentafluorobenzonitrile and benzylamine
Graphene can also be synthesized by exfoliation of have shown improved stable dispersions of graphene with up to
GICs with the aid of solvents, sonication, or thermal treat- a concentration of 0.5 mg/mL [43,44]. A significant improve-
ment. For example, solvent-assisted exfoliation of GICs was ment in exfoliation of pristine graphite has been achieved by
reported in which the exfoliation of alkali metal GICs in using drastically longer sonication times or using a sonic probe
N-methylpyrrolidone yielded large-size graphene flakes [36]. [45]. Aiming to avoid using expensive and harmful solvents, a
The intercalation of solvent molecules may help expanding strategy has been developed to disperse and exfoliate graph-
the layers [37]; interaction with the solvent may cause gas to ite to give graphene suspended in aqueous surfactant solutions
be expelled that aids exfoliation. Besides alkali metal, iron [35]. They found that the dispersed phase was mainly consist-
chloride, nitromethane, and ionic liquid (IL) crystals have ing of graphene flakes, with more than 40% of these flakes
been intercalated with graphite to promote exfoliation [41,42]. having <5 layers and ~3% of flakes consisting of monolayer.
Generally, these methods are labor consuming, involving sev- Atomic resolution transmission electron microscopy (TEM)
eral steps, and need a long time to allow the intercalants and shows the monolayer to be generally free of defects. The
organic solvents to fully insert into the graphitic layers [37]. adsorbed surfactant played an important role to stabilize the
In contrast to the above oxidation and multistep methods, dispersed graphitic flakes against the aggregation due to their
direct exfoliation of natural graphite via sonication in suitable Coulomb repulsion [35]. However, those surfactants are also
organic solvent has been developed. High-quality single-layer difficult to remove and can affect the electrical and electro-
graphene sheets have been prepared by direct exfoliation in chemical properties of graphene. Furthermore, graphene can
an organic solvent such as N-methylpyrrolidone under mild also be obtained by sonication of graphite in IL. The method
sonication [38] (Figure 15.1). Coleman and coworkers have will be discussed in detail in the later part of this chapter.
developed a method to effectively exfoliate graphite to pro- Although this direct exfoliation method shows its advantage
duce single-layer graphene sheets and few-layer graphene over oxidation and reduction methods, the amount of exfoli-
sheets (FLGS) without the use of intercalants. The proper sol- ated product is usually very limited relative to the starting
vents have surface tension closely matching the surface energy material, typically <1% yield. Therefore, it is still challenging
of graphite, which is a key prerequisite of solvents for direct to develop a high-yield, environmentally benign method that
exfoliation of graphite. This method could produce chemi- can directly exfoliate graphite into graphene sheets.
cally unmodified, pristine graphene with a concentration of There have been several reports on environment-friendly
~0.01 mg/mL with a monolayer yield of ~1 wt.%. This mono- methods for direct production of graphene sheets with high
layer yield could potentially be improved to ~7–12 wt.% by yield. For example, reduction of graphite oxide to graphene can
serial re-sonication and re-centrifugation of the unexfoliated be processed by a green and controllable approach using hydro-
material. The advantage of this approach is that the obtained thermal dehydration method [46]. The hydrothermal route was
product is high-quality graphene without any defects or oxides applied by tuning the engineering parameters of temperature
on the surface, which is completely different from that of the and pressure toward the supercritical point of water, which
other solution methods such as graphite oxidation approach. change the reduction chemistry and facilitate the removal of the
As a result, the films made from these sheets have a superior surface oxide groups to produce graphene sheets. As a result,
conductivity of 6500 S/m. Since this initial work, attempts the thickness of the graphene sheet, observed by AFM, is
have been made on finding suitable solvent and technique about 0.8 nm which is smaller than that of as-used GO of about
for high-yield production and highly concentrated dispersion 1.5 nm. Zheng et al. [47] has developed a solvothermal route to
of graphene sheets. It was found that N-methylpyrrolidone prepare graphene by direct exfoliation using oleylamine as a
(NMP) is the best solvent for the production of high percentage solvent as well as a stabilizing agent. The exfoliated graphene
(a) (b) (c)
30
Counts
20
10
N = 100
0
0 2 4 6 8 10 12 14
Number of layers per sheet
FIGURE 15.1 (a) SEM image of pristine graphite (scale bar: 500 mm). (b) Bright-field TEM image of monolayer graphene flakes deposited
from N-methylpyrrolidone (scale bars: 500 nm). (c) Histogram of the number of visual observations of flakes as a function of the number of
monolayer per flake for N-methylpyrrolidone dispersions. (Reprinted by permission from Macmillan Publishers Ltd. Nat. Nanotechnol.,
Y. Hernandez et al. 3, 563–568, copyright 2008.)
sheets are of high quality and have few defects. The graphene density of the fluid may be tuned from “gas-like” to “liquid-
dispersion, predominantly composed of monolayer graphene, like” values simply by varying pressure and temperature. The
had a high concentration of 0.15 mg/mL, and can be stabilized other advantageous properties including low interfacial ten-
against re-aggregation [47]. Thermal quenching of graphite has sion, excellent wetting of surfaces, and high diffusion coeffi-
been used as an effective route to separate layers in graphite. cients make supercritical fluids potentially superior solvent for
This technique involved quenching hot graphite in an aqueous diffusion in between the layers of graphite and its expansion
solution of hydrogen carbonate, ammonia, or hydrazine hydrate [50]. Honma et al. have developed a supercritical fluid method
[48,49]. The process yielded sheets with thicknesses between for the exfoliation of graphite to obtain the high-quality and
0.4 and 2 nm and diameters in the range 1–80 mm [48]. Jiang high-yield pristine graphene sheets [50,51]. This fast and fac-
et al. [49] reported 15 wt.% yield of nanosheets <5 layers thick ile one-pot exfoliation method resulted in a highly conduc-
after one round of quenching and an 80 wt.% yield of such tive pristine graphene sheet maintaining its original nature.
sheets after five sequential rounds of quenching, with 10% of Using ethanol, N-methylpyrrolidone, and dimethylformamide
this product consisting of monolayers. (DMF) as solvent, we have succeeded in the conversion of
Graphite layer can be also exfoliated by intercalating them graphite to graphene sheets with a high yield under supercriti-
with a supercritical fluid. The supercritical fluids have been cal fluid conditions. This method led to well-dispersed solu-
considered to be unique reaction media for nanomaterials syn- tions at 2–4 mg/mL, and all sheets had <10 layers, including
thesis owing to their unique properties such as gas-like diffu- 10% monolayers (Figure 15.2). When 1-pyrene sulfonic acid
sivity, low viscosity, and a density closer to that of a liquid. The sodium salt surfactant was used, the percentage of monolayers
(a)
: Super critical fluid (EtOH, NMP, and DMF)
SCF
Treatment
Graphite
flakes (GF)
Exfoliation by
SCF diffusion through GF Exfoliated graphene
(b)
8.6
C 3.7 nm 4.1 nm
A B 2.9 nm
Height (nm)
500 nm
3
0 Distance along x axis (nm) 3820
(c) (d)
200 nm
FIGURE 15.2 (a) Scheme showing the supercritical fluid exfoliation of graphite crystals to graphene. (b) AFM image of graphene sheets
exfoliated by ethanol and the corresponding height profile. (c) HRTEM image of graphene sheets exfoliated by ethanol and (d) corre-
sponding selected area electron diffraction. (D. Rangappa et al.: Rapid and direct conversion of graphite crystals into high-yielding, good-
quality graphene by supercritical fluid exfoliation, Chem. Eur. J. 2010. 16. 6488–6494. Copyright Wiley-VCH Verlag GmbH & Co. KGaA.
Graphene 273
improved from 10% to 60%. Thus, these supercritical fluid been proposed to tackle this challenge, including oxidation,
techniques are very promising due to their resulting significant chemical intercalating reactions, solvent immersion, and so on.
number of pristine monolayers and the ease of processing. The external forces such as stirring, shaking, ultrasonication,
When a voltage is applied to graphite rods serving as elec- microwave irradiation, and other thermal treatment have been
trodes in an electrolyte solution, the rods may corrode and applied to overcome the attractive van der Waals interactions
graphene nanosheets were produced with functionalization between layers. After exfoliation of graphite into graphene
by the various organic/inorganic chemistries. This technique sheets, stabilization of graphene against the reassembly is also
presents a unique way for simultaneous exfoliation and func- a crucial task. The functionalization of graphene with surfac-
tionalization of graphene with different functional groups, tants, polymers, and organic molecules and other inorganic
while potentially lessening the number of lattice defects intro- molecules could stabilize graphene dispersions in solvents.
duced by common oxidation methods such as the Hummers These functions induce repulsive electric double layers or
method. Furthermore, this rapid and facile one-pot exfoliation repulsive steric forces, preventing aggregation or reassembly.
method enables to produce monolayer graphene sheets with When a voltage is applied to graphite rods serving as elec-
very high yield, at low cost and large scale. trodes in an electrolyte solution, graphene can be formed due
to the expansion and exfoliation of the graphite rods. The van
der Waals attractions between the adjacent layers are overcome
15.3 ELECTROCHEMICAL EXFOLIATION
by the introduction of functional groups such as hydroxyl and
Graphite has a layered, planar structure and the distance oxo at the edge and by immersion of various organic or inor-
between planes is 0.335 nm. In each layer, the carbon atoms are ganic matter. As a result, graphene nanosheets were function-
arranged in a honeycomb lattice with separation of 0.142 nm. alized by various kinds of organic/inorganic groups depending
The layers in graphite are held together by an interplanar on the electrolytes used. Therefore, this approach open a new
van der Waals attraction force. The van der Waals attraction route for exfoliating graphite into pristine graphene as well as
is strong enough to stabilize the system and to make separa- functionalizing graphene with different functional groups with
tion into graphene sheets difficult. Thus, in order to obtain high yield and high quality. Moreover, the approach represents
graphene from graphite, one should reduce the van der Waals the advantages over the common chemically reduced from
attraction between the adjacent layers and applied additional graphene oxide or oxidation methods in terms of density of
external forces to exfoliate these layers. Various methods have lattice defects and resulting conductivity (Table 15.2).
TABLE 15.2
Summary of Electrochemical Exfoliation Methods for Producing Graphene
Electrolyte Working Electrode Voltage Thickness, Size References
[C8mim][PF6] Graphite rod 15 V 1.1 nm thick, 500 nm [52]
[C1bim][BF4] Graphite rod/HOPG 1.5–15 V Monolayer or bilayer, 200 nm [53]
[Bmim][BF4] Flexible graphite −2 –3 V Graphene flakes [54]
[Bmim][Br] Graphite rod 15 V 1.1 nm thick, 800 nm [55]
Sulfonyl imide Pencil graphite 0–8 V 0.8 nm thick, flakes [56]
Emim/Bmim Graphite rod 7V Monolayer or bilayer, 500 nm [57]
PC electrolyte Graphite −15 ± 5 V Monolayer or bilayer, flakes [58]
PC/butylammonium Graphite foil 5 V – [59]
Dimethylsulfoxide Graphite rod 5 V 3.1 nm, flakes [60]
DMSO/alkylNH Graphite −7–3 V 10 layers, 1–20 µm [61]
LiOH/LiCl Graphite rod 4.5 V 2–4 nm, 200 nm [62]
KNO3 Graphite rod – 600 nm [63]
KPF6 Graphite rod 15 V 1.1 nm thick, 500 nm [64]
(NH4)2SO4 Graphite flake 10 V Monolayer or bilayer, 20 µm [65]
Perchloric acid Graphite foil – Four layers, 5 µm [66]
Sulfuric acid Graphite 1–10 V 1.5 nm, 1–40 µm [67]
Sulfuric acid Artificial graphite 1–10 V 2.2 nm, 1–40 µm [68]
Sulfuric acid Graphite 10 V Monolayer or bilayer, 20 µm [69]
Phosphoric acid Graphite rod 6−8 V 6 nm, 5 µm [70]
Oxalic acid Graphite rod 40 V Three layers, 10 µm [71]
PSS surfactant Graphite rod 5 V 0.8 nm, 1 µm [72]
PSS surfactant Graphite foil – 3 nm, 1 µm [73]
SDS surfactant Graphite 1–2 V 1 nm, 500 nm [74]
TSCuPc Graphite rod 12 V Five layers, 500 nm [75]
SDS surfactant HOPG 5 V 0.8 nm, flakes [76]
15.3.1 Exfoliation in IL mostly with <5 layers, by direct exfoliation of pristine graph-

ite in Bmim bis(trifluoro-methane-sulfonyl)imide with the aid
Ionic liquids (ILs), described as a broad class of low-melt-
of sonication. The resulting colloidal suspension had a con-
ing semi-organic salts, have attracted extraordinary research
centration of 0.95 mg/mL of exfoliated graphene. It was found
interest in the past decade. ILs possess unique properties such
that optimizing the initial graphite concentration and increas-
as low vapor pressure, nonflammability, and high chemical
ing the sonication time up to 24 h could produce a few-layer
and thermal stability. ILs have been proposed to replace con-
graphene concentration as high as 5.33 mg/mL. Nuvoli et al.
ventional organic solvents in many technological applications
[82] have ground and sonicated graphite with a commercial IL
such as chemical synthesis, nanocatalysis, and batteries [77].
1-hexyl-3-methylimidazolium hexafluorophosphate, as a con-
Recently, ILs have been used for direct chemical exfoliation
venient solvent medium for graphite exfoliation. This method
of graphite owing to the fact that their surface tension and
resulted in few layers of graphene with a concentration as high
surface energy matches with those of graphite [78,79]. Owing
as 5.33 mg/mL, which is the highest value reported so far in
to its unique structure, ILs can adsorb onto the graphene
any solvent. They proposed that the high-yield exfoliation of
surface through the noncovalent interactions with graphene.
graphite in ILs is due to the surface tension of ILs close to that
Both the anion and cation in hydrophobic ILs are orientated
of graphite, and due to the ionicity of ILs, which stabilizes
at the graphene surface, for example, the imidazolium ring
the exfoliated graphene sheets. It has also been reported that
is slightly tilted to the surface plane with the alkyl chain
the stable dispersions of reduced graphene oxide (rGO) sheets
extended parallel to the surface plane [80,81]. On the contrary,
in hydrophilic imidazolium- and pyridinium-based ILs are of
there is no immediate interaction between the anion of hydro-
relatively high concentrations of up to 7.0 mg/mL [83]. The
philic ILs and graphene surface. It was also found that there
resulting graphene–IL dispersions can be dispersed in various
is some charge transfer between the ions of ILs and graphene.
organic solvents. Furthermore, graphene sheets can also form
Consequently, versatile methods have been proposed to syn-
a gel hybrid by grinding or sonicating a mixture of graphene
thesize graphene sheets using ILs with the aid of external
and ILs [84,85].
forces such as mechanical grinding, sonication, microwave
Inspired by the electrochemical reactions of negative
irradiation, and electrolysis.
graphite electrodes in rechargeable LIBs, various electrolytic
For example, Shang et al. synthesized graphene nanosheets
methods have been developed for the exfoliation of graphite.
of 2–5 layer thickness, by mechanical grinding exfoliation
When a voltage is applied to graphite rods serving as elec-
of natural graphite in a small quantity of ILs. The result-
trodes in an electrolyte solution, graphene can be formed due
ing graphene sheets are solvent free with low levels of natu-
to the expansion and exfoliation of graphite rods. In early
rally absorbed oxygen, retaining its nature from the starting
studies, the setup for electrochemical exfoliation of graphite
graphite (Figure 15.3) [78]. Wang et al. [79] fabricated FLGS,
involved using graphite as a sacrificial electrode and IL as
a conductive solution (Figure 15.4) [52]. After static poten-
– tials (10–20 V) were applied for a period of time, the exfoli-
(a) 1010
–
ated graphene can be collected from the electrolyte solution.
0110 Typically, the electrochemical exfoliation-functionalized
graphene with ILs such as octyl methyl-imidazolium hexa-
fluorophosphate ([C8mim]+[PF6]−) was obtained. The imid-
azolium-functionalized graphene were easily dispersed in
organic solvents such as DMF via sonication.
The AFM analysis reveals that the obtained graphene
were flattened nanosheets of 1.1 nm thick on average, which
is three times thicker than the theoretical value of a single
100 nm layer due to the functionalization of hydrocarbon chains
above and below the graphene plane [52]. Microscopic
(b) (c) (d) observation reveals that the average size of graphene sheets
was up to 700 nm × 500 nm (Figure 15.4). Although the
qualitative TEM and AFM analysis show the presence of
monolayers, the quantitative investigation on the product
yield, thickness of sheets, and dispersible ability was not
2 nm 2 nm 2 nm provided. Based on the x-ray photoelectron spectroscopy
(XPS), Raman, and FTIR (Fourier transform infrared spec-
troscopy) investigation, it was speculated that during the
FIGURE 15.3 (a) A typical low-magnification TEM image of gra-
phene sheets; the inset shows the corresponding electron diffrac- electrochemical reaction process, the cation of the IL is
tion pattern. HRTEM images of (b) bilayer, (c) triple layer, and (d) reduced on the cathode, leading to the formation of 1-octyl-
4–5 layer graphene sheets. (N. G. Shang et al. Chem. Commun., 3-methylimidazolium free radical. The radical can combine
48, 1877–1879. Copyright 2012. Reproduced by permission of The with one of the electrons of the p-bond of graphene, under
Royal Society of Chemistry.) the formation of IL–graphene.
Graphene 275
(a) Anode graphile rod (d)

A
Water
Graphite rods ionic
liquid
N N+ +N N + e
– N N
Ionic liquid and water Mixture
z +
z + z
(b) (c) z
z
+
z
z
z + z
+ z
z
+
z
+ z
Acc V Spot Magn Dot WD Exp 200 nm
500 nm 15.0 kV 3.0 78739x SE 9.6 0
FIGURE 15.4 (a) Experimental diagram of electrochemical synthesis of graphene, (b) TEM, (c) FESEM images of graphene, and (d)
schematic illustrations of the formation of 1-octyl-3-methylimidazolium radical and how the ILs are attracted to graphene and polymer.
(N. Liu et al. Adv. Funct. Mater. 2008. 18. 1518–1525. Copyright 2010 John Wiley & Sons. With permission.)
The electrolytic exfoliation of graphite was then thor- former have less steric hindrance and can be intercalated into
oughly investigated by tailoring amounts of water with water- the graphite more efficiently [54].
miscible ILs for electrolytes [53]. IL-functionalized graphene Ling et al. have developed IL-assisted electrolysis using
sheets were generated similar to the previous work by Liu various ILs including 1-butyl-3-methylimidazolium bro-
et al. when an electrolyte solution with <10% water content mide ([Bmim][Br]), 1-butyl-3-methylimidazolium hexafluo-
was used. On the contrary, when the voltage was applied to rophosphate ([Bmim][PF6]), 1-butyl-3-methylimidazolium
a solution with >10% water content, a variety of oxidized acetate ([Bmim][Ac]), and 1-octyl-3-methylimidazolium hexa
graphene products were produced, including 8–10 nm thick fluorophosphate ([C8mim]+[PF6]−) [55]. The functionalized
nanosheets, graphene nanoribbons, and graphene nanosheets. graphene strongly depends on the charged properties of the
On the basis of the time-dependent experiment, it was pro- electrolyte solution, viscosity, and the nature of the ILs. TEM
posed that the generated hydroxyl and oxygen radicals from images reveal that graphene sheets exhibit a typical wrinkled
the anodic oxidation of water play the role of an electrochem- structure with corrugation and scrolling like crumpled silk
ical unzipping agent that attacked the edges of individual veil waves. The AFM analysis reveals the presence of flat-
sheets. The oxidation of edge planes opened the edge sheets tened nanosheets with an average thickness of 1.00 nm, which
for the IL to intercalate, leading to depolarization and expan- are consistent with the result of a previous report by Liu et al.
sion of the graphite anode [53]. [52]. The graphene sheet is three times thicker than the theo-
Although there is an ongoing debate regarding the mecha- retical value of a single layer, indicating the presence of less
nism of electrolytic exfoliation of graphite, the IL-assisted than three graphene layers or the functionalization of IL ions on
method has been further developed for the synthesis of graphene. Differently, Singh et al. [56] report greener, one-pot
high-quality graphene. For example, graphite was also elec- electrochemical synthesis of graphene sheets in neat IL [triethyl
trochemically exfoliated in aqueous solvents containing sulfonium bis (trifluromethyl sulfonyl) imide] (Figure 15.5). By
hydrophilic IL, 1-butyl-3-methylimidazolium tetrafluorobo- changing the electrolytic program and with the aid of sonica-
rate ([Bmim]+[BF4]−) [54]. It is found that the addition of lith- tion, they also successfully obtained graphene oxide sheets.
ium salts could facilitate exfoliation due to the improved ionic
conductivity of the electrolyte. DI (deionized water) water is
15.3.2 Exfoliation in Common Electrolyte
necessary to catalyze the sheet unzipping performed by the IL
anions (e.g., BF4− from [Bmim][BF4]) as they found that exfo- At the same time, electrochemical strategies have included
liation of graphite in the absence of water turned the end prod- expanding the graphite electrode in situ using propylene car-
uct into a “bucky gel” [54]. The anion size strongly affects bonate rather than IL. The exfoliation of the electrode was
the intercalation of IL ions into graphite. Smaller inorganic facilitated using further sonication [58]. The work was moti-
anions such as BF4− may be more preferable in the exfolia- vated by the fact that during electrochemical charging in a
tion reaction to bulky anions such as [N(CN)2]− because the graphite electrode, cointercalation of propylene carbonate
(a) (b) electrochemical exfoliation of the graphite [88]. Further soni-

cation acts to exfoliate and cut the graphene layers by injecting
thermal shock into the material through ultrasonic cavitation
[89]. Sonication of lithium-salt-intercalated graphite in an
LiCl dissolved in solvent mixture of DMF and propylene car-
bonate has been reported to give >70% few-layer graphene
[58], which is the highest yield ever reported for the chemi-
cal exfoliation method. The obtained sheets have an average
6/23/2011 Det HV WD Mag HFW 2.0 μm 100 nm
lateral dimension of 1–2 μm. On the basis of electron diffrac-
tion investigation, induced by diffraction intensity from the
FIGURE 15.5 (a) SEM image. (b) Low-magnification TEM image (0110) plane than from the (1210) plane, the layer thickness
of graphene nanosheets, resembling crumpled silk. The featureless was found to be 1–3 layers. The statistical sampling of the
regions indicated by the arrows are monolayer graphene nanosheets. graphene flakes using AFM reveals that 50% of the graphene
The inset is the SAED, which confirms the crystalline nature of the sheets comprised 2–3 layers of graphene (Figure 15.6).
graphene nanosheets. (V. V. Singh et al.: Greener electrochemical Similarly, Zhong and Swager [59] demonstrated a two-step
synthesis of high quality graphene nanosheets directly from pen- process synthesis including first activation in Li+ -containing
cil and its spr sensing application, Adv. Funct. Mater. 2012. 22. propylene carbonate electrolytes and further activated/
2352–2362. Copyright Wiley-VCH Verlag GmbH & Co. KGaA.
expanded in tetra-n-butylammonium electrolytes. It was
speculated that positively charged tetra-n-butylammonium
cations penetrate into the graphite lattice by cation exchange
with Li+ leads to the formation of a ternary graphite inter- with the intercalated lithium ions [90]. Furthermore, an addi-
calation compound (GIC) into graphite interlayers, which tional critical factor is the electrodecomposition of the inter-
exfoliates the graphite due to expansion [86,87]. The decom- calated tetra-n-butylammonium which partially neutralizes
position products of propylene carbonate intercalation com- the positive charges within the graphite lattice and thereby
pounds formed upon its reduction were not able to form an continually maintaining a driving force for intercalation of
effective solid electrolyte interface that passivates the graph- tetra-n-butylammonium cations [91]. This improved function-
ite from further solvent cointercalation, thus resulting in alization is attributed to the presence of the electrochemically
(a) e– e– e–
e– e–
Negatively + Li+
charging (–) e– Li+/DMF/PC
e–
e– Sonication
e–
+ +
Graphite Dispersible
O O
O O
Li+ Li+
O O
Li+/PC
complexes
Expansion
(d) Size of graphene sheet (μm)
(b) (c)
2 × 0.34 nm >10 8 6 4 2 0
50 0
40
Percentage (%)
Percentage (%)
10
2L >10 L
30 3L 20
4–10 L
20 30
10 1 L 40
0 50
500 nm 0 1 2 3 4–10 >10
500 nm
Number of graphene layer per sheet
FIGURE 15.6 (a) Schematic illustration of exfoliation of graphite into few-layer graphene flakes via intercalation of Li+ propylene carbon-
ate complexes. (b) TEM images and electron diffraction pattern for few-layer graphene flakes. (c) AFM image of few-layer graphene flakes
spin-coated onto an Si substrate. (d) Thickness and size distribution histograms of the few-layer graphene flakes produced, as estimated
from AFM analysis of the few-layer graphene flakes. (Reprinted with permission from J. Wang et al. J. Am. Chem. Soc. 133, 8888–8891.
Graphene 277
generated solid electrolyte interface layers within the graph- into the graphite cathode destabilizes the layered graphite
ite lattices, which acted as stable spacers [59]. The enhanced structure, which instigates the formation of carbon-based
expansion of the graphite further allows for functionalization nanostructured species. This formation process is somewhat
of individual graphene sheets with aryl diazonium salts as a similar to the exfoliation of graphene sheets from bulk graph-
model of scalable route to new graphene and the composites. ite through an oxidation-derived intercalation–expansion
Zhou et al. [60] studied a similar method to synthesize process [94]. Inspired by this approach, Huang et al. [62] dem-
few-layer graphene by electrochemical intercalation of graph- onstrated an electrochemical strategy expanding the graphite
ite cathode using Na+/dimethyl sulfoxide (DMSO) complexes cathode in situ using moten salt. It should be noted that in
as an intercalation agent. By adding thionin acetate salt into the above-mentioned IL electrolytes, exfoliation of graphite
the electrolyte, the exfoliated graphite is stabilized and further takes place in the anodic graphite. Thus, the oxidation pro-
exfoliated into few-layer graphene by mild sonication. Thionin cess of the graphite inevitably occurs, which will damage the
ion has a planar aromatic structure with two hydrophilic –NH2 honeycomb lattices of graphene and introduce defects in the
symmetrically distributed on each side, thus, enabling to sta- exfoliated graphene sheets [62]. Huang et al. reported, for the
bilize aqueous graphene dispersion due to its amphiphilic first time, the electrolytic exfoliation of the graphite cathode
nature. By electron microscopy and AFM they confirmed that (rather than graphite anode) in LiOH molten electrolyte for the
most of the graphene flakes have large lateral size and macro- synthesis of graphene. This expansion has been achieved by
pores are present. The graphene flakes that have an average intercalation of lithium salts to form GICs to exert mechani-
thickness of 3.1 nm is composed of seven layers of graphene. cal stress on the lattice of graphite. Sonication of lithium-salt-
intercalated graphite in water could assist exfoliation due to
microexplosion reaction between lithium and water producing
15.3.3 Exfoliation in Alkali Metal Salt
hydrogen gas. As a result, graphene with 80% conversion effi-
Since 1995, carbon nanotubes (CNTs) have been prepared by ciency could be obtained through these lithium intercalation–
molten-salt-assisted electrolysis [92]. The electrolytic produc- expansion–microexplosion processes (Figure 15.7). SAED
tion of pure CNTs has been realized using various molten (selected-area electron diffraction) and AFM images indicate
alkali chlorides LiCl, NaCl, KCl, and LiBr as the electro- that the sheets are composed of multilayers, mostly 2–4 layer
lyte [93]. In this process, the intercalation of the alkali metal graphene sheets.
Li+
(a) Intercalation Expansion
Graphite LixCy Expanded graphite
Disintegration
H2
Microexplosion Sonication
Graphene layers
H2O
Graphene
Graphene stacks
6.5 nm
(b) 2 nm 3
(c) 2 (d)
1
Height
0
–1
+ –2
–3
1 nm 0.2 0.4 0.6 0.8 1.0 1.2 1.4 μm
+ 4
+ + 2 (e)
Height
0
500 nm –2
0 1 2 nm –4
–5.2 nm nm 0.2 0.4 0.6 0.8 1.0 1.2 1.4 μm
FIGURE 15.7 (a) Schematic illustration of the formation steps of graphene sheets through Li intercalation–expansion–microexplosion
processes. (b) TEM image. (c) AFM image of a 2 mm × 2 mm scan of graphene sheets overlaid onto a silicon surface. (d and e) Two height
profiles acquired in different locations. (H. Huang et al. J. Mater. Chem., 2012, 22, 10452–10456. Reproduced by permission of The Royal
Society of Chemistry.)
At the same time, encouraged by the preparation of gra- same time, anodic oxidation of water produced hydroxyl and
phene through an electrolyzing graphite rod, Ai et al. prepared oxygen radicals. The hydroxylation or oxidation of graphite
the mixture of graphene nanosheets and carbon nanospheres by these radicals results in the dissolution of carbon nano-
(GNS–CNS) by electrolyzing graphite rob using KNO3 solu- crystals from the anode [96]. The SAED indicates that the
tion as the electrolyte [63,95]. They proposed that water can pattern was obtained from the overlapping graphene sheets, in
be oxidized to form strong oxidants of hydroxyl (⋅OH) and which there were about 2–5 layers of graphene. By TEM, they
oxygen radicals (⋅O), which can attack the graphite edge confirmed that the graphene sheets are eventually decorated
planes to generate hydroxylation or carboxylation of graphite by MoO3 nanoparticles with diameters of 5–10 nm and the
according to the mechanism proposed by Loh et al. [53]. The MoO3 nanoparticles are entrapped inside the graphene sheets
corrosion process occurs initially at edge sites, grain boundar- (Figure 15.8). The tapping mode AFM image of the electro-
ies, or defect sites and takes place before the expansion of the chemically exfoliated graphene also reveals that the thickness
graphite electrode [63,95]. In addition, NO3− can be reduced at of the graphene sheets is about 1.0 nm, corresponding to the
cathode to form NO2− and OH−. It was proposed that the H+ or 2–5 layers.
OH− or NO2− or NO3− in solution can intercalate the interlami- Zhang et al. [97] reported the synthesis of Fe2O3/graphene,
nation of graphite anode to facilitate expansion. Finally, some Co3O4/graphene, and V2O5/graphene by the electrolysis method.
expanded graphite precipitate as graphene sheets or carbon It was proposed that the sulfate anions were driven by the elec-
nanospheres with some hydroxyl and carboxyl decorated on tric field into the interlayer between the graphene layers, leading
the surface. Owing to the presence of these hydrophilic func- the exfoliation. At the same time, metal oxides are electro-
tional groups, graphene can easily be suspended in water by chemically deposited onto the surface of the graphite electrodes
ultrasonication. Such homogeneous suspension is fairly stable and remained in the resultant graphene. Na2SO4, Li2SO4, and
against the coagulation for more than 30 days of storage. K2SO4, were adopted as the source of sulfate anions for success-
Similarly, functionalized graphene with PF6 was fabri- ful synthesis of metal oxides/graphene hybrids. The formation
cated by electrolysis with potassium hexafluorophosphate of the hybrids with metal oxide particles in nanoscale size was
(KPF6) solution as an electrolyte [64]. The characterization confirmed by SEM (scanning electron microscopy) and TEM. It
results of TEM, AFM, XPS, Raman spectroscopy, and TGA was found that the metal oxides (Co3O4 nanoplates) with diame-
(thermogravimetric analysis) indicate that graphite rod was ters of 20 nm and thicknesses of 1–2 nm are uniformly attached
completely exfoliated to graphene layer containing 30 wt.% on the graphene sheets.
PF6 − with an average thickness of ca. 1.0 nm. After this Recently, Mullen et al. reported a prompt electrochemi-
work, Luo et al. introduced an interesting work to synthesize cal exfoliation of graphene sheets into aqueous solutions of
MoO3-decorated graphene sheets from graphite by molyb- different inorganic salts ((NH4)2SO4, Na2SO4, K2SO4) [65].
date-assisted electrochemical exfoliation [96]. During the Exfoliation in these electrolytes leads to graphene with a high
electrochemical reaction process, the anions MoO42– moved yield (>85%, ≤3 layers), large lateral size (up to 44 μm), low
to the positive graphite electrode under the influence of an oxidation degree (a C/O ratio of 17.2) with the lowest defect
applied electric field and inserted the graphite layers; subse- density, and a remarkable hole mobility of 310 cm2/V s.
quently MoO42– is converted to MoO3 nanoparticles under Among the different types of aqueous inorganic salt elec-
strong acid conditions (the graphite of the anode interface), trolyte solutions, sulfate-containing salts such as (NH4)2SO4
leading to electrolytic exfoliation of the graphite rod. At the exhibited the best exfoliation efficiency.
(a) Inset (b)
Converted H+
Exfoliation
: Anodic graphite rod : MoO42– : Graphite sheet

50 nm
: Oxidation of graphite sites : α-MoO3
FIGURE 15.8 (a) Schematic illustration of the formation steps of MoO3/graphene by electrolytic route. (b) TEM image of MoO3/graphene.
(J. B. Hu et al. J. Mater. Chem., 2011, 21, 15009–15014. Reproduced by permission of The Royal Society of Chemistry.)
Graphene 279
The mechanism of electrochemical exfoliation was pro- anion intercalation in graphite. It was found that the HOPG
posed as follows: (i) The reduction of water at the cathode interlayer spacing increases by 32% due to the formation
when applying bias voltage, creating hydroxyl ions (OH−) that of GICs when perchlorate anions intercalate and the anion
act as a strong nucleophile in the electrolyte. (ii) Oxidation at intercalation can subsequently damage the sp2 lattice due to
the edge sites and grain boundaries due to the nucleophilic side reactions such as the formation of graphite oxide and
attack of graphite by OH− ions then leads to depolarization carbon dioxide. Recently, Morales et al. [66] employed con-
and expansion of the graphite layers, thereby facilitating the centrated perchloric acid and with a careful selection of the
intercalation of sulfate ions (SO42−) within the graphitic lay- intercalation potential, for controlled electrolytic exfoliation
ers. (iii) Reduction of SO42− anions and self-oxidation of water of graphite. By TEM, they confirmed that the thickness of
produce gaseous species such as SO2, O2, as the vigorous gas the graphene flakes is well below four layers, evidenced from
evolution during the electrochemical process. the edge image and the difference in intensity obtained at the
STEM (scanning transmission electron microscope)-Z con-
trast images (Figure 15.9). Furthermore, number layers of
15.3.4 Exfoliation in Acid
graphene sheets could be determined by the analysis of the
Since 1999, the electrochemical intercalation of anions into second-order 2D peaks in Raman spectra which arise from
highly oriented pyrolytic graphite (HOPG) in aqueous acid the zone-boundary phonons. Particularly, they analyzed a
media has been investigated [98]. Siegenthaler et al. used a false-color image of the baseline to 2680 cm−1 intensity and
highly oriented pyrolitic graphite electrode in perchloric acid, Raman spectra observed at different points marked on the
as a model to elucidate the mechanism of electrochemical sample in the high-frequency region corresponding to the 2D
(a)
1
1
2
5 nm
2
(b) (d)
D
1000 1500 2000 2500 3000 3500
Raman shift (cm–1)
C C
(c)
D
B
B
A
A
2500 2600 2700 2800 2900

FIGURE 15.9 (a) HRTEM image of graphene sheet with the fringes indicating the maximum stacking of four layers (left). The filtered
images indicate a thickness well below four monolayer (right): top corresponds to defaulted and bottom to free of defects areas. (b) Raman
spectrum observed from exfoliated graphene deposited on a quartz slide. (c) Optical image of a typical deposit on the quartz slide and false-
color map of the zone indicated. (d) Representative spectra from the points indicated of the 2D region of graphene. (Reprinted from Carbon,
49, G. M. Morales et al. 2809–2816. Copyright 2011, with permission from Elsevier.)
modes (Figure 15.9). The Raman results suggest the forma- vertical cell configuration, enabling multiple exfoliation pro-
tion of graphene sheets composed of 3–6 graphene layers. cess to improve both the quality and yield of graphene sheets
At the same time, Li et al. found that sulfuric acid was a from electrochemical exfoliation of graphite. The graphite rod
good electrolyte for graphite exfoliation [67]. [SO4]2− ions is (anode) and platinum wire (cathode) were placed vertically at
thought to be a good intercalation agent, which were often the bottom and top of the electrochemical cell. A plastic tube
detected between the graphite layers after the processing as an insulator container ensured that only the top surface
[97]. When sulfuric acid was used as an electrolyte, gra- of it was exposed to the electrolyte solution. The exfoliation
phene nanosheets with large lateral size, ranging from several processes occurred very fast as few minutes, which is much
micrometers to tens of micrometers, were produced. For suc- shorter than that of the conventional method [67]. The exfoli-
cessful intercalation of sulfate ions and exfoliation of graph- ated graphite sheets/particles leaving from the graphite rod
ite, they used a two-step electrolysis, that is, applying static may undergo more than one round of electrochemical exfo-
bias at low voltage and ramping bias to high voltage. Since liation, resulting in a more efficient and higher exfoliation.
H2SO4 caused a high level of defect in the resultant graphene, Optical and TEM images reveal that the product consists of a
KOH was added to the electrolyte solution to reduce acidity large amount of exfoliated graphene with lateral size ranging
and minimize oxidation. All sheets were thinner than 3 nm, from 1 to few mm [70]. On the basis of statistical sampling of
and more than 65% of the sheets were thinner than 2 nm exfoliated graphene flakes using AFM they found that approx-
with the lateral in the range from 1 to 40 μm. The total yield imately 32% of the graphene flakes have thicknesses in the
of the graphene sheets was found to be 5–8 wt.% and more range 4–6 nm while nearly 80% of them are in the thickness
than 60% of the sheets were bilayers with AB stacking. More range of 4–8 nm. The total yield of graphene flakes dispersed
importantly, they were found to have significantly higher elec- in DI water excluding the large graphite pieces is nearly 50%.
tron mobility than most rGO and these graphene films formed Recently, Nanda et al. [71] demonstrated an electrochemi-
on a quartz substrate are promising for replacing currently cal technique for the synthesis of high-quality FLGS directly
used high-cost ITO (indium tin oxide) electrodes [67]. After from graphite using oxalic acid (a weak acid) as the electro-
that, Li et al. used commercially available LiFePO4 powders lyte. Wu et al. also used electrochemical intercalation exfolia-
(300 nm)/thin amorphous carbon layer (cLFP) to produce a tion (EIE) of expanded graphite in sulfuric acid electrolyte in
composite with graphene flakes [99]. The DMF solution of which H2SO4-GIC is used as a key intermediate to realize effi-
graphene flakes (concentration 250 ppm) was dropwise added cient exfoliation. The yield of the FLGS (<7 layers) with large
to cLFP powders under gentle stirring at 180°C to achieve a lateral sizes (tens of microns) is more than 75% relative to the
homogeneous coating of few-layer graphene flakes on the sur- total amount of starting expanded graphite. A low degree of
face of cLFP. The graphene flakes spread in the mixture with oxygen functionalization existing in the prepared FLG flakes
the aid of DMF solvent and wrapped the surfaces of cLFP enables them to disperse effectively in the solvent [100].
particles through van der Waals interaction [99]. By SEM, it
was illustrated that the graphene flakes nicely cover the cLFP
15.3.5 Exfoliation in Surfactants
particles after being added to cLFP and the wrinkles of the
graphene layers could be clearly observed. The hydrophobic nature of graphene makes the dispersion
The same group demonstrated that massive graphene difficult. The surfactant may help to stabilize the dispersed
sheets can be successfully obtained by exfoliating artificial graphitic flakes against aggregation due to their Coulomb
graphite, instead of high-cost graphite flakes using the same repulsion. Therefore, a strategy has been developed to dis-
electrochemical method. The product after the electrochemi- perse and exfoliate graphite to give graphene suspended in
cal reaction is substantially composed of wrinkled or folded aqueous surfactant solutions. For example, Lotya et al. [35]
thin sheets [68]. The thickness of the graphene sheet, mea- reported on sonication-based exfoliation of graphite in a solu-
sured by the AFM, is 2.2 nm, which can be considered as a tion of surfactant dodecyl benzene sulfonate. They found that
bilayer graphene. Meanwhile, the transparent electrodes can the dispersed phase was mainly consisting of thin graphene
be simply made by spraying graphene dispersions onto a flex- flakes, in which more than 40% of these flakes have <5 layers
ible polymer substrate. with ~3% of flakes consisting of monolayers. The graphene
Similarly, Mullen et al. used electrochemical exfoliation could be dispersed at a concentration up to 1.5 mg/mL [101].
of graphite in sulfuric acid solution for the fabrication of gra- A variety of alternative surfactants have been used to aid soni-
phene sheets in high quality and high yield. The as-prepared cation-based exfoliation and dispersion of graphene, including
graphene has a large sheet size, low oxygen content, and/or cationic surfactants cetyltrimethylammonium bromide, natu-
high C/O ratio as well as excellent electronic properties com- ral surfactants sodium cholate and sodium deoxycholate bile
parable to CVD graphene [69]. They introduced a vacuum salts and a bolaamphiphile [102]. The ionic surfactants adsorb
filtration method in association with the dry transfer method onto graphene and induce dispersion of graphene in water by
to produce highly conductive graphene films on various sub- providing electrostatic repulsion. On the other hand, nonionic
strates such as PTFE (polytetrafluoroethylene) membranes, surfactants, having a hydrophobic tail and a long hydrophilic
glass, or polyethylene terephthalate (PET). Lin et al. used part, provide steric repulsions that stabilize graphene [101].
various protonic acids (i.e., H2SO4, H3PO4, or H2C2O4) in Therefore, surfactants are promising agents to serve as elec-
aqueous solution as the electrolyte [70]. They developed the trolytes for electrochemical synthesis of graphene. Surfactants
Graphene 281
may prevent reagglomeration of graphene, with the hydropho- FESEM (field emission scanning electron microscopy), they
bic groups interacting with the p-orbitals of graphene, and the confirmed the presence of a thin crumpled sheet structure on
hydrophilic groups stabilizing the sheets in water [101,102]. the surface of the graphite foil. They also proposed that the
Wang et al. first reported electrolytic exfoliation of graphite intrinsic ripples of the graphene sheets might develop into
using surfactant poly(sodium-4-styrenesulfonate) (PSS) [72]. wavy structures within the macroscopic scale. An edge view
TEM analysis reveals that the obtained sheets are few flat gra- image by HRTEM (high-resolution transmission electron
phene sheets in larger sizes of a few square micrometers, in microscopy) reveals that the sheets were a few layers of gra-
which about 2–3 layers of graphene overlap. The thickness of phene, with a thickness of 3 nm, corresponding to an approxi-
the graphene flake is measured to be about 0.8 nm, which is mate 1–10 layer stacking of the monatomic graphene sheets.
similar to that of graphene sheets synthesized by other chemi- Furthermore, crystalline nonuniformities were observed, sug-
cal methods (Figure 15.10). It was proposed that the edge-to- gesting the ripple and turbostratic structure due to the incom-
face interaction between the graphene surface and aromatic mensurate and disordered layer stacking. Interestingly, they
rings of PSS is primarily responsible for the binding (p–p provided an autocorrelation function image, obtained to fur-
interaction). Particularly, PSS is highly hydroscopic. When it ther improve HRTEM image quality for statistical analysis.
is dissolved in water, it will dissociate into Na+ cations and The autocorrelation function image was found to match well
polystyrenesulfonate anions. During the electrolytic process, with the atomic nature of graphene, additionally confirmed
polystyrenesulfonate anions were forced to move to the posi- by overlaid hexagonal structural representation on top of the
tive graphite electrode under the electric force and interact image [103]. It was estimated that the respective hexagon
with graphite, leading to electrolytic exfoliation of the graph- width and C–C bond length were 0.247 and 0.142 nm, close
ite rod [72]. to the expected values (0.25 and 0.142 nm). This fascinating
Kim and coworkers prepared graphene sheets on the HRTEM characterization indicates the production of high-
graphite surface by electrolysis in PSS solution [73,103]. By quality graphene sheets by the electrolytic process.
(a) Galvanostat
– +
Cathode Anode
SO3–
Electrolyte
SO3–
SO3–
(b) (c)
100 nm 200 nm
(d) 400
400 200
Height
300
200 0
nm
pm
200 0 –200
–200 –400pm
100
–400
0 0 0 100 200 300 400
0 100 200 300 400 nm
nm
FIGURE 15.10 (a) Diagram of the apparatus for the synthesis of graphene via electrolytic exfoliation. (b) FESEM image of bulk graphene
powders. (c) TEM image of many graphene sheets. (d) AFM image of spray-deposited graphene flakes and a line scan taken horizontally
through the image as marked with a gray line, from which the height of a small graphene flake and a large graphene flake was determined
to be about 0.8 nm. (Reprinted from Carbon, 47, G. Wang et al. 3242–3246. Copyright 2009, with permission from Elsevier.)
A three-electrode cell system with controlled potential order of 1 − 20 μm [61]. The exfoliation was induced by the
of working electrode has been developed for controlled syn- intercalation of Et3NH+ ions and the subsequent formation
thesis of graphene by Alanyalioglu et al. [74]. In the work, of triethylamine (Et3N). The intercalation and decomposition
sodium dodecyl sulfate (SDS) was used as the surfactant in a of the solvent cations was found to play a critical role in the
two-step synthesis: electrochemical intercalation of SDS into complete exfoliation of graphene layers [61]. DMSO has a
graphite anode followed by electrochemical exfoliation of an wide electrochemical window and its surface tension is close
SDS–intercalated graphite cathode [74]. These electrochemi- to the surface energy of graphite, preventing the detached
cal processes yield a stable colloidal graphene/SDS suspen- graphene from restacking. Mixture solution of Na2SO4 and
sion. As found in a previous work [103], the potential value SDS surfactant was used as an electrolyte to produce con-
for SDS intercalation into graphite also plays an important trolled thickness graphene flakes [76]. When using the ultra-
role in determining the structural order, size, and number sound assistance, bilayer sheets (0.8 nm thick) are dominant
of layers of synthesized graphene sheets. Alanyalioglu et al. with an yield of about 54% of the total, whereas without
found that relatively high intercalation potentials led to the ultrasound four-layer (1.6 nm thick) sheets (52% of the total)
production of graphene sheets with the highest structural are obtained.
order and lowest number of layers. Average size and thick- The above-mentioned synthetic approaches have been
ness of graphene sheets prepared at high potentials for SDS developed for graphene synthesis. Other carbon materials,
intercalation into graphite were measured to be about 500 such as graphene platelets and graphene dots have also been
and 1 nm, respectively, indicating the presence of graphene synthesized by electrochemical exfoliation using flame-retar-
sheets as thin as a monolayer. Similarly, graphene modified dant electrolyte or direct electrolysis [106,107]. Inspired by
with SDS or sodium dodecylbenzenesulfonate (SDBS) was the work of Loh et al. using propylene carbonate-based elec-
prepared by the electrochemical method [104,105]. Most trolyte [58], Xiang et al. [106] developed a method employ-
of the obtained graphene had a thickness of <2 nm and the ing trimethyl phosphate, flame-retardant electrolyte in Li-ion
resulting graphene/SDBS composites showed very good dis- battery (LIB), as the conductive solution. After applying the
persion stability both in water and in organic solvents ethanol charge–discharge program, the original graphite was exfo-
or acetone. liated into graphitic platelets with different dimensions and
Mensing et al. [75] reported electrochemical method shapes, such as disk shape, square, triangular platelets with
for the production of a stable aqueous dispersion of a gra- thickness in the range 0.5–1 µm, 1–2 µm, depending on the
phene–metal phthalocyanine hybrid material. The material temperature of the solution. Direct electrolysis of graphite
has been prepared by electrolytic exfoliation of graphite in pure water resulted in the production of high-quality carbon
an electrolyte containing copper phthalocyanine-tetrasul- nanodots (C-dots) with high purity [107]. By the dynamic light
fonic acid tetrasodium salt (TSCuPc). Single-layer graphene scattering histogram, they confirmed that the C-dots are well
sheets and FLGS, decorated with metal phthalocyanine dispersed with particle sizes ranging from 3 to 6 nm. TEM
molecules, are generated during the electrolysis and sta- image of C-dots revealed that C-dots with a diameter of about
bly dispersed in the electrolyte with no further chemical 4.5 nm consisted of several layers of graphene. Importantly,
treatment. Scanning electron/atomic force microscopic C-dots showed peroxidise mimetic function and visible-light
characterization shows that the TSCuPc–graphene hybrid sensitive photocatalytic activity for methyl orange degrada-
structure has a sharp-edged particle morphology with tion and a novel photocatalyst (TiO2/C-dots) exhibited an
thicknesses ranging from 2 to 6 nm, corresponding to 1–6 excellent visible-light photocatalytic activity.
graphene stacked layers and largely varied lateral dimen- It should be noted that besides direct electrochemical exfo-
sions from a few tens to several hundreds of nanometers. liation of graphite, electrochemical reduction of graphene
They proposed that the TSCuPc–graphene hybrid is formed oxide has been developed for the synthesis of graphene.
by noncovalent p–p interactions between graphene sheets Applying cathodic potential is an effective tool to reduce
and metal phthalocyanine. electronic states due to the change in the Fermi energy level
Recently, the combination of different electrolytes enables of the materials surface under the external power source
to produce high-quality FLGS [57,61,76]. When IL was used [108,109]. Thus, electrochemical reduction could provide an
with acetonitrile electrolyte with a much lower content of ILs extremely rapid and green chemical process for synthesizing
(1:50 IL/acetonitrile vol. ratio), up to 86% of exfoliation with graphene sheets [108,109].
the majority of the products as graphene flakes in addition Overall, electrochemical exfoliation represents a novel
to smaller quantities of carbonaceous particles and rolled and promising method of graphite exfoliation into graphene
sheets was achieved [57]. The approach has superior advan- sheets. This method was implemented quickly, in some cases
tages in terms of cost efficiency due to the low IL content, within 1 h, low temperature process, and yield graphene
extended electrochemical stability in a nonaqueous electro- sheets of different sizes that are functionalized with differ-
lyte, and high exfoliation yield by effective anionic intercala- ent chemistries. The functionalized graphene could be soluble
tion within the graphitic layers. When the electrolyte used in solvent; therefore, it is readily processed on the substrate
was DMSO containing lithium and small alkylammonium with flexibility. Furthermore, the method has the potential for
ions (triethylammonium, Et3NH+), the produced material was large-scale production at low cost as graphite is abundant in
predominantly few-layer graphene with particle sizes on the nature (800 MT).
Graphene 283
15.4 APPLICATIONS The cations in IL were simultaneously reduced, leading to

the formation of 1-butyl-3-methylimidazolium free radicals,
A myriad of potential applications for graphene have been which could combine with either one of the paired electrons
proposed based on its outstanding properties. The quantity in the π-bond of the graphite [52]. The complex interactions
and functionality of graphene required varies in accordance between IL and graphite include electrostatic interaction and
with the application. For example, some applications such π–π interactions, eventually exfoliating graphite into gra-
as sensors [11], solar cells [15,16] , and counterelectrodes in phene sheets (Figure 15.11). The strong binding between IL
dye-sensitized solar cells [17,18] require thin films of gra- and graphene led to the formation of Bmim-functionalized
phene with high conductivity; other applications such as graphene. The efficient doping of nitrogen from Bmim ions
polymer composites [12], electrocatalysts for oxygen reduc- was realized upon thermal annealing at 400°C.
tion in fuel cells [19,20], photocatalysts in water splitting XPS is a powerful tool for the characterization of surface
[21,22], electrodes in supercapacitors [23], and LIBs [24,25] chemical composition. As shown in Figure 15.12, the pres-
require relatively large quantities of graphene nanosheets ence of N1s in the XPS survey spectra evidences the forma-
or platelets with functionality. Furthermore, each appli- tion of NG with estimated N content of 15.6%–22.1%. The
cation type required different specific properties of gra- N content was further confirmed using TGA and elemental
phene such as conductivity, surface area, doping, and so analysis [55]. Notably, the N content in NG synthesized using
on. Particularly, graphene synthesized by electrochemical ILs in this study is much higher than NG synthesized using
exfoliation has modified the surface (basal planes or edges) melamine (10%). The HR-XPS (high-resolution x-ray pho-
by bringing foreign atoms or molecules onto the surface toelectron spectroscopy) spectra of C1s exhibited tailing at
through chemical bonding, van der Waals forces, or p–p higher binding energies (Figure 15.12b), suggesting the pres-
interactions. This functionality can modulate the proper- ence of N and O heteroatom. The spectral deconvolution of
ties of graphene for various applications such as conductive C1s HR-XPS spectra reveals the chemical environment of
paper [58,100], water purification [64], electrochemical sen- C. The corresponding binding energy can be ascribed for
sors [63,95], energy storage, and energy conversion devices C–C/C = C (284.4–285 eV), C = N (285.5 eV), C–N (286.9–
[54,96,97,73,103,100,76]. 288.6 eV), C–OH (285.9–286.7 eV), and C–C = O/C = O
(287.1–289 eV). The spectral deconvolution of N1s HR-XPS
15.4.1 Precursor for the Synthesis spectra (Figure 15.12c) indicates the presence of pyridinic N
(398.2–398.4 eV), pyrrolic N (400.1–400.5 eV), and graphitic
of N-Doped Graphene
N (401.3–401.5 eV). The NG synthesized using [Bmim]Br
Nitrogen-doped graphene (NG) has attracted a great deal of showed the presence of pyrrolic N as a major configuration and
scientific interest in recent years owing to its unique proper- graphitic N as a minor one. In contrast, the NG synthesized
ties such as excellent electric conductivity, high binding abil- using [Bmim]Ac showed the presence of dominant pyridinic
ity, as well as great potential as an alternative electrocatalyst N, whereas using [Bmim][PF6] exhibited equal contribution
to the expensive platinum-based materials [110,111]. Various from pyridinic and pyrrolic forms [55]. To our surprise, the
techniques have been proposed for the synthesis of NG includ- NG with dominant graphitic N functionality, which is respon-
ing CVD [112], arc discharge of graphite electrode in pyri- sible for the catalytic activity of NG, can also be obtained by
dine vapor [113,114], annealing with NH3 and nitrogen plasma tailoring the pH of the electrolyte solution. The N bonding
treatments [115]. Clearly, these methods are costly, compli- configuration strongly depends on the charged properties of
cated, and utilize highly toxic nitrogen precursors. Therefore, electrolyte solution, viscosity, and the nature of ILs.
it is of great importance to develop a facile, eco-friendly, and The morphology and crystalline nature of NG were inves-
scalable approach for the synthesis of NG. In our recent work, tigated using TEM. HR-TEM image and selected area elec-
we demonstrated a facile synthesis of NG by IL-assisted electron diffraction indicate the presence of well-crystalline NG.
trolysis with subsequent thermal annealing of the resultant The SAED pattern does not show perfect hexagonal pattern,
IL-functionalized graphene sheet [55]. The method allows presumably due to the partial replacement of carbon atoms in
the synthesis of NG directly from graphite with a controlled graphene by nitrogen atoms, demonstrating that the nitrogen
N-doping configuration obtained by tailoring the ILs. atoms in IL cations have replaced the carbon atom in gra-
Briefly, the graphene was prepared by IL-assisted electroly- phene upon thermal annealing. In addition, we used AFM
sis method in which the electrolytes consist of IL, 1-butyl-3- to characterize the morphology and determine the number
methylimidazolium brom/hexafluorophosphate/acetate ([Bmim] of layers in NG. A uniform monolayer NG with an average
Br, [Bmim][PF6], [Bmim]Ac), and water. After 3-h electrolysis at thickness of <1 nm is clearly observed (Figure 15.12e and f).
a static potential of 15 V, the anode graphite rod was exfoliated The virgin graphene is known to possess low carrier density,
and a black precipitate gradually appeared at the bottom of the and thus a high sheet resistance. We expect the synthesized NG
reactor. The precipitate was washed off by distilled water, dried to exhibit excellent conductivity. To investigate the electrical
at 60°C, and subjected to thermal annealing under N2 atmo- properties of NG, we fabricated a back-gated electrical device.
sphere at 400°C for 4 h to yield NG. The volume resistivity was determined to be 1.4 × 10−5Ω m
Typically, hydroxyl and oxygen radicals were produced and the conductivity was about 7.15 × 104 S/m, which is sig-
by the oxidation of water on anode after applying voltage. nificantly higher than currently reported graphene and NG.
(a) (b)
van der Wall’s forces and/or

N
N+ Br+ electrostatic force
Exfoliation by the sum force
Electric force 3.350Å

1.420Å
van der wall’s forces and/or

electrostatic force
3.350Å
1.420Å
Electric field
(d) (c) van der Wall’s forces and/or

electrostatic force
Graphitic N
Pyridinic N
3.350Å
1.420Å
Pyrrolic N Exfoliation by the sum force
van der Wall’s forces and/or

electrostatic force
FIGURE 15.11 Schematic of the synthesis of N-doped graphene from graphite. (a) Anions of ionic liquids were pulled by anode, and
ionic liquids cations were also pulled to anode; (b) ionic liquids cations were inserted into each layer of graphite and exfoliated it by the
sum force of electric force, Van der Wall’s forces and electrostatic force; (c) ionic-liquid-functionalized graphene was obtained; (d) thermal
annealing to produce nitrogen-doped graphene with different N-doping configuration. (J. Y. Liu et al. J. Mater. Chem. C, 2013, 1, 1713–1716.
Reproduced by permission of The Royal Society of Chemistry.)
We attributed this advantageous property to its high N/C per- Traditionally, printable conducting inks are composed of met-
centage. Hopefully, the NG might possess properties similar to als such as Au, Ag, and Cu and conductive polymers [116–
those of carbon nitrides, that is, the incorporation of nitrogen 118]. Although these metals have outstanding conductivities,
atoms into the carbon matrix can significantly enhance their most of them are too expensive to be used in large quantities.
conducting, field emission, catalytic, and storage properties. Based on its unique properties, solution-dispersible graphene
Lin et al. also prepared the N-doped graphene from materials can be used for many solution-processing-based
electrolysis-obtained graphene. A certain amount of the as- applications including field-effect transistors, gas sensors,
prepared graphene flakes was annealed at 800°C for 1 h in and conductive paper [119,120]. Loh et al. dispersed of the
the flow of 400 sccm Ar/NH3 (95/5, v/v). The incorporation graphene flakes in several solvents (e.g., 10 mg/mL in dichlo-
of N in the exfoliated graphene oxide and the reduction in robenzene) to produce a highly dispersible carbon “ink”
oxygen-containing groups was confirmed by XSP measure- and performed brush painting on commercial paper (Figure
ment. The N/C atomic ratio was calculated to be ca. 5.4 at.% 15.13b) [58]. The conformal coating on paper could be readily
based on XPS spectra. The high-resolution XPS N1s spec- achieved, transforming the paper into a conductive sheet [58].
trum reveals the presence of pyridinic N (398.2 eV), pyrrolic It was found that the paper conductivity was propositional
N (400.5 eV), quantury N (401.1 eV), and N-oxides of pyri- to the graphene loading amount. A sheet resistance could be
dinic N (402.1 eV). The N-doped graphene flakes processes achieved as low as 15 Ω/◻, which is better than that of rGO
excellent electrocatalytic activity, stability, and toxicity toler- paper [121] and comparable to that of high-quality carbon
ance for oxygen reduction reaction in alkaline solution [70]. nanotube (CNT)-treated paper [122].
Li et al. also developed applications of graphene sheets
synthesized by electrolysis in transparent conducting films
15.4.2 Conductive Paper
[68]. The sheets were dispersed in DMF and then coated on
Printable conducting inks play important role in inkjet heated quartz by airbrush spraying method. The direct spray-
printing which are competitive alternatives to conventional ing deposition is easy to scale up for large-area surfaces, and
photolithography for the production of electronic devices. also well compatible with flexible substrates such as PET.
Graphene 285
(a) (b)
C ls
N ls
Intensity (a.u.)
Intensity (a.u.)
0 100 200 300 400 500 600 700 800 900 1000
284 286 288 290 292
Binding energy (eV)
Binding energy (eV)
(c) (d)
Pyridinic N Pyrrolic N
Intensity (a.u.)
Graphitic N
396 397 398 399 400 401 402 403 404

Binding energy (eV)
(e) (f )
2.5
1.8
3.0
2.0
1.6
2.5
Height (nm)
1.4
2.0
1.5
μm
1.2
1.5
1.0
1.0
1.0
0.8
0.5
0.5
0.6
0
0
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
μm Phase (μm)
FIGURE 15.12 XPS spectra of N-doped graphene synthesized using [Bmim]Br in HBr at pH 2.0. (a) Survey, (b) high-resolution C1s spectra,
(c) high-resolution N1s spectra, (d) schematic illustration of N-doping configurations, (e) AFM image, and (f) thickness profile of NG synthe-
sized using [Bmim]Br. (J. Y. Liu et al. J. Mater. Chem. C, 2013, 1, 1713–1716. Reproduced by permission of The Royal Society of Chemistry.)
The sample of size 1.5 × 1.5 cm showed a sheet resistance date, graphene-modified electrodes have been successfully
of 1.35 × 105 Ω/◻ with 70% light transmittance and suitable applied to detect organic and inorganic molecules such as glu-
use for flexible displays. Freestanding graphene paper can be cose [124], hydroquinone, catechol [125], guanine and adenine,
readily prepared by vacuum filtration of highly dispersible ascorbic acid, dopamine [126], ethanol [127], trinitrotoluene
graphene “ink” followed by peeling off the membrane filter (TNT) [128] azithromycin [129], and nitrate [130]. Graphene
[100]. The mean conductivity of the graphene paper without synthesized by electrochemical exfoliation is expected to
any thermal annealing treatment measured by a four-point improve the electron transfer between the target species and
probe technique is 24,500 S/m at room temperature. graphene [63,95]. Ai et al. demonstrated that graphene sheets
displayed sensitive and fast response to glucose. It was found
that the charge-transfer coefficient was 0.51, the electron trans-
15.4.3 Electrochemical Sensors
fer rate constant was 2.64 s−1, and the surface coverage was
Graphene has emerged as a new material replacing CNTs for 3.18 × 10 −10 mol/cm2. They further used the immobilized
fabricating electrochemical sensing and biosensing devices. glucose oxidase (GOD) to catalyze the reduction of dissolved
Particularly, its large specific surface area, good biocompat- oxygen and the glucose biosensor has a linear range from 0.4 to
ibility, and ability to promote electron transfer between elec- 20 mM with a detection limit of 0.1 mM. Moreover, the biosen-
troactive species and electrodes are beneficial for detecting sor exhibits acceptable reproducibility and storage stability. The
individual molecules on its surface [123]. In the past, graphene glucose in human plasma sample could be determined by the
has been studied for potential applications in the field of elec- fabricated biosensor with recoveries from 96.83% to 105.52%.
trochemistry, particularly in the area of sensing [123–127]. To Therefore, GOD/graphene holds promise to determine blood
(a) (b) (c)
(d) (e)
8000
Paper resistance (Ω/ )

4000
800
600
400
200
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Loading amount of graphene (mg/cm2)
FIGURE 15.13 Few-layer graphene powder and its derived ink that can be used to make highly conductive paper. (a) Photograph of few-layer
graphene powder (15 g). (b) Ink consisting of few-layer graphene flakes in dichlorobenzene (10 mg/mL) for brushing and writing on common
paper. (c, d) Commercial A4 printing paper coated with FLG. (e) Relationship between the resistance of the paper and the graphene loading
amount. (Reprinted with permission from J. Wang et al. J. Am. Chem. Soc., 8888–8891. Copyright 2011 American Chemical Society.)
sugar concentration in practical clinical analysis. Ai et al. also properties of graphene for heavy metal ions were investigated
performed the catalytic and adsorptive test toward the redox [64]. Graphene sheets were synthesized by IL-assisted elec-
of hydroquinone on the graphene modified glassy carbon electrolysis. The obtained graphene sheets synthesized with KPF6
trode [95]. Under optimal conditions, the oxidation peak cur- and IL C8P, ([C8mim]+[PF6]−) were used for the removal of
rent was proportional to hydroquinone concentration in the Pb(II) or Cd(II) ions from water. The result shows that the
range of 0.1–100 and 100–600 µM and the detection limit was graphene functionalized with PF6 could absorb 406.6 mg/g
estimated to be 0.01 µM (S/N = 3). (pH = 5.1) for Pb(II) and 73.42 mg/g (pH = 6.2) for Cd(II)
which is much higher than that by IL-functionalized graphene.
They speculated that the cation groups of ILs occupied some
15.4.4 Water Purification
sites of the GNSC8P surface, and competed with metal ions
Many approaches have been employed to remove metal ions in the adsorption process. On the other hand, the groups of
from contaminated water including precipitation, ion exchange, OH−, –COOH, and PF6 on GNSPF6 can synergize with metal
membrane filtration, and adsorption. However, adsorption is ions by ion exchange or ionic binding.
the most promising approach for the removal of metal ions The adsorption capacity to Pb2+ is higher than that of Cd2+
from water owing to its simplicity, cost-effectiveness, and ions. The specific cation formed an electric double-layer com-
enabling large-scale application. Therefore, adsorption method plex on the surface of graphene, which blocked the adsorp-
is widely used for processes of the magnitude of municipal tion of other ions. The different adsorption capacities to the
water supplies to small domestic water filters, particularly as metal ions is mainly depended on their ionic nature. The
packed bed filters. For practical use, there are several materials adsorption capacity to Pb2+ being slightly higher than that to
which could absorb heavy metals effectively, such as activated Cd2+ ions is presumably due to the larger atomic size of the
carbon, fly ash, prawn shell, and peanut hull pellet [131–133]. Pb(II) cation. The adsorption capacities of graphene sheets
In fact, people in the South Asia, where groundwater enrich- increased sharply versus pH value in the range of 1.0–6.0 and
ment by arsenic is a widespread problem, are currently using reached a maximum value of 406.6 mg/g (pH = 5.1) for Pb(II)
handmade filter containing natural sand and activated carbon and 73.42 mg/g (pH = 6.2) for Cd(II), respectively (Figure
for cleaning domestic water. The current research has focused 15.14). The adsorption competition of H+ with metal ions for
and developed new adsorbents with better absorptive capac- adsorption sites resulted in low adsorption capacity at low
ity for water purification. To date, the adsorption of metal pH condition. It was reported that the isoelectric point (IEP)
ions onto CNTs has been investigated extensively. Recently, of graphene where the zeta potential equals zero is 3.7 [64].
it has been suggested that graphene oxide and graphite oxide This indicates that at pH > IEP, graphene has negative sur-
can be practically usable for the separation of arsenic and face charge, which benefits the adsorbing cations. Thus, the
heavy metal from water [134,135]. In recent works, adsorption increase of the adsorption capacity in the pH range of 1.0–6.0
Graphene 287
processes reach equilibrium in 40 min and the adsorption iso-

100 therms could be described well by Langmuir and Freundlich
(1) GNSPF6-Pb (1)
(2) GNSPF6-Cd classical isotherm models [64].
80 (3) GNSC8P-Pb
(4) GNSC8P-Cd 15.4.5 Energy Storage and Conversion Devices
60 (2)
E (%)
Graphene-based electrodes have attracted much attention for

40 (3) applications in energy storage and conversion devices such
as rechargeable LIBs, supercapacitors, and solar cells [15,16,
20 (4) 23–25]. The LIBs function is based on the reversible shuttling
of Li+ ions and the difference of electrochemical properties
between the cathode and the anode. Therefore, the inherent
0
properties of the electrode materials would be a crucial fac-
1 2 3 4 5 6 7 tor that largely determines the overall performance of bat-
pHe teries. Discovering novel electrode materials or designing
novel nanostructures of electrode materials is a challenging
FIGURE 15.14 Effect of pH on the sorption (qe) of Pb(II) and task in order to meet the demand of next-generation batteries
Cd(II) on graphene/PF6, and graphene/C8P. (Reprinted from J. Hazar. for potential applications in high power systems. The key to
Mater., 183, L. Lu et al. 923–930. Copyright 2010, with permission increase the energy densities and cycling performance of LIBs
from Elsevier.) is first directed to optimize the performance of anode mate-
rials. Therefore, graphene-based electrode materials such as
may be attributed to the electrostatic interaction between the metal oxide/graphene composites have been designed in order
negatively charged surface of graphene and metal cations. The to improve the energy density and power density of LIBs.
highest adsorption of Pb2+ and Cd2+ on graphene sheets at pH The specific capacity of GNS for the anode was found to
5.0–6.0 is presumably due to the negatively charged surface be 540 mA h/g, which was much larger than that of graphite.
of graphene. Thus, it can be concluded that the π electron-rich Zhang et al. [97] reported the simultaneously exfoliation of
surfaces of graphene sheets offered the strongest attraction to graphite and deposition of metal oxides for electrode applica-
metal cations at pH 5.0–6.0. It was found that the adsorption tion. It was proposed that metal oxides are electrochemically
(a) (b)
1200
3.0 Discharge
Specific capacity (mAh/g)
2.5 1000 Charge

2.0 800
Voltage (V)
1.5
600
1.0
400
0.5
0.0 200
2nd 1st
0
0 200 400 600 800 1000 0 10 20 30 40
Specific capacity (mAh/g) Cycle number
(c)
1200
Co3O4/graphene
1000 0.3 A/g Discharge

0.6 A/g Charge
800
1.2 A/g
2.4 A/g
600
400
Discharge
200 Charge 4.8 A/g
Co3O4-acetylene black
0
0 5 10 15 20 25
Cycle number
FIGURE 15.15 (a) Voltage profiles of selected cycles at a current density of 0.3 C between 0.05 and 3 V, (b) cycling performance at 0.3 C,
and (c) rate capabilities of Co3O4/graphene and pure Co3O4 with acetylene black. (W. Y. Zhang et al. J. Mater. Chem., 2012, 22, 8455–8461.
deposited onto the surface of graphite electrodes and remained than those of some of the previously reported metal oxides/
in the resultant graphene, resulting in the formation of Fe2O3/ rGO [137]. The increase in the specific capacities is attributed
graphene, Co3O4/graphene. They found that Fe2O3/graphene to the activation process, for example, the gradual release of
electrodes exhibited a reversible capacity of 925 mA h/g dur- the trapped lithium ions inside the metal oxides crystal frame-
ing the 50th cycle at 0.3 C with a corresponding Coulombic work [97]. Kim et al. designed novel lightweight anodes for
efficiency of 96.6%. The result indicates its drastically an LIB by direct assembly of graphene nanosheets onto metal
enhanced performance compared with that of Fe2O3/acetylene current collectors and by the fabrication of freestanding gra-
black (743 mA h/g). The performance is comparable to that phene nanosheets paper electrodes (Figure 15.16) [73,103].
of Fe3O4/graphene synthesized by chemical exfoliation with Using the electrophoretically deposited graphene nanosheets
in situ reduction of iron hydroxide [136]. Besides that, Co3O4/ electrode, the discharge and charge capacities were initially
graphene composites were reported to exhibit large reversible found to be 932 and 255 mA h/g, respectively (Figure 15.16b).
capacity (880 mA h/g after 40 cycles), excellent cyclic per- The stable reversible capacities were attained at 200 mA h/g
formance, high Coulombic efficiency (above 99%), and good after 20 cycles. More interestingly, the freestanding graphene
rate capability (Figure 15.15). Such rate capabilities are better nanosheets electrode without a conductive metallic current
e–
(a) Conventional casting
With binder and

conductive carbon
H F
C C O
H
H F
F CH3
Electrophoretic deposition
Without binder and

conductive carbon
IPA
Ni2+
SO–3
Vacuum filtration
Without binder and

conductive carbon
Current collector - free
2.0
(b) (c) (d) 2.0 300
250
1st cycle Charge
1.5 2nd cycle 1.5
200 Discharge
5th cycle 150
C/100
Voltage (V)
Voltage (V)
10th cycle 100

20th cycle 50 C/5
1.0 1.0 0
0 10 20
Cycle (No.)
1st cycle 2nd cycle

0.5 0.5 5th cycle 10th cycle
20th cycle
0.0 0.0
0 200 400 600 800 1000 0 50 100 150 200 250
Specific capacity (mAh/g) Specific capacity (mAh/g)
FIGURE 15.16 (a) Illustration of the electrolytic exfoliation process for the synthesis of graphene sheets and schematic electrode fabrica-
tion approaches using graphene sheets for the LIB: conventional casting, electrophoretic deposition, and vacuum filtration. (b) Charge–dis-
charge curves of the electrophoretically deposited graphene sheets electrode at a C/5 rate. Inset shows a digital camera image of graphene
sheets electrode deposited over the entirety of the stainless steel. (c) Digital camera images of a freestanding graphene paper electrode. (d)
Charge–discharge curves of the graphene sheets paper electrode at a C/5 rate. Inset shows specific capacity versus cycle number of the gra-
phene sheets paper electrode at C/5 and C/100 rates. (Reprinted from Mater. Chem. Phys., 135, S.-H. Lee et al. 309–316. Copyright 2012,
with permission from Elsevier.)
Graphene 289
collector showed a capacity of 40 mA h/g at C/5 after 20 graphene hybrid electrode was fairly good with the second-
cycles, which were better electrochemical performances than cycle discharge capacities 270,245,224,202, and 156 mA h/g
the as-received graphite foil electrode (Figure 15.16d). The at current densities of 0.3 (1.0 C), 0.6 (2.0 C), 1.2 (4.1 C), 2.4
electrolytically prepared graphene nanosheets are potential (8.2 C), and 4.8 A/g (16.3 C), respectively (Figure 15.17f).
for lightweight anode applications in LIBs [103]. After changing the current density to 0.3 A, the electrode can
Wei et al. employed exfoliated graphene from IL electro- deliver a discharge capacity as high as 232 mA h/g, exhibit-
lyte intercalated with Li+ as an electrode material for battery. ing a good rate performance. Recently, Wang et al. [138] also
The specific capacity was about 640 mA h/g with a voltage synthesized V2O5/graphene by thermal decomposition of
plateau at 2 V when exfoliated graphene from IL was used graphene oxide; however, their cathode performance, that is,
[54]. Specific capacity improved to about 750 mA h/g with capacity is lower than V2O5/graphene synthesized by the elec-
a voltage plateau at 2.8 V when using the graphene exfoli- trochemical method [97].
ated from an IL–DI water mixture containing 0.5 M LiClO4. The graphene coated on LiFePO4 cathodes provided a
Graphene obtained from multiple electrochemical exfoliation capacity from 187 to 208 mA h/g, depending on the weight
was also used as the carbon material for LIB anode [70]. It percentage of graphene (from 0.8 to 2 wt.%) which are higher
was found that graphene was superior to commercial graph- than the reported values of 120–160 mA h/g for commercially
ite. The specific capacity measure data current density of available or synthetic LiFePO4 materials. The values are even
200 mA/g on the as-prepared was 1046 mA h/g for the first in excess of the theoretical value of 170 mA h/g for LiFePO4.
cycle while the reversible specific capacity remains as high Li et al. speculated that the extra capacity is mainly attributed
as 312 mAh/g after 200 cycles, which is 32% higher than to the reversible redox reaction between the lithium ions of the
that of raw and as-prepared graphene was capable of deliver- electrolyte and graphene flakes [99].
ing an impressive 234 mA h/g at 500 mA/g and 173 mA h/g Supercapacitors, a system that involve the electrochemi-
at1000 mA/g. These results indicate that the as-prepared gra- cal processes of either physical adsorption or redox reactions
phene is an excellent carbon material with high electrochemi- of ions, have attracted great interest as potential electrical
cal performance for LIBs [70]. storage technologies. Redox-based supercapacitors with
V2O5/graphene and LiFePO4/graphene have been synthe- nanostructured electrode materials have shown the poten-
sized by electrochemical exfoliation and used for cathode tial to combine the high energy density of conventional
materials [97,99]. Zhang et al. tested the cathode performance batteries and the high power capabilities of electrostatic.
of V2O5/graphene and the electrode delivered a second- Nanostructured metal oxide–graphene hybrids such as
cycle discharge capacity of 270 mA h/g, which decreased to SnO2, ZnO, MnO2/graphene were widely used for superca-
208 mA h/g during the 100th cycle with a capacity reten- pacitor applications [139–141]. Hu et al. reported that MoO3
tion of 99.74% per cycle. The rate performance of the V2O5/ nanoparticles directly grown on graphene by electrochemical
(a) (b) (e) 4.0

V2O5/gaphene
3.5
Voltage (V)
001
110
Intensity (a.u.)
011 400
3.0
200
211 310
101
600
2.5
022
411
012
102
401
2.0
10 20 30 40 50 0 50 100 150 200 250
2 θ (deg) 1 μm
(c) (d) Graphene (f )
Graphene
300 Discharge
Charge
250
200
150
0.438 nm
(001) 100
V2O5 nanobelt 50
0
0 20 40 60 80 100
50 nm 5 nm
Cycle number
FIGURE 15.17 (a) XRD pattern, (b) SEM image, (c) TEM image, and (d) HRTEM image of V2O5/graphene. (e) Voltage profiles of the
third cycle at a current density of 1.0 C between 2 and 4 V. (f) Cycling performance at 6.8 C. (W. Y. Zhang et al. J. Mater. Chem., 2012, 22,
8455–8461. Reproduced by permission of The Royal Society of Chemistry.)
(a) 0.4 (b)

4
0.3 B
3
0.2 2
B
Current (A/g)
Current (A/g)
0.1 1
0
0.0
–1 A
A
0.1
–2
–0.2 –3
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Voltage (V) Voltage (V)
(c) 4.0 (d) 1.0

3.5 Scan rate 100 mV s–1
3.0
50
2.5 0.8
20
2.0
1.5 10
Current (A/g)
1.0 0.6
Voltage (V)
5
0.5
0.0
–0.5 0.4
–1.0
–1.5
–2.0 0.2
–2.5
–3.0
–3.5 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0 500 1000 1500 2000 34,500 35,000 35,500 36,000
Voltage (V) Time (s)
FIGURE 15.18 (a) CV curves of (A) MoO3/graphene and (B) heat-treated MoO3/graphene samples at 5 mV/s in 6 M KOH solution. (b) CV
curves of (A) MoO3/graphene and (B) heat-treated MoO3/graphene samples at 100 mV/s in 6 M KOH solution. (c) CVs for heat-treated MoO3/
graphene electrodes at different scan rates using 6 M KOH aqueous electrolyte. (d) Galvanostatic charge/discharge curve of heat-treated
MoO3/graphene electrodes (5 mg each) in the potential window of 0–1 V at a constant current density of 100 mA/g at room temperature,
using 6 M KOH electrolyte. (J. B. Hu et al. J. Mater. Chem., 2011, 21, 15009–15014. Reproduced by permission of The Royal Society of
Chemistry.)
exfoliation could exhibit a high specific capacitance of 86 F/g routes to obtain high-quality graphene. The electrochemical
at a charge and discharge current density of 100 mA/g with exfoliation method has proved to be a promising method of
excellent cycling ability [96]. Although the specific capaci- graphite exfoliation into graphene sheets. This method can
tance is lower than other reported materials, the result is very be implemented quickly, is a low temperature process, and
encouraging to develop novel graphene composite electrode yields graphene sheets of different sizes that are functional-
materials for supercapacitors (Figure 15.18). ized with different chemistries. So far, various experimental
designs have been established to investigate the electroly-
sis process aiming to understand the fundamental science
15.5 CONCLUDING REMARKS
of exfoliation. Different electrolyte solutions are currently
The large-scale synthesis of high-quality graphene at low used for the synthesis of graphene and its composite and the
cost in an environmentally friendly manner still remains a resulting material exhibits its advantages in many applica-
challenge. A deep insight into the chemical and technologi- tions. Single-layer graphene sheets have been produced with
cal processes is crucial for further design of the synthetic increased percentage; however, the major products of most
Graphene 291
(a) (d)
Stencil mask
8 in.
Screen
39 in.
printer
(b) (e)
After
heating
Before
heating
(c) (f )
1st
30 2nd
in.
FIGURE 15.19 Photographs of the roll-based production of graphene films. (a) Copper foil wrapping around a 7.5-in. quartz tube to be
inserted into an 8-in. quartz reactor. The lower image shows the stage in which the copper foil reacts with CH4 and H2 gases at high tem-
peratures. (b) Roll-to-roll transfer of graphene films from a thermal release tape to a PET film at 120°C. (c) A transparent ultralarge-area
graphene film transferred on a 35-in. PET sheet. (d) Screen printing process of silver paste electrodes on graphene/PET film. The inset
shows 3.1-in. graphene/PET panels patterned with silver electrodes before assembly. (e) An assembled graphene/PET touch panel showing
outstanding flexibility. (f) A graphene-based touchscreen panel connected to a computer with control software. (Reprinted by permission
from Macmillan Publishers Ltd. Nat. Nanotechnol., S. Bae et al. 5, 574–578, copyright 2010.)
methods are few-layer graphene or thin graphene flakes with the flexible electronics and replacing ITO in displays which
improved dispersion in the solution which is beneficial for facilitate the chances of appearance in commercial devices
further solution processing. Graphene synthesized with IL (Figure 15.19) [142]. Future development of graphene syn-
was found to be a useful material for water purification and thesis, on a large scale and low cost, with the desired elec-
a very promising precursor for the synthesis of N-doped gra- trical properties could create an intriguing opportunity for
phene [52–56,60,104,105]. When molten salts were used as the development of next-generation electronic devices with
electrolyte, the method is very powerful for in situ coating higher speed, chemical stability, improved sensitivity, and
of metal oxide nanostructures on exfoliated graphene which functionality.
has potential for high energy density LIBs [62,96,97,99]. On
the other hand, graphene obtained with the surfactant was
able to disperse in organic/aqueous solvents and suitable REFERENCES
for solution processing, that is, the assembly of graphene 1. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y.
nanosheets onto metal current collectors and the fabrica- Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov,
tion of freestanding graphene nanosheets paper electrodes Science, 2004, 306, 666–669.
[73,103]. Graphene flakes have shown promise in applica- 2. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
3. A. K. Geim, Science, 2009, 324, 1530–1534.
tions such as conducting inks in which graphene has filled 4. M. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2010,
the gap between traditional poorly conducting polymer and 110, 132–145.
carbon-based films and expensive complicated silver inks 5. C. Lee, X. Wei, J. W. Kysar and J. Hone, Science, 2008, 321,
[58,68]. The future direction of graphene could be forming 385–388.
6. A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, 38. Y. Hernandez, V. Nicolosi, M. Lotya, F. M. Blighe, Z. Sun, S.
F. Miao and C. N. Lau, Nano Lett., 2008, 8, 902–907. De, I. T. McGovern et al. Coleman, Nat. Nanotechnol., 2008,
7. S. Park and R. S. Ruoff, Nat. Nanotechnol., 2009, 4, 217–224. 3, 563–568.
8. Y. Zhang, Y. -W. Tan, H. L. Stormer and P. Kim, Nature, 2005, 39. Y. Hernandez, M. Lotya, D. Rickard, S. D. Bergin and J. N.
438, 201–204. Coleman, Langmuir, 2009, 26, 3208–3213.
9. S. Latil and L. Henrard, Phys. Rev. Lett., 2006, 97, 036803. 40. D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z.
10. P. Avouris, Z. Chen and V. Perebeinos, Nat. Nanotechnol., Sun, A. Slesarev, L. B. Alemany, W. Lu and J. M. Tour, ACS
2007, 2, 605–615. Nano, 2010, 4, 4806–4814.
11. F. Schedin, A. K. Geim, S. V. Morozov, E. M. Hill, P. Blake, M. I. 41. W. Fu, J. Kiggans, S. H. Overbury, V. Schwartz and C. Liang,
Katsnelson and K. S. Novoselov, Nat. Mater., 2007, 6, 652–655. Chem. Commun., 2011, 47, 5265–5267.
12. S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. 42. A. Safavi, M. Tohidi, F. A. Mahyari and H. Shahbaazi, J.
Kohlhaas, E. J. Zimney, E. A. Stach, R. D. Pinen, S. T. Nguyen Mater. Chem., 2012, 22, 3825–3831.
and R. S. Ruoff Nature, 2006, 442, 282–286. 43. A. B. Bourlinos, V. Georgakilas, R. Zboril, T. A. Steriotis and
13. E. Y. Choi, T. H. Han, J. Hong, J. E. Kim, S. H. Lee, H. W. A. K. Stubos, Small, 2009, 5, 1841–1845.
Kim and S. O. Kim, J. Mater. Chem., 2010, 20, 1907–1912. 44. S. P. Economopoulos, G. Rotas, Y. Miyata, H. Shinohara and
14. A. S. Wajid, S. Das, F. Irin, H. S. Tanvir Ahmed, J. L. N. Tagmatarchis, ACS Nano, 2010, 4, 7499–7507.
Shelburne, D. Parviz, R. J. Fullerton, A. F. Jankowski, R. C. 45. U. Khan, A. ONeill, M. Lotya, S. De and J. N. Coleman,
Hedden and M. J. Green, Carbon, 2012, 50, 526–534. Small, 2010, 6, 864–871.
15. X. Wang, L. J. Zhi and K. Mullen, Nano Lett., 2008, 8, 323–327. 46. Y. Zhou, Q. Bao, L. A. L. Tang, Y. Zhong and K. P. Loh,
16. Q. Liu, Z. F. Liu, X. Y. Zhang, N. Zhang, L. Y. Yang, S. G. Yin Chem. Mater., 2009, 21, 2950–2956.
and Y. S. Chen, Appl. Phys. Lett., 2008, 92, 223303. 47. J. Zheng, C. Di, Y. Liu, H. Liu, Y. Guo, C Du, T. Wu, G. Yu
17. W. J. Hong, Y. X. Xu, G. W. Lu, C. Li and G. Q. Shi, and D. Zhu, Chem. Commun., 2010, 46, 5728–5730.
Electrochem. Commun., 2008, 10, 1555–1558. 48. Y. B. Tang, C. S. Lee, Z. H. Chen, G. D. Yuan, Z. H. Kang, L.
18. Y. X. Xu, H. Bai, G. W. Lu, C. Li and G. Q. Shi, J. Am. Chem. B. Luo, H. S. Song, Y. Liu, Z. B. He and W. J. Zhang, Nano
Soc., 2008, 130, 5856–5857. Lett., 2009, 9, 1374–1377.
19. B. Seger and P. V. Kamat, J. Phys. Chem. C, 2009, 113, 49. B. J. Jiang, C. G. Tian, L. Wang, Y. X. Xu, R. H. Wang, Y. J. Qiao,
7990–7995. Y. G. Ma and H. G. Fu, Chem. Commun., 2010, 46, 4920–4922.
20. S. J. Guo, S. J. Dong and E. W. Wang, ACS Nano, 2010, 4, 50. D. Rangappa, K. Sone, M. Wang, U. K. Gautam, D. Golberg,
547–555. H. Itoh, M. Ichihara and I. Honma, Chem. Eur. J., 2010, 16,
21. T. F. Yeh, J. M. Syu, C. Cheng, T. H. Chang and H. S. Teng, 6488–6494.
Adv. Funct. Mater., 2010, 20, 2255–2262. 51. J. -H. Jang, D. Rangappa, Y. -U. Kwon and I. Honma, J. Mater.
22. X. Y. Zhang, H. P. Li, X. L. Cui and Y. H. Lin, J. Mater. Chem., 2011, 21, 3462–3466.
Chem., 2010, 20, 2801–2806. 52. N. Liu, F. Luo, H. Wu, Y. Liu, C. Zhang and J. Chen, Adv.
23. M. D. Stoller, S. J. Park, Y. W. Zhu, J. H. An and R. S. Ruoff, Funct. Mater., 2008, 18, 1518–1525.
Nano Lett., 2008, 8, 3498–3502. 53. J. Lu, J. X. Yang, J. Z. Wang, A. L. Lim, S. Wang and K. P.
24. S. M. Paek, E. Yoo and I. Honma, Nano Lett., 2009, 9, 72–75. Loh, ACS Nano, 2009, 3, 2367–2375.
25. D. H. Wang, D. W. Choi, J. Li, Z. G. Yang, Z. M. Nie, R. Kou, 54. D. Wei, L. Grande, V. Chundi, R. White, C. Bower, P. Andrew
D. H. Hu et al. ACS Nano, 2009, 3, 907–914. and T. Ryhanen, Chem. Commun., 2012, 48, 1239–1241.
26. G. Ruan, Z. Sun, Z. Peng and J. M. Tour, ACS Nano, 2011, 5, 55. J. Y. Liu, H. Y. Chang, Q. D. Truong and Y. C. Ling, J. Mater.
7601–7607. Chem. C, 2013, 1, 1713–1716.
27. H. Ji, Y. Hao, Y. Ren, M. Charlton, W. H. Lee, Q. Wu, H. Li 56. V. V. Singh, G. Gupta, A. Batra, A. K. Nigam, M. Boopathi,
et al. ACS Nano, 2011, 5, 7656–7661. P. K. Gutch, B. K. Tripathi et al. Adv. Funct. Mater., 2012, 22,
28. K. S. Kim, Y. Zhao, H. Jang, S. Y. Lee, J. M. Kim, K. S. Kim, J. 2352–2362.
H. Ahn, P. Kim, J. Y. Choi and B. H. Hong, Nature, 2009, 457, 57. A. T. Najafabadi and E. Gyenge, Carbon, 2014, 71, 58–69.
706–710. 58. J. Wang, K. K. Manga, Q. Bao and K. P. Loh, J. Am. Chem.
29. P. Sutter, Nat. Mater., 2009, 8, 171–172. Soc., 2011, 133, 8888–8891.
30. K. V. Emtsev, A. Bostwick, K. Horn, J. Jobst, G. L. Kellog, L. 59. Y. L. Zhong and T. M. Swager, J. Am. Chem. Soc., 2012, 134,
Ley, J. L. McChesney et al. Nat. Mater., 2009, 8, 203–207. 17896–17899.
31. A. Dato, V. Radmilovic, Z. Lee, J. Phillips and M. Frenklach, 60. M. Zhou, J. Tang, Q. Cheng, G. J. Xu, P. Cui and L. G. Qin,
Nano Lett., 2008, 8, 2012–2016. Chem. Phys. Lett., 2013, 572, 61–65.
32. C. R. Herron, K. S. Coleman, R. S. Edwards and B. G. Mendis, 61. A. M. Abdelkader, I. A. Kinloch and R. A. W. Dryfe, ACS
J. Mater. Chem., 2011, 21, 3378–3383. Appl. Mater. Interfaces, 2014, 6, 1632 − 1639.
33. X. Huang, X. Qi, F. Boey and H. Zhang, Chem. Soc. Rev., 62. H. Huang, Y. Xia, X. Tao, J. Du, J. Fang, Y. Gan and W. Zhang,
2012, 41, 666–686. J. Mater. Chem., 2012, 22, 10452–10456.
34. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 63. H. S. Yin, Y. L. Zhou, X. M. Meng, K. Shang and S. Y. Ai,
80, 1339–1339. Biosens. Bioelectron., 2011, 30, 112–117.
35. M. Lotya, Y. Hernandez, P. J. King, R. J. Smith, V. Nicolosi, L. 64. X. J. Deng, L. L. Lu, H. W. Li and F. Luo, J. Hazard. Mater.,
S. Karlsson, F.M. Blighe et al. J. Am. Chem. Soc., 2009, 131, 2010, 183, 923–930.
3611–3620. 65. K. Parvez, Z. S. Wu, R. Li, X. Liu, R. Graf, X. Feng and K.
36. C. Valles, C. Drummond, H. Saadaoui, C. A. Furtado, M. He, Mullen, J. Am. Chem. Soc., 2014, 136, 6083–6091.
O. Roubeau, L. Ortolani, M. Monthioux and A. Penicaud, J. 66. G. M. Morales, P. Schifani, G. Ellis, C. Ballesteros, G. Martınez,
Am. Chem. Soc., 2008, 130, 15802–15804. C. Barbero and H. J. Salavagione, Carbon, 2011, 49, 2809–2816.
37. X. Li, G. Zhang, X. Bai, X. Sun, X. Wang, E. Wang and H. 67. C. Y. Su, A. Y. Lu, Y. P. Xu, F. R. Chen, A. N. Khlobystov and
Dai, Nat. Nanotechnol., 2008, 3, 538–542. L. J. Li, ACS Nano, 2011, 5, 2332–2339.
Graphene 293
68. C. M. Gee, C. C. Tseng, F. Y. Wu, H. P. Chang, L. J. Li, Y. P. 97. W. Y. Zhang, Y. Zeng, N. Xiao, H. H. Hng and Q. Yan, J.
Hsieh, C. T. Lin and J. C. Chen, Displays, 2013, 34, 315–319. Mater. Chem., 2012, 22, 8455–8461.
69. K. Parvez, R. J. Li, S. R. Puniredd, Y. Hernandez, F. Hinkel, 98. D. Alliata, P. Haring, O. Haas, R. Kotz and H. Siegenthaler,
S. Wang, X. L. Feng and K. Mullen, ACS Nano, 2013, 7, Electrochem. Commun., 1999, 1, 5–9.
3598–3606. 99. L. H. Hu, F. Y. Yu, C. T. Lin, A. N. Khlobystov and L. J. Li,
70. J. L. Liu, C. K. Poh, D. Zhan, L. F. Lai, S. H. Lim, L. Wang, Nat. Commun., 2013, 4, 1687–1694.
X. X. Liu et al. Nano Energy, 2013, 2, 377–386. 100. L. Wu, W. Li, P. Li, S. Liao, S. Qiu, M. Chen, Y. Guo, Q. Li,
71. P. Mahanandia, F. Simon, G. Heinrich and K. K. Nanda, C. Zhu and L. Liu, Small 2014, 10, 1421–1429.
Chem. Commun., 2014, 50, 4613–4615. 101. L. Guardia, M. J. Fernandez-Merino, J. I. Paredes, P. S.
72. G. Wang, B. Wang, J. Park, Y. Wang, B. Sun and J. Yao, Fernandez, S. V. Rodil, A. M. Alonso and J. M. D. Tascon,
Carbon, 2009, 47, 3242–3246. Carbon, 2011, 49, 1653–1662.
73. S. -H. Lee, S. –D. Seo, Y. -H. Jin, H. -W. Shim and D. -W. 102. M. Lotya, P. J. King, U. Khan, S. De and J. N. Coleman, ACS
Kim, Electrochem. Commun., 2010, 12, 1419–1422. Nano, 2010, 4, 3155–3162.
74. M. Alanyalioglu, J. J. Segura, J. Oro-Sole and N. Casan- 103. S. -H. Lee, S. -D. Seo, Y. -H. Jin, H. -W. Shim and D. -W. Kim,
Pastor, Carbon, 2012, 50, 142–152. Mater. Chem. Phys., 2012, 135, 309–316.
75. J. P. Mensing, T. Kerdcharoen, C. Sriprachuabwong, A. 104. P. Li, S. H. Bae, Q. Y. Zan, N. H. Kim and J. H. Lee, Adv.
Wisitsoraat, D. Phokharatkul, T. Lomas and A. Tuantranont, Mater. Res., 2010, 123–125, 743–746.
J. Mater. Chem., 2012, 22, 17094–17099. 105. K. Kakaei, Carbon, 2013, 51, 195–201.
76. J. Liu, M. Notarianni, G. Will, V. T. Tiong, H. Wang and N. 106. H. F. Xiang, J. Y. Shi, X. Y. Feng, X. W. Ge, H. H. Wang and
Motta, Langmuir, 2013, 29, 13307 − 13314. C. H. Chen, Electrochim. Acta, 2011, 56, 5322–5327.
77. M. Tariq, M. G. Freire, B. Saramogo, J. A. P. Coutinho, J. 107. H. Ming, Z. Ma, Y. Liu, K. Pan, H. Yu, F. Wang and Z. Kang,
N. C. Lopes and L. P. N. Rebelo, Chem. Soc. Rev., 2012, 41, Dalton Trans., 2012, 41, 9526–9531.
829–868. 108. H. -L. Guo, X. -F. Wang, Q. -Y. Qian, F. -B. Wang and X. -H.
78. N. G. Shang, P. Papakonstantinou, S. Sharma, G. Lubarsky, Xia, ACS Nano, 2009, 3, 2653–2659.
M. Li, D. W. McNeill, A. J. Quinn, W. Zhou and R. Blackley, 109. Y. Shao, J. Wang, M. Engelhard, C. Wang and Y. Lin, J. Mater.
Chem. Commun., 2012, 48, 1877–1879. Chem., 2010, 20, 743–748.
79. X. Q. Wang, P. F. Fulvio, G. A. Baker, G. M. Veith, R. R. 110. S. S. Yu, W. T. Zheng, C. Wang and Q. Jiang, ACS Nano, 2010,
Unocic, S. M. Mahurin, M. F. Chi and S. Dai, Chem. 4, 7619–7629.
Commun., 2010, 46, 4487–4489. 111. X. L. Li, H. L. Wang, J. T. Robinson, H. Sanchez, G. Diankov
80. S. Baldelli, J. Bao, W. Wub and S. Pei, Chem. Phys. Lett., and H. J. Dai, J. Am. Chem. Soc., 2009, 131, 15939–15944.
2011, 516, 171–173. 112. D. C. Wei, Y. Q. Liu, Y. Wang, H. L. Zhang, L. P. Huang and
81. M. H. Ghatee and F. Moosavi, J. Phys. Chem. C, 2011, 115, G. Yu, Nano Lett., 2009, 9, 1752–1758.
5626–5636. 113. L. S. Panchokarla, K. S. Subrahmanyam, S. K. Saha, A.
82. D. Nuvoli, L. Valentini, V. Alzari, S. Scognamillo, S. B. Bon Govindaraj, H. R. Krishnamurthy, U. V. Waghmare and C. N.
and M. Piccinini, J. Mater. Chem., 2011, 21, 3428–3431. R. Rao, Adv. Mater., 2009, 21, 4726–4730.
83. B. Zhang, W. Ning, J. Zhang, X. Qiao, J. Zhang and J. He, J. 114. K. S. Subrahmanyam, L. S. Panchakarla, A. Govindaraj and
Mater. Chem., 2010, 20, 5401–5403. C. N. R. Rao, J. Phys. Chem. C, 2009, 113, 4257–4259.
84. C. X. Guo, Z. S. Lu, Y. Lei and C. M. Li, Electrochem. 115. Y. Shao, S. Zhang, M. H. Engelhard, G. Li, G. Shao, Y. Wang,
Commun., 2010, 12, 1237–1240. J. Liu, I. A. Aksay and Y. Lin, J. Mater. Chem., 2010, 20,
85. B. N. Reddy, M. Deepa, A. G. Joshi and A. K. Srivastava, J. 7491–7496.
Phys. Chem. C, 2011, 115, 18354–18365. 116. H. M. Nur, J. H. Song, J. R. G. Evans and M. J. Edirisinghe, J.
86. M. R. Wagner, J. H. Albering, K. C. Moeller, J. O. Besenhard Mater. Sci.: Mater. Electron., 2002, 13, 213–219.
and M. Winter, Electrochem. Commun., 2005, 7, 947–952. 117. J. Perelaer, B. J. de Gans and U. S. Schubert, Adv. Mater. 2006,
87. T. Abe, Y. Mizutani, N. Kawabata, M. Inaba and Z. Ogumi, 18, 2101–2104.
Synth. Met., 2001, 125, 249. 118. K. P. Bong, K. Dongjo, J. Sunho, M. Jooho and S. K. Jang,
88. H. Buqa, A. Wursig, J. Vetter, M. E. Spahr, F. Krumeich and Thin Solid Films, 2007, 515, 7706–7711.
P. Novak, J. Power Sources, 2006, 153, 385–390. 119. S. Wang, P. K. Ang, Z. Q. Wang, A. L. L. Tang, J. T. L. Thong
89. K. S. Suslick and G. J. Price, Annu. Rev. Mater. Sci., 1999, 29, and K. P. Loh, Nano Lett. 2010, 10, 92–98.
295–326. 120. V. Dua, S. P. Surwade, S. Ammu, S. R. Agnihotra, S. Jain, K.
90. W. Sirisaksoontorn, A. A. Adenuga, V. T. Remcho and M. M. E. Roberts, S. Park, R. S. Ruoff and S. K. Manohar, Angew.
Lerner, J. Am. Chem. Soc., 2011, 133, 12436–12438. Chem. Int. Ed., 2010, 49, 2154–2157.
91. G. Eggert and J. Heitbaum, J. Electrochim. Acta, 1986, 31, 121. D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H.
1443–1448. B. Dommett, G. Evmenenko, S. T. Nguyen and R. S. Ruoff,
92. W. K. Hsu, J. P. Hare, M. Terrones, H. W. Kroto, D. R. M. Nature, 2007, 448, 457–460.
Walton and P. J. F. Harris, Nature, 1995, 377, 687–688. 122. L. Hu, J. W. Choi, Y. Yang, S. Jeong, F. L. Mantia, L. F.
93. C. Schwandt, A. T. Dimitrov and D. J. Fray, J. Electroanal. Cui and Y. Cui, Proc. Natl. Acad. Sci. U.S.A., 2009, 106,
Chem., 2010, 647, 150. 21490.
94. W. Lv, D. M. Tang, Y. B. He, C. H. You, Z. Q. Shi, X. C. Chen, 123. T. Gan and S. Hu, Microchim. Acta, 2011, 175, 1–19.
C. M. Chen, P. X. Hou, C. Liu and Q. H. Yang, ACS Nano, 124. C. Shan, H. Yang, J. Song, D. Han, A. Ivaska and L. Niu, Anal.
2009, 3, 3730–3736. Chem., 2009, 81, 2378–2382.
95. J. Wang, H. S. Yin, X. M. Meng, J. Y. Zhu and S. Y. Ai, J. 125. Z. Liu, Z. Wang, Y. Cao, Y. Jing and Y. Liu, Sens. Actuators
Electroanal. Chem., 2011, 662, 317–321. B, 2011, 157, 540–546.
96. J. B. Hu, A. Ramadan, F. Luo, B. Qi, X. J. Deng and J. Chen, 126. F. Li, J. Chai, H. Yang, D. Han and L. Niu, Talanta, 2010, 81,
J. Mater. Chem., 2011, 21, 15009–15014. 1063–1068.
127. C. Shan, H. Yang, D. Han, Q. Zhang, A. Ivaskac and L. Niu, 136. G. M. Zhou, D. W. Wang, F. Li, L. L. Zhang, N. Li, Z. S. Wu,
Biosens. Bioelectron., 2010, 25, 1504–1508. L. Wen, G. Q. Lu and H. M. Cheng, Chem. Mater., 2010, 22,
128. S. Guo, D. Wen, Y. Zhai, S. Dong and E. Wan, Biosens. 5306–5313.
Bioelectron., 2011, 26, 3475–3481. 137. J. X. Zhu, Y. K. Sharma, Z. Y. Zeng, X. J. Zhang, M.
129. J. Y. Peng, C. T. Hou, X. X. Liu, H. B. Li and X. Y. Hu, Srinivasan, S. Mhaisalkar, H. Zhang, H. H. Hng and Q. Y.
Talanta, 2011, 86, 227–232. Yan, J. Phys. Chem. C, 2011, 115, 8400–8406.
130. K. Liu, J. Zhang, G. Yang, C. Wanga and J. Zhu, Electrochem. 138. Z. L. Wang, D. Xu, Y. Huang, Z. Wu, L. M. Wang and X. B.
Commun., 2010, 12, 402–405. Zhang, Chem. Commun., 2012, 48, 976–978.
131. Y. J. Xu, G. L. Weinberg, X. Liu, O. Timpe, R. Schlogl and D. 139. F. H. Li, J. F. Song, H. F. Yang, S. Y. Gan, Q. X. Zhang, D.
S. Su, Adv. Funct. Mater., 2008, 18, 3613–3619. X. Han, A. Ivaska and L. Niu, Nanotechnology, 2009, 20,
132. J. S. Hu, L. S. Zhong, W. G. Song and L. J. Wan, Adv. Mater., 455602.
2008, 20, 2977–2982. 140. Y. Zhang, H. Li, L. Pan, T. Lu and Z. Sun, J. Electroanal.
133. Y. Liu, Y. Li and X.P. Yan, Adv. Funct. Mater., 2008, 18, Chem., 2009, 634, 68–71.
1536–1543. 141. S. Chen, J. Zhu, X. Wu, Q. Han and X. Wang, ACS Nano,
134. V. Chandra, J. Park, Y. Chun, J. W. Lee, I. C. Hwang and K.S. 2010, 4, 2822–2830.
Kim, ACS Nano, 2010, 4, 3979–3986. 142. S. Bae, H. K. Kim, Y. B. Lee, X. F. Xu, J. S. Park, Y. Zheng, J.
135. W. Gao, M. Majumder, L. B. Alemany, T. N. Narayanan, M. Balakrishnan et al. Nat. Nanotechnol., 2010, 5, 574–578.
A. Ibarra, B. K. Pradhan and P. M. Ajayan, ACS Appl. Mater.
Interfaces, 2011, 3, 1821–1826.
16 Modification of Graphene with
Polymers via Addition Chemistry
Horacio J. Salavagione
CONTENTS
Abstract...................................................................................................................................................................................... 295
16.1 Introduction...................................................................................................................................................................... 295
16.2 General Considerations on Graphene Reactivity.............................................................................................................. 296
16.3 General Aspects of Addition Reactions on Graphene: The Simple Molecules Case....................................................... 297
16.3.1 Nitrene.................................................................................................................................................................. 297
16.3.2 Free-Radical Additions......................................................................................................................................... 297
16.3.3 Click Reactions..................................................................................................................................................... 300
16.3.3.1 CuAAC Reaction................................................................................................................................... 300
16.3.3.2 Thiol-Radical Click Reactions............................................................................................................... 301
16.3.4 Other Additions.................................................................................................................................................... 302
16.4 Coupling Graphene with Polymers by Means of Cycloaddition Reactions...................................................................... 302
16.4.1 Polymer Chemistry Considerations...................................................................................................................... 304
16.4.2 Nitrene Chemistry................................................................................................................................................ 304
16.4.3 Free-Radical Addition.......................................................................................................................................... 306
16.4.4 Click Chemistry.................................................................................................................................................... 307
16.4.4.1 Coupling Graphene and Polymers by CuAAC Reaction....................................................................... 307
16.4.4.2 Coupling Graphene and Polymers by Thiol-Radical Click Reactions...................................................310
16.4.5 Other Less Common Addition Reactions..............................................................................................................311
16.5 Conclusions........................................................................................................................................................................311
Acknowledgment........................................................................................................................................................................311
References...................................................................................................................................................................................311
ABSTRACT make this happen. Currently the main problems associated

with graphene are related to the lack of a reliable method for
The covalent modification of graphene with polymers, two mass-scale production of pristine graphene and to limitations
families of materials with very different but complementary for its integration in different devices or multicomponent sys-
properties, constitutes a powerful strategy for the preparation tems due to some chemical inertness [1–6]. Although gra-
of new advanced materials. phene possesses per se outstanding electronic and structural
This chapter focuses in the modification of graphene with properties, these alone do not necessarily guarantee good
polymers by different addition reactions including the cou- “performance.” The key to obtain superior graphene-con-
pling of nitrene, free-radicals and click reactions. The general taining materials lies in the selection of the correct approach
aspects of these reactions on graphene are presented consider- to integrate graphene in these systems. With this in mind,
ing the cases of simple molecules. While the first two types the chemistry of graphene or, in other words, the possibil-
of reactions proceed by direct coupling of radicals to the gra- ity to functionalize it with different substituents, covalently
phene surface, the click reactions may require the previous or through supramolecular interactions, is a necessary step in
functionalization of graphene with clickable groups. order to broaden the scope of its applications [7–10]. Thus, the
The main interest of this chapter relies in reactions directed chemistry of graphene is currently not receiving a great deal
to pure sp2 carbon network, comparing the graphene’s possibili- of attention as a route to create an electronic gap and to make
ties with older carbon nanosctructures such as carbon nanotubes the handling of graphene easier, but also plays an important
and fullerene. However, other reactions using graphene deriva- role in developing novel materials with better performance,
tives such as graphene oxide are also included for comparison. apart from that related to electronics. Covalent chemical
functionalization has been employed to control the electronic
properties, that is, for bandgap opening [11] and electrical iso-
16.1 INTRODUCTION
lation of graphene [12]. However, one of the fields in which
Despite the fact that graphene holds much promise for count- the chemistry of graphene can be decisive is that of the poly-
less applications, some problems must be resolved in order to mer nanocomposites [13–16].
295
Nowadays the concept of synergy through the combination can be viewed as 0D (zero-dimensional) and 1D (one-dimen-
of different materials is one of the most successful approaches sional) structures of a 2D graphene sheet, respectively.
at the frontiers of materials technology. In this respect, the Thus, concerning the reactivity of [60] fullerene in com-
combination of graphene with polymers, two families of mate- parison with the parent graphene there are some differences
rials with very different but complementary properties, consti- which may be taken into account. First, [60] fullerene does
tutes a powerful route for the creation of new materials with not have edge planes and the π-orbital alignments of their car-
superior properties. This chapter will focus on the different bon atoms are almost perfect with a π-orbital misalignment
chemical routes described to date for the covalent modification angle of ϕ = 0°. Second, in contrast with planar graphene,
of graphene with polymers by addition reactions. The chapter the sp2 C atoms in [60] fullerene are pyramidalized and this
will focus specifically on reactions directed toward a pure sp2 deviation from planarity introduces a large strain energy
carbon network, and compare the possibilities in graphene with (Figure 16.1). The major driving force for addition reactions
other carbon nanosctructures such as carbon nanotubes (CNTs) leading to adducts bound to C60 is the resulting relief of strain
and fullerenes. The state of the art of nitrene, free-radical addi- in the fullerene cage, which is the major reactivity principle of
tion from diazonium salts and click chemistry with graphene fullerenes. Thus, nucleophilic addition, cycloadditions, radi-
will be revised and balanced alongside other less explored addi- cal additions, hydrogenation, and halogenation reactions have
tion reactions. Although the chapter aims at reactions on pris- been reported for [60] fullerene [18] without the need for acti-
tine graphene, other reactions using graphene derivatives such vation strategies or the presence of heteroatoms necessary for
as graphene oxide will be used for comparison. the bulk functionalization of graphene [7].
While fullerene reactivity is primarily driven by strain
16.2 GENERAL CONSIDERATIONS relief, which arises from pyramidalization at the carbon atom,
it turns out that in single-walled carbon nanotube (SWCNT)
ON GRAPHENE REACTIVITY
sidewalls the π-orbital misalignment between adjacent carbon
Since graphene is entirely composed of sp2 carbon atoms, atoms plays a larger influence in determining the overall reactiv-
the extensive previous work developed with its structurally ity (Figure 16.1). This misalignment, associated with bonds at
similar older cousins, CNTs and fullerenes, has paved the an angle to the tube circumference (i.e., bonds that are neither
way for the chemical modification of graphene. Therefore, parallel nor perpendicular to the tube axis), is the origin of tor-
the chemical modification and reactivity of graphene needs sional strain in the nanotubes, and the relief of this strain con-
to be discussed in the context of the reactivity of its closest trols the extent to which addition reactions occur with nanotubes
carbon relatives—CNTs (carbon nanotubes) and fullerenes [19]. Since π-orbital misalignment as well as pyramidalization
(Figure 16.1). In comparison with the functionalization of the scale inversely with tube diameter [20], smaller diameter tubes
curved π-surface of fullerenes which can be carried out very are expected to be more reactive than larger diameter tubes.
easily, the characteristic aromatic sp2-bonded carbon frame- Therefore, multiwalled carbon nanotubes (MWCNTs) with an
work of graphene is chemically more inert. While fullerenes outer tube diameter in the range of 5–50 nm generally exhibit
are curved in two dimensions, CNTs are curved only in one a much lower tendency toward sidewall addition reactions than
dimension and their reactivity lies somewhere intermediate SWCNT with diameter distributions in the range of 0.7–2 nm.
between that of fullerenes and graphene. A variety of reactions similar to that reported for fullerene
To rationalize the differences in reactivity among these derivatives have been described for the sidewall functional-
carbon nanostructures, different variables have to be consid- ization of pristine SWCNT including nucleophilic additions,
ered, that is, (i) the local sites for reactions (edges or basal radical additions, carbene additions, hydrogenation, ozonation,
planes), (ii) the strains generated by the geometry of the car- diazotization, and a variety of cycloadditions, among others [21].
bon nanostructures, and (iii) the π-orbital alignments of car- In addition to this “fullerene-like” reactivity it is important
bon atoms [17]. Ideally graphene is a two-dimensional (2D) to point out that other kinds of reactions are observed in CNT,
flat monolayer of carbon atoms in a honeycomb-like struc- which are more similar to that observed in graphene. As a
ture having basal and edge planes that produce fullerenes and consequence of the purification methods used to isolate CNT,
CNTs when appropriately closed. Thus, fullerenes and CNT the removal of amorphous carbon and catalyst impurities is
Fullerene Carbon nanotube Graphene

Curved basal plane Edges Flat basal plane
Curved basal
plane
No edges
Edges
No π-misalignment θp = 11.6° π-misalignment θp(5,5)~6° θp = 0°
FIGURE 16.1 Crystal structures of different carbon allotropes.

Modification of Graphene with Polymers via Addition Chemistry 297
accompanied by the introduction of carboxylic acid function- most important), each with specific chemistry that improves
alities and other oxygen-bearing groups at defect sites of the the versatility of graphene for further chemistry. However,
CNT framework, especially at the edges. This kind of edge such conditions may damage the graphitic lattice of the gra-
functionalities that contain heteroatoms with specific chemis- phene sheets and may be incompatible with some functional
try are the primary target locations for other functionalization groups. To face these challenges, pristine graphene sheets
reactions similar to that observed in graphene. modified by acetylene or azide moieties have recently been
Regarding graphene, the mechanical exfoliation method shown to provide facile platforms for attaching various func-
developed by Geim and coworkers provides single-layer gra- tional molecules to graphene by the Cu(I)-catalyzed azide–
phene which can be identified by atomic force microscopy alkyne cycloaddition (CuAAC) reaction.
(AFM) [22]. Graphene sheets on a centimeter scale have
also been prepared by chemical vapor deposition (CVD) of 16.3 GENERAL ASPECTS OF ADDITION
low-molecular-weight hydrocarbon precursors on heated REACTIONS ON GRAPHENE: THE
transition-metal surfaces [23]. While the mechanical method
SIMPLE MOLECULES CASE
produces “perfect” graphene, other methods have been devel-
oped for solution-based approaches, although they produce Before considering the possibilities of nitrene, free-radical
low-quality graphene [24]. The most extensively used method addition of diazonium salts, and click reactions in connecting
to this purpose involves the Hummers and Offeman treatment graphene with polymers, a brief description on the fundamen-
that relies on the oxidation of graphite [25] followed by the tals of these reactions and their use in some cases involving
reduction of the resulting graphite oxide (GO) to yield stable graphene and discrete molecules will be given. The following
and processable colloidal suspensions of graphene [26,27]. section will discuss in detail those reactions that have been
Considering all-carbon-composed nanostructures (i.e., extended to polymers.
without any heteroatoms) graphene is seen, a priori as the least
reactive of this family of carbon nanostructures [28]. However,
16.3.1 Nitrene
different chemical reactions at both edge and basal planes of
graphene have been successfully attempted, the edges being Strictly speaking, nitrene is a very reactive species analogous
considered as the more reactive sites [28,29]. Edge carbons to carbene. It is known that optical (light) or thermal activa-
can adopt tetrahedral geometries more freely (without caus- tion of the azide group produces its dissociation into molecu-
ing extra strain) than the corresponding carbon atoms in basal lar nitrogen and highly reactive singlet nitrene radicals that
planes. Therefore, edge carbon atoms are preferred in covalent in the presence of sp2 networks undergo cycloaddition reac-
addition reactions, the zigzag edges being particularly reactive. tions to give aziridine adducts (Scheme 16.1a). As mentioned
Concerning the basal plane reactivity of graphene, it must be above, nitrene addition was first demonstrated for fullerene
taken into account that graphene is chemically unsaturated, [37,38], CNTs [39–43], carbon nano-onions [44] and has been
composed of sp2 carbon atoms protected within a π-conjugated subsequently extended to graphene [45–50].
system whose motion is constrained by the surrounding carbon The list of single molecules connected to graphene is given
atoms. Therefore, basal plane covalent addition usually encoun- in Scheme 16.1b. Perfluorophenyl azides (PFPA) have been one
ters large energy barriers, and reactive chemical groups, such of the preferred compounds to react by means of nitrene chem-
as atomic hydrogen, fluorine, and precursors of other chemical istry with fullerene [51], CNTs [52], and graphene [46,47,53].
radicals, are usually needed as reactants. So far, the chemical The para-functionalized PFPA can be seen as a heterobifunc-
modification of graphene with these reactive chemical groups tional reagent composed of a functional group R with specific
cannot be fully controlled, and can also take place on graphene chemistry in the para-position to the light-activated azido
edges [28]. However, recently internal carbons of graphene group [54]. Moreover, other moieties such as azidosilanes [45],
have been modified by cycloaddition reactions [30–32]. This phenylalanine [48], azidoethanol [55], and other nonaromatic
enhanced reactivity may result from the influence of structural [56] compounds have also been bonded to graphene.
distortion, where the origin is a spontaneous rippling (of ca.
1 nm [33]) causing some local strain [34] as predicted [35].
16.3.2 Free-Radical Additions
In summary, the feasibility of reactions occurring at the
edges of graphitic planes decreases in passing from graphene This type of reaction involves aryldiazonium salts, peroxides
to CNT and to fullerenes as a consequence of the diminu- [57,58] Bergman cyclization [59], and Kolbe electrosynthesis
tion of active sites, while the possibility of reactions on basal [60]. The major part of the studies in the literature corresponds
planes roughly follows the order fullerenes > CNT > gra- to the addition of aryl radical formed by decomposition of
phene. Regarding unmodified/untreated carbon nanostruc- certain aryldiazonium salts. Diazonium chemistry is perhaps
tures, that is, all-C-atom-composed materials, the reactivity the most widely employed method to functionalize carbon
is dominated by the reactions on the basal planes and is higher surfaces [61–63].
in nonplanar structures than in planar 2D laminates [36]. Contrary to other reactions such as cycloaddition (nitrene,
Moreover, as occurred with CNTs the strong oxidation of carbene, etc.) where the reaction is triggered by light, heat,
graphene generates a series of functional moieties on the 2D or specific chemical reagents, graphene itself takes part in
sheets (carboxylic groups, epoxides, hydroxyls, among the the mechanism of aryldiazonium coupling (Scheme 16.2).
+ – – + Δ / hν
(a) R N N N R N N N R N N N
N2
Graphene
(b) Nonaromatic compounds PFPA
F F
OH O
R= O NH Si(OCH3)3
R=
N F F
NH2 F F H
O
O F F
OH F F
O O
O O O
14 O O OH O O
O F F nS 2
Br F F
O
F F
O O F F
O (CF2)9 CF3
F F
SCHEME 16.1 (a) Mechanism of graphene functionalization by nitrene from the decomposition of azide. (b) Chemical structure of the
different families of compound connected to graphene by nitrene addition.
+ –
N N X
e–
R R
N2
SCHEME 16.2 Mechanism of diazonium salt coupling to graphene laminates.

The reaction is initiated when one delocalized electron on nitrite under mild, warm conditions. The in situ method was
graphene is transferred to the diazonium cation. Then, this first reported by Zhu et al. [66]. In this study, they demon-
intermediate loses a molecule of N2 generating the aryl radi- strated that both procedures proceed reasonably well high-
cal, which immediately attacks the graphene lattice forming lighting that the in situ procedure broadens the possibilities
a covalent bond [64]. of functionality since anilines are used instead of previously
Although there is a consensus on this general mechanism, prepared diazonium salts. However, it has been recently pro-
with the same reagents the results obtained are very differ- posed that the efficiency of the in situ approach depends on
ent depending on the reaction conditions. In fact, countless several parameters such as the diazonium salt concentration,
synthetic protocols have been addressed to modify graphene the participation of surfactants, and the degree of solvent par-
with aryldiazonium compounds [7,65–89]. The methods can ticipation (with solvent or without solvent) [90].
be divided into those that use previously prepared diazonium A list of the most important studies on the functional-
salts (ex situ) and those where they are generated in situ in ization of graphene by free-radical addition using diazo-
the presence of the specific aromatic amine and an organic nium salts is given in Table 16.1, where it can be seen that
TABLE 16.1
List of the Different Experimental Approaches Used to Functionalize Graphene by Free-Radical Addition of
Diazonium Salts
Diazonium Substituent Groups on
Preparation Method the Aromatic Ring Anion Experimental Remarks Graphene Source References
Ex situ ─NO2 BF4 − Tetrabutylammonium hexafluorophosphate Epitaxial graphene [65]
([Bu4N]PF6)/CAN, r.t.
Ex situ methoxyprop-1-yne BF4− Water, 0.5% sodium dodecylsulfate (SDS), Dispersed and CVD-grown [67]
45°C graphene
Ex situ tert-butyl BF4− 1,2-dimethoxyethane (DME) Potassium intercalated graphite [7]
Ex situ and in situ ─Cl, ─Br, ─NO2, ─OMe, BF4− Ex situ: (i) hydrazine, SDS; (ii) diazonium Graphene nanoribbons [66]
coupling in water in situ: (i) hydrazine,
SDS (ii) isoamyl nitrite, 70°C
Ex situ ─COOH Cl− Water (pH < 6.0) r-GO [68]
In situ ─C2H5OH Isoamyl nitrite, 80°C r-GO [83]
Ex situ ─Cl, ─Br, ─NO2, ─OMe BF4− (i) Hydrazine, water (pH = 10), 80°C; GO [69]
(ii) Diazonium coupling in water, r.t.
Ex situ ─NO2 BF4− Electrochemical reduction in sulfuric acid CVD–graphene [70]
(0.1 M)
In situ ─OH, Isoamyl nitrite, 80°C r-GO [85]
In situ ─C≡CH Isoamyl nitrite, 80°C r-GO [84]
Ex situ ─NO2 BF4− ACN graphene nanoribbons [72]
Ex situ ─Cl, ─Br, ─I, ─NO2, BF4− K2CO3, 1-octyl-3-methylimidazolium Tr–GO [73]
─COOH, ─C≡CH, tetrafluoroborate (OMIBF4) ionic
C≡N liquid (IL)
Ex situ ─C8H17 Cl− (i) NaBH4, pH = 10, 80°C; (ii) diazonium GO [74]
coupling, r.t; (iii) hydrazine, water, 100°C
Ex situ Anthracene Cl− HCl, NaNO2, 0°C GO [75]
Ex situ ─NO2 BF4− Water, r.t. Scotch-tape peeled graphene [76]
Ex situ ─SO3− (i) NaBH4 pH = 10; (ii) diazonium GO [79]
coupling pH = 10.
In situ ─Br NaNO2, AIBN, chlorosulphonic acid, Expanded graphite [81]
80–90°C
In situ ─SO3− (i) Hydrazine, 90°C (ii) water (pH > 8), GO [77]
isoamyl nitrite
Ex situ ─Br Water, r.t. Graphene from mechanical [82]
exfoliation of graphite
Ex situ ─NO2 SDS 0.5 wt.%, 35°C CVD-grown graphene on [91]
different substrates
In situ ─SH (i) NaBH4; (ii) NaNO2, HCl, 0–5°C GO [86]
tetrafluoroborates of aryldiazonium salts are preferred, while occurrence of physisorption and its extent is of fundamental
for in situ generation isoamyl nitrite is the most used organic importance [93].
nitrite.
Apart from the wet chemistry, free-radical addition reac-
16.3.3 Click Reactions
tions have also been addressed on the graphene surface. In
this case the underlying substrate plays an important role. Researchers in the biotechnology field, as well as in (nano)
Strano and colleagues observed remarkably different rates materials and polymer science, are continuously searching for
for electron-transfer reactions with organic diazonium salts well-defined ligation strategies that can be effectively used in
on monolayer graphene supported on various substrates [91]. the presence of a wide range of different functional groups typi-
Furnishing graphene with acetylenic groups is especially cally encountered in these fields. Important requirements for
interesting because it allows further reactions, as illustrated successful ligation strategies are high selectivity, orthogonality
below. The acetylenic groups can click with azides and thiol to other functional groups, compatibility with water and other
compounds opening a wide range of functionalities that can protic solvents and, if possible, with almost quantitative yields.
be subsequently connected to graphene (see Section 16.3.3). In this respect, Sharples and coworkers developed the approach
Recently, efforts have been devoted to the functionaliza- known as “click chemistry” that was defined as a set of stringent
tion of graphene with 4-ethynylaniline in order to prepare criteria that a process must meet to be useful in this context [94].
a 2D platform for click coupling with single molecules and
polymers [71,84,92]. However, for these systems, in addi- 16.3.3.1 CuAAC Reaction
tion to covalent coupling, π–π stacking of aromatic moieties Among the entire collection of click reactions currently pro-
on the basal planes of the carbon nanostructures may occur posed, the Huisgen 1,3-dipolar azide–alkyne cycloaddition
[7,64,93], hindering washing and subsequent post-modifi- is probably the most important exponent [95]. This chemical
cation. These adsorbed molecules, once the material is re- transformation is characterized, in its classical thermal version,
dispersed, can be more reactive than those covalently bonded by high reliability and broad tolerance to diverse reaction condi-
to the nanostructure where steric effects are decisive. This tions and functional groups. However, it was not until the discov-
can result in very low overall yields of the click reaction and ery of the Cu(I)-catalyzed variant of this reaction (Scheme 16.3)
consequently to products with properties almost identical to by the groups of Meldal [96] and Fokin and Sharpless [97] that
the starting materials. Therefore, the determination of the the actual potential of this coupling technology was shown. The
Cu(I)
+ N3–R'
N
R N
R N
R'
Cu(I)
+ R'
N
N3 N
N
R'
SCHEME 16.3 CuAAC reactions with graphene derivatives furnished with acetylenic or azide functionalities.
TABLE 16.2
List of the Procedures Employed to Furnish Graphene with Alkyne or Azide Groups for CuAAC Click Reaction
Graphene Source Clickable Group Principal Chemical Reaction Experimental Conditions References
GO ─N3 Acylation with –OH groups in GO (i) 2-bromo-2-methylpropanoyl bromide, triethylamine [100]
(TEA), DMF; (ii) NaN3, DMF
GO ─C≡C─ Esterification with COOH groups in GO (i) SO2Cl; (ii) propargylalcohol, TEA, Cl3CH [149,101]
GO ─C≡C─ Amidation with COOH groups in GO Propargylamine, N,N′-dicyclohexylcarbodiimide [102]
(DCC)/4-dimethylaminopyridine (DMAP), DMF
GO ─N3 Amidation with COOH groups in GO (i) 2-chloroethyl isocyanate, DMF; (ii) NaN3, dimethyl [109]
sulfoxide (DMSO)
Graphene ─C≡C─ Coupling with diazonium salts Isoamyl nitrite, free-solvent, 60°C [67,104]
GO ─C≡C─ Coupling with diazonium salts N2H2, water, 90°C; (ii) isoamyl nitrite, free-solvent, 60°C [71,84]
Graphene ─N3 Hassner-type addition ICl/NaN3, acetonitrile (ACN) [108]
GO ─N3 Nucleophilic attack NaN3, water/ACN [107]
main characteristics of the Cu(I)-catalyzed reaction are related subsequent nucleophilic substitution of the available bromine
to the fact that it is quantitative, very robust, insensitive, general, with sodium azide in DMF [109]. In another study, the reaction
and orthogonal. The triazole formed is essentially chemically of an isocyanate derivative bearing azide moieties with –COOH
inert to reactive conditions, for example, oxidation, reduction, and –OH groups on GO leading to the formation of ester carba-
and hydrolysis, and has an intermediate polarity. For a more mates have been reported [110]. A summary of these synthetic
detailed discussion on this reaction the reader is directed to a approaches developed to provide graphene with alkyne or azide
comprehensive review by Meldal and Tornøe [98]. functionalities is shown in Table 16.2.
To be involved in CuAAC reactions, graphene must be fur- The use of the CuAAC reaction with graphene derivatives
nished with either acetylenic or azide groups. These groups can has been addressed using the reaction conditions already devel-
be introduced on pristine graphite as well as on its chemically oped for the other parent carbon nanostructures thus allowing
richer cousin GO, which is produced by the strong oxidation the incorporation of different moieties to graphene sheets. In
of graphite. This harsh method produces a highly defective this respect, polymers are the most extensively explored enti-
material since a disruption to the extended 2D π-conjugation ties clicked to graphene and will be discussed later.
of graphene sheets occurs, leading to nanoscale graphitic sp2
islands surrounded by a disordered, highly oxidized sp3 sea, 16.3.3.2 Thiol-Radical Click Reactions
as well as other defects such as carbon vacancies [99]. While The discovery of the click features of this reaction goes back
this treatment provokes insulating behavior of the GO sheets, to 2007 [111]. Radical-thiol click reactions are based on the
conductivity may be partially restored by convenient thermal photo or thermal formation of a thiol radical that attacks dou-
or chemical reduction treatment, producing chemically modi- ble or triple bounds via a typical chain process with initiation,
fied graphene sheets. propagation, and termination steps (Scheme 16.4) [112–114].
Despite the graphene sources the strategies to modify gra- Initiation involves the irradiation of a thiol mixed with a pho-
phene with click-reactive groups as well as the conditions of the toinitiator (or heating of a thiol/thermal initiator) resulting in
CuAAC click reaction itself are limited to a few. In fact, acety- the formation of a thiyl radical along with other byproducts.
lenic groups have been incorporated to the 2D laminates by Propagation is a two-step process involving first the direct
using well-known procedures such as the esterification or ami- addition of the thiyl radical to unsaturated bonds yielding an
dation of carboxylic groups at the edges with certain alcohols intermediate carbon-centered radical followed by chain trans-
or amines under typical catalytic conditions [100–103] or the fer to a second molecule of thiol to give the thiol-ene addition
coupling of reactive aryl radicals generated from arenediazo- product with the simultaneous generation of a new thiyl radi-
nium salts [67,71,84,104–106]. In the case of the azide moiety, cal. The termination step could occur by means of radical–
the most employed procedure is the typical nucleophilic attack radical coupling processes (Scheme 16.4).
with sodium azide to the epoxide groups in GO producing the This reaction is divided into thiol-ene and thiol-yne click
ring opening [107]. Meanwhile, other somewhat more complex reactions when the target is a double or triple bond, respec-
procedures have been also addressed. One involves a Hassner- tively. Both reactions meet the requirements of a click reac-
type addition of iodine azide to the double bonds at graphitic tion as they are highly efficient, simple to execute with no
edges and elimination of hydrogen iodide [108]. The edge side products, and proceed rapidly in high yield. Further, two
functionalization of pristine graphene is ideal since it improves very important features of these reactions are: (i) they can be
processability without altering its outstanding electronic prop- externally triggered by the action of light or heat and (ii) they
erties. Another approach involves two steps, that is, esterifica- do not require any catalyst. When photoinitiation is employed,
tion of 2-bromoisobutyryl bromide with hydroxyl groups in it allows spatial and temporal control and their associated
GO in triethylamine/DMF (N,N-dimethylformamide) and the benefits with regard to the ability to manipulate the reaction
Thiol-ene product
RS
R' R SH
Photoinitiator / hν
RS
or
R SH R S
Thermoinitiator /Δ R'
Thiol-yne product
R'
RS SR
R S
RS
R SH
R'
R'
Photoinitiator / hν
RS
or
R SH R S
R SH
Thermoinitiator /Δ R' RS SR
R'
R'
SCHEME 16.4 Mechanism of photo- or thermally induced free-radical thiol-ene and thiol-yne coupling.
further by controlling the reaction rate and extent simply by [115] and diazirines [116] as carbene precursors. Furthermore,
adjusting the light intensity, the dose, or the time of exposure. other cycloaddition reactions widely used on fullerene and
Further, these reactions can also be controlled on surfaces CNTs including [2 + 2] cycloadditions of aryne or benzyne
leading to patterning. intermediaries [31], [3 + 2] cycloadditions via azomethine
Both of these radical-mediated processes display fast reac- ylide [30,32,117,118] and Diels–Alder [4 + 2], whereby gra-
tion rates, chain-like reactions wherein a single radical causes phene behaves as both diene or dienophile [119–121], have
hundreds to tens of thousands of chemical reaction events to been recently reported. Furthermore, there is one example
occur, and high reaction degrees. of nucleophilic addition (in this case the Bingel reaction) of
With respect to the thiol-yne reaction the distinguishing cyclopropane onto graphene [122].
feature, when compared with the thiol-ene reaction, is the
ability for an yne-bond to react with two equivalents of thiol, 16.4 COUPLING GRAPHENE WITH
that is, form a double addition product with 1,2-regioselectiv- POLYMERS BY MEANS OF
ity. However, mechanistically, the two thiol addition reaction
CYCLOADDITION REACTIONS
steps each proceed in a manner similar to the radical thiol-ene
reaction (Scheme 16.4). Before discussing the cycloaddition reactions on graphene
However, despite its potentiality these reactions have only case by case, it is worth mentioning the general features of
been addressed in a few studies involving polymers, but in covalent coupling of graphene with polymers. When graphene
the case of carbon nanostructures (graphene, CNTs, etc.) the and certain polymers are covalently connected the classical
use of this strategy has only just started. The few examples on concept of the interface changes from a traditional view of
graphene will be discussed later. molecular interactions between components at a polymer–
filler interface (e.g., van der Waals, hydrogen bonding, halo-
gen bonding, etc.) to the concept of a single compound where
16.3.4 Other Additions
graphene forms an integral part of the polymeric chain, or
Other less common addition reactions have been addressed polymer brushes emerge from the graphene surface [13,15].
with graphene. These are listed in Table 16.3. Carbene addi- This effect is shown in Figure 16.2, where the differences in
tion to graphene has been recently reported using chloroform the interface of the polymer with pristine graphene, graphene
TABLE 16.3
Summary List of the Less Common Cycloaddition Reactions Addressed with Graphene
Graphene
Cycloaddition Reaction Reactive Species Source Experimental Remarks References
Carbene attack Dichlorocarbene r-GO Mixture of CHCl3, phase transfer catalyst [115]
(PTC) and NaOH under reflux conditions
Aryl-3-(triflforomethyl)diazirine Tr–GO Decoration of graphene with gold [116]
nanoparticles protected with a 3-aryl-3-
(trifluoromethyl)diazirine by irradiation-
induced formation of carbene
2 + 2 Arynes Graphene 2-(Trimethylsilyl)phenyl triate , CsF, in [31]
CH3CN, 45°C, 24 h
3 + 2 Azomethine ylides Graphene Condensation of paraformaldehyde with [30,118]
modified α-amino acids in the presence of
graphene colloidal dispersion, ~120–130°C
Azomethine ylides Graphene Pyridine/DMF-dispersed graphene, N-methyl- [117]
glycine, 3,4-dihydroxybenzaldehyde,
145–150°C, 96 h
Tetraphenylporphyrin, Graphene Sarcosine, porfirine, 160°C, N2 atm, 1 week [32]
Pd-tetraphenylporphyrin
Diels–Alder Tetracyanoethylene (graphene as diene) Graphene 1,4-dioxane/dichloromethane, r.t. [119,120]
2,3-dimethoxy-1,3-butadiene, or p-xylene, 50 or 130°C [119,120]
9-methylanthracene (graphene as dienophile)
Cyclopentadienes Graphene PEG, THF/H2O, pressure with AFM tip [121]
Nucleophilic addition Bingel cyclopropanation with diethylmalonate Geraphene suspensions in benzylamine, [122]
and malonate modified-tetrathiafulvalene 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU),
CBr4), microwave irradiation
functionalized with small molecules, and with polymers are coupling of graphene with polymers can be attempted either
illustrated. Owing to the fact that graphene is formed only by on pristine graphene or on modified graphene depending on
sp2 carbon atoms in a honeycomb-lattice structure, polymers the nature of the polymer as well as the synthetic procedure
are expected to interact with graphene only by π–π and van employed. When graphene is modified with functional groups
der Waals interactions. Thus, the existence of an imaginary bearing heteroatoms (oxygen, nitrogen, halogen, sulfur, etc.)
line dividing graphene from the polymer can be assumed. The the possibilities of interaction with polymers is enhanced
Small-molecules functionalized Polymer-functionalized

Pristine graphene
graphene graphene
Two-component material with Two-component material with Pure graphene-based

limited supramolecular enhnced supramolecular compound with covalent
interactions: interactions: π–π, van der Waals, linkage.
π–π, van der Waals. hydrogen bonding, halogen bonding
electrostatics, etc.
FIGURE 16.2 Drawing illustrating the differences on the nature of the interface in graphene-based nanocomposites depending on the
graphene source and the type of graphene/polymer interaction.
because such moieties have specific chemistry and are capa- the main chain. Therefore, the success of using graphene as
ble of generating other supramolecular interactions such as 2D platforms for growing polymer brushes is only governed
hydrogen bonding, halogen bonding, electrostatics, etc. In this by the advantages and drawbacks of the polymerization pro-
case, we can envisage a picture where the modifying groups cess itself in terms of the type of monomer, initiator, catalyst,
penetrate into the polymer network acting as compatibilizers. solvent, additives (if used), and experimental conditions as
This new layer formed on the graphene surface because of its pressure and temperature.
functionalization is responsible for improving the interaction The grafting-from method depends fundamentally on
with the matrix. Covalent linking between graphene and the the feasibility of immobilizing the specific initiators on
matrix results in a blurring of the interface, where a single the graphitic sheets, which may not be possible in certain
phase appears since the two materials are no longer blended, cases, where the covalent linkage between the pre-synthe-
but a unique gradient material is produced (Figure 16.2). In sized polymer and graphene emerges as the only alterna-
the case of polymer-modified graphene the product is no lon- tive, expanding the type of polymers that can be bound to
ger a two-component material but a single compound where graphene. Grafting-to methods allow the incorporation of
graphene and the polymer are covalently connected. In this graphene either at the end of the chain (terminal) as well as
case, the interface is the strongest. in the middle. In the latter case graphene forms part of the
The absence of this “interface barrier” would be expected main chain and higher influence on the final properties of
to lead to a full transfer of the properties of graphene to the the material is expected than in the case of grafting-from. In
matrix and, consequently, to achieve maximum improve- the grafting-from method usually the amount of graphene in
ments in the final properties. In other words, the strategy of the final product is low and depends on the molecular weight
covalent functionalization of graphene with polymers would as well as on the polydispersity of the polymer. Meanwhile,
allow homogeneous dispersions of graphene and adequate in grafting-to the degree of modification does not strongly
control of the microstructure of the nanocomposites to be depend on these factors since the linkage sites are distributed
achieved. Furthermore, the approach has the additional ben- along the polymeric chain. However, due to the huge size of
efit of preventing restacking of the graphitic sheets if further the graphene sheets, steric factors may determine the degree
processing is needed. of modification. While in grafting-from approaches graphene
So far, the modification of graphene with polymers has is seen as the platform from where the polymer grows, in
been addressed by two principal methodologies. The first grafting-to methods it is a microparticular pendant of the
involves the use of graphene as the initiation site to grow poly- polymer chain.
mer brushes from its surface, known as the “grafting-from” While the number of initiators that can be linked to the
method, whereas the second approach consists in combining graphitic surface in grafting-from is limited to a few, the wide
graphene and polymers through a simple chemical reaction, chemistry covered by diverse families of polymers makes
known as the “grafting-to” method. possible the development of lots of synthetic strategies to link
The former approach is based on the growth of polymer it with graphene in the case of grafting-to. Chemists have
chains from the surface of graphene, which can be considered two main tools; that is, the chemistry of polymers and the
a macro-initiator. In this strategy, initiator-modified graphene chemistry of graphene and especially its derivatives. To graft
is a 2D platform from which polymer chains are grown, and polymers to graphene it should be furnished with adequate
the method is principally related to the chemistry of gra- functional groups to react with specific polymers, or the poly-
phene with discrete organic molecules rather than chemical mer can be endowed with functionalities capable of reacting
reactions of graphene with polymeric matrices. Therefore, with the graphene derivatives.
the wide background on the chemistry of graphene, GO and
reduced GO (r-GO) [7,8,123–125] is useful to design the syn-
16.4.1 Polymer Chemistry Considerations
thetic protocol for anchoring appropriate initiators to the 2D
sheets, where the initiator molecules are fixed either to the The extension of the aforementioned reactions for simple
basal planes or at the edges of graphene by simple chemical molecules to the chemistry of polymers is not straightfor-
reactions. ward and some additional aspects must be considered, more
Regarding the real possibilities of this approach, from a so when the compound to react with the polymer is a huge
chemical standpoint it is widely viable because the reaction nano/microparticle such as graphene. To simplify, we can
of graphene laminates with simple molecules is easier and say that from the standpoint of polymer factors other than
can be better controlled than in the case of modification with those associated with simple molecules are related to poly-
polymeric macromolecules. In addition, from the point of mer configuration and conformation, supramolecular interac-
view of the type of polymerization, despite the huge dimen- tions (more important as the chain lengths increase), degree
sions of the graphene or GO laminates, in principle no steric of crystallinity, etc.
hindrance is expected by grafting-from since the graphene is
the 2D platform and polymer chains grow from it. In other
16.4.2 Nitrene Chemistry
words, as the polymer grows from graphene, and hence away
from the surface the steric effect is minimized, a very differ- In the case of polymers, the simplest case of nitrene chemis-
ent situation to that occurring when graphene forms part of try is that where the azide reactive groups are located at the
end of the polymer chains. The highly reactive species gener- produce nanohybrid materials for thermoresponsive nanode-
ated by thermolysis or irradiation of these azides give [2 + 1] vices and hydrophobic drug-delivery vehicles, respectively.
cycloaddition to π-electron systems such as graphene. The The benefits of nitrene chemistry have been also explored
nitrene radical can attack the sp2 carbon of pristine graphene for transferring graphene grown by CVD on a Cu surface
as well as that of chemically modified graphene, including to PS [128]. The procedure comprises a two-step procedure
the extensively oxidized, GO. This is an important feature of PS modification. The polymer is first modified with oxy-
because while graphene is hydrophobic, GO is water soluble, gen-containing groups (hydroxyl, carbonyl, carboxyl, etc.)
providing access to a wide range of families of polymers that by carbon dioxide plasma treatment. This modified poly-
can bind to graphene via reactions in solution. mer was treated with a PFPA derivative, N-ethylamino-4-
The first example of nitrene addition to graphene was azidotetrafluorobenzoate (TFPA), by two different approaches:
reported by He and Gao who showed the usefulness of nitrene covalent chemistry or supramolecular interactions. While the
chemistry for grafting polymers onto graphene as well as the covalent reaction was carried out by an amidation reaction
versatility of the procedure [56]. They used GO as a graphene in the presence of an N-ethyl-N′-(3-(dimethylamino)-propyl)
precursor and demonstrated that nitrene cycloaddition occurs carbodiimide (EDC)/N-hydroxysuccinimide (NHS) catalytic
simultaneously to the reduction of GO enabling the produc- system, hydrogen bonding between amine groups in TFPA
tion of large amounts of functionalized graphene (GO can be and hydroxyl and carboxylic groups in PS were spontaneously
produced in mass scale by oxidation of graphite, an abundant formed by simple mixing. Finally, in both cases the azide-
natural allotrope of carbon) and the recovery of the electrical modified PS was mixed with the graphene layer on copper
conductivity lost in the oxidation step. Furthermore, polymer and the mixture heated up to induce thermolysis of the azide.
functionalization avoids reaggregation of the graphene sheets The importance of this approach lies in the fact that the whole
assuring good solubility in different solvents. process is conducted in solid state; the copper substrate is not
With respect to the nature of the polymers, polystyrene eliminated by etching and can be reutilized. In addition, it
(PS) and poly(ethylene glycol) (PEG) have been linked to proves the efficiency of nitrene chemistry in solid-state reac-
graphene by using azido-terminated polymers. In the case of tions with graphene.
PS the authors addressed both grafting-from and grafting-to Polyethylene (PE) was recently linked to graphene oxide
approaches. In fact the authors reacted GO with an azide- by two different routes involving polymeric macroradicals
terminated PS as well as using graphene-bearing bromine [129]. While one of the routes uses a PE radical generated by
atoms—introduced by nitrene chemistry—to polymerize the action of the well-known radical inducer benzoil peroxide,
styrene by atom transfer radical polymerization (ATRP). the other one uses nitrene radicals generated by thermolysis
The combination of graphene chemistry (in this case nitrene) of the azide-terminated PE. In the former, the reaction was
with ATRP polymerization constitutes a powerful tool for the conducted in the presence of suspended GO in 1,4-dioxane/o-
preparation of new materials with better properties. ATRP is dichlorobenzene (DCB) and benzoyl peroxide at 100°C fol-
probably the most practical technique for the preparation of lowed by a subsequent solvothermal reduction at 150°C. In
functional polymers, because the terminal alkyl halide can the latter, the reaction between GO and azido-terminated
be converted to a wide variety of functionalities using con- PE was carried out at 190°C in order to ensure the thermal
ventional organic synthetic procedures [126]. However, at this cleavage of azide groups and reduction of GO. The authors
point it is worth mentioning that in ATRP the amount of the compare both methods and determine, by thermogravimetric
initiator in the process determines the final molecular weight analysis, that the grafting density was higher when the ben-
of the polymer at full monomer conversion. Therefore, deter- zoyl radical was used, but they do not provide any hypoth-
mining the number of sites on graphene to anchor the initiator eses on the reasons for these differences. Although the PE
molecules and, more importantly, the proportion that effi- grafting densities are quite low (5 × 10−3–7.6 × 10−3), this com-
ciently reacted with the initiator is fundamental. Regarding parison could be useful as it suggests that the benzoyl radi-
the properties of the polymer-modified graphene, the reduc- cal route is more efficient that the nitrene one. Furthermore,
tion of GO during nitrene cycloaddition leads to materials this reaction requires lower temperatures and the polymer is
with conductivity values between 270 and 680 S m−1 for gra- used as such and does not need previous modification steps.
phene modified with PEG and 370 S m−1 for graphene modi- Besides these differences and despite the low grafting den-
fied with PS of Mn = 2515 g mol−1. These high conductivities sities, the conductivities of the materials were significantly
are logical since although the polymer is linked to graphene low (4.9 × 10−5–3.2 × 10−4 S cm−1), mainly due to the partial
by the ends, the molecular weights of the polymers are low restoration of the sp2 network under the low-temperature con-
making the mass percentage of graphene high. ditions used for GO reduction. A much higher conductivity
This strategy involving simultaneous nitrene radical addi- has been obtained for PE brushes-modified graphene by thiol-
tion and GO reduction has also been employed to graft cyclo- radical click chemistry where graphene (not GO) was used as
dextrin (CD) from graphene [127] The CD-grafted graphene the starting material (see below).
was further functionalized with poly(N-isopropylacrylamide- Finally, the nitrene approach has also been employed to
co-vinylferrocene(II)) (PNIPAM-Fc) and ferrocene-modified connect graphene with a polyphenylacetylene derivative. In
hyperbranched polyglycerol (HPG-Fc) via host–guest inclu- this case, the azides are pendant groups distributed through-
sion complexation of CD and ferrocene (Fc) moieties to out the polymer chain rather than at the end of this [130].
TABLE 16.4
Summary List of the Different Experimental Conditions Used to Modify Graphene by
Diazonium Chemistry and Subsequent Connection to Polymers
Modifying Group Coupled to Approach for Polymer–
Graphene by Diazonium Chemistry Graphene Connection Type of Polymerization Polymer References
2-(4-aminophenyl)ethanol Grafting-from ATRP PS [80,83]
p-aminophenol Grafting-from ATRP PNIPAM [85]
2-(4-aminophenyl)ethanol Grafting-from ATRP PEMA [132]
2-(4-aminophenyl)ethanol Grafting-from ATRP PMMA [133]
Propargyl p-aminobenzoate Grafting-from RAFT PNIPAM [134]
p-aminobenzoic acid Grafting-from RAFT PAA [135]
2-(4-aminophenyl)ethanol Grafting-from SET-LRP PtBA [136]
4-bromoaniline Grafting-from Polycondensation P3HT [137]
4-bromoaniline Grafting-to Suzuki coupling PFTB [138]
2-(4-aminophenyl)ethanol Grafting-to Alcohol/isocyanate reaction PU [139,140]
16.4.3 Free-Radical Addition confined, in particular for segments near the laminate sur-
face. Actually, changes in molecular dynamics of the poly-
This type of addition reactions, mainly focused on diazonium mer due to confinement of the polymer chains by r-GO as
compounds, have not been used to directly link graphene with well as a dependence of the glass transition temperature, Tg
polymers but have been employed to provide graphene with on the grafting density were observed. Besides, improvement
groups capable of reacting with polymers or to act as polym- of Young’s modulus by 57% and tensile strength by 70% for
erization initiators. As summarized in Table 16.4, this section PS-grafted r-GO were reported, as well as improved thermal
discusses the cases where diazonium coupling to graphene conductivity in nanocomposites of PS containing PS-grafted
has been subsequently used for connecting polymers either r-GO. Finally, the confinement effect was accompanied by an
via polymerization (grafting-from) or by reaction with poly- enhancement in thermal conductivity of the final material by
mers (grafting-to). more than 2.5 times that of pure PS.
In the former case, diazonium coupling has been widely A procedure very similar to the aforementioned has been
employed to decorate 2D graphene laminates with typi- also accomplished for grafting poly(N-isopropyl acrylamide)
cal polymerization initiators, the propionyl bromide family (PNIPAM) from graphene [85] The method consisted in the
being the preferred one. The first example of polymeriza- functionalization of graphene surfaces with aminophenol
tion from graphene functionalized with this type of initiator groups by the diazonium reaction in water and the subse-
was based on the ATRP of styrene [83]. Here, graphene was quent attachment of bromoisobutyrate groups to the phenol-
furnished with 2-bromopropionyl bromide by a combination functionalized graphene (G-OH) surface by esterification of
of coupling reactive aryl radicals of the diazonium salt of 2-bromoisobutyrate with the hydroxyl groups, forming bro-
2-(4-aminophenyl)ethanol to r-GO followed by acylation moisobutyrate-functionalized graphene (G-Br). Finally, the
of hydroxyl groups with 2-bromopropionyl bromide. Then polymer was grown from G-Br via ATRP.
the styrene monomer was polymerised in the presence of Another example of polymers grafting-from initiator-
CuBr and N,N,N′,N′,N″-pentamethyldiethylenetriamine modified graphene by diazonium coupling is the case of
(PMDETA) catalytic system. This example is highly relevant poly(2-(ethyl(phenyl)amino)ethyl-methacrylate) (PEMA)
because it combines ATRP and diazonium coupling to mod- [132]. PEMA brushes grafted from r-GO have been prepared
ify graphene that allows the control of both the chain length, and subsequently modified with azo compounds to produce
a feature of ATRP, and the grafting density (the number of a light-responsive material. Here, r-GO was reacted with the
polymer chains emerging from graphene), a characteristic of diazonium salt of 2-(4-aminophenyl) ethanol through a dia-
diazonium coupling to carbon nanostructures [131]. In fact, zonium addition reaction and the pendant hydroxyls of the
in the case of graphene it has been demonstrated that the modified material were subjected to a Schötten–Baumann
grafting density and the chain length of PS were controlled reaction with 2-bromoisobutyl bromide. Subsequently, poly
by regulating the concentrations of diazonium compound and merization was conducted using a mixture of initiators
monomer during the grafting reaction of the initiator and the composed of initiator-modified r-GO and methyl bromoiso-
subsequent polymerization [80]. It has been proved that in butyrate, which provided a better control of polymerization.
samples with high graft density the distribution of grafting Finally, the light-sensitive group was anchored by coupling
polymer chains on r-GO was more uniform in comparison to with the diazonium salt 4-aminobenzonitrile. Taking advan-
those with low graft density. Moreover in these two studies tage of the improved solubility in organic solvents as well as
[80,83] it has been observed that relaxation of the polymer better dispersibility in polymers of the graphene nanosheets,
chains covalently bonded to the r-GO surface was strongly
the authors studied the effect on the diffraction efficiency improved solubility in organic solvents as well as optical-lim-
(DE) of photoinduced surface-relief gratings (SRGs) of a iting effect with a 532-nm laser beam. In addition, diazonium
star-shaped azo compound, Tr–Az–CN doped with azo- coupling has been employed to furnish graphene with click-
modified PEMA/r-GO. able groups capable to react with polymers by click reactions,
The coupling of 4-(2-hydroxyethyl)benzenediazonium as discussed in Section 16.4.4.
tetrafluoroborate to graphene has also been electrochemi- Graphene functionalized by diazonium chemistry has been
cally driven [133]. The reaction was addressed on graphene very recently used in another grafting-to approach [139,140].
electrodes having low and high quality and important find- Specifically graphene functionalized with the diazonium salt
ings were taken from this. Thus, it was concluded that for of 2-(4-aminophenyl)ethanol has been reacted with appro-
low-quality electrodes the grafting occurs directly at defect priately modified polyurethane (PU) and the functionalized
sites, while for high-quality graphene the modifying group graphene/PU composites showed good shape memory proper-
attaches to the basal plane. In addition, the diazonium com- ties, as well as mechanical, thermal, and dynamic mechanical
pounds also attach to already grafted molecules. Similar properties.
to other studies with 2-bromoisobutyl bromide initiator
was attached to the modified graphene, and in this case
16.4.4 Click Chemistry
poly(methyl methacrylate) (PMMA) was grafted from gra-
phene by ATRP. This is perhaps one of the most powerful tools in the chemist’s
Finally, other varieties of controlled/living radical polym- toolbox for creating new advanced materials because it allows
erization such as reversible addition-fragmentation chain combining graphene and polymers, two families with very dif-
transfer (RAFT) polymerization and single electron-transfer ferent but complementary properties in one step. In order to
living radical polymerization (SET-LRP) have been employed click react, graphene must be provided with clickable groups
to graft polymers from graphene modified by diazonium and several approaches using different chemical reagents, con-
chemistry. For instance, PNIPAM polymer brushes have been ditions, and/or graphene sources have been studied.
grafted from the surfaces of r-GO sheets by RAFT [134].
The RAFT initiator was linked to the r-GO surface via “click 16.4.4.1 Coupling Graphene and Polymers
chemistry,” an approach that in the case of polymers will be by CuAAC Reaction
discussed in Section 16.4.4. RAFT polymerization has been As described in Section 16.3.3.1, several synthetic protocols
also employed to grow poly(acrylic acid) (PAA) from gra- exist that can provide graphene with clickable groups for
phene surfaces [135]. In this case, the diazonium chemistry CuAAC (Table 16.2). The source of graphene used in most
was employed to modify graphene with benzoic acid moieties of these cases was GO, whose main limitation is its higher
that were then reacted with the RAFT initiator by esterifica- degree of structural defects and vacancies that disrupt the sp2
tion. The posterior RAFT polymerization in the presence of a network and dramatically worsen the electronic properties
cross-linking agent produced a graphene/PAA hydrogel with [141]. When the GO sheets are reduced by chemical [142–144]
better mechanical properties. and thermal [145] methods or by a the combination of both
In another study, poly(ter-butyl methacrylate) (PtBMA) [146], the conductivity is partially restored, but some defects
has been grafted from the surface of r-GO by SET-LRP [136]. or vacancies are irreversible and the conductivity remains
Here, the Br-containing initiator was anchored to the r-GO significantly lower than that of pristine graphene. Therefore,
sheets by treatment of hydroxyl-modified r-GO with 2-bro- using GO is clearly not the best choice if the ultimate goal is to
mopropionyl bromide. Previously, the hydroxyl groups were obtain materials with high electrical or thermal conductivity.
added by treating r-GO by diazonium coupling with 2-(4-ami- Regarding the specific reaction conditions, the CuAAC
nophenyl) ethanol and isoamyl nitrite at 80°C. The PtBMA- reaction with graphene derivatives has been addressed using
modified r-GO by SET-LRP (71.7 wt.% grafting efficiency) the same protocols already developed for the other parent car-
displayed improved dispersibility. bon nanostructures thus allowing the incorporation of differ-
A conjugated polymer, poly(3-hexylthiophene) (P3HT) has ent moieties to graphene sheets [92,147,148].
also been grown from the graphene surface [137]. In this case, A list of the polymers connected to graphene by CuAAC
graphene was reacted with the diazonium salt of 4-bromoani- reaction can be found in Table 16.5, where information on
line and the polymer was grown from brominated graphene other features including synthetic conditions and the most
by Kumada catalysis transfer polycondensation. important remarks can also be found.
Bromobenzene-modified graphene obtained by diazonium One of the most employed polymers in this type of
coupling was also used to graft another conjugated polymer reactions is PS, which was the first polymer connected to
based on fluorene, thiophene, and benzothiadazole (PFTB) graphene by CuAAC reactions [149]. Here, almost mono-
(donor–spacer–acceptor triad) to graphene [138]. In this case disperse PS (polydispersity index, PDI = 1.04) bearing a ter-
the bromine atoms react with a pre-synthesized polymer minal azide group was prepared by ATRP, with controlled
(grafting-to method). The approach involves the function- molecular weight (in this case the MW obtained was 4600
alization of graphene via diazonium coupling with phenyl g mol−1). On the other side, GO was provided with acetyle-
bromide and the posterior coupling with the polymer by nic moieties by acylation reaction with propargyl alcohol.
the Suzuki reaction. These polymer–graphene hybrids show Finally, both materials GO and PS were reacted using CuBr
TABLE 16.5
Summary List of the Different Experimental Conditions Used to Click Polymers and Graphene
Clickable Clicked
Graphene Group on Method to Provide Graphene Polymer/ Click Reaction Remarks on the Final
Source Graphene with Clickable Entities Approach Conditions Materials References
Cu(I)-Catalyzed Azide–Alkyne Cycloaddition

GO –C≡C– (i) SOCl2,70°C; (ii) Propargyl PS/graft to CuBr/PMDETA, Good dispersibility in DMF, [149]
alcohol, TEA, Cl3CH, r.t. DMF, r.t. THF, CH2Cl2, toluene
GO –N3 (i) 2-bromoisobutyryl bromide, PS/graft to CuBr/PMDETA, Good dispersibility in THF, [100]
TEA, DMF, 0°C; (ii) NaN3, r.t. DMF, r.t. DMF CHCl3
r-GO –C≡C– p-aminophenyl propargyl ether, PMMA, PMAA, CuI/DBU, DMF, r.t. Full study on the synthetic [90]
isoamyl nitrite, 80°C PS, P4VP, conditions and parameters
PDMA/graft to
and graft from
GO –C≡C– (i) SOCl2, 70°C, (ii) propargyl SEBS/graft to CuBr/PMDETA, Excellent compatibility with PS. [101]
alcohol, TEA, CHCl3, r.t. DMF, r.t. Improved mechanical
properties and thermal stability
of nanocomposites
GO –C≡C– Propargylamine, DCC/DMAP, PNIPAM/graft to CuBr/PMDETA/ Potential biomedical: drug [102]
DMF, r.t. sodium l-ascorbate, delivery and killing cancer cell
DMF, r.t.
r-GO –C≡C– (i) Propargyl p-aminobenzoate, PNIPAM/graft CuBr/PMDETA, Water solubility, changes in [134]
NO2BF4 ,CH3CN, −30°C, (ii) from DMF, r.t. lower critical solution
r-GO, r.t. temperature
Pristine –C≡C– Propargyloxybenzenediazonium PEG/graft to Sodium bicarbonate/ Higher reactivity on CVD-grown [67]
graphene Tetrafluoroborate, water, 45°C CuSO4/THPTA/ samples. Edges and defects are
sodium ascorbate preferential sites of reactions
Graphene –C≡C– 4-ethynylaniline, isoamyl nitrite, PE/graft to CuI/DBU, DMF, Filler for PE matrices [151]
80°C 60°C
GO –C≡C– Propargylamine, DCC/DMAP, PCL/graft to CuBr/PMDETA, Enhancement of the mechanical [152]
DMF DMF, r.t. properties, thermal stability,
thermal conductivity, and
thermoresponsive
shape memory properties
r-GO –C≡C– Route I: (i) SOCl2, o-DCB, (ii) PFA/graft to CuBr/PMDETA, Improved solubility. Solvent- [84]
propargylamine, DCC/DMAP, DMF, r.t. dependent electronic properties
o-DCB, r.t.
Route II: 4-ethynylaniline,
isoamyl nitrite, 80°C
GO –N3 APTMS, DMF P3HT/graft to CuI/DIPEA, DMF Red-shifted optical absorption [153]
maximum
GO –C≡C– (i) SOCl2, DMF, 70°C; (ii) PMPCS/graft to CuBr/PMDETA, Improved mechanical properties [156]
propargyl alcohol, TEA, 60°C DMF, 90°C
GO –C≡C– Ethynylaniline, DCC, DMF, PEDOT:PSS/graft CuSO4, sodium Good electrical conductivity [155]
90°C, N2, 2 h to ascorbate, r.t. 48 h
Thiol-Radical Click Reactions
r-GO –C≡C– 4-Ethynylaniline, isoamyl PE/graft to AIBN, o-DCB, 70°C Better thermal, mechanical, and [151,157]
–C≡C– nitrite, 60°C electrical properties
r-GO –C≡C– – PMMA/graft to (AIBN), sodium Better electrical conductivity [158]
methoxide, Cl2CH2,
60°C, inert
atmosphere
and N,N,N′,N″,N″ -PMDETA as the catalyst in DMF, giv- for preparing well-defined nanomaterials with desired func-
ing products with good dispersibility and full exfoliation in tionalities [150]. Here ARTP has two important advantages:
common organic solvents. It is very important to note that First, the terminal alkyl halide can be easily converted to
this study combines click chemistry and a type of controlled azide moiety for click reactions and second, the amount of
radical polymerization (CRP) that constitutes a powerful tool initiator in the process determines the final molecular weight
of the polymer at full monomer conversion allowing for the host polymer, and observed remarkable improvements in the
control of the chain length. mechanical properties and thermal stability of the resulting
In another example including PS, the roles were inverted and composite films.
the PS derivative carrying alkyne functionalities was clicked A thermoresponsive, PNIPAM has also been clicked
to azide-modified GO [100]. GO with azide groups involves to modified GO [102]. In this study, the azide-terminated
acylation of hydroxyl groups on GO with 2-bromoisobutyryl PNIPAM was clicked with alkyne-modified GO obtained by
bromide followed by nucleophilic substitution of the pendant amidation reaction with propargylamine. The most outstand-
bromine with NaN3. Then acetylene-terminated PS was pre- ing finding of this work is related to the potential biomedical
pared by ATRP of styrene using propargyl 2-bromoisobutyr- applications of the graphene–polymer conjugate, which can
ate as the initiator. The click reaction was conducted under be seen as an amphiphilic material composed of a hydropho-
the same conditions as in the earlier example and the products bic graphene core and hydrophilic PNIPAM arms. That is,
displayed the same solubility as before. the hydrophobic graphene absorbs the aromatic water-insol-
PS and other polymers such as PMMA, poly(methacrylic uble drug, while the water-soluble PNIPAN transports it into
acid) (PMAA), poly(4-vinylpyridine) (P4VP), and the cells. The graphene–PNIPAM conjugate was used as an
poly((dimethylamino)ethyl methacrylate) (PDMA) were effective vehicle to internalize into the cells a water-insolu-
grown from or attached to graphene by combining RAFT ble anticancer drug, camptothecin (CPT). Interestingly, the
polymerization and click chemistry [90]. As mentioned pre- graphene–PNIPAM conjugate was able to load CPT with a
viously, grafting-from involves clicking the suitable initia- superior loading capacity of 15.6 wt.%, while not exhibiting
tor by click reaction with the alkyne-modified graphene and significant toxicity. Thus, the PNIPAM–graphene–CPT com-
subsequent polymerization, whereas grafting-to comprises plex appears as an excellent system with a high potency of
the click reaction between alkyne-modified graphene with killing A-5RT3 cancer cells in vitro.
an azide-terminated polymer. In this very detailed study both Brushes of the same polymer, PNIPAM have also been
grafting methods were considered separately. Regarding the grafted from the surface of r-GO by combining CuAAC reac-
grafting-to approach, PMMA was used as a model to deter- tion and RAFT polymerization [134]. The r-GO was reacted
mine the optimal synthesis conditions (time, catalytic system, with propargyl p-aminobenzoate by diazonium coupling.
reagent ratios, etc) as shown in Table 16.5. For both methods Then the acetylene-modified compound was clicked to the
the preferred click conditions employed CuI and 1,8-diaz- RAFT agent S-1-dodecyl-S’-(α,α′-dimethyl-α″-3- azido-
abicyclo[5.4.0] undec-7-ene (DBU). The effect of the chain 1-propyl acetate) trithiocarbonate followed by NIPAM RAFT
length was also studied and the authors concluded that the polymerization with azobisisobutyronitrile (AIBN) at 60°C.
higher the molecular weight the lower the grafting density. Another water-soluble polymer PEG, which is one of the
This result was explained by a decrease in polymer reactiv- most employed polymers for biomedical applications for cell
ity of the terminal azide due to increased hindrance as the internalization has been also clicked to graphene [67]. The
polymer adopts a more random coil structure. In addition, the click reaction was studied on bulk graphene dispersions as
decrease in grafting density results in a decrease in solubil- well as on CVD-grown graphene. Graphene was furnished
ity. In other words, although long-chain polymer arms emerg- with alkyne groups by diazotization in the presence of sodium
ing from graphene are better for improving solubility, the low dodecylsulfate (SDS) surfactant and the CuAAC reaction with
degree of functionalization (low grafting density) achieved is azide-terminated short-chain PEG was conducted in the pres-
detrimental to solubility, the latter having the greatest influence of sodium bicarbonate, CuSO4, tris(3-hydroxypropyltri-
ence on this parameter. This issue can be resolved using a azolylmethyl)amine (THPTA), and sodium ascorbate. It was
grafting-from protocol, where the grafting density can be tai- demonstrated that the CVD–graphene is more reactive than
lored by controlling the concentration of the initiator linked solution-dispersed graphene and that edges and defects are
to the graphene surface. Once the desired grafting density is preferentially reacted. In addition this is one of the few where
achieved the polymer molecular weight can be increased by neither GO nor r-GO are used as the graphene source, but
using higher polymerization times and/or varying the mono- highly ordered pyrolytic graphite (HOPG) and CVD growth
mer/initiator ratio, which results in better solubility. The only graphene were used.
problem of this procedure is the broadening of the polydisper- Very recently the most popular thermoplastic polymer PE
sity index (PDI). has been click-linked to graphene by several click chemistry
Elastomeric polymers such as poly(styrene-b-ethylene-co- approaches. In the specific case of CuAAC reaction short-
butylene-b-styrene) (SEBS) triblock copolymers have also chain PE brushes were coupled to alkyne-modified graphene
been grafted to GO employing the same synthetic protocols prepared using CuI/DBU catalyst in DMF at 60°C [151]. The
as in the earlier examples (CuBr/PMDETA) [101]. The SEBS PE brushes were also linked to graphene by thiol-radical click
copolymer carried the azide groups in the styrene segments reactions such as thiol-ene and thiol-yne. A more detailed dis-
while the acetylene moieties were in the GO. The product cussion will be given in Section 16.4.4.2.
was soluble in a variety of organic solvents that allowed Another polymer clicked to graphene is poly
their homogeneous incorporation into polymer matrices. The (ε-caprolactone) (PCL) [152]. The alkynyl-modified GO
authors investigated the compatibility and reinforcing effect obtained by esterification of GO with propargylamine was
of the clicked material in polymer composites, using PS as a clicked with PCL bearing side azide groups under typical
synthetic conditions (Table 16.5). The click products displayed weight remained amorphous after being heated, while the
good interaction with a PU matrix, leading to remarkable larger molecular-weight PMPCS could go through a thermo-
improvements in the mechanical properties, thermal stability, tropic liquid crystalline transition on heating. In agreement
thermal conductivity, and thermoresponsive shape memory with the results reported in an aforementioned work [90], this
properties of the resulting nanocomposites. For instance, the study shows that the lower the molecular weight the higher
breaking stress, Young’s modulus, elongation at break, and the grafting density. It was demonstrated by Raman and XPS
thermal stability of nanocomposites containing 2% of the (x-ray photoelectron spectroscopy) measurements that the
click material displayed improvements of 109%, 158%, 28%, higher molecular-weight sample (lower grafting density) had
and 71°C, respectively. better interactions with the GO sheets due to the increased
The CuAAC protocol has also been used to prepare aromaticity that results in a higher density of π–π interactions
hybrids of graphene and conjugated polymers [84,153–155]. (more intimate contact between PMPCS and GO). However,
In the first example, an azide-polyfluorene derivative was when using the click product as a nano-filer for a neat PMPCS
clicked to alkynyl-modified r-GO under typical catalytic matrix, in solution blending the rheological measurements
conditions [84]. The r-GO functionalization was addressed showed that the sample with higher grafting density had a
by using two protocols, that is, amidation and diazotiza- better reinforcing effect due to improved compatibility with
tion, the latter being the most efficient. The clicked products the PMPCS matrix.
were soluble in organic solvents such as DMF, N-methyl-2-
pyrrolidone (NMP), and o-DCB. However striking differ- 16.4.4.2 Coupling Graphene and Polymers
ences were observed in the absorption and emission spectra by Thiol-Radical Click Reactions
of the graphene–polyfluorene conjugate depending on the Despite the advantages of thiol-radical click reactions and its
solvent, due to the different effects on the electronic behav- demonstrated utility in polymer chemistry [113], the interest
ior of polyfluorene by the presence of the electron-acceptor of applications in CNTs and graphene has only just started.
graphene. While the photoluminescence of polyfluorene was The first and only study on this topic so far involves the cova-
slightly affected in NMP and o-DCB, it was totally inhibited lent modification of graphene with PE [157], perhaps one of
in DMF. These changes were not related to the polarity of the polymers that has demonstrated most difficulties when
the solvents (for instance, DMF and NMP has similar polari- preparing graphene/PE nanocomposites. This study used
ties) but to their affinity by graphene in terms of surface a short-chain PE linked to pristine graphene by a thiol-ene
energy. reaction and to alkyne-modified graphene by thiol-yne and
Poly(3-hexylthiophene) brushes have been attached to CuAAC click reactions. For thiol-ene and thiol-yne reactions
GO surfaces [153]. The GO was first furnished with azide the SH-terminated PE was reacted with graphene and alkyne-
groups by a silylation reaction with 3-azidopropyltrime- modified graphene, respectively in the presence of AIBN ther-
thoxysilane (APTMS) and the click reaction with the ethy- mal initiator at 70°C. The case of PE is one of the best examples
nyl-terminated P3HT carried out in the presence of CuI and to highlight the importance of the interface in graphene-based
N,N-diisopropylethylamine (DIPEA) in DMF. The clicked polymer nanocomposites. As PE has a somewhat low melting
products were compared with materials prepared by the ami- point it offers the possibility for rapid sample preparation of
dation reaction between GO and P3HT; the grafting efficiency nanocomposites by melt blending. However, melt mixing does
of the click approach is much higher because of the greater not assure that an efficient polymer/filler interface is obtained
quantity of hydroxyl and epoxide groups than the carboxylic and the final properties of the nanocomposites are quite simi-
acid groups on GO sheets. In addition, the electronic prop- lar to those of the neat polymer. Therefore, other strategies
erties changed with respect to P3HT due to some ordering/ oriented to tailor the interface properties are needed. So, the
crowding of the P3HT grafted on GO. The most famous strategy of functionalizing graphene with short-chain poly-
polythiophene, poly(3,4-ethylenedioxythiophene):polystyr mers and their subsequent use in a masterbatch as a filler for a
ene sulfonate (PEDOT:PSS) has been also clicked to r-GO higher molecular-weight polymer can be extended to the case
[155]. Terminal alkyne-modified r-GO sheets were prepared of PE. Using this masterbatch approach, it has been noted that
by amidation of GO with 4-ethynylaniline followed by chem- improvements in the final electrical, mechanical, and thermal
ical reduction. The click reaction with azide-functionalized properties of high-density polyethylene (HDPE) depend on
PEDOT:PSS (PEDOT-N3:PSS) was carried out in aqueous the synthetic approach addressed to furnish graphene with PE
solution of CuSO4 and sodium ascorbate using the PEDOT- brushes. For instance, significant improvements on the elec-
N3:PSS aqueous dispersion as the solvent. trical properties were achieved only for graphene modified
In the last example of this type of click reaction a thermo- by thiol-ene chemistry. This is reasonable since the graphene
tropic liquid crystalline polymer has been linked to graphene source is graphene itself and no new defects are generated
[156] The azido-functionalized liquid crystalline poly- by premodification steps. However, the mechanical proper-
mer, poly{2,5-bis[(4-methoxyphenyl)oxycarbonyl] styrene} ties varied in a different manner. For instance, the elongation
(PMPCS), prepared by ATRP was coupled to the alkyne-mod- at break of all clicked-reinforced samples is slightly higher
ified GO using the common catalyst (CuBr/PMEDTA) under than that of the neat polymer, the value of the material pre-
higher than usual temperature (90°C). Two different molecu- pared by thiol-ene being the highest. In addition, the Young’s
lar-weight PMPCS were utilized; that with lower molecular modulus of the samples using thiol-yne and CuAAC-modified
graphene is remarkably lower than that for the neat polymer, general, versatile, and very promising for the modification of
while that for the thiol-ene modified graphene is similar. graphene with polymers.
The click thiol-ene reaction has been extended to acrylic
polymers and copolymers [158]. As against the case of
16.5 CONCLUSIONS
PMMA homopolymer having a thiol end-group, in the case
of the copolymers the thiol moieties were located along the A series of addition reactions leading to covalent function-
polymer chain, allowing for multiple bonds to the graphene alization of graphene with polymers have been discussed.
surface. This comparative study shows that the copolymers Although in this chapter the covalent modification of gra-
form a thin layer covering the graphene surface, contrary to phene has been analyzed from a perspective based on the
the case for typical polymer brushes obtained for thiol-ter- compatibilization of graphene with polymeric matrices, these
minated PMMA. The modification of graphene with polymer reactions are of paramount importance for other applications
brushes renders it easier to disperse than in the case of the including opening bandgap for electronic applications, spe-
copolymer. However, when thin layer-modified graphene is cific modification for sensing or biosensing, etc. With respect
used as a filler of pure PMMA, the PMMA/graphene inter- to the specifics of each reaction studied, we can say that in the
face is much stronger leading to nanocomposites with higher case of nitrene addition high temperature is required, while
conductivity than in the case of brush-modified graphene. the free-radical addition of diazonium compounds and the
click reactions proceed under mild conditions. In these lat-
ter cases, although there is still no general synthetic protocol,
16.4.5 Other Less Common Addition Reactions
there are optimized reaction conditions which are more suitable
Another class of methods for coupling polymers to graphene for the different cases. A particular characteristic of click reac-
laminates involves the grafting of growing radical polymers tions is their enormous potentiality for providing graphene with
to the graphene surface after a certain stage of polymeriza- functional groups or polymers of very different nature. Among
tion, where the coupling to graphene can be envisaged as a these, in the case of graphene the thiol-radical, thiol-ene, and
type of termination step [159–161]. Although this approach thiol-yne click reactions are in their initial stages. However, the
does not comprise a specific chemical reaction between gra- potential in this field has recently been demonstrated and will
phene and a pre-synthesized polymer, it can probably be clas- undoubtedly be a very powerful tool to expand the field of
sified as grafting-to because the polymer does not grow from chemistry of graphene.
the surface of graphene but is attached to it after reaching a
certain molar mass. However, the approach has some features
of grafting-from methods since the graphene is located at the ACKNOWLEDGMENT
end of the polymer chain. By means of this strategy, polymer The author acknowledges the Spanish government for fund-
brushes are attached to the flat 2D graphene surface, in anal- ing (MINECO; project MAT2013-47898-C2-2-R) as well as
ogy to work done with CNTs [162,163]. Importantly, neither for a Ramon y Cajal Senior Research Fellowship.
graphene nor the polymers need to be previously modified.
Using this approach, several families of polymers have been
attached to graphene including acrylates (poly(methyl acry- REFERENCES
late), poly(hydroxyethyl acrylate) and poly(acrylic acid)) [160],
methacrylates (PMMA, poly(butyl methacrylate, poly(t- 1. Novoselov, K. S., Falko, V. I., Colombo, L., Gellert, P.R.,
Schwab, M.G., and Kim, K. 2013. A roadmap for graphene.
butyl acrylate, poly(hydroxyethyl methacrylate), poly(2- Nature 490:192–200.
(dimethylamino)ethyl methacrylate), and poly(methacrylic 2. Luo, J., Kim, J., and Huang, J. 2013. Material processing of
acid (MAA)[160], styrenics (PS and poly(sodium p-styrene- chemically modified graphene: Some challenges and solu-
sulfonate) [160], acrylamides (polyacrylamide (PAM) and tions. Acc. Chem. Res. 46:2225–34.
PNIPAM) [160], P4VP [160], nylon 6 [159], and PS–PAM 3. Wassei, J. and Kaner, R.B. 2013. Oh, the places you’ll go with
copolymer [161]. However, certain aspects related to the graphene. Acc. Chem. Res. 46:2244–53.
polymerization mechanism remain unclear and further stud- 4. Koehler, F.M. and Stark, W.J. 2013. Organic synthesis on gra-
phene. Acc. Chem. Res. 46:2297–306.
ies are required to fully understand the process. The authors 5. James, D.K. and Tour, J.M. 2013. Graphene: Powder, flakes,
stated that during propagation macromolecular radicals were ribbons, and sheets. Acc. Chem. Res. 46:2307–18.
formed, some of which add to the sp2 framework of GO pro- 6. Ciesielski, A. and Samori, P. 2014. Graphene via sonica-
ducing brushes and generating new radicals capable of further tion assisted liquid-phase exfoliation. Chem. Soc. Rev.
reaction. Therefore, the 2D sheets have many reactive sites 43:381–98.
onto which the brushes can bind. However, the mechanism 7. Englert, J.M., Dotzer, C., Yang, G. et al. 2011. Covalent bulk
by which these macroradicals bind to the GO sheets remains functionalization of graphene. Nat. Chem. 3:279–86.
8. Loh, K.P., Bao, Q., Ang, P.K., and Yang, J. 2010. The chemis-
uncertain and thus far has only been described statistically. It
try of graphene. J. Mater. Chem. 20:2277–89.
has been also suggested that the grafting density as well as the 9. Georgakilas, V., Otyepka, M., Bourlinos, A.B. et al. 2012.
chain length can be tuned by controlling the reaction time and Functionalization of graphene: Covalent and non-cova-
conversion. However, although much effort is still required to lent approaches, derivatives and applications. Chem. Rev.
clarify the mechanism, this in situ polymerization method is 112:6156–214.
10. Malig, J., Jux, N., and Guldi, D.M. 2013. Toward multifunc- 31. Zhong, X., Jin, J., Li, S. et al. 2010. Aryne cycloaddition:
tional wet chemically functionalized graphene—Integration Highly efficient chemical modification of graphene. Chem.
of oligomeric, molecular and particulate building blocks Commun. 46:7340–42.
that reveal photoactivity and redox activity. Acc. Chem. Res. 32. Zhang, X., Hou, L., Cnossen, A. et al. 2011. One-pot function-
46:53–64. alization of graphene with porphyrin through cycloaddition
11. Park, J., Jo, S.B., Yu, Y.J. et al. 2012. Single-gate bandgap reactions. Chem. Eur. J. 17:8957–64.
opening of bilayer graphene by dual molecular doping. Adv. 33. Meyer, J.C., Geim, A.K., Katsnelson, M.I., Novoselov, K.S.,
Mater. 24:407–11. Booth, T.J., and Roth, S. 2007. The structure of suspended
12. Lee, W.H., Suk, J.W., Chou, H. et al. 2012. Selective-area fluo- graphene sheets. Nature 446:60–3.
rination of graphene with fluoropolymer and laser irradiation. 34. Ryu, S., Han, M.Y., Maultzsch, J. et al. 2008. Reversible basal
Nano Lett. 12:2374–8. plane hydrogenation of graphene. Nano Lett. 8:4597–602.
13. Salavagione, H.J., Martínez, G., and Ellis, G. 2011. Recent 35. Fasolino, A., Los, J.H., and Katsnelson, M.I. 2007. Intrinsic
advances in the covalent modification of graphene with poly- ripples in graphene. Nat. Mater. 6:858–61.
mers. Macromol. Rapid Commun. 32:1771–89. 36. Lin, T., Zhang, W.D., Huang, J., and He, C. 2005. A DFT study
14. Kuilla, T., Bhadra, S., Yao, D., Kim, N.H., Bose, S., and Lee, of the amination of fullerenes and carbon nanotubes: Reactivity
J.H. 2010. Recent advances in graphene based polymer com- and curvature. J. Phys. Chem. B 109:13755–60.
posites. Prog. Polym. Sci. 35:1350–75. 37. Prato, M., Li, Q.C., Wudl, F., and Lucchini, V. 1993. Addition
15. Salavagione, H.J. 2013. Innovative strategies to incorporate of azides to fullerene C60: Synthesis of azafulleroids. J. Am.
graphene in polymer matrices: Advantages and drawbacks Chem. Soc. 115:1148–50.
from an applications viewpoint. In Innovative Graphene 38. Nakahodo, T., Okada, M., Morita, H. et al. 2008.
Technologies: Developments and Characterisation. Ed. A. [2 + 1] Cycloaddition of nitrene onto C60 revisited:
Tiwari, pp. 175–221. Shawbury, UK: Smithers Rapra. Interconversion between an aziridinofullerene and an azaful-
16. Potts, J.R., Dreyer, D.R., Bielawski, C.W., and Ruoff, R.S. leroid. Angew. Chem., Int. Ed. 47:1298–1300.
2011. Graphene-based polymer nanocomposites. Polymer. 39. Gao, C., He, H., Zhou, L., Zheng, X., and Zhang, Y. 2008.
52:5–25. Scalable functional group engineering of carbon nanotubes
17. Niyogi, S., Hamon, M.A., Hu, H. et al. 2002. Chemistry of by improved one-step nitrene chemistry. Chem. Mater.
single-walled carbon nanotubes. Acc. Chem. Res. 35:1105–13. 21:360–70.
18. Hirsch, A. and Brettreich, M. 2005. Fullerenes: Chemistry 40. Tasis, D., Tagmatarchis, N., Bianco, A., and Prato, M. 2006.
and Reactions, Weinheim: Wiley-VCH. Chemistry of carbon nanotubes. Chem. Rev. 106:1105–136.
19. Peng, X. and Wong, S.S. 2009. Functional covalent chemistry 41. Holzinger, M., Vostrowsky, O., Hirsch, A. et al. 2001. Sidewall
of carbon nanotube surfaces. Adv. Mater. 21:625–42. functionalization of carbon nanotubes. Angew. Chem., Int.
20. Chen, Z., Thiel, W., and Hirsch, A. 2003. Reactivity of the con- Ed. 40:4002–5.
vex and concave surfaces of single-walled carbon nanotubes 42. Holzinger, M., Abraha, J., Whelan, P. et al. 2003.
(SWCNT) towards addition reactions: Dependence on the car- Functionalization of single-walled carbon nanotubes with (R-)
bon-atom pyramidalization. Chem. Phys. Chem. 4:93–7. oxycarbonyl nitrenes. J. Am. Chem. Soc. 125:8566–80.
21. Banerjee, S., Hemraj-Benny, T., and Wong, S.S. 2005. 43. Holzinger, M., Steinmetz, J., Samaille, D. et al. 2004. [2 + 1]
Covalent surface chemistry of single-walled carbon nano- cycloaddition for cross-linking SWCNTs. Carbon 42:941–7.
tubes. Adv. Mater. 17:17–29. 44. Zhou, L., Gao, C., Zhu, D. et al. 2009. Facile functionalization
22. Novoselov, K.S., Geim, A.K., Morozov, S.V. et al. 2004. of multilayer fullerenes (carbon nano-onions) by nitrene chem-
Electric field effect in atomically thin carbon films. Science istry and “grafting from” strategy. Chem. Eur. J. 15:1389–96.
306:666–9. 45. Choi, J., Kim, K.J., Kim, B., Lee, H., and Kim, S. 2009.
23. Wei, D., Liu, Y., Wang, Y., Zhang, H., Huang, L., and Yu, Covalent functionalization of epitaxial graphene by azidotri-
G. 2009. Synthesis of N-doped graphene by chemical vapor methylsilane. J. Phys. Chem. C 113:9433–5.
deposition and its electrical properties. Nano Lett. 9:1752–58. 46. Liu, L.H., Lerner, M.M., and Yan, M. 2010. Derivitization of
24. Rao, C.N.R., Sood, A.K., Subrahmanyam, K.S., and pristine graphene with well-defined chemical functionalities.
Govindaraj, A. 2009. Graphene: The new two-dimensional Nano Lett. 10:3754–6.
nanomaterial. Angew. Chem. Int. Ed. 48:7752–77. 47. Liu, L.H., Zorn, G., Castner, D.G., Solanki, R., Lerner, M.M.,
25. Hummers, W.S. and Offeman, R.E. 1958. Preparation of gra- and Yan, M. 2010. A simple and scalable route to wafer-size
phitic oxide. J. Am. Chem. Soc. 80:1339. patterned graphene. J. Mater. Chem. 20:5041–6.
26. Li, D. and Kaner, R.B. 2008. Graphene-based materials. 48. Strom, T.A., Dillon, E.P., Hamilton, C.E., and Barron, A.R.
Science 320:1170–1. 2010. Nitrene addition to exfoliated graphene: A one-step
27. Chua, C.K. and Pumera, M, 2014. Chemical reduction of gra- route to highly functionalized graphene. Chem. Commun.
phene oxide: A synthetic chemistry viewpoint. Chem. Soc. 46:4097–9.
Rev. 43:291–312. 49. Xu, X., Lv, W., Huang, J. et al. 2012. Functionalization of gra-
28. Yan, L., Zheng, Y.B., Zhao, F. et al. 2012. Chemistry and phene by tetraphenylethylene using nitrene chemistry. RSC
physics of a single atomic layer: Strategies and challenges for Advances 2:7042–7.
functionalization of graphene and graphene-based materials. 50. Snell, K.E. Ismaili, H. and Workentin, M.S. 2012.
Chem. Soc. Rev. 41:97–114. Photoactivated nitrene chemistry to prepare gold nanopar-
29. Jiang, D., Sumpter, B.G., and Dai, S. 2007. Unique chemical ticle hybrids with carbonaceous materials. ChemPhysChem.
reactivity of a graphene nanoribbon’s zigzag edge. J. Chem. 13:3185–93.
Phys. 126:134701. 51. Yan, M., Cai, S.X., and Keana, J.F.W. 1994. Photochemical
30. Quintana, M., Spyrou, K., Grzelczak, M., Browne, W.R., and thermal-reactions of C-60 with N-succinimidyl 4-azido-
Rudolf, P., and Prato, M. 2010. Functionalization of graphene 2,3,5,6-tetrafluorobenzoate—A new method for functional-
via 1,3-dipolar cycloaddition. ACS Nano 4:3527–33. ization of C-60. J. Org. Chem. 59:5951–4.
52. Pastine, S.J., Okawa, D., Kessler, B. et al. 2008. A facile and 72. Sinitskii, A., Dimiev, A., Corley, D.A., Fursina, A.A.,
patternable method for the surface modification of carbon Kosynkin, D.V., and Tour, J.M. 2010. Kinetics of diazonium
nanotube forests using perfluoroarylazides. J. Am. Chem. Soc. functionalization of chemically converted graphene nanorib-
130:4238–9. bons. ACS Nano 4:1949–54.
53. Liu, L.H. and Yan, M. 2009. Simple method for the covalent 73. Jin, Z., Lomeda, J.R., Price, B.K., Lu, W., Zhu, Y., and Tour,
immobilization of graphene. Nano Lett. 9:3375–8. J.M. 2009. Mechanically assisted exfoliation and functional-
54. Liu, L.H. and Yan, M. 2010. Perfluorophenyl azides: New ization of thermally converted graphene sheets. Chem. Mater.
applications in surface functionalization and nanomaterial 21: 3045–7.
synthesis. Acc. Chem. Res. 43:1434–43. 74. Kamada, S., Nomoto, H., Fukuda, K., Fukawa, T., Shirai, H.,
55. Wu, L., Zhang, B., Lu, H., and Liu, C.Y. 2014. Nanoscale and Kimura, M. 2011. Noncovalent wrapping of chemically
ionic materials based on hydroxylfunctionalized graphene. modified graphene with π-conjugated disk-like molecules.
J. Mater. Chem. A. 2:1409–17. Colloid Polym. Sci. 289:925–32.
56. He, H. and Gao, C. 2010. General approach to individually dis- 75. Lu, Y., Jiang, Y., Wei, W. et al. 2012. Novel blue light emitting
persed, highly soluble, and conductive graphene nanosheets graphene oxide nanosheets fabricated by surface functional-
functionalized by nitrene chemistry. Chem. Mater. 22:64. ization. J. Mater. Chem. 22:2929–34.
57. Liu, H., Ryu, S., Chen, Z., Steigerwald, M.L., Nuckolls, C., 76. Koehler, F.M., Jacobsen, A., Ensslin, K., Stampfer, C., and
and Brus, L.E. 2009. Photochemical reactivity of graphene. Stark, W.J. 2010. Selective chemical modification of graphene
J. Am. Chem. Soc. 131:17099–101. surfaces: Distinction between single- and bilayer graphene.
58. Hamilton, C.E., Lomeda, J.R., Sun, Z., Tour, J.M., and Barron, Small 6:1125–30.
A.R. 2010. Radical addition of perfluorinated alkyl iodides to 77. Coskun, E., Zaragoza-Contreras, E.A., and Salavagione, H.J.
multi-layered graphene and single-walled carbon nanotubes. 2012. Synthesis of sulfonated graphene/polyaniline compos-
Nano Res. 3:138–45. ites with improved electroactivity. Carbon, 50:2235–43.
59. Ma, X., Li, F., Wang, Y., and Hu, A. 2012. Functionalization of 78. Liu, J., Wang, R., Cui, L. et al. 2012. Using molecular level
pristine graphene with conjugated polymers through diradical modification to tune the conductivity of graphene papers.
addition and propagation. Chem. Asian J. 7:2547–50. J. Phys. Chem. C 116:17939−46.
60. Sarkar, S., Bekyarova, E., and Haddon, R.C. 2012. Reversible 79. Li, Q., Dong, L., Sun, F., Huang, J., Xie, H., and Xiong, C.
grafting of α-naphthylmethyl radicals to epitaxial graphene. 2012. Self-unfolded graphene sheets. Chem. Eur. J. 18:7055–9.
Angew. Chem. 124:4985–8. 80. Fang, M., Wang, K., Lu, H., Yang Y., and Nutt, S. 2010. Single-
61. Pinson, J. and Podvorica, F. 2005. Attachment of organic lay- layer graphene nanosheets with controlled grafting of polymer
ers to conductive or semiconductive surfaces by reduction of chains. J. Mater. Chem. 20:1982–92.
diazonium salts. Chem. Soc. Rev. 34:429–39. 81. Sun, Z., Kohama, S., Zhang, X., Lomeda, J.R., and Tour, J.M.
62. Mahouche-Chergui, S., Gam-Derouich, S., Mangeney, C., and 2010. Soluble graphene through edge-selective functionaliza-
Chehimi, M.M. 2010. Aryl diazonium salts: A new class of tion. Nano Res. 3:117–25.
coupling agents for bonding polymers, biomacromolecules 82. Dong, X., Long, Q., Wei, A. et al. 2012. The electrical prop-
and nanoparticles to surfaces. Chem. Soc. Rev. 40:4143–66. erties of graphene modified by bromophenyl groups derived
63. Huang, P., Jing, L., Zhu, H., and Gao, X. 2013. Diazonium from a diazonium compound. Carbon, 50:1517–22.
functionalized graphene: Microstructure, electric, and mag- 83. Fang, M., Wang, K., Lu, H., Yang, Y., and Nutt, S. 2009. Covalent
netic properties. Acc. Chem. Res. 46:43–52. polymer functionalization of graphene nanosheets and mechani-
64. Paulus, G.L.C., Wang, Q.H., and Strano, M.S. 2013. Covalent cal properties of composites. J. Mater. Chem. 19:7098–105.
electron transfer chemistry of graphene with diazonium salts. 84. Castelaín, M., Martínez, G., Merino, P. et al. 2012. Graphene
Acc. Chem. Res. 46:160–70. functionalisation with a conjugated poly(fluorene) by click
65. Bekyarova, E., Itkis, M.E., Ramesh, P. et al. 2009. Chemical coupling: Striking electronic properties in solution. Chem.
modification of epitaxial graphene: Spontaneous grafting of Eur. J. 18:4965–73.
aryl groups. J. Am. Chem. Soc. 131:1336–7. 85. Ren, L., Huang, S., Zhang, C., Wang, R., Tjiu, W.W., and
66. Zhu, Y., Higginbotham, A.L., and Tour, J.M. 2009. Covalent Liu, T. 2012. Functionalization of graphene and grafting of
functionalization of surfactant-wrapped graphene nanorib- temperature-responsive surfaces from graphene by ATRP “on
bons. Chem. Mater. 21:5284–91. water”. J. Nanopart. Res. 14:940.
67. Jin, Z., McNicholas, T.P., Shih, C.J. et al. 2011. Click chemis- 86. Debguptaa, J. and Pillai, V.K. 2013. Thiolated graphene—A
try on solution-dispersed graphene and monolayer CVD gra- new platform for anchoring CdSe quantum dots for hybrid
phene. Chem. Mater. 23:3362–70. heterostructures. Nanoscale 5:3615–19.
68. Wei, G., Yan, M., Dong, R. et al. 2012. Covalent modification 87. Jayasundara, D.R., Cullen, R.J., and Colavita, P.E. 2013. In situ
of reduced graphene oxide by means of diazonium chemistry and real time characterization of spontaneous grafting of aryl-
and use as a drug-delivery system. Chem. Eur. J. 18:14708–16. diazonium salts at carbon surfaces. Chem. Mater. 25:1144–52.
69. Lomeda, J.R., Doyle, C.D., Kosynkin, D.V., Hwang, W.F., and 88. Park, O.K., Hahm, M.G., Lee, S. et al. 2012. In situ synthe-
Tour, J.M. 2008. Diazonium functionalization of surfactant- sis of thermochemically reduced graphene oxide conducting
wrapped chemically converted graphene sheets. J. Am. Chem. nanocomposites. Nano Lett. 12:1789–93.
Soc. 130, 16201–6. 89. Hadad, C., Ke, X., Carraro, M., Sartorel, A., Bittencourt, C., Van
70. Gan, L., Zhang, D., and Guo, X. 2012. Electrochemistry: An Tendeloo, G., Bonchio, M., Quintana, M., and Prato, M. 2014.
efficient way to chemically modify individual monolayers of Positive graphene by chemical design: Tuning supramolecular
graphene. Small 8:1326–30. strategies for functional surfaces. Chem. Commun. 50:885–7.
71. Castelaín, M., Salavagione, H.J., and Segura, J.L. 2012. 90. Ye, Y.S., Chen, Y.N., Wang, J.S et al. 2012. Versatile graft-
“Click”-functionalization of [60]fullerene and graphene with ing approaches to functionalizing individually dispersed
an unsymmetrically functionalized diketopyrrolopyrrole graphene nanosheets using RAFT polymerization and click
(DPP) derivative. Org. Lett. 14:2798–801. chemistry. Chem. Mater. 24:2987–97.
91. Wang, Q.H., Jin, Z., Kim, K.K. et al. 2012. Understanding and 109. Wang, Z., Ge, Z., Zheng, X. et al. 2012. Polyvalent DNA-
controlling the substrate effect on graphene electron-transfer graphene nanosheets “click” conjugates. Nanoscale 4:394–9.
chemistry via reactivity imprint lithography. Nat. Chem. 4:724–32. 110. Meng, F., Ishida, H., and Liu, X. 2014. Introduction of benzox-
92. Segura, J.L. and Salavagione, H.J. 2013. Graphene in copper azine onto the graphene oxide surface by click chemistry and
catalyzed azide-alkyne cycloaddition reactions: Evolution the properties of graphene oxide reinforced polybenzoxazine
from [60] fullerene and carbon nanotubes strategies. Curr. nanohybrids. RSC Adv. 4:9471–5.
Org. Chem. 17:1680–93. 111. Gress, A., Volkel, A., and Schlaad, H. 2007. Thio-click modi-
93. Castelaín, M., Shuttleworth, P.S., Marco, C., Ellis, G., and fication of poly[2-(3-butenyl)-2-oxazoline]. Macromolecules
Salavagione, H.J. 2013. Comparative study of the covalent 40:7928–33.
diazotization of graphene and carbon nanotubes using ther- 112. Hoyle, C.E., Lowe, A.B., and Bowman, C.N. 2010. Thiol-click
mogravimetric and spectroscopic techniques. Phys. Chem. chemistry: A multifaceted toolbox for small molecule and
Chem. Phys. 15:16806–11. polymer synthesis. Chem. Soc. Rev. 39:1355–87.
94. Kolb, H.C., Finn, M.G., and Sharpless, K.B. 2001. Click 113. Lowe, A.B. 2010. Thiol-ene “click” reactions and recent appli-
chemistry: Diverse chemical function from a few good reac- cations in polymer and materials synthesis. Polym. Chem.
tions. Angew. Chem. Int. Ed. 40:2004–21. 1:17–36.
95. Huisgen, R. 1963. 1,3-Dipolar cycloadditions. Past and future. 114. Hoyle, C.H. and Bowman, C.N. 2010. Thiol-ene click chemis-
Angew. Chem. Int. Ed. 2:565–98. try. Angew. Chem. Int. Ed. 49:1540–73.
96. Tornøe, C.W., Christensen, C., and Meldal, M. 2002. 115. Chua, C.K., Ambrosi, A., and Pumera, M. 2012. Introducing
Peptidotriazoles on solid phase: [1,2,3]-triazoles by regiospe- dichlorocarbene in graphene. Chem.Commun. 48:5376–8.
cific copper(I)-catalyzed 1,3-dipolar cycloadditions of termi- 116. Ismaili, H., Geng, D., Sun, A.X., Kantzas, T.T., and
nal alkynes to azides. J. Org. Chem. 67:3057–64. Workentin, M.S. 2011. Light-activated covalent formation of
97. Rostovtsev, V.V., Green, L.G., Fokin, V.V., and Sharpless, K.B. gold nanoparticle-graphene and gold nanoparticle-glass com-
2002. A stepwise Huisgen cycloaddition process: Copper(I)- posites. Langmuir 27:13261–8.
catalyzed regioselective ligation of azides and terminal 117. Georgakilas, V., Bourlinos, A.B., Zboril, R. et al. 2010.
alkynes. Angew. Chem. Int. Ed. 41:2596–9. Organic functionalisation of graphenes. Chem. Commun.
98. Meldal, M. and Tornøe, C.W. 2008. Cu-catalyzed azide- 46:1766–8.
alkyne cycloaddition. Chem. Rev. 108:2952–3015. 118. Quintana, M., Montellano, A., del Rio Castillo, A.E., Van
99. Erickson, K., Erni, R., Lee, Z., Alem, N., Gannett, W., and Tendeloo, G., Bittencourt, C., and Prato, M. 2011. Selective
Zettl, A. 2010. Determination of the local chemical structure organic functionalization of graphene bulk or graphene edges.
of graphene oxide and reduced graphene oxide. Adv. Mater. Chem. Commun. 47:9330–2.
22:4467–72. 119. Sarkar, S., Bekyarova, E., Niyogi, S., and Haddon, R.C. 2011.
100. Yang, X., Ma, L., Wang, S., Li, Y., Tu, Y., and Zhu, X. 2011. Diels–alder chemistry of graphite and graphene: Graphene as
“Clicking” graphite oxide sheets with well-defined polysty- diene and dienophile. J. Am. Chem. Soc. 133:3324–7.
renes: A new strategy to control the layer thickness. Polymer 120. Sarkar, S., Bekyarova, E., and Haddon, R.C. 2012. Chemistry
52:3046–52. at the Dirac point: Diels–alder reactivity of graphene. Acc.
101. Cao, Y., Lai, Z., and Feng, J. 2011. Graphene oxide sheets Chem. Res. 45:673–82.
covalently functionalized with block copolymers via click 121. Bian, S., Scott, A.M., Cao, Y. et al. 2014. Covalently patterned
chemistry as reinforcing fillers. J. Mater. Chem. 21:9271–8. graphene surfaces by a force-accelerated diels–alder reaction.
102. Pan, Y., Bao, H., Sahoo, N.G., Wu, T., and Li, L. 2011. Water- J. Am. Chem. Soc. 135:9240–3.
soluble poly(N-isopropylacrylamide)–graphene sheets synthe- 122. Economopoulos, S.P., Rotas, G., Miyata, Y., Shinohara, H.,
sized via click chemistry for drug delivery. Adv. Funct. Mater. and Tagmatarchis, N. 2010. Exfoliation and chemical modifi-
21:2754–63. cation using microwave irradiation affording highly function-
103. Ryu, H.J., Mahapatra, S.S., Yadav, S.K., and Cho, J.W. 2013. alized graphene. ACS Nano 4:7499–507.
Synthesis of click-coupled graphene sheet with chitosan: 123. Dreyer, D.R., Park, S., Bielawski, C.W., and Ruoff, R.S. 2010.
Effective exfoliation and enhanced properties of their nano- The chemistry of graphene oxide. Chem. Soc. Rev. 39:228–40.
composites. Eur. Polym. J. 49:2627–34. 124. Chua, C.K. and Pumera, M. 2013. Covalent chemistry of pris-
104. Wang, H.X., Zhou, K.G., Xie, Y.L. et al. 2011. Photoactive gra- tine graphene. Chem. Soc. Rev. 42:3222–33.
phene sheets prepared by “click” chemistry. Chem. Commun. 125. Rodríguez-Pérez, L., Herranz, M.A., and Martín, N. 2013.
47:5747–9. The chemistry of pristine graphene. Chem. Commun. 49:
105. Ragoussi, M.E., Casado, S., Ribeiro-Viana, R., de la Torre, 3721–35.
G., Rojo, J., and Torres, T. 2013. Selective carbohydrate–lectin 126. Matyjaszewski, K. and Xia, J. 2001. Atom transfer radical
interactions in covalent graphene- and SWCNT-based molec- polymerization. Chem. Rev. 101:2921–90.
ular recognition systems. Chem. Sci. 4:4035–41. 127. Xu, L.Q., Yee, Y.K., Neoh, K.G., Kang, E.T., and Fu, G.D.
106. Ragoussi, M.E., Katsukis, G., Roth, A., Malig, J., de la Torre, 2013. Cyclodextrin-functionalized graphene nanosheets, and
G., Guldi, D.M., and Torres, T. 2014. Electron-donating behav- their host–guest polymer nanohybrids. Polymer 54:2264–71.
ior of few-layer graphene in covalent ensembles with electron- 128. Lock, E.H., Baraket, M., Laskoski, M. et al. 2012. High-
accepting phthalocyanines. J. Am. Chem. Soc. 136:4593–8. quality uniform dry transfer of graphene to polymers. Nano
107. Salvio, R., Krabbenborg, S., Naber, W.J.M., Velders, A.H., Lett. 12:102–107.
Reinhoudt, D.N., and van de Wiel, W.G. 2009. The formation 129. Guimont, A., Beyou, E., Cassagnau, P. et al. 2013. Grafting of
of large-area conducting graphene-like platelets. Chem. Eur J. polyethylene onto graphite oxide sheets: A comparison of two
15:8235–40. routes. Polym. Chem. 4:2828–36.
108. Devadoss, A. and Chidsey, C.E.D. 2007. Azide-modified gra- 130. Xu, X., Luo, Q., Lv, W.Y. et al. 2011. Functionalization of
phitic surfaces for covalent attachment of alkyne-terminated graphene sheets by polyacetylene: Convenient synthesis and
molecules by “click” chemistry. J. Am. Chem. Soc. 129:5370–1. enhanced emission. Macromol. Chem. Phys. 212:768.
131. Nair, N., Kim, W.J., Usrey, M.L., and Strano, M.S. 2007. A 148. Clavé, G. and Campidelli, S. 2011. Efficient covalent function-
structure-reactivity relationship for single walled carbon alisation of carbon nanotubes: The use of “click chemistry”.
nanotubes reacting with 4-hydroxybenzene diazonium salt. Chem. Sci. 2:1887–96.
J. Am. Chem. Soc. 129:3946–54. 149. Sun, S., Cao, Y., and Feng, J. 2010. Click chemistry as a route
132. Wang, D., Ye, G., Wang, X., and Wang, X. 2011. Graphene func- for the immobilization of well-defined polystyrene onto gra-
tionalized with azo polymer brushes: Surface-initiated polymer- phene sheets. J. Mater. Chem. 20:5605–7.
ization and photoresponsive properties. Adv. Mater. 23:1122–5. 150. Fu, R. and Fu, G. D. 2011. Polymeric nanomaterials from
133. Lillethorup, M., Kongsfelt, M., Ceccato, M., Jensen, B.B.E., combined click chemistry and controlled radical polymeriza-
Jørgensen, B., Pedersen, S.U., and Daasbjerg, K. 2014. High- tion. Polym. Chem. 2:465–75.
versus low-quality graphene: A mechanistic investigation of 151. Castelaín, M., Martínez, G., Ellis, G., Marco, C., and
electrografted diazonium-based films for growth of polymer Salavagione, H.J. 2013. The effect of the chemical approaches
brushes. Small 10:922–34. to modify graphene on the electrical, thermal and mechani-
134. Yang, Y., Song, X., Yuan, L. et al. 2012. Synthesis of PNIPAM cal properties of graphene/polyethylene nanocomposites.
polymer brushes on reduced graphene oxide based on click Macromolecules 46:8980–7.
chemistry and RAFT polymerization. J. Polym. Sci. A: 152. Yadav, S.K., Yoo, H.Y., and Cho, J.W. 2013. Click coupled gra-
Polym. Chem. 50:329–37. phene for fabrication of high-performance polymer nanocom-
135. Liu, J., Cui, L., Kong, N., Barrow, C.J., and Yang, W. 2014. posites. J. Polym. Sci. B: Polym. Phys. 51:39–47.
RAFT controlled synthesis of graphene/polymer hydrogel 153. Meng, D., Sun, J., Jiang, S. et al. 2012. Grafting P3HT brushes
with enhanced mechanical property for pH-controlled drug on GO sheets: Distinctive properties of the GO/P3HT com-
release. Eur. Polym. J. 50:9–17. posites due to different grafting approaches. J. Mater. Chem.
136. Chen, X., Yuan, L., Yang, P., Hu, J., and Yang, D. 2011. 22:21583–91.
Covalent polymeric modification of graphene nanosheets 154. Wang, H.X., Wang, Q., Zhou, K.G., and Zhang, H.L. 2013.
via surface-initiated single-electron-transfer living radical Graphene in light: Design, synthesis and applications of
polymerization. J. Polym. Sci. A: Polym. Chem. 49:4977–86. photo-active graphene and graphene-like materials. Small
137. Yong, S., Ming, F., and Hongbin, L. 2012. Poly(3- 9:1266–83.
hexylthiophene) brushes grown from graphene nanosheets. 155. Deetuam, C., Samthong, C., Thongyai, S., Praserthdam, P.,
Acta Polym. Sin. 12:223–30. and Somwangthanaroj, A. 2014. Synthesis of well dispersed
138. Midya, A., Mamidala, V., Yang, J.X. et al. 2010. Synthesis and graphene in conjugated poly(3,4-ethylenedioxythiophene):pol
superior optical-limiting properties of fluorene-thiophene- ystyrene sulfonate via click chemistry. Compos. Sci. Technol.
benzothiadazole polymer-functionalized graphene sheets. 93:1–8.
Small 6:2292–300. 156. Jing, Y., Tang, H., Yu, G., and Wu, P. 2013. Chemical modi-
139. Han, S. and Chun, B.C. 2014. Preparation of polyurethane nano- fication of graphene with a thermotropic liquid crystalline
composites via covalent incorporation of functionalized graphene polymer and its reinforcement effect in the polymer matrix.
and its shape memory effect. Composites: Part A 58:65–72. Polym. Chem. 4:2598–607.
140. Yadav, S.K. and Cho, J.W. 2013. Functionalized graphene 157. Castelaín, M., Martínez, G., Ellis, G., and Salavagione, H.J.
nanoplatelets for enhanced mechanical and termal properties 2013. Versatile chemical tool for the preparation of conductive
of polyurethane nanocomposites. Appl. Surf. Sci. 266:360–7. graphene-based polymer nanocomposites. Chem. Commun.
141. Park, S. and Ruoff, R.S. 2009. Chemical methods for the pro- 49:8967–9.
duction of graphenes. Nat. Nanotechnol. 4:217–24. 158. Liras, M., García, O., Quijada-Garrido, I., Ellis, G., and
142. Stankovich, S., Dikin, D.A., Piner, R.D. et al. 2007. Synthesis Salavagione, H.J. 2014. Homogenous thin layer coated
of graphene-based nanosheets via chemical reduction of exfo- graphene via one pot reaction with multidentate thiolated
liated graphite oxide. Carbon 45:1558–65. PMMAs. J. Mater. Chem. C 2:1723–9.
143. Tung, V.C., Allen, M.J., Yang, Y., and Kaner, R.B. 2009. High- 159. Xu, Z. and Gao, C. 2010. In situ polymerization approach to
throughput solution processing of large-scale graphene. Nat. graphene-reinforced nylon-6 composites. Macromolecules
Nanotechnol. 4:25–9. 43:6716–23.
144. Wang, G., Yang, J., Park, J. et al. 2008. Facile synthesis and 160. Kan, L., Xu,, Z., and Gao, C. 2011. General avenue to individu-
characterization of graphene nanosheets. J. Phys. Chem. C ally dispersed graphene oxide-based two-dimensional molec-
112:8192–5. ular brushes by free radical polymerization. Macromolecules
145. McAllister, M.J., Li, J.L., Adamson, D.H. et al. 2007. Single 44:444–52.
sheet functionalized graphene by oxidation and thermal 161. Shen, J., Hu, Y., Li, C., Qin, C., and Ye, M. 2009. Synthesis of
expansion of graphite. Chem. Mater. 19:4396–404. amphiphilic graphene nanoplatelets. Small 5:82–5.
146. Gao, W., Alemany, L.B., Ci, L., and Ajayan, P.M. 2009. New 162. Shaffer, M.S.P. and Koziol, K. 2002. Polystyrene grafted
insights into the structure and reduction of graphite oxide. multi-walled carbon nanotubes. Chem. Commun. 38:2074–5.
Nat. Chem. 1:403–8. 163. Qin, S., Qin, D., Ford, W.T. et al. 2004. Solubilization and
147. Segura, J.L., Castelaín, M., and Salavagione, H. 2013. purification of single-wall carbon nanotubes in water by in
Synthesis and properties of [60]fullerene derivatives function- situ radical polymerization of sodium 4-styrenesulfonate.
alized through copper catalyzed huisgen cycloaadition reac- Macromolecules 37:3965–7.
tions. Curr. Org. Synth. 10:724–36.
17 Molecular Theory of Graphene
Chemical Modification
Elena F. Sheka
CONTENTS
Abstract.......................................................................................................................................................................................317
17.1 Introduction.......................................................................................................................................................................317
17.2 Grounds of Graphene Molecular Chemistry.....................................................................................................................318
17.2.1 Criteria of Odd Electrons Correlation...................................................................................................................318
17.2.2 Quantitative Description of Graphene Molecule Radicalization..........................................................................319
17.2.3 Cause of Odd Electrons Correlation in sp2 Nanocarbons..................................................................................... 320
17.3 Molecular Theory of Graphene Hydrogenation............................................................................................................... 320
17.3.1 Two-Side Atomic Adsorption of Hydrogen on Fixed Membrane........................................................................ 321
17.3.2 One-Side Atomic Adsorption of Hydrogen on Fixed Membrane......................................................................... 322
17.4 Molecular Theory of Graphene Oxidation....................................................................................................................... 324
17.4.1 The First-Stage Oxidation of the (5,5) NGr Molecule.......................................................................................... 325
17.4.1.1 Homo-Oxidant Action........................................................................................................................... 325
17.4.1.2 Hetero-Oxidant Action.......................................................................................................................... 328
17.4.2 Second-Stage Oxidation of the (5,5) NGr Molecule............................................................................................. 328
17.4.2.1 Oxidation of Hydroxyl-Framed GOs..................................................................................................... 328
17.4.2.2 Oxidation of Hetero-Oxidant-Framed GOs........................................................................................... 329
17.4.3 Main GO Characteristics and Properties Based on the Computational Experiment........................................... 330
17.4.4 Oxidation-Induced Structure Transformation.......................................................................................................331
17.4.5 Some Comments Concerning GO Reduction....................................................................................................... 333
17.5 How Does Chemical Modification of Graphene Molecules Affect Electronic Structure?.............................................. 334
17.5.1 Size Dependence of Chemical Reactions Final Products..................................................................................... 334
17.5.2 Chemically Induced Destruction of Graphene Samples....................................................................................... 335
17.5.3 Chemically Affected Topological Mechanochemistry of Graphene.................................................................... 336
17.6 Conclusive Remarks......................................................................................................................................................... 338
Notes in Proofs........................................................................................................................................................................... 340
References.................................................................................................................................................................................. 340
ABSTRACT 17.1 INTRODUCTION
The chapter is devoted to the chemical uniqueness of gra- The nickname of a “miracle material” has been widely
phene. In view of molecular theory, the uniqueness is a conse- accepted by the graphene community. It normally means the
quence of benzenoid packing of carbon atoms, based on three material superior properties such as high electron mobility,
neighbors to each carbon atom leaving the atom fourth valence high thermal conductivity, high value of the Young’s modu-
electron (the odd one) on its own. Due to rather large C─C lus, peculiar structure of its electron bands, etc. However, all
distances, the odd electrons of graphene molecules are corre- these properties are only the outward manifestation of the
lated whereupon the molecules are radicalized, and molecular graphene’s wonderful nature. The real miracle of graphene is
chemistry of graphene is the chemistry of “dangling bonds.” that it is a union of two entities: the physical and the chemical,
The chapter presents the description of the molecular-theory- each of which is unique in its own way. The abovementioned
dangling-bond view on graphene molecule hydrogenation and superior properties are characteristics of the graphene physi-
oxidation. Specific peculiarities of the graphene molecular cal entity. Much less has been told about its chemical unique-
chemistry related to size-dependent topochemical features ness. Moreover, physicists often refer to graphene as “highly
are discussed, as well. inert” material [1]. However, it is not the case. First of all,
317
we must give an account of how much attractive sound of the electrons. The latter turns out to play the governing role. As
word “carbon” is to anyone. “Carbon” implies practically the will be shown below, such an approach occurs very efficient
countless amount of scientific knowledge, both empirical and in describing chemical properties of graphene.
theoretical-computational. Obviously, this aspect has mark-
edly contributed to the unprecedented enthusiasm about gra-
17.2.1 Criteria of Odd Electrons Correlation
phene. However, the uniqueness of the graphene chemistry is
generated, not by carbon atoms themselves but their packing In spite of formally two-atomic unit cell of crystalline gra-
in a flat honeycomb structure. The structure, based on benze- phene, its properties are evidently governed by the behavior of
noid units, offers three neighbors to each carbon atom leaving odd electrons of the hexagonal benzenoid units. The only thing
the atom fourth valence electron on its own. These electrons that we know about the behavior for sure is that the interac-
form a pool of odd electrons whose behavior might change tion between odd electrons is weak; nevertheless, how weak
from the covalent bonding, characteristic for π electrons, to is it? Is it enough to provide a tight covalent pairing when two
free electrons of radicals when the interaction between the electrons with different spins occupy the same place in space
electrons becomes weaker and weaker. The two electron states or, oppositely, is it too weak for this and the two electrons are
belong to different limit cases in terms of the electron corre- located in different places thus becoming spin-correlated?
lation: π electrons are not correlated while radical electrons This supremely influential molecular aspect of graphene can
are strongly correlated so that two electrons with different be visualized on the platform of the molecular quantum theory.
spins occupy different places in the space. The odd electrons To exhibit a trend, a system computational experiment
of benzenoid units of graphene in contrast with π electrons of should be carried out meaning that a vast number of com-
the benzene molecule are correlated, which is caused by the putations are to be performed as well as a great number of
difference in the C─C bond lengths [2]. atoms are to be considered. At the same time, when speaking
The odd electrons correlation and the extreme tight con- about electron correlation, one should address the problem to
nection between the correlation degree and benzenoid bond the configuration interaction (CI). However, neither full CI
structure make graphene material highly sensitive to any kind nor any of its truncated versions, clear and transparent con-
of external action such as morphological changing, chemical ceptually, can be applied for the computational experiments,
modification, mechanical loading and fixation, application valuable for graphene nanoscience. Owing to this, techniques
of electric and magnetic field, and so forth, thus making it based on single determinants become an acceptive alterna-
structurally and electronically nonstable. The “fluid” elec- tive. Unrestricted Hartree–Fock (UHF) and unrestricted den-
tronic structure [3] accompanied by the flexible space struc- sity functional theory (spin polarized, UDFT) approaches
ture aggravated with equienergy topological phase transitions form the techniques ground and both are sensitive to the elec-
are the main reasons for failures of, say, stable working tech- tron correlation, but differently due to different dependence
nologies for converting graphene from semimetal to semi- of their algorithms on electron spins [5,6]. The approach
conductor [4] as well as for postponing the application of the application raises two questions: (1) what are criteria that
high-performance graphene to 2035 or later [1]. Since until indicate the electron correlation in the studied system and (2)
now chemical modification of graphene has been considered how much informative are the solutions of single-determinant
as the main way for achieving its best physical properties it is approaches for a system of correlated electrons.
necessary to analyze all pro and contra on the way. Meeting To answer the first question, three criteria, which highlight
the requirement, the current chapter is devoted to peculiarities the electron correlation at the single-determinant level of the-
of chemical modification of graphene in view of its chemi- ory, can be suggested [7]. This concerns the following char-
cal uniqueness. The chapter is organized as follows. Section acteristic parameters related to a benzenoid-packed molecule:
17.2 describes the grounds of the computational chemistry
of graphene on the basis of molecular theory. Section 17.3 Criterion 1: Misalignment of energy
concerns the graphene molecule hydrogenation. Oxygenation
and reduction of the molecules are described in Section 17.4. ∆E RU ≥ 0, where ∆E RU = E R − E U (17.1)

Some peculiar effects caused by the chemical modification
are considered in Section 17.5. Section 17.6 summarizes the
Here, ER and EU are the total energies calculated by using the
most important points of the graphene molecular chemistry.
restricted and unrestricted versions of a software in use.
17.2 GROUNDS OF GRAPHENE Criterion 2: Misalignments of squared spin

MOLECULAR CHEMISTRY
The following is concentrated on the molecular essence of ∆Sˆ 2 ≥ 0; ∆Sˆ 2 = SÛ2 − S (S + 1). (17.2)
graphene considered from the viewpoint of the molecular
theory of sp2 nanocarbons. The theory is based on two main Here, ŜU2 is the squared spin calculated within the applied
concepts, which involve (1) the odd electron origin of the unrestricted technique while S(S + 1) presents the exact value
graphene electron system and (2) the correlation of the odd of Ŝ 2 .
Molecular Theory of Graphene Chemical Modification 319
Criterion 3: Appearance of effectively unpaired relation N D = 2∆SÛ2 , which is characteristic of spin-contam-

electrons inated solutions in the singlet state [9], is rigidly kept over
ND ≠ 0 all the fragments. Therefore, the odd electrons correlation

for the studied molecules is quite significant. Generally, the
Here, ND is the total number of effectively unpaired electrons. conclusion can be addressed to nanographenes of any shape.
The value is determined as However, the numeric identifying parameters will obviously
depend on the species shape due to a particular role of cir-

N D = trD(r | r ′ ) ≠ 0 and ND = ∑ A
DA

(17.3)
cumference atoms, which will be shown next. Nevertheless, it
is possible to state that any nanographene—that is graphene
molecule—is considerably radicalized, not depending on its
Here, D(r|r′) [8] and DA [9] present the total and atom-frac- size and shape.
tioned spin density caused by the spin asymmetry due to the
location of electrons with different spins in different spaces. 17.2.2 Quantitative Description of Graphene
Criterion 1 follows from the well-known fact that the
Molecule Radicalization
electron correlation, if available, lowers the total energy
[10]. Criterion 2 is the manifestation of the spin contamina- The UHF computational scheme leads to spin-contaminated
tion of unrestricted single-determinant solutions [9,11]; the solutions leading to the question: how reliable and informa-
stronger electron correlation, the bigger spin contamination tive are the obtained solutions? Thoroughly analyzed in
of the studied spin state. Criterion 3 highlights the fact that Reference 14, the following answers to the questions have
the electron correlation is accompanied by the appearance of been obtained: (1) the broken symmetry UHF approach
effectively unpaired electrons that provide the molecule radi- allows obtaining the exact energy of pure spin states; (2) the
calization [8,9,11]. approach provides exact determination of the magnetic con-
Table 17.1 presents sets of the three parameters evaluated stant; and (3) the approach forms the grounds for computa-
for a number of graphene molecules presented by rectangular tional chemistry of electron-correlated graphene molecules.
(na, nz) fragments of graphene (na and nz count the benzenoid Criterion 3 lays the foundation for the basic concepts of
units along armchair (AC) and zigzag (ZZ) edges of the frag- the computational chemistry of graphene. First shown by
ment, respectively [12]), (na, nz) nanographenes (NGrs) below, Takatsuka et al. [8], the correlation of weakly interacting elec-
by using the AM1 version of the semiempirical UHF approach trons is manifested through the density matrix D(r|r′), named
implemented in the CLUSTER-Z1 codes [13]. To our knowl- as the distribution of “odd” electrons. The function was
edge, only this software allows for getting all the above three proven to be a suitable tool to describe the spatial separation
parameters within one computational session. Throughout the of electrons with opposite spins, and its trace ND = trD(r|r′)
current chapter, the referred results are obtained within the was interpreted as the total number of these electrons [8,15].
framework of this approach. The authors suggested ND to manifest the radical character
As seen in the table, the parameters are certainly not zero, of the species under investigation. Staroverov and Davidson
obviously greatly depending on the fragment size while their [9,16], 20 years later, changed the term by the “distribution of
relative values are practically size independent. The atten- effectively unpaired electrons” emphasizing that not all the
tion should be called to rather large ND values, both absolute odd electrons may be taken off the covalent bonding. Even
and relative, indicating a considerable radicalization of the Takatsuka et al. [8] mentioned that the function D(r|r′) can
studied nanographenes. It should be added as well that the be subjected to the population analysis within the framework
TABLE 17.1
Identifying Parameters of the Odd Electrons Correlation in the Rectangular Graphene Molecules
(na, nz) Odd electrons, ΔERUa
∆ŜU2
Nanographenes Nodd (kcal/mol) δERU (%)b ND (e–) δND (%)b
(5,5) 88 307 17 31 35 15.5
(7,7) 150 376 15 52.6 35 26.3
(9,9) 228 641 19 76.2 35 38.1
(11,10) 296 760 19 94.5 32 47.2
(11,12) 346 901 20 107.4 31 53.7
(15,12) 456 1038 19 139 31 69.5
Source: E. F. Sheka, Int. J. Quantum Chem. 2012, 121, 3076.

a AM1 version of UHF codes of CLUSTER-Z1 [13]. Presented energy values are rounded off to an integer.
b Percentage values are related to δERU = ΔERU/ER(0) and δN = N /N , respectively.

D D odd
of the Mulliken partitioning scheme. In the case of a single 1825. However, only a hundred years later Hückel suggested
Slater determinant, ND can be determined as [9] the explanation of the deficiency of hydrogen atoms in the
molecule to complete the valence ability of its carbon atoms.
N D = trDS (17.4) Extra, or odd, electrons were named as π electrons that, in
contrast to trigonal σ electrons, interact much weaker while
where providing the additional covalent coupling between neighbor-
ing atoms. The two electrons are located in the same space,
and their spins are subordinated to the Pauli law. Formally,
DS = 2 PS − ( PS )2 (17.5)
this π electrons view, which lays the foundation of the aro-
maticity concept, has been expanded over all sp2 nanocarbons
Here, D is the spin density matrix D = Pα − Pβ while and has been shared by a number of material scientists in
P = Pα + Pβ is a standard density matrix in the atomic orbital the field until now. However, the concept does not take into
basis and S is the orbital overlap matrix (α and β mark differ- account a crucial role of the distance between two neighbor-
ent spins). The population of effectively unpaired electrons on ing odd electrons. As seen in Figure 17.1, which presents a
atom A is obtained by partitioning the diagonal of the matrix plotting of the total number of effectively unpaired electrons
DS as ND as a function of the C─C distance in ethylene, benzene,
and hexamethylbenzene molecules, the bond stretching
N DA = DA = ∑ (DS)
µ ∈A
µµ (17.6)
below Rcov = 1.395 Å does not cause the appearance of the
unpaired electrons so that the relevant π electrons are fully

covalently bound. However, above Rcov the number ND grad-
ually increases up to clearly vivid knees, which evidence a
so that
radicalization of the previous π electrons. Further stretching
concerns mainly two σ electrons that, once fully covalently

ND = ∑ A
N DA

(17.7) bound until R ~ 2 Å, gradually become unpaired just repeat-
ing the fortune of π electrons. In the figure, a horizontal bar
circumscribes the region of the C─C bond lengths typical for
Staroverov and Davidson [9] showed that the atomic popu- fullerenes, carbon nanotubes, and graphene. As follows from
lation NDA is close to the Mayer free valence index FA in a this picture, the radicalization of π electrons of these mole-
general case while in the singlet state NDA and FA are identi- cules is still nearly at the beginning. Nevertheless, the number
cal. Thus, plotting NDA over atoms gives a visual picture of the of unpaired electrons becomes quite significant to determine
actual radical electrons distribution that, in its turn, exhibits specific molecule chemistry.
atoms with the enhanced chemical reactivity. In spite of clear explanation of where unpaired electrons
The UHF computational algorithm realized in the neglect are coming from, the question about their existence still
of diatomic differential overlap approximation (NDDO) [17] remains due to suspicion of their attribution to an artifact
(the basis for the AM1/PM3 semiempirical techniques in caused by the limitations of the single-determinant calcu-
CLUSTER-Z1) suggests the exact determination of both ND lations. However, recent examinations have convincingly
and NDA as follows [18]: shown that unpaired electrons are a physical reality [19] (see
NAT NORBS NAT NORBS
comments in Reference 14).
ND = ∑A
N DA = ∑∑∑∑D
A =1 i ∈A B =1 j ∈B
ij (17.8)
17.3 MOLECULAR THEORY OF

GRAPHENE HYDROGENATION
Here, Dij are elements of the spin density matrix D, NORBS, According to UHF calculations, C─C bond length of graphene
and NAT mark the number of orbitals and atoms, respectively. molecules fill an interval that depends on the molecule size. It
The expression can be easily implemented into UHF codes, should be mentioned that the application of the restricted ver-
as was done for CLUSTER-Z1, thus opening a possibility to sion of the same program results in practically non-dispersive
quantitatively describe the extent of radicalization of gra- value of the C─C bond length of 1.42 Å. Thus, for the (5,5)
phene molecules and thereby predetermining their chemi- NGr the interval constitutes 1.322–1.462 Å while for (11,11)
cal behavior, which will be shown in the following sections. NGr it is 1.284–1.468 Å. The relative number of bonds whose
Before going to these topics, let us turn our attention to the length exceeds Rcrit is 62% and 81% in the two cases, respec-
cause of the correlation of odd electrons in graphene. tively. This leads to a considerable amount of the effectively
unpaired electrons, whose total numbers are listed in Table 17.1
17.2.3 Cause of Odd Electrons Correlation for different nanographenes. From the chemical viewpoint, ND
values describe the molecules chemical susceptibility [2] while
in sp2 Nanocarbons
NDA presents atomic chemical susceptibility (ACS) of atom A.
The odd electrons story of nanocarbons is counted from the These two quantities become the main parameters of the com-
discovery of the benzene molecule by Michael Faraday in putational chemistry of graphene molecules.
3.5
Effectively unpaired electrons (ND, e)

3
Hexamethylbenzene
2.5
1.5
Ethylene
Benzene
1
0.5 Rcov Bond length (Å)

0 *
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0
FIGURE 17.1 Total numbers of effectively unpaired electrons ND accompanying a stretching of C–C bond in ethylene, benzene, and hexa-
methylbenzene. Rcov marks the extreme distance that corresponds to the completion of the covalent bonding. A horizontal bar marks the
interval of the C–C bond lengths characteristic of sp2 nanocarbons.
Figure 17.2 exhibits a typical ACS image map presented we shall refer to the two structures as free-standing and fixed
by the NDA distribution over atoms of the (5,5) NGr. The mol- membrane, respectively.
ecule edges are not terminated, and the NDA image map has The chemical portraits of the structures shown in Figure
a characteristic view with a distinct framing by edge atoms 17.3b and d are quite similar and reveal the transformation of
since the main part of the unpaired electrons is concentrated brightly shining edge atoms in Figure 17.2b into dark spots.
in this area. This map presents a typical chemical portrait The addition of two hydrogen atoms to each of the edge ones
of any graphene molecule and exhibits the exceptional role saturates the valence of the latter completely, which results
of the circumference area, which looks like a typical “dan- in zeroing NDA values as is clearly seen in Figure 17.3e. The
gling bonds” icon [20]. At the same time, the ACS image map chemical activity is shifted to the neighboring basal atoms and
intensity in the basal plane is of ~0.3e in average so that the retains higher in the vicinity of zg edges, however, differently
basal plane should not be discounted when it comes to chemi- in the two cases. The difference is caused by the redistribution
cal modification of the molecule thus disproving the com- of the C─C bond lengths of the free-standing membrane when
mon idea of its chemical inertness. Moreover, basing on the it is fixed over the perimeter, thus providing different starting
NDA value and choosing the largest of them as a quantitative conditions for the hydrogenation of the two membranes.
pointer of the target atom for the coming chemical attack, one Besides the two types of initial membranes, the hydroge-
can suggest the algorithmic “computational syntheses” of the nation depends on other factors such as (1) the hydrogen spe-
molecule polyderivatives [21]. cies in use and (2) the accessibility of the membrane sides to
The absolute NDA values of the (5,5) NGr, shown in the hydrogen. Even these circumstances evidence the hydro-
Figure 17.2c according to the atom numbering in the output genation of graphene to be a complicated chemical event
file, clearly exhibit 22 edge atoms involving 2 × 5 zg and 2 × 6 that strongly depends on the initial conditions, once divided
ach ones have the highest NDA, thus marking the perimeter as into eight adsorption modes in regard to atomic or molecu-
the most active chemical space of the molecule. The molecule lar adsorption; one- or two-side accessibility of membranes;
hydrogenation starts on atom 14 (star-marked in Figure 17.2c) and free or fixed state of the membranes perimeter. Only two
according to the largest NDA in the output file. The next step of of the latter correspond to the experimental observation of
the reaction involves the atom from the edge set as well, and hydrogenated specimens discussed in Reference 23, namely,
this continues until all the edge atoms are saturated by a pair two-side and one-side atomic hydrogen adsorption on the
of hydrogen atoms each since all 44 steps are accompanied fixed membrane. Stepwise hydrogenation of the (5,5) NGr
by the high-rank NDA list where edge atoms take the first posi- molecule was considered in details in Reference 22.
tion [22]. Thus, obtained hydrogen-framed (5,5) NGr mole-
cule is shown in Figure 17.3 alongside with the corresponding 17.3.1 Two-Side Atomic Adsorption of
ACS image map. Two equilibrium structures are presented.
Hydrogen on Fixed Membrane
The structure in panel (a) corresponds to the optimization of
the molecule structure without any restriction. In the second The hydrogenation concerns the basal plane of the fixed
case, positions of the edge carbon atoms and framing hydro- hydrogen-framed membrane shown in Figure 17.3c that is
gen atoms under optimization were fixed. In what follows, accessible to hydrogen atoms from both sides. As seen in
(a) (b) 2
0 1.3
1.2
1.1
–2
Y coordinate (Å)
1
0.9
–4 0.8
0.7
0.6
–6 0.5
0.4
–8 0.3
0.2
–10 0.1
–12
–4 –2 0 2 4 6 8 10
X coordinate (Å)
(c) 1.4
*
1.2
Atomic chemical susceptibility (NDA, e)
0.8
0.6
0.4
0.2
0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64
Atom number
FIGURE 17.2 Top and side views of the equilibrium structure of (5,5) NGr molecule (a), NDA image map (b), and NDA distribution over
atoms according to atom numbers in the output file (c). (Adapted from E.F. Sheka and N.A. Popova. J. Mol. Model. 2012, 18, 3751.)
Figure 17.3e, the first hydrogenation step occurs on basal structure H44, including framing hydrogen atoms, is shown at
atom 13 marked by a star. Since the membrane is accessi- the bottom of the figure and presents a computationally syn-
ble to hydrogen from both sides, a choice in favor of depo- thesized fully saturated chair-like (5,5) nanographane (NGra)
sition of the hydrogen atom above the carbon plane (“up”) that is in full accordance with the experimental observation
was done following the lowest-total-energy (LTE) criterion. of the graphane crystalline structure [23]. A complete set of
After the first deposition, the NDA map revealed carbon atom the two-side-adsorption products, which involves both a free-
46 for the next deposition (see H1 NDA map in Figure 17.4). standing and fixed membranes, forms polyhydrides family I
The LTE criterion favors the “down” position of structure shown in Figure 17.6.
H2 shown in the figure. The second deposition highlighted
next targeting carbon atom 3 (see NDA map of H2 hydride), the 17.3.2 One-Side Atomic Adsorption of
third adsorbed hydrogen atom activated target atom 60, the
Hydrogen on Fixed Membrane
fourth did the same for Atom 17, and so forth. Checking up
and down depositions in view of the LTE criterion, a choice Starting from the first step of the hydrogenation described
of the best configuration was performed and the correspond- in Section 17.3, the reaction proceeded further with the sec-
ing equilibrium structures for a selected set of the (5,5) NGr ond and all the next steps of the up deposition only. The
polyhydrides from H1 to H11 are shown in Figure 17.4. A set choice of the target atom at each step was governed by the
of the next structures from the 15th step up to the final 44th high-rank NDA values. Figure 17.6c presents the saturated
step is presented in Figure 17.5. A random distribution of the graphene polyhydride related to the final 44th step over the
adsorbed hydrogen atoms, presented by the set, well corre- basal atoms. A peculiar canopy shape of the carbon skele-
lated with the experimentally observed picture [24]. The set ton of the hydride is solely provided by the formation of the
is completed by the regular structure at the 44th step. This table-like cyclohexanoid units. However, the unit packing is
(a) (b) 2
0 0.6
–2 0.5
Y coordinate (Å)
0.4
–4
0.3
–6
0.2
–8 0.1
–10
–12
–4 –2 0 2 4 6 8 10
X coordinate (Å)
(c) (d) 2
0
0.8
–2 0.7
Y coordinate (Å)
0.6
–4 0.5
0.4
–6
0.3
–8 0.2
0.1
–10
–12
–4 –2 0 2 4 6 8 10
X coordinate (Å)
(e) 1.4
1.2
Atomic chemical susceptibility (NDA, e)
0.8
0.6
0.4
0.2
0
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64
Atom number
FIGURE 17.3 Equilibrium structures of free-standing (a) and fixed (c) H2-terminated (5,5) NGr membrane; NDA image maps (b, d), and NDA
distribution over atoms according to atom numbers in the output file (e). (Adapted from E.F. Sheka and N.A. Popova. J. Mol. Model. 2012,
18, 3751.) Gray and dark gray balls mark carbon and hydrogen atoms, respectively. Light gray histogram plots ACS data for the pristine (5,5)
NGr molecule. The curve and black histogram are related to membranes in panels (a) and (c), respectively.
quasi-regular which may explain the amorphous character graphene polyhydrides. In both cases, the distribution con-
of the polyhydrides formed at the outer surface of graphene sists of two parts, the first of which covers 44 C─H bonds
ripples observed experimentally [23]. A complete set of the formed at the molecule skeleton edges. Obviously, this part is
one-side-adsorption products, which involves both a free- identical for both hydrides since the bonds are related to the
standing and fixed membranes forms polyhydrides family II framing atoms. The second part covers C─H bonds formed
shown in Figure 17.6. by the hydrogen atoms attached to the basal plane. As seen
As for the hydrogen coverage, Figure 17.7 presents the dis- in the figure, in the case of polyhydride 1, C─H bonds are
tribution of C─H bond lengths of the two discussed saturated practically identical with the average length of 1.126 Å and
H1 H5
H2 H6
H3 H10
H4 H11
FIGURE 17.4 Equilibrium structures (left) and NDA image maps (right) of graphene hydrides related to the initial stage of the basal-plane
hydrogenation. HKs denote hydrides with K hydrogen atoms deposited on the membrane basal plane. (Adapted from E.F. Sheka and N.A.
Popova. J. Mol. Model. 2012, 18, 3751.) Framing hydrogen atoms are not shown to simplify the structure image presentation. Atoms mark-
ing as shown in caption of Figure 17.3.
only slightly deviate from those related to framing atoms. at final steps the hydrogen adsorption and desorption become
This is just a reflection of the regular graphane structure of competitive.
the hydride H44 shown in Figure 17.5 similar to highly sym-
metric fullerene hydride C60H60 [25]. In contrast, C─H bonds
17.4 MOLECULAR THEORY OF
on a canopy-like carbon skeleton of hydride II are much lon-
ger than those in the framing zone and significantly oscillate GRAPHENE OXIDATION
around the average value of 1.180 Å. In spite of the values In view of the molecular theory, the hot points of the GO
markedly exceed a “standard” C─H bond length of 1.11 Å, (graphene oxide) chemistry can be presented as the following
typical for benzene, those are still among the chemical C─H questions:
bonds, whilst stretched, since the C─H bond rupture occurs
at the C─H distance of 1.72 Å [26]. A remarkable stretching • What is the role of individual oxidants, such as O,
of the bonds points to a considerable weakening of the C─H OH, and COOH, in the oxidation of a graphene
interaction for polyhydride II in comparison with polyhydride molecule?
I, which is supported by the energetic characteristics of the • How does the coupling of the oxidants with the mol-
hydrides, as well [22]. The total energies of both hydrides are ecule depend on the local place of the addition?
negative by sign and gradually increase by the absolute value • What is the highest degree of the derivatization
when the number of adsorbed atoms increases. However, the under graphene molecule oxidation?
absolute value growth related to polyhydrides II is slowing • What is the structure of the carbon skeleton of GO
down starting at step 11 in contrast to the continuing growth and can the regular structure be achieved?
for polyhydrides I [22]. This retardation obviously shows that • What chemical composition of GO can be obtained
the one-side addition of hydrogen to the fixed membrane of under conditions close to experimental ones?
polyhydrides II at the coverage higher than 30% is more dif-
ficult than in the case of the two-side addition of polyhydrides Answering each of the questions presents a topic for a
I, for which the reaction of the chemical attachment of the valuable computational study. A temptation to simultaneously
hydrogen atoms is thermodynamically profitable through over answer all the questions raises a problem of a system approach,
the covering up to the 100% limit. In contrast, the large cover- or, in other words, of an extended computational experiment.
age for polyhydrides II becomes less and less profitable so that The computational molecular theory of graphene is the very
H15 (a) (b)
Family I
H16
(c) (d)
H17 Family II
H25
FIGURE 17.6 Equilibrium structures (top and side views) of fixed

(a, c) and free-standing (b, d) (5,5) NGr membranes under two-side
(a, b) (polyhydrides I) and one-side (c, d) (polyhydrides II) hydrogen
H26 adsorption. Framing hydrogen atoms are not shown to simplify the
structure image presentation. Atoms marking as shown in caption
of Figure 17.3.
hydrogenation described above. In the background of a tight

similarity in both processes, in general, important difference of
H40 H44 the events concerns the fact that instead of atomic hydrogens,
which were attacking agents in the first case, a set of oxidants
consisting of oxygen atoms O, hydroxyls OH, and carbox-
yls COOH had to be considered in the latter case. A detailed
description of the molecule oxidation is given in Reference 27.
A comparable study of the results related to different configura-
tions of GOs has led the foundation for the suggested conclu-
FIGURE 17.5 Equilibrium structures (left) and NDA image maps sions. The oxidation was considered as a two-stage reaction,
(right) of graphene hydrides from H15 to H44. (Adapted from E.F. the first stage covered first 22 steps that involve edge atoms of
Sheka and N.A. Popova. J. Mol. Model. 2012, 18, 3751.) Framing
the molecule only, on one hand, and all the oxidants, on the
hydrogen atoms are shown for H44 only. Atoms marking as shown in
caption of Figure 17.3. other hand. The second stage combines results related to both
edge and basal atoms for oxidation by OH and COOH and basal
feasible method in the framework in which the approach can atoms only when attaching atomic oxygen.
be realized. The latter means that not the selected individ-
ual computations but a solution of a number of particularly 17.4.1 The First-Stage Oxidation of the
arranged computational problems is the aim of the computa- (5,5) NGr Molecule
tional experiment. Evidently, the bigger the set, the more col-
ors participate in drawing the image of the GO chemistry. The 17.4.1.1 Homo-Oxidant Action
first attempt of such system computational approach including Figure 17.8 presents equilibrium structures of the O─, OH─,
about 400 computational jobs has been recently undertaken and COOH─ multifold GOs at the first 5 and the 22nd steps
[27]. The pristine (5,5) NGr molecule was the basis of the per- of oxidation. For comparison, a similar first stage of the (5,5)
formed computational experiment. NGr molecule hydrogenation is added. In all the cases, each
Stepwise oxidation of the (5,5) NGr molecule in the form next step of a subsequent addition was governed by the high-
of free-standing membrane was considered similarly to the est rank NDA value at a preceding step. As seen in the figure,
1.3
1.28
1.26
1.24
C–H bond length (Å)
1.22 Basal-plane atoms
1.2 2
1.18
1.16
Edge atoms 1
1.14
1.12
1.1
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88
C–H bond number
FIGURE 17.7 C–H bond length distribution for saturated graphene polyhydrides Ia (1) and IIc (2). (Adapted from E.F. Sheka and N.A.
Popova. J. Mol. Model. 2012, 18, 3751.)
Step H-framing O-framing OH-framing COOH-framing
22
22-fold hydride 22-fold GO I 21-fold GO II 22-fold GO III
FIGURE 17.8 (a, b) Equilibrium structures of (5,5) NGr polyhydrides and polyoxides related to the 1st, 2nd, 3rd, 4th, 5th, and 22nd (21st
in the case of OH-framing, see text) steps of first-stage reactions. Roman figures mark the obtained GOs. Dark gray and light gray balls
mark hydrogen and oxygen atoms, respectively. (Adapted from E.F. Sheka and N.A. Popova. Phys. Chem. Chem. Phys. 2013, 15, 13304.)
the difference in the starting points at the second step has to the end. Thus, carbonyls formed over GO I provide prac-
actually led to the different progress of the hydrogenation and tically full suppression of the chemical activity of the edge
oxidation processes in space. atoms. The further oxidation of the molecule, starting at the
In the case of O─GOs, the obtained 22-fold GO (GO I) 23rd step, occurs at the basal plane. In contrast, the chemical
presents a completed framing of the molecule. In the case activity at the circumference of GO II and GO III is not fully
of OH─GOs, a successive OH-framing continued up to the suppressed; this is why both edge and basal atoms should be
12th step. After optimization of the 13-fold OH─GO, its taken into account when considering further oxidation.
equilibrium structure revealed the dissociation of one previ- The analysis of energetic regularities completes the con-
ously attached OH group and the formation of C─H and C═O sideration of homo-oxidant framing of the (5,5) NGr mole-
bonds instead. All the next steps of oxidation up to the 21st cule. Figure 17.9a presents a set of plottings related to per step
did not cause any similar transformation and the final GO II coupling energies (PCEs). At each step, the PCE was deter-
shown in the last row in the figure was obtained. The mol- mined as
ecule did not become flat practically since the third step. Next
steps increased the corrugation, but the skeleton deformation pst
Ecpl (GOn ) = ∆H (GOn ) − ∆H (GOn−1 ) − ∆H (Oxd ) (17.9)

became more smoothed by the end of framing. In the case of
COOH-oxides, sterical constraints additionally complicated Here, ΔH(GOn), ΔH(GOn−1), and ΔH(Oxd) are heats of for-
framing as seen in the case of the 22-fold COOH-framed GO mation of the considered GOs at the nth and (n − 1)th steps of
III shown at the bottom of Figure 17.8. oxidation, as well as LTE of oxidant itself, respectively.
The first-stage framing considerably suppresses the chemi- Analyzing the data presented in Figure 17.9a, one can con-
cal activity of the GO molecule edge atoms, but evidently not clude the following:
(a) 0 5 10 15 20 25
60
Per step coupling energy (kcal/mol)
–20
3
3
–100 2 1
H 2, H
1
–180
–260
Step number
(b) 0 5 10 15 20 25
0
3
–80
2
3
2
1 1
–160
–240
Step number
FIGURE 17.9 Per step coupling energy versus step number for families of GOs obtained in the course of the first-stage oxidation. (a) Homo-
oxidant reactions, per step addition of O (curve 1, GO I), OH (curve 2, GO II), COOH (curve 3, GO III); and H denotes hydrides. (b) Hetero-
oxidant reactions for GO V, per step addition of O (curve 1), OH (curve 2), and COOH (curve 3). Ovals mark intersections of curves 2 and 1
when OH-addition becomes more profitable than O-addition.
pst
1. The dependences of Ecpl (GOn ) on step number have
much in common for all the oxidants. Thus, they
show non-regular oscillating behavior that is char-
acteristic of the hydrogenation as well once provided
supposedly by a particular topology of the (5,5) NGr
molecule that reflects a changeable disturbance of
the molecule carbon skeleton in the course of the
chemical modification.
2. The molecule framing by carbonyls is evidently
the most energetically favorable while oppositely,
COOHs provide the least stable configurations. The
coupling characteristic for OH─GOs is in between
these two limit cases and only at the 13th step of the
OH─GO is the largest one among the others due to
dissociation of the hydroxyl group.
FIGURE 17.10 Equilibrium structure of free-standing (5,5) NGr
membrane subjected to hetero-oxidant framing (equilibrium struc-
According to this finding, one should expect that carbon- ture of GO V). Atom marking as shown in caption of Figure 17.8.
yls play the dominant role in the framing of GO molecules 2. Hetero-oxidant computational experiment exhibits
in practice. In contrast, the most spread opinion supports quite undoubtedly that COOHs are not favorable for
models of the GO chemical composition suggested by Lerf the (5,5) NGr molecule framing.
and Klinovski [28,29] where not carbonyls but carboxyls and 3. The molecule framing is predominantly provided by
epoxy groups frame the GO platelets. However, as shown in carbonyls and partially by hydroxyls so that the final
Reference 27, the epoxy-GOs are not only the least stable by framing composition is double-oxidant correspond-
energy but also even become energetically nonprofitable since ing to the 22-fold GO V (the oxide numeration fol-
PCE becomes positive from the third step. lows [27]) whose equilibrium structure is presented
in Figure 17.10.
17.4.1.2 Hetero-Oxidant Action 4. The carbonyl : hydroxyl ratio of 20:2, as well as
In practice, the homo-oxidant treatment of graphene has so the absence of COOH ending groups are evidently
far been observed for the graphene hydroxylation only [30]. non-axiomatic and quite strongly depend on the pris-
The 21-fold GO II presented at the bottom of Figure 17.8 is tine graphene molecule size, as it will be shown in
well consistent with the reported behavior of graphene oxide Section 17.5.1. However, the dominance of carbonyls
[30] and can be considered as a reliable model of the oxide in the circumference area of GO molecules should
structure. In all other cases, graphene oxidation predomi- be expected and seems realized in practice.
nantly occurs in the multi-oxidant media [3] so that the final
products might present hetero-oxidant GOs. As suggested
[27], products synthesis can be traced in such a way. Basing 17.4.2 Second-Stage Oxidation of
the ACS-governing algorithm, the first-attack target atom the (5,5) NGr Molecule
is determined and successively subjected to the addition of
either O or OH and COOH oxidant. GO, which meets the The data presented in Section 17.4.1, even once limited to
requirement of the largest PCE criterion, is selected, and its the first steps of oxidation, convincingly show how compli-
ASC map is used to determine the second-step attack atom. cated is the chemical modification of graphene in this case.
Afterward, the procedure of the successive addition of the Surely, the process does not stop at this level and continues
three oxidants is repeated, and new GO with the largest PCE further involving both edge atoms, if they are still not valence
is chosen. Its ACS map shows the target atom for the next compensated, and basal ones thus becoming more and more
attack and so forth. Figure 17.9b exhibits PCE plottings versus complicated.
step number for the three oxidants. Comparing the data with
those related to a homo-oxidant action shown in Figure 17.9a 17.4.2.1 Oxidation of Hydroxyl-Framed GOs
one can conclude the following: Figure 17.11 summarizes structural results typical for the
second-stage homo-oxidant reaction related to a complete
hydroxylation of the pristine (5,5) NGr molecule. The 74-fold
1. Plottings related to each of the considered oxidants GO VII shown in the figure corresponds to the oxide equi-
are quite different in the two cases. This is a conse- librium structure after termination of the 52nd step of the
quence of the difference of the chemical composition second-stage reaction. All additions occurred at the basal
at the edge atoms during homo- and hetero-oxidant plane and were considered for up and down hydroxyls posi-
action at the current step. tions. The GO best configuration was chosen according to
(a) (b)
(c) 0 10 20 30 40 50 60
50 Step number
Per step coupling energy (kcal/mol) ND* 3, e
3
0
2
–50
–100 1
–150
–200
FIGURE 17.11 (a) Equilibrium structure of GO VII after 52 steps of the second-stage homo-oxidant reaction. (b) Core structure of GO
VII. (c) Per step coupling energy versus step number for GO VII family in the course of the first-stage (1) and second-stage (2) oxidation.
(Adapted from E.F. Sheka and N.A. Popova. Phys. Chem. Chem. Phys. 2013, 15, 13304.) Top: Evolution of the total number of effectively
unpaired electrons of GO VII during the second-stage oxidation.
the largest PCE criterion [27]. Dense covering of the basal addition to the molecule edge atoms in comparison with either
plane by hydroxyls has resulted in a considerable distortion the double additions to the edge atoms or single addition to
of the carbon skeleton structure (Figure 17.11b) due to the atoms of the basal plane.
reconstruction of flat benzenoid-packed configuration by cor- The plotting on the top of Figure 17.11c exhibits the evolu-
rugated cyclohexanoid-packed one. A large family of cyclo- tion of the correlation of the (5,5) NGr odd electrons in the
hexane isomorphs may explain a fanciful character of the course of the oxidation presented by changing the molecu-
skeleton structure. In contrast to the (5,5) NGr hydrogenation lar chemical susceptibility ND. The plotting shows that as the
[22], it was unable to distinguish particular kinds of cyclohex- number of the attached hydroxyls increases, the number of
ane isomorphs like AC and boat-like that were observed at the effectively unpaired electrons, a total of 17.62e for the pristine
hydrogenation of two-side accessible free-standing molecular GO II, gradually decreases, thus highlighting gradual deple-
membrane. None can exclude that the cyclohexane isomorph tion of the molecular chemical ability. At the 52nd step, ND
definition is not suitable for the chemically modified graphene becomes zero and keeps the value afterward, which means
that offers a limitless number of possible accommodation to stopping the reaction at this step.
minimize energy losses.
Evolution of PCE in due course of the 52-step reaction 17.4.2.2 Oxidation of Hetero-Oxidant-Framed GOs
at the second stage is shown in Figure 17.11c. As seen, the Equilibrium structure of the final configuration of the 43-fold
oscillating behavior observed for PCE during the first stage GO X obtained at the 21st step of the second-stage subse-
of oxidation is retained in this case, as well. When comparing quent hetero-oxidant treatment is shown in Figure 17.12a. The
the data with the PCE plotting related to GO II (curve 2), one final structure involves 20 carbonyls terminating the molecule
can see a drastic difference in the absolute PCE values in the edge atoms; 4 hydroxyls doing the same job, 2 of which are
two cases. Actually, the average values for GO II and GO VII added during the second stage of the oxidation; and 15 epoxy
constitute −102.15 and −44.79 kcal/mol, which indicate two- and 4 hydroxyl groups randomly distributed over the basal
and-half strengthening of hydroxyl coupling at their single plane. A particular attention should be given to a considerable
(a) GO X (b)
(c) 0 5 10 15 20 25
0
2
–80
–160
–240
Step number
FIGURE 17.12 (a) Equilibrium structure of GO X obtained in the course of hetero-oxidant second-stage oxidations. (b) Core structure of
GO X. (c) Per step coupling energy versus step number for GO X family under O (curve 1)- and OH (curve 2)-additions in the course of the
second-stage oxidation. Curve 3 plots the same data for GO V family during the first-stage hetero-oxidant reaction.
curving of the carbon skeleton (Figure 17.12b) caused by the the second zone involves all basal atoms as well
sp2–sp3 transformation of the carbon valence electrons due as those edge atoms with much lower ACS whose
to chemical saturation of their odd electrons. The structure chemical reactivity was not suppressed in the course
transformation is similar to that one observed under one-side of the first-stage reactions.
hydrogenation of (5,5) NGr [22], albeit not so drastic. The 2. The first-stage reaction concerns 22-step framing of
considered configuration well suits real conditions, when one- the (5,5) NGr molecule by oxidants among which
side accessibility of the basal plane is the most favorable due addressed oxygen atoms, hydroxyls, and carboxyls.
to a top-down manner of graphite layer exfoliation [3,31]. The obtained fully framed GOs involve both homo-
The PCE evolution, which accompanies the second-stage oxidant and hetero-oxidant GOs. All the GOs cor-
oxidation of GO V, is presented in Figure 17.12c. As seen in respond to one-point contact of oxidants with the
the figure, the O addition dominates and is characterized by molecule edge atoms thus providing the formation
an average PCE of −48,35 kcal/mol. The average PCE char- of C═O, C─OH, and C─COOH ending groups.
acteristic for hydroxyls constitutes −38,67 kcal/mol. For com- Table 17.2 summarizes average PES and oxidant-
parison, the average PCE for carbonyl–hydroxyl framing of induced chemical bonds data that characterize the
GO V, presented by curve 3, is −133,76 kcal/mol. obtained GOs.
3. The energy needed for each step of the second-
17.4.3 Main GO Characteristics and Properties stage oxidation decreases about three times for all
the oxidants indicating that the oxidant coupling
Based on the Computational Experiment
with the basal atoms is much weaker in compari-
The discussed computational oxidation of the (5,5) NGr mol- son with the interaction occurred at edge atoms.
ecule shows the following: The latter is supported by the elongation of all the
second-stage formed C─OH bonds and the forma-
1. Two-zone chemical reactivity of the graphene mole- tion of long O─C bonds of epoxy groups in com-
cule causes the two-stage character of the oxidation. parison with rather short C═O bonds of ending
The first reaction zone covers circumference area carbonyls. Two-stage character of graphene oxida-
that includes edge atoms with high ACS values while tion was confirmed experimentally [32].
TABLE 17.2
Average Characteristics of the (5,5) NGr Molecule Oxidation
PCE (kcal/mol) C─O Bonds Lengths (Å)
GOs a O OH COOH C─OH O─C─O C═O C–H Bonds (Å)
First-Stage Oxidation
I −135,065 – – – – 1,234 –
II – −102,15b – 1,372 – 1,249c 1, 101c; 0,971
III – – −77,34 1,361 – 1,233 0,972
V −133,14 −101,08 −82,77 1,348 – 1,233 0,978
Second-Stage Oxidationd
VII (IV) e – −44.79 – 1,371 – – 0,969
1,417
X (V) −48,35 −38,67 – 1,413 1,436 1,223 0,971
a Oxides under the corresponding numbers are presented in Figures 17.8, 17.10, and 17.12.
b The value is slightly overestimated due to one hydroxyl dissociation in due course of the reaction.
c Solitary C═O and C–H bonds, see Figure 17.8.
d The PCE values are averaged over steps of the second stage only.
e Here and below figures in brackets point the reference to GOs of the first-stage reaction.
4. The obtained results allowed suggesting a reliable valence angles between the corresponding C─C bonds and
vision of the GO composition that meets the require- the bond lengths are noticeably different in the two cases, the
ment of the top-down oxidation and is energetically structure of the carbon skeleton of the pristine (5,5) NGr mol-
profitable: GO X presented in Figure 17.12a seems ecule loses its flatness and becomes pronouncedly distorted.
to be one of most reliable model structures for GOs Comparing the pristine bond length diagram with those
obtained by one of the versions of the Hummers belonging to a current oxidized species makes it possible to
method [3]. Computations have shown that carbo- trace changes in the molecule skeleton structure.
carboxyl contacts are the least probable at both the Figure 17.13a and b demonstrate the transformation of the
first and the second stages of oxidation so that if even skeleton structure in the course of both first-stage and second-
present they constitute a definite minority in the GO stage oxidation exemplified by the changes within a fixed set
final structures. of the C─C bonds of homo-oxidant hydroxylated GOs II and
5. Concerning GO morphology, it should be noted that VII. As seen in Figure 17.13a, the OH framing of GO II causes
the performed calculations have not revealed either a definite regularization of the C─C bonds distribution. None
regularly structured GOs or even those close to regu- of single C─C bonds is formed since the chemical activity of
larly structured ones. Moreover, the GO carbon skele- edge atoms is not completely suppressed due to which sp3–sp2
ton has lost its planarity at the early stage of oxidation transformation does not occur. Contrary to this, the second-
and becomes remarkably curved when the oxidation stage oxidation is followed by the formation of single C─C
is terminated. Both features are due to sp2 –sp3 trans- bonds that evidently prevail in Figure 17.13b thus exhibiting
formation of the carbon electron system and the sub- a large-scale sp3–sp2 transformation. The appearance of the
stitution of flat benzenoids of graphene by corrugated elongated C─C bonds, the number of which increases when
cyclohexanoids of GOs. A large variety of cyclohex- oxidation proceeds, is naturally expected. However, to keep
anoid structures greatly complicates the formation of the skeleton structure integrity, this effect, as well as changes
regularly structured graphene derivatives, in general, in valence angles, should be compensated. At the level of
which is why the regular structure of graphene poly- bonds, this compensation causes squeezing a part of pristine
derivatives should be considered as a very rare event. bonds. The plotting in Figure 17.13b shows seven C─C bonds
So far, it has been observed for a particularly arranged that are extremely short and whose lengths are in the interval
graphene hydride named graphane [22]. of 1.345–1.355 Å. These lengths point to nonsaturated valence
state of the relevant carbon atoms, which are not attached by
oxygen, on one hand, however, on the other hand, are under
17.4.4 Oxidation-Induced Structure Transformation
the critical Rcrit value of 1.395 Å that put a lower limit for
The stepwise oxidation is followed by the gradual substitution observation of the odd electrons correlation. Under this value,
of the sp2-configured carbon atoms by sp3 ones. Since both the bond length provides a complete covalent bonding of odd
(a) 1.6
TABLE 17.3
1.5 Average Bond Lengths of the (5,5) NGr Oxides and
Bond length (Å)
Their Dispersion (%)

1.4 C─O Bonds Lengths (Å)
GOsa C─OH O─C─O C═O
1.3
First-Stage Oxidation
I – – 1,234
1.2
(5.30; −0,90)
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61
II 1,372 – 1,249b
Bond number
(0.70; −0,30)
(b) 1.6 III 1,362 – 1,233
(3.79; −0.99) (0.58; −0.47)
1.5 V 1,348 – 1,233
Bond length (Å)
(0.43; −0.43) (1.04; −0.83)

1.4 Second-Stage Oxidation
VII (IV)c 1,371 – –
1.3 (0.32; 0.18)
1,417
(1.98; −1.68)
1.2
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 X (V) 1,413 1,436 1,223
(1.08; −0.46) (0.24; −0.46) (0.63; −0.30)
Bond number
(c) 1.6 Source: E.F. Sheka and N.A. Popova. Phys. Chem. Chem. Phys. 2013,
C–O(H) bonds 15, 13304.
a Oxides under the corresponding numbers are presented in Figures
1.4
17.8, 17.10, 17.11a, and 17.12b.
Bond length (Å)
b A solitary C═O bond, see Figure 17.9.
1.2 c Here and below the figures in brackets point the reference to GOs of
the first-stage reaction.

C–H bonds
1
Edge atoms Basal plane atoms

0.8
basal plane atoms are longer than those related to OH fram-
0 20 40 60 80 100 120 140 160 ing of the molecule during the first stage of oxidation (1.417
Bond number and 1.371 Å, respectively). Important to note that previously
short C─OH bonds related to edge atoms elongate during the
FIGURE 17.13 The sp2 –sp3 transformation of the (5,5) NGr mole- second stage of oxidation as well and approach the average
cule skeleton structure in the course of the successive homo-oxidant length of 1.417 Å. At the same time, parts of these bonds,
oxidation; the C–C bond length distribution after (a) the first-stage which remain untouched during this stage, preserve their
oxidation, GO II and (b) the second-stage oxidation, GO VII. (c) shorter lengths.
Light gray histogram is related to the pristine molecule. Oxidant The transformation of the molecule skeleton structure in the
given C–O(H) and C–H bonds at the first-stage (black) and the sec-
case of the hetero-oxidant reaction is shown in Figure 17.14.
ond-stage (gray) oxidation.
Figure 17.14a presents the skeleton distortion of GO V by
the end of the first-stage oxidation. Since 20 carbonyls and
electrons due to which these electrons are noncorrelated so 2 C─OH groups form the molecule framing, the dominant
that the total number of effectively unpaired electrons ND is elongation of C─C bonds due to sp3–sp2 transformation pro-
zero [2]. The formation of these short bonds explains stopping vided by carbonyls is clearly seen. Two C─OH groups leave
of the oxidation reaction at the 52th step of the second-stage the corresponding edge atoms in the sp2 configuration. The
reaction. second-stage oxidation concerns 2 edge and 38 basal atoms.
Figure 17.13c presents the bond length distribution related Figure 17.14b exhibits the skeleton distortion by the 21st step
C─OH and O─H oxidant groups. Black plotting is related of the second-stage oxidation for GO X. The dominant major-
to GO II and describes the bonds of framing atoms while ity of bonds become single due to the large-scale sp3–sp2
gray plotting shows the similar distribution in the case of the transformation. The remaining untouched carbon atoms form
74-fold GO VII. As seen in the figure, the bonds are quite bonds of 1.343–1.358 Å in length similarly to the case shown
regularly distributed in both the cases. The relevant disper- in Figure 17.13b. Reasons for these bonds creation are dis-
sion of the bonds is listed in Table 17.3. As seen in the fig- cussed above. As previously, the bonds length provides a com-
ure, all the C─OH bonds related to hydroxyl attached to the plete covalent bonding of the remaining odd electrons thus
(a) 1.6 after terminating of the second-stage oxidation for GO X at

the 21st step. As seen, previously much shorter C─OH bonds
1.5 of 1.348 Å lengthen up to 1.413 Å and are comparable by
length with all the newly formed C─OH bonds; C═O bonds
Bond length (Å)
differ slightly, and C─O─C bonds of epoxide groups are

1.4
added. The average length values are listed in Table 17.3
alongside with the bond length dispersion. As seen in the
1.3 table, the bond distributions are quite homogeneous, and their
dispersion does not exceed 1%.
1.2
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61
Bond number 17.4.5 Some Comments Concerning GO Reduction
(b) 1.6 The reduction of massively produced GOs is the second
chemical reaction in the way of the transformation of dis-
1.5 persed graphite into a powder consisting, in an ideal case, of
one-layer graphene flakes. Looking at GO VII (Figure 17.11a)
Bond length (Å)
and GO X (Figure 17.12a) as reliable models of GO in the case

1.4
of homo- and hetero-oxidant processes and aiming at their
returning to the graphene molecule shown in Figure 17.2a,
1.3 one should answer the following questions:
1.2 1. Is it possible to return a drastically deformed and

1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 curved carbon skeleton structures of GO VII and GO
Bond number
X to the planar structure of the (5,5) NGr molecule?
(c) 1.6 2. Is it possible to take out all the oxidant atoms from
C–O–C and C–O(H) bonds GO VII and GO X making them free of dopants?
1.4
Core structures X and Xfr shown in Figure 17.15 correspond
Bond length (Å)
C O bonds
to GO X subjected to optimization after removing all the oxi-
1.2
dant atoms, in the first case, and oxidant atoms attached to
C–H bonds
1 GO core
Edge atoms Basal plane atoms
0.8
0 10 20 30 40 50 60 X
Bond number
FIGURE 17.14 The sp2 –sp3 transformation of the (5,5) NGr mole-
cule skeleton structure in the course of the successive hetero-oxidant
oxidation. The C─C bond length distribution after (a) the first-stage
oxidation, GO V and (b) the second-stage oxidation, GO X. (c) Light Xfr
gray histogram is related to the pristine molecule. Oxidant given
C═O, C─O─C, C─OH, and O─H bonds at the first-stage (black)
and the second-stage (gray) oxidation.
conserving sp2 configuration of the relevant carbon atoms and VII

terminating further oxidation. The appearance of shortened
C─C bonds in GO VII and GO X, lengths of which are less
than the critical value of 1.395 Å, is quite similar to that takes
place during the stepwise hydrogenation [25] and fluorination
[21] of fullerene C60.
Figure 17.14c discloses events that concern attached oxi- VIIfr
dants. As said previously, the first-stage oxidation is termi-
nated by a complete framing of the pristine molecule by 20
carbonyls and 2 hydroxyls. Consequently, the black curve in
Figure 17.14c exhibits 20 C═O bonds as well as 2 C─OH and FIGURE 17.15 Core structures of GO X (X and Xfr) and GO VII
2 O─H bonds of GO V. Gray curve presents the bond structure (VII and VIIfr) after (b) structure optimization (see text).
the molecule during the second stage of oxidation only, in the but play a significant role in the chemical behavior of the sub-
second. Skeleton structures VII and VIIfr in the figure demon- stance. The latter concerns the size-dependent chemical com-
strate the GO VII comes after similar removing concerning position of final products of chemical reactions, destruction of
OH groups. As seen in the figure, the optimization/reduc- graphene samples in the course of chemical reactions, chang-
tion of cores X (Figure 17.12b) and VII (Figure 17.11b) fully ing the character of mechanical behavior of graphene as well
restores the flat structure of the (5,5) NGr molecule. The dif- as of other topochemical reactions, and much more. Next, a
ference in the total energy of thus restored structures consti- few examples will be considered.
tutes 0%, 3%, and 1.1% the energy of the (5,5) NGr molecule
relating to cores X and VII. Therefore, a drastic deformation 17.5.1 Size Dependence of Chemical
of the GOs carbon skeletons is no obstacle for the restoration
Reactions Final Products
of the initial flat graphene pattern in spite of large deforma-
tional energy, namely, 874 and 1579 kcal/mol, incorporated As shown in the preceding sections, the circumference of
in the deformed X and VII core structures. This is an exhibi- any graphene molecule governs any chemical reaction deter-
tion of the extreme flexibility of the graphene structure that mining the reaction flowing from the very beginning. At the
results in a sharp response by structure deformation on any same time, the chemical reactivity of the area changes in the
external action, on one hand, and provides a complete restora- course of reaction due to a highly sensitive response of the
tion of the initial structure, on the other hand. molecule carbon skeleton structure to each act of chemical
It should be noted, however, that not pure graphene but slightly addition. In its turn, the structure reconstruction results in
oxidized product, called as rGO (reduced graphene oxide) (see a redistribution of C─C bonds over length thus promoting
References 3, 33–36 and references within), is obtained after the redistribution of the ACS over the molecule atoms. Due
the GO reduction in practice. As follows from the plottings in to high restructure ability, both progress of the reaction, and
Figure 17.13c, GOs are characterized by two regions of chemi- final chemical composition of its product obviously depend
cal bonding of oxidants with graphene body. While the edge on the molecule size. Intensively studied chemistry of gra-
atoms region should be attributed to that one of strong chemical phene oxides is a strong evidence for the correctness of this
bonding, the basal plane, for which the average coupling energy statement. Enough to mention ongoing debates about the
is three times less, is evidently related to the area of a weak chemical composition of the GO circumference area, con-
coupling. The finding is crucial for the GO reduction showing cerning the presence of carboxyl units, in particular. One of
that the latter concerns the basal plane in the first instance while the reasons for the question remaining unsolved until now is
oxidants located in the framing area may not be removed under that experimental reports on FTIR (Fourier transform infra-
conditions of the convenient reduction without destruction of red spectroscopy), XPS (x-ray photoelectron spectroscopy),
the carbon skeleton. This finding explains the residual oxygen Raman scattering, optical spectra, and so forth, presented by
content in reduced rGOs of 5%–10% [31, 37–39]. The equilib- different groups of chemists are not identical in spite of using
rium structure of core Xfr in Figure 17.15 may give a reliable similar technological procedures for the GOs production.
presentation of the product while core VIIfr may be attributed to The ambiguity may be understood if suppose that the final
the main product obtaining during reduction of the hydroxylated product in each case presents dispersion of GOs differing by
graphene [30]. Recent extended neutron scattering study con- the size and chemical structure. In its turn, size distribution
vincingly confirmed the topochemical nature of large polyvari- profiles of dispersions are obviously sensitive to the reaction
ance of both GO and rGO products [40–42]. conditions that might be different. It is obvious that what is
said can be attributed to the reduced graphene as well. As
for the latter, the optical spectroscopy of graphene quantum
17.5 HOW DOES CHEMICAL MODIFICATION dots, which present dispersions of rGOs, gives a straight evi-
OF GRAPHENE MOLECULES AFFECT dence of the structural and electronic inhomogeneity of the
dots (see review [44] and references therein).
ELECTRONIC STRUCTURE?
Figure 17.16 presents results of quantum chemical calcu-
The physical impact of the graphene chemical modification has lations performed for the current chapter. The equilibrium
been largely discussed (see [4,39] and references within). The structure of the (5,5) rGO molecule shown in panel (a) has
main issues concern opening the gap in the energy spectrum been discussed in Section 17.4.5. The structure of the (11,11)
of graphene crystal and peculiar edge atom states in the vicin- rGO molecule in panel (b) was obtained in the course of step-
ity of the Fermi energy. The first is caused by the violation of wise hetero-oxidant addition of three oxidants to the pristine
the crystal planarity due to the sp2 –sp3 transformation of car- (11,11) NGr molecule (see Figure 17.17a) that is twice and four
bon atoms. The second, according to Hoffmann [43, p. 101] is times larger than the former one by linear size and atom num-
“the origin of the reactivity of the terminated but not passivated ber, respectively. At each reaction step, the choice in favor of
point, line, or plane” but “may be mistaken for real states in the oxidant, which contributes to the final product, was made on
band gap, important electronically.” In spite of physical con- the basis of the largest PCE, as was considered in Section 17.4
sequences, the chemical nature of both effects is clearly vivid. in details.
Serious consequences caused by chemical modification are As seen in the figure, the chemical composition of the both
not limited to peculiar physical properties of graphene crystal molecules framing is quite different. Twenty carbonyls and
(b) oxidation cuts a large graphene sheet into pieces of ~1 nm in

(a) size [34]. Important that further prolongation of the oxidation
does not cause decreasing the size thus stabilizing them at the
1 nm level. This finding has allowed the author to put for-
ward a revolutionary idea about the real structure and origin
of the well-known natural allotrope of carbon—shungite—
and present it as loosely packed fractal nets of graphene-based
(rGO) quantum dots [45,46].
The feature is evidently connected with the role of the
(c)
100
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 edge zone for the chemical interaction. If there is any topo-
Step number
logical reason for distinguishing a piece of the carbon atomic
structure as an edge zone, strong chemical interaction with
0 addend immediately fixes the area thus creating conditions

for the zone enlarging due to anisotropy of the reconstructed
–100
odd electron cloud, on one hand, and mechanical stress, on
the other hand.
A high ability of the graphene molecule to follow this way
–200 is distinctly revealed via its ACS map. It should be reminded
that changes in the ACS maps image reflects the changes in the
–300
C─C bond length distribution caused by the relevant chemical
modification. Figure 17.17 presents a comparable view of the
FIGURE 17.16 Equilibrium structure of (a) (5,5) rGO and (b) map image changing caused by one-atom hydrogenation of
(11,11) rGO molecules obtained in the course of hetero-oxidant first- edge atoms of (5,5) NGr, (11,11) NGr, and (15,12) NGr mol-
stage oxidations. (c) Per step coupling energy versus step number for ecules. Carbon core of the latter is more than three and six
GO XII family under O (curve 1)-, OH (curve 2)-, and COOH (curve times bigger than the former one by the linear size and atom
3)-additions. Histogram presents the PCE plotting that corresponds number, respectively. As seen in the figure, the hydrogen addi-
to maximum PCE values at each step and determines the formation tion created a vivid, highly anisotropic topological changing
of the (11,11) rGO final structure. of all the molecules. However, if in the case of the first mole-
cule, these changes do not seem to destroy the molecule, in the
two hydroxyls compose the framing in the first case while case of the other two molecules, new topological channels of
30 carbonyls, 9 hydroxyls, and 5 carboxyls frame the second drastically reduced chemical activity for some atoms and sig-
molecule. The difference is well explained by PCE behavior nificantly enhanced on the other, extending through the mid-
in both cases (see Figures 17.9b and 17.16c). In all the cases, dle of the molecules, undoubtedly indicate a possible division
the PCE plottings of both molecules for all oxidants oscillate of the latter into two parts due to their chemical modification.
around a mean value. In spite of that the average PCE values Apparently, this feature of the ACS maps lays the foun-
differ considerably for the oxidants, large amplitudes of the dation of graphene molecules fragmentation during chemi-
plottings cause their intersection due to which carbonyls are cal reactions. The channels are characteristic of large
substituted by not only hydroxyls as in the case of the (5,5) molecule while small molecules can retain their integrity.
rGO molecule but also by carboxyls as well in the case of Unfortunately, the picture typical for hydrogen termination
(11,11) rGO. This substitution can be traced when following cannot be expanded over other chemical additions due to the
along the histogram in Figure 17.16c. Therefore, the oxidant extremely strong dependence of the ACS map image on the
composition of the circumference area of both GO and rGO chemical addend. This is well seen in Figure 17.8 showing
molecules depends on the molecule size. However, there are how significantly chemical addend influences the target-atom
serious reasons to expect the carbonyl contribution to be the ordering in the course of stepwise reactions. Similar effects
largest by both absolute and relative values while the carboxyl are seen in Figures 17.4 and 17.5 exhibiting a considerable
one to be relatively much smaller and absolutely the least. reconstruction of the C─C bond distribution at each addi-
tion of one hydrogen atom. The odd electrons correlation,
17.5.2 Chemically Induced Destruction which provides the collective behavior of the latter, forms the
grounds for the feature. The latter makes difficult a prediction
of Graphene Samples
of the molecule fragmentation at different chemical reactions.
The chemistry of graphene oxidation has revealed one more Thus, Figure 17.18 presents the ACS image maps of the (5,5)
size-dependent effect. As follows from reviews [3,29,35,37], NGr and (11,11) NGr molecules subjected to complete first-
the size of chemically produced GO is always less than that of stage oxidation. As seen from Figure 17.18b, the molecule
the pristine graphite. Moreover, the GO size reduces when O: edge hydroxylation that, apparently, was to be similar to mon-
C at.% ratio increases. As shown, oxidation causes the destruc- atomic hydrogen termination shown in Figure 17.17b, leads
tion of the pristine graphite and graphene sheets just cutting to absolutely different ACS map not showing any indication
them into small pieces [34,38]. Thus, 900 s of continuous of topological changes over the molecule carbon skeleton. No
Bare edges H-Terminated edges

(a) (b)
6 6
1.3 0.4
4 4
1.1
0.35
2 0.9 2
Y axis (Å)
0.3
Y axis (Å)
0 0.7 0
0.25
0.5
–2 –2 0.2
0.3
–4 –4 0.15
0.1
–6 –6
–6 –4 –2 0 2 4 6 –6 –4 –2 0 2 4 6
X axis (Å) X axis (Å)
(c) (d)
10 1.3 10
1.15 0.45
5 1 5
0.85 0.35
Y axis (Å)
Y axis (Å)
0 0.7 0
0.25
0.55
–5 0.4 –5 0.15
0.25
–10 –10 0.05
–10 –5 0 5 10 –10 –5 0 5 10
(e) 15 (f )
25
10 0.5
1.6
1.4 20
5 0.4
1.2
Y axis (Å)
Y axis (Å)
15
0 1 0.3
0.8
10 0.2
–5 0.6
0.4 5 0.1
–10
0.2
0
–15
–15 –10 –5 0 5 10 15 5 10 15 20 25 30 35
FIGURE 17.17 ACS image maps of the bare (left column) and terminated by hydrogen atoms per each edge atom (right column) of (5,5)
NGr (a and b), (11,11) NGr (c and d), and (15,12) NGr (e and f) molecules.
similar indications follow from the ACS map of the (11,11) time-consuming now. Some new algorithms might be needed
rGO molecule that corresponds to the (11,11) NGr molecule to perform the job.
subjected to the hetero-oxidant first-stage oxidation. Since all
the studies show that the odd electrons form an easily change- 17.5.3 Chemically Affected Topological
able, live system and since the mentioned topological channels
Mechanochemistry of Graphene
are obviously created in the course of a collective process, it
is difficult to predict at which, namely, step of reaction, say, Graphene molecule is a very interesting topological object
oxidation, the disruption of graphene molecule will start. It is [47,48], and its chemical modification allows for making the
obvious that computationally the problem solution is greatly feature more bright [49]. Next, we will consider a particular
(a) Bare edges (b) Terminated edges
6 6
1.3
4 4
1.1 0.36
2 0.9 2
0.32
Y axis (Å)
Y axis (Å)
0 0.7 0
0.28
0.5
–2 –2
0.3 0.24
–4 0.1 –4
0.2
–6 –6
–6 –4 –2 0 2 4 6 –6 –4 –2 0 2 4 6
(c) (d)
10 1.3 10
1.15 0.7
5 1 5 0.6
0.85 0.5
Y axis (Å)
Y axis (Å)
0 0.7 0 0.4
0.55 0.3
–5 0.4 –5 0.2
0.25 0.1
–10 –10
–15
–10 –5 0 5 10 –10 –5 0 5 10
FIGURE 17.18 ACS image maps of the bare (left column) and framed (right column) (5,5) NGr (a and b) and (11,11) NGr (c and d) mol-
ecules. The framed molecules are presented by (5,5) rGO (see VIIfr in Figure 17.15) and (11,11) rGO in Figure 17.16b.
topological effect caused by the influence of both the load- respect to two deformational modes. Second exhibits a strong
ing direction and the graphene molecule edge termination dependence of the deformation on the chemical composition
on the inherited topology of the molecule. As turned out, the of the molecule edge atoms. As seen in the figure, the defor-
graphene deformation under external mechanical loading is mation behavior is the most complex for the naked molecule.
extremely sensitive to the state of the edge atoms and makes The mechanical behavior of the (5,5) NGr molecule is similar
it possible to disclose a topological nature of this sensitiv- to that of a tricotage sheet when either the sheet rupture has
ity. Consider the example of the effect of uniaxial tension both commenced and completed by the rupture of a single
graphene that is a complicated mechanochemical reaction stitch row (AC mode) or rupture of one stitch is “tugging at
[46,47]. thread” the other stitches that are replaced by still elongated
Figure 17.19 presents the (5,5) NGr molecule, prepared for one-atom chain of the carbon atoms (ZZ mode). In the former
uniaxial tension in two orthogonal directions that correspond case, the deformation is one stage and is terminated on the
to the ZZ and AC modes of deformation. Straight lines exhibit 17th step of the deformation. In contrast, the ZZ mode is mul-
mechanochemical internal coordinates fixed by black atoms tistage and consists of 250 consequent steps with elongation
in the pristine molecule. The deformation occurs as a stepwise of 0.1 Å at each step [50,51].
elongation of these coordinates under fixed increment. Details As turned out, the character of the deformation strongly
of the reaction description are given elsewhere [46,47]. depends on the chemical situation at the molecule edges. As
The equilibrium structures of the (5,5) NGr molecule seen in Figure 17.20b, the addition of one hydrogen atom to
before and after uniaxial tension, which was terminated by the each of the molecule edge atoms does not change the general
rupture of the last C─C bond coupling two fragments of the character of the deformation: it remains a tricotage-like one
molecule, are shown in Figure 17.20. Looking at the picture, so that there is still a large difference between the behavior
two main peculiarities of the molecule deformation should be of ZZ and AC modes. At the same time, the number of the
notified. First concerns the anisotropy of the deformation with deformation steps in the first case reduces to 125.
Zigzag mode Armchair mode molecules, the dynamic parameters of the stretching C─C
F6 F6 vibrations of the units are mainly responsible in the case
F5
of the uniaxial tension. Change in Young’s modules means
F5
change in the force constants (and, consequently, frequencies)
F4 F4 of these vibrations. The latter are attributed to the G-band
F3 of graphene that lays the foundation for a mandatory test-
F3
ing of any graphenium system by the Raman spectroscopy. In
F2
F2 numerous cases, the relevant band is quite broad, which might
F1 F1 indicate that the chemical modification of the edge zone of
the graphene objects under investigation, may by one of the
reasons of the band broadening.
FIGURE 17.19 Six mechanochemical internal coordinates of the
Since the deformation-induced molecule distortion mainly
uniaxial tension of the molecule (5,5) NGr for two deformation
modes. F1, F2, F3, F4, F5, and F6 are forces of response along these concerns the basal atoms, so drastic changes in the deforma-
coordinates. Black atoms fix the coordinates starting lengths. tion behavior point to a significant influence of the chemi-
cal state of the edge atoms on the electronic properties in the
Even more drastic changes for this mode are caused by basal plane. The observed phenomenon can be understood if,
the addition of the second hydrogen atoms to the edge ones suggest that, (1) the deformation and rupture of the molecule
(Figure 17.20c). Still, the AC mode is quite conservative while are a collective event that involves the electron system of the
the ZZ one becomes practically identical to the former. The molecule as a whole; (2) the electron system of the graphene
tricotage-like character of the deformation is completely lost, molecule is highly delocalized due to extreme correlation of
and the rupture occurs at the 20th step. the odd electrons; and (3) the electrons correlation is topo-
Figure 17.21 presents a set of the “stress–strain” rela- logically sensitive due to which the chemical termination of
tions that fairly well highlight the difference in the mechani- the edge atoms strongly influences the behavior of the entire
cal behavior of all the three molecules. Table 17.4 presents molecule.
Young’s modules that were defined in the region of the elastic
deformation. As seen from the table, Young’s modules depend
17.6 CONCLUSIVE REMARKS
on the type of the edge atom chemical modification. As shown
in Reference 52, elastic properties of extended molecules such A “solid-and-molecule” dualism lays the foundation of the
as polymers [53] and nanographenes [54] are determined by graphene uniqueness, and chemical modification of graphene
dynamic characteristics of the objects, namely, by force con- is the best platform for exhibiting its molecular essence. As
stants of the related vibrations. Since benzenoid units provide follows from the molecular theory of graphene, due to rather
the determining resistance to any deformation of the graphene long C─C chemical bonds, odd electrons of carbon atoms of
Before deformation Zigzag Armchair
(a)
(b)
(c)
FIGURE 17.20 Equilibrium structures of the (5,5) NGr with different chemical modification of edge atoms before and after completing
tensile deformation in two modes of deformation. Bare edges (a); H1-terminated edges (b); and H2-terminated edges (c). White balls mark
hydrogen atoms. (Adapted from E.F. Sheka et al. J. Exp. Theor. Phys. 2011, 112, 602.)
(a) Zigzag mode Armchair mode

160
160
Stress σ (N/m2 × 109)

120
120
80 80
40 40
0 0
0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 0.0 0.1 0.2 0.3 0.4
Strain ε Strain ε
(b)
160 160

120 120
80 80
40 40
0 0
0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 0.0 0.1 0.2 0.3 0.4
Strain ε Strain ε
(c)
160 Stress σ (N/m2 × 109) 160
120 120
80 80
40 40
0 0
0.0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 0 0.1 0.2 0.3 0.4
Strain ε Strain ε
FIGURE 17.21 Stress-versus-strain dependences of tensile deformation of the (5,5) NGr molecule with different chemical modification in
two deformation modes. (a) Nake molecule; (b) H1-terminated edges; (c) and H2-terminated edges. (Adapted from E.F. Sheka et al. J. Exp.
Theor. Phys. 2011, 112, 602.)
TABLE 17.4
Young’s Modules for (5,5) NGr with Different Configuration of
Edge Atoms, TPa
Mode Bare Edges HI-Terminated Edges H2-Terminated Edges
Zigzag 1.05 1.09 0.92
Armchair 1.06 1.15 0.95
graphene molecules are correlated and electrons with different by Hoffmann [43]); second, any chemical modification of
spins are located in different sites of the space. The latter gen- graphene molecules is a reaction that produces a number of
erates the molecule radicalization with a peculiar distribution polyderivatives; third, the graphene polyderivative structure
of ACS over the molecule atoms. The spatial ACS map exhib- can be convincingly predicted in the course of calculations
its a “chemical portrait” of the molecule selecting atoms with following the ACS-pointer algorithms; and fourth, the odd
the largest ACS as targets for addition reactions. The ACS map electrons correlation explains extremely strong influence of
of naked graphene has a characteristic image showing that the small changes in the molecule structure on properties as well
graphene molecule of any size and shape is divided into two as sharp response of the molecule behavior on small action
zones that cover circumference and basal-plane carbon atoms, of external factors. Taken together, the theory facilities have
respectively, whose ACS values differ by four times in favor allowed for getting a clear, transparent, and understandable
of the former while all the atoms of the molecule are chemi- explanation of hot points of the graphene chemistry and sug-
cally active. Due to the latter, first, the molecular chemistry gesting reliable models of the final products such as chemi-
of graphene is the chemistry of dangling bonds (as termed cally produced and chemically reduced graphene oxides.
NOTES IN PROOFS 19. L. Gross et al. Science 2009, 325, 1110.

20. E.F. Sheka. Int. J. Quantum Chem. 2007, 107, 2935.
This chapter was written in 2013. Since that, the main issue 21. E.F. Sheka. J. Exp. Theor. Phys. 2010, 111, 395.
of practical molecular theory, presented in this chapter, has 22. E.F. Sheka and N.A. Popova. J. Mol. Model. 2012, 18, 3751.
not been changed and the governing role of chemical modifi- 23. D.C. Elias et al. Science 2009, 323, 610.
cation for the graphene science has been strengthened [55,56]. 24. R. Balog et al. Nat. Mater. 2010, 9, 315.
25. E.F.Sheka. J. Mol. Model. 2011, 17, 1973.
However, a drastic change has affected the theory grounds. As
26. E.F. Sheka, Adv. Quant. Chem. 2015, 70, 111–161.
shown recently [57], the theory features, laying the foundation of 27. E.F. Sheka and N.A. Popova. Phys. Chem. Chem. Phys. 2013,
criteria that reflect the electron correlation at the single-determi- 15, 13304.
nant level of the unrestricted Hartree–Fock approach (Section 28. H. He, T. Riedl, A. Lerf, and J. Klinowski. J. Phys. Chem.
17.2), are in fact the manifestation of spin-orbit coupling in par- 1996, 100, 19954.
ticular open-shell molecules one of which is graphene. The spin- 29. A. Lerf, H. He, M. Foster, and J. Klinowski. J. Phys. Chem. B
orbit story of graphene is still at the beginning while promising 1998, 102, 4477.
a number of fascinating conceptual discoveries. 30. L. Yan et al. Prog. Mater. Sci. 2012, 57, 1061.
31. J.I. Paredes et al. Langmuir 2009, 25, 5957.
32. Z. Xu et al. ACS Nano 2011, 5, 4401.
ACKNOWLEDGMENTS 33. J. Zhao, S. Pei, W. Ren, L. Gao, and H.-M. Cheng. ACS Nano
2010, 4, 5245.
The author is grateful to N. A. Popova and L. Kh. 34. S. Pan and I.A. Aksay. ACS Nano 2011, 5, 4073.
Shaymardanova for performing calculations. The finan- 35. K.-H. Liao et. al. ACS Nano 2011, 5, 1253.
cial support provided by the Ministry of Science and High 36. A.M. Dimiev, L. B. Alemany, and J.M. Tour. ACS Nano 2013,
Education of the Russian Federation grant 2.8223.2013 as 7, 576.
37. H. Wang and I.H. Hu. Ind. Eng. Chem. Res. 2011, 50, 6132.
well as the Russian Science Foundation grant 14-08-91376 are 38. S.V. Tkachev et al. Inorg. Mater. 2012, 48, 796.
highly acknowledged. 39. T. Enoki and T. Ando (Eds). Physics and Chemistry of
Graphene. Pan Stanford Publishing, Singapore, 2013.
40. E.F. Sheka et al. Nanosyst. Phys. Chem. Math. 2014, 5, 659.
REFERENCES 41. I. Natkaniec et al. J. Phys. Chem. C 2015, 119, 18650.
1. K.S. Novoselov et al. Nature 2012, 490, 192. 42. E.F. Sheka et al. Nanosyst. Phys. Chem. Math. 2015, 6, 378.
2. E.F. Sheka. Fullerenes. Nanochemistry, Nanomagnetism, 43. R. Hoffmann. Angew. Chem. Int. Ed. 2013, 52, 93.
Nanomedicine, Nanophotonics. CRC Press, Taylor & Francis 44. L. Li et. al. Nanoscale 2013, 5, 4015.
Group, Boka Raton, FL, 2011. 45. E.F. Sheka and N.N. Rozkova. Int. J. Smart Nano Mater. 2014,
3. D.R. Dreyer et al. Chem. Soc. Rev. 2010, 39, 228. 5, 1.
4. G. Lu et al. Nanoscale 2013, 5, 1353–68. 46. B.S. Razbirin et al. JETP 2014, 118, 735.
5. E. Davidson. Int. J. Quantum Chem. 1998, 69, 214. 47. E.F. Sheka and L.K. Shaymardanova. J. Mater. Chem. 2011,
6. I. Kaplan. Int. J. Quantum Chem. 2007, 107, 2595. 21, 17128.
7. E.F. Sheka. Int. J. Quantum Chem. 2012, 121, 3076. 48. E.F. Sheka. 2013. Topological Modelling of Nanostructures
8. K. Takatsuka, T. Fueno, and K. Yamaguchi. Theor. Chim. and Extended Systems. Series: Carbon Materials: Chemistry
Acta 1978, 48, 175. and Physics. Vol. 7, Eds. A.R. Ashrafi, F. Cataldo, A.
9. V.N. Staroverov and E.R. Davidson. Chem. Phys. Lett. 2000, Iranmanesh, O.Ori. Springer, Berlin, p. 137
330, 161. 49. E.F. Sheka. In Advances in Quantum Methods and
10. M.J. Benard, J. Chem. Phys. 1979, 71, 2546. Applications in Chemistry, Physics, and Biology. Series:
11. L. Lain, A. Torre, and D.R. Alcoba, Theor. Chem. Acc. 2011, Progress in Theoretical Chemistry and Physics, Vol. 27, Eds.
128, 405. M. Hotokka, E.J. Brändas, J. Maruani and G. Delgado-Barrio,
12. X. Gao et al. J. Phys. Chem. C 2008, 112, 12677. G. 2013. Springer, Berlin, p. 285.
13. V.A. Zayets. CLUSTER-Z1: Quantum-Chemical Software for 50. E.F. Sheka et al. J. Exp. Theor. Phys. 2011, 112, 602.
Calculations in the s,p-Basis. Institute of Surface Chemistry, 51. E.F. Sheka et al. J. Mol. Model. 2011, 17, 1121.
National Academy of Science of Ukraine, Kiev, 1991. 52. E.A. Nikitina et al. J. Phys. Chem. A 1999, 103, 11355.
14. E.F. Sheka. In Advances in Quantum Methods and 53. K. Tashiro, M. Kobayashi, and H. Tadacoro. Polym. J. 1992,
Applications in Chemistry, Physics, and Biology. Series: 24, 899.
Progress in Theoretical Chemistry and Physics, Vol. 27, Eds. 54. N.A. Popova and E.F. Sheka. J. Phys. Chem. C 2011, 115,
M. Hotokka, E. J. Brändas, J. Maruani and G. Delgado-Barrio, 23745.
G. 2013. Springer, Berlin, p. 249. 55. E.F. Sheka. In Advances in Quantum Methods and
15. K. Takatsuka and T.J. Fueno. J. Chem. Phys. 1978, 69, 661. Applications in Chemistry, Physics, and Biology. Series:
16. V.N. Staroverov and E.R. Davidson. J. Am. Chem. Soc. 2000, Progress in Theoretical Chemistry and Physics. Vol. 27. Eds.
122, 186. M. Hotokka, E.J. Brändas, J. Maruani and G. Delgado-Barrio.
17. M.J.S. Dewar and W. Thiel. J. Am. Chem. Soc. 1977, 99, 4899. 2013. Springer, Berlin, p. 249.
18. E.F. Sheka and V.A. Zayets. Russ. J. Phys. Chem. 2005, 79, 56. E.F. Sheka. Int. J. Quant. Chem. 2014, 114, 1079.
2009. 57. E.F. Sheka. arXiv 1511.05480 [physics.chem-ph].
Section III
Recent Progress
18 Low-Cost and Simple Method
for Graphene Synthesis
Isaiah Owusu Gyan, Haoyu Zhu, and I. Francis Cheng
CONTENTS
Abstract...................................................................................................................................................................................... 343
18.1 Introduction...................................................................................................................................................................... 343
18.1.1 Discovery of GUITAR......................................................................................................................................... 344
18.2 GUITAR Synthetic Route................................................................................................................................................. 344
18.2.1 Starting Material and Synthesis Conditions......................................................................................................... 344
18.2.2 Proposed Synthesis Mechanism........................................................................................................................... 346
18.3 Conventional Synthesis of Multilayer Graphene and HOPG........................................................................................... 346
18.4 Physical Characteristics of GUITAR............................................................................................................................... 347
18.4.1 GUITAR Has Grains of Nanosized Dimensions.................................................................................................. 347
18.4.2 GUITAR Is Nearly Pure sp2 Carbon.................................................................................................................... 347
18.4.3 GUITAR Layered with a Flat Surface Morphology............................................................................................. 349
18.4.4 Summary of Physical Characteristics................................................................................................................... 349
18.5 Electrochemical Characteristics of GUITAR................................................................................................................... 349
18.5.1 GUITAR Has High Anodic Stability from Lack of Electrolyte Intercalation..................................................... 350
18.5.2 GUITAR Has High Cathodic Potential Limits.....................................................................................................351
18.5.3 Electron Transfer and Density of Electronic States.............................................................................................. 353
18.6 GUITAR as a Dimensionally Stable Anode..................................................................................................................... 355
18.6.1 Degradation of Methylene Blue............................................................................................................................ 355
18.7 GUITAR Relative to Other Graphitic/Graphenic and Related Carbon Materials........................................................... 355
18.8 Future Studies and Applications of GUITAR.................................................................................................................. 356
18.8.1 Scanning Tunneling Microscopy.......................................................................................................................... 356
18.8.2 Ultracapacitor Development................................................................................................................................. 356
18.8.3 Fuel Cells and Flow Batteries............................................................................................................................... 356
18.8.4 Waste Water Treatment......................................................................................................................................... 356
18.9 Conclusion........................................................................................................................................................................ 357
References.................................................................................................................................................................................. 357
ABSTRACT more structural disorder in each plane than either graphene or

most graphites. The grains are nanocrystalline (La = 5 nm) as
In the search for a low-cost and simple method for synthesis indicated by the D and G bands at 1354 and 1594 cm−1, respec-
of graphene, a closely related material, GUITAR (graphenic/ tively, and ID/IG = 0.93. All GUITAR electrodes share excel-
graphitic material from the University of Idaho Thermolyzed lent electron transfer characteristics. This is in contrast with
Asphalt Reaction), was discovered. The synthetic method is ordered graphites and graphene (La = 100–1000 microns)
inexpensive, simple, rapid, and low temperature (600–800°C), electrodes. These studies strongly suggest that disorder gives
and does not require a crystalline template. Starting materials unrecognized beneficial properties that sets itself apart from
include any organic with boiling and melting points between graphene and ordered graphites. On the basis of GUITAR’s
80°C and 180°C, and sulfur, either in its elemental form or wide potential windows and facile electron transfer kinetics,
as a compound. Other starting materials have included roof- proposed applications which include water purification, ultra-
ing tars, and food products. The final products have visual capacitors, and redox flow batteries are discussed.
and microscopic characteristics that are similar to other forms
of graphene and graphite in that all are atomically flat and
18.1 INTRODUCTION
have layered 2D morphologies, yet recent results indicate that
GUITAR has different chemical and physical characteristics. Below is a summary of published studies on GUITAR (gra-
GUITAR, from all starting materials, shares the same physi- phenic/graphitic material from the University of Idaho
cal features. Raman spectroscopy indicates that GUITAR has Thermolyzed Asphalt Reaction).
343
Year Brief Description of Publication Reference

SEM. On the other hand, r-GO has a mottled, wavy surface.
Another related material, highly oriented pyrolytic graphite
2011 Discovery and physical characterization Cheng et al. (2011)
of GUITAR
(HOPG) has morphological features that include a flat surface
2012 The role of sulfur in the mechanism of Xie et al. (2012) with layered graphene planes. In determining as to whether
GUITAR formation the new material was r-GO paper or HOPG, we began to call
2013 Electrochemical characterization of Cheng et al. (2013) it GUITAR. In the physical and chemical characterization of
GUITAR in terms of electron transfer this new material, it was realized that GUITAR was neither
kinetics and potential windows r-GO paper nor HOPG, and has features that indicate that it is
in a class by itself.
18.1.1 Discovery of GUITAR 18.2 GUITAR SYNTHETIC ROUTE

In the spring of 2008, an interesting carbon film was
18.2.1 Starting Material and Synthesis Conditions
encountered during sample preparation of a Piceance Basin
(Colorado) oil shale for atomic absorption analysis. In an Sulfur is an important cofactor in the GUITAR process.
effort to remove the crude oil matrix, flame heat was applied Asphalt is a mixture of high molecular weight hydrocarbons.
to the sample in a crucible (Figure 18.1a). Upon cooling, it was Observations from our laboratories indicate that many start-
discovered that under conditions of partial enclosure, a film ing materials produced similar results. These include crude or
with a metallic sheen coated the interior of the ceramic cruci- refined oils, taco chips, and candy bars. Besides the hydrocar-
ble (Figure 18.1b). Careful removal of the material resulted in bons, another common feature is sulfur, which ranged from
flakes (Figure 18.1c). Because of its layered characteristics, it 0.1% to 5% (m/m) in these materials. Sulfur is important in
was surmised that the material was a form of graphite. Optical the growth of graphitic planes in carbon steel, and nanotubes,
microscopy indicates that the material has a flat and layered vulcanization, and in the dehydrogenation and dehydration
morphology. Figure 18.2a is that material in a water suspen- of organics and coal (Fu and Harvey 1978; Morrison 1984;
sion. The micrograph shows that it is flat and featureless. It is Morrison and Porter 1984; Double and Hellawell 1995; Ci
hydrophobic and tends to curl upon itself in aqueous solutions. et al. 2002; Kowata et al. 2002; Jusino and Schobert 2006;
The scanning electron micrograph (SEM) in Figure 18.2b Semmelhack et al. 2006; Vallés et al. 2006; Ghosh et al. 2008;
and c show the flat and layered characteristics of the material. Huang et al. 2008a,b; Bazdar et al. 2009; Palyanov et al.
The transmission electron micrograph (Figure 18.2d) dem- 2009). These processes all occur at relatively low tempera-
onstrates that this layered feature extends to the nanoscale. tures (100–300°C). The sum of these effects may indicate a
In many ways, the described material resembles reduced crucial role for sulfur in GUITAR formation (Figure 18.3).
graphene oxide (r-GO) paper (Poh et al. 2012). Comparison A deposit is evident in the left-hand crucible when reagents
of Figure 18.2c with the literature underscores this observa- consisted of 6.0 g of cyclohexanol with 5.0 g of sulfur. Physical
tion. Both have flat layered structures. In r-GO paper, these characterization indicates that it is a form of multilayer gra-
planes consist of sp2 hybridized carbon, that is, graphene. phene (Xie et al. 2012). When sulfur is absent, no graphenic or
An important difference is that GUITAR’s surface appears graphitic film was observed. Similar results are obtained with
to be almost atomically flat, that is, flat to the resolution of other hydrocarbons. Over 20 organics were tested in GUITAR.
(a) (b) (c)
01/12/
FIGURE 18.1 Discovery of GUITAR: (a) pyrolysis of oil shale, (b) metallic-like product in crucible after pyrolysis, and (c) a photograph
of a flake of GUITAR (approximately 25 mm in diameter) exfoliated from crucible in B. ((c) Reprinted from Carbon, 49, Cheng, I. F. et al.
Synthesis of graphene paper from pyrolyzed asphalt. 2852–2861, Copyright 2011, with permission from Elsevier.)
Low-Cost and Simple Method for Graphene Synthesis 345
(a) (b)
1 μm* WD = 4 mm Signal A = Inlens

Mag = 9.45 K × EHT = 1.00kV
(c) (d)
12.5 nm
200 nm WD = 5 mm Signal A = SE2
Mag = 33.08 K × EHT = 1.00kV
FIGURE 18.2 GUITAR: (a) an optical micrograph (400×) of GUITAR dispersed in water. (b) 9.45 K x SEM of microtomed layers on Si.
(c) 23.08 K x SEM showing layered characteristics. (d) A TEM showing layered characteristics on the nanometer scale. (Reprinted from
Carbon, 49, Cheng, I. F. et al. Synthesis of graphene paper from pyrolyzed asphalt. 2852–2861, Copyright 2011, with permission from
Elsevier.)
6.0 g cyclohexanol 6.0 g cyclohexanol

+ 5.0 g S only
(a) (b)
10/29/2009 500 μm
FIGURE 18.3 Evidence of sulfur in the synthesis of GUITAR: (a) A photograph of GUITAR film formation. Starting materials in the left
crucible 6.0 g of cyclohexanol and 5.0 g of sulfur, the right crucible omitted sulfur. (b) A photomicrograph of exfoliated GUITAR flakes
from the left crucible. (Xie, Y. et al. Sulfur as an important co-factor in the formation of multilayer graphene in the thermolyzed asphalt
reaction. J. Mater. Chem. 22, 2012: 5723–5729. Reproduced by permission of The Royal Society of Chemistry.)
Structural considerations did not follow any trends. The only in their pure state. The temperature was ramped at 10°C/min to
trend that is important is that the organic be in the liquid state 600°C. Pure substances boiled away at their expected boiling
between 100°C and 180°C. If the organic boiled off before points. When each is present, a sequence of reactions occurs
100°C, for example, ethanol, or if was solid under 180°C, for between 140°C and 240°C and 288–600°C. The curve drops to
example, anthracene, it failed. Addition of elemental sulfur to near-complete mass loss at 450°C; however, as observed previ-
organics was successful, as well as cases of sulphur-containing ously, this is low-yield reaction (Cheng et al. 2011).
compounds, for example, thiophenol (BP 169°C). In addition Final temperature affects Raman spectra of the TGA prod-
to optical and electron microscopies, X-ray photoelectron spec- uct as shown in Figure 18.4b. The characteristic G band at
troscopy (XPS) and atomic force microscopy (AFM) indicate 1594 cm−1 and D band at 1350 cm−1 were obtained with ter-
multilayer graphenic materials (Alliata et al. 1999; Banerjee mination at 600°C (Ferrari and Robertson 2000). Positions of
et al. 2005; Panigrahi et al. 2011). these bands indicate the material is crystalline to nanocrys-
Thermogravimetric analyses (TGA); Figure 18.4a shows talline, sp2 carbon using the Ferrari amorphization trajectory
TGAs of cyclohexanol (40 mg) with sulfur (1 mg) and with each (Ferrari and Robertson 2000). The ID/IG is 0.93 which lies
(a) 100 sulfur between room temperature and 140°C can be attributed
90 to evaporation; however, mass stabilizes between 140°C and
80 240°C. Between these limits, a series of reactions are possible
70 Element sulfur
in Figure 18.5. Sulfur dehydrates cyclohexanol to cyclohex-
ene in Step 1 (Silverwood and Orchin 1962; Bajus and Veselý
Weight loss (%)
60 Combination
1980; Wang et al. 2007). It then reacts with cyclohexene form-
50 140 Cyclohexanol
240 ing bridging polysulfide linkages followed by degradation
40 to a monosulfide (steps 2–3) (Morrison 1984; Morrison and
30 Porter 1984; Granata et al. 2003). Above 240°C, a sequence of
20 288 dehydrogenation steps continue up to 400°C (Fu and Harvey
170 350 450
10 1978; Jusino and Schobert 2006). At that temperature, the
0 1450 cm−1 intermediate forms. In the range 400–600°C, step-
0 100 200 300 400 500 wise dehydrogenation produces H2S and GUITAR. Failures
Temperature (°C) of anthracene, fluoranthene, and pyrene as reagents for the
GUITAR process are based on melting points that are too
(b) 1450
high to undergo sulfur cross-linking, that is, steps II and III
between 100°C and 140°C.
i
400°C
18.3 CONVENTIONAL SYNTHESIS OF
1444
MULTILAYER GRAPHENE AND HOPG
D 1350 G 1594 ii
Monolayer graphene is one atom thick sp2 hybridized carbon
Intensity
550°C
and represents the ideal form. On the other hand, GUITAR
D 1350 G 1594 iii
is multilayered, and is most analogous to r-GO paper and
HOPG. It is most appropriate to compare GUITAR synthesis
600°C
with these two carbon materials. The highly ordered form of
graphite, HOPG, is made by high-temperature decomposition
1000 1500 2000 2500 of gaseous hydrocarbons, for example, acetylene, followed
Wavenumber (cm–1) by hot pressing at high pressure (>10 GPa) and temperature
(>2000°C) (McCreery 2008).
FIGURE 18.4 (a) Thermogravimetric analysis of cyclohexanol Synthesis of the r-GO paper begins with the treatment of
(40 mg), sulfur (1 mg), and combination of cyclohexanol and sulfur. graphite in a mixture of solvents under strong oxidizing con-
The starting temperature is 23°C, with a 10°C/min ramp to 600°C ditions to create and disperse graphene oxide. The latter is
under N2 flow. Major inflection points are labeled. (b) Raman spec- subsequently reduced chemically, thermally (1000°C) (Poh
tra of products obtained from the TGA of cyclohexanol + sulfur run et al. 2012), and/or photolytically to produce r-GO (Li et al.
at (i) 400°C, (ii) 550°C, (iii) 600°C final temperatures with a ramp 2008; Tung et al. 2009; Dubin et al. 2010; Strong et al. 2012).
from room temperature at a rate of 10°C/min. (Xie, Y. et al. Sulfur as The product may contain a few residual oxygen species along
an important co-factor in the formation of multilayer graphene in the
the basal plane and edges, and the process includes intercala-
thermolyzed asphalt reaction. J. Mater. Chem. 22, 2012: 5723–5729.
tion of ions into the layers of graphite, forming graphite inter-
calation compounds (GICs) (Zhi and Mullen 2008; Eswaraiah
et al. 2011; Wassei and Kaner 2012). Characteristics of the
with the literature range for other multilayer and disordered
product depend on the oxidation bath used (Poh et al. 2012).
graphenes (Makarov and Han 2007; Pan et al. 2009; Wang
The chemical vapor deposition (CVD) of producing multi-
et al. 2009; Gao et al. 2010; Chen et al. 2010). Crystal grain
layer graphene results in layers that are randomly oriented
size based on the Tuinstra–Koenig equation is 5 nm similar
with respect to each other (Diaz-Pinto et al. 2012). This
to multilayer graphene prepared by other methods (Matthews
method requires expensive apparatus, and high temperatures
et al. 1999; Li et al. 2001; Subrahmanyam et al. 2008). At
(>1000°C) (Brownson et al. 2012), which affords selectivity in
400°C, an intermediate to GUITAR is evident by a single
substrates on which the product could be deposited. Solvent-
sharp peak at 1450 cm−1. At 550°C, D and G Raman bands,
based methods described above require as long as 96 hours
along with the intermediate appear. At 600°C, the 1450 cm−1
processing time (Guo et al. 2009), are prone to the danger of
Raman band intermediate is not evident, with only D and G
sudden increment in temperature and the formation of explo-
bands present.
sive chlorine dioxide gas (Poh et al. 2012), and eventually
result in noncrystalline products (Wassei and Kaner 2012).
18.2.2 Proposed Synthesis Mechanism
These disadvantages have called for a low-cost, simple
A hypothesis for the formation of GUITAR has been devel- method to synthesize graphene and graphenic analogs.
oped based on interpretation of the TGA and Raman spectra. Furthermore, graphite exfoliation, as a source of graphene
The initial loss of mass of cyclohexanol in the presence of will probably not be a reliable method for large-scale
OH
I S8
+ H2O
100–140 °C
II + S8
100–140 °C
(S)n
III
100–140 °C
(S)n S
+ S8
IV 1450 cm–1 Intermediate
S – H2S
240–400 °C
V 1450 cm–1 Intermediate

400–600 °C
FIGURE 18.5 The proposed overall scheme for GUITAR formation from sulfur and cyclohexanol based on thermogravimetric analyses
(see above). Steps I–III occur concurrently between 100°C and 140°C. Above 240°C a sequence of sulfur based dehydration steps occur to
give the 1450 cm−1 (as detected by Raman) intermediate. That intermediate rearranges to give GUITAR between 450°C and 600°C. (Xie,
Y. et al. Sulfur as an important co-factor in the formation of multilayer graphene in the thermolyzed asphalt reaction. J. Mater. Chem. 22,
2012: 5723–5729. Reproduced by permission of The Royal Society of Chemistry.)
graphene production for technological applications (Avouris and intensity ratio (I D/IG = 0.93), indicates 100% sp2 hybrid-
and Dimitrakopoulos 2012). ization with nanocrystalline features (Ferrari and Robertson
The GUITAR process is a form of CVD that affords a sim- 2000). The Tuinstra–Koenig relationship where I D/IG = C(λ)/
ple, low-cost method for synthesizing a graphenic analog, a La indicates that GUITAR has a grain size (La) of 5.3 nm
material that is very unique and demonstrates electrochemical (Matthews et al. 1999). On the other hand, CVD-grown
properties superior to graphene and other carbon materials. graphene has grain sizes of 250 nm (Huang et al. 2011a,b).
This process requires simple carbon precursors such as roof- HOPG has grain sizes that vary from 1 to 10 µm (Banks et al.
ing tar, casserole crucible, common laboratory apparatuses, 2005; Luning et al. 2011). Both CVD graphene and HOPG
and a synthesis temperature as low as about 600°C. In this have Raman G-bands with very little D-band characteristics,
process, little skill is required, no flow or restriction of an that is, I D/IG ≅ 0. r-GO papers have I D/IG characteristics very
inert or reactive gas is necessary, and there is no need for pres- similar to GUITAR (Li et al. 2001; Subrahmanyam et al.
sure control; however, bulk synthesis of the product can be 2008; Poh et al. 2012; Wang et al. 2012). The much smaller
achieved. The relatively low temperature has made it possible grain sizes of GUITAR and of GP indicate a more disor-
to deposit GUITAR on a wide range of substrates, including dered structure in these materials.
glass fibers, nanosprings, and diatomaceous earth (Figure
18.6) (Cheng et al. 2011). The ability to produce graphenic 18.4.2 GUITAR Is Nearly Pure sp2 Carbon
nanostructures is a unique characteristic of the GUITAR
process. The Fourier transform infrared analysis of GUITAR depos-
ited on germanium exhibits two absorption bands at 868 and
1599 cm−1 (Figure 18.7). The position of these bands, respec-
18.4 PHYSICAL CHARACTERISTICS OF GUITAR tively, matches those of the A1U out-of-plane and E1U stretch of
18.4.1 GUITAR Has Grains of the intralayer bonds of graphene (Reich and Thomsen 2004;
Gupta et al. 2006; Jiang et al. 2008; Kostic et al. 2009). On the
Nanosized Dimensions
other hand, there are no peaks that match the expected posi-
GUITAR exhibits characteristic D (1354 cm−1) and G tions for C–O or C=O stretches (1715–1740 and 1050 cm−1)
(1594 cm−1) Raman bands expected for graphitic materi- (Subrahmanyam et al. 2008; Guo et al. 2009). The Raman
als as seen in Figure 18.7 (Ferrari and Robertson 2000). and IR data indicate that GUITAR is mostly sp2 carbon, with
Ferrari’s amorphization trajectory, based on band position minimal oxides.
(a) (b)
10 μm 10 μm
10 μm EHT = 5.00kV Date: 20 Apr 2006
Mag = 700 X 20-30 10 μm* Signal A = SE2
WD = 7 mm Signal A = SE2 WD = 5 mm
Mag = 878 X EHT = 2.00kV
(c) (d)
200 nm 200 nm
200 nm* WD = 5 mm Signal A = SE2 200 nm WD = 6 mm Signal A = SE2

Mag = 24.47 K X EHT = 2.00kV Mag = 25.62 K X EHT = 5.00kV
FIGURE 18.6 GUITAR deposits on various substrates: SEM of untreated diatomaceous earth. (a) Au sputter coating was required to pre-
vent charging, diatomaceous earth (b and c) and silica nanosprings (d) coated with GUITAR. (Reprinted from Carbon, 49, Cheng, I. F. et al.
Synthesis of graphene paper from pyrolyzed asphalt. 2852–2861, Copyright 2011, with permission from Elsevier.)
(a) (b)
G 1593 80
D 1354
70
60
1599
50
Intensity
%T
40
30
20
10
868
0
2000 1500 1000
Wavenumber (cm–1) Wavenumber (cm–1)
FIGURE 18.7 (a) Raman spectrum for GUITAR acquired in ambient air using 532 nm excitation; G mode peak at 1593 cm−1, D mode peak
at 1354 cm−1 at maximum intensity. Intensity ratio (ID/IG) is 0.93. (b) IR spectrum of GUITAR on a germanium disk. The 868 and 1599 cm−1
bands correspond to the A1u and E1u intralayer graphene stretches. (Reprinted from Carbon, 49, Cheng, I. F. et al. Synthesis of graphene
paper from pyrolyzed asphalt. 2852–2861, Copyright 2011, with permission from Elsevier.)
(a) (b)
284.24 eV C 1s
C1s GUITAR
hν = 1253 eV
hν=1253 eV
Intensity (arb. units)

285.15 eV
O1s
N1s
286.15 eV
Si 2p
Si 2s
0 200 400 600 800 1000 1200 280 282 284 286 288
Binding energy (eV) Binding energy (eV)
FIGURE 18.8 (a) A wide XPS scan of cleaved GUITAR. (b) XPS spectrum of the C 1 s core level state of graphene. The peaks at 284.24,
285.15, and 286.15 eV are assigned to sp2 C-C, sp3 C-H or C=N sp2, and C-O sp3 or C-N sp2 respectively. (Reprinted from Carbon, 49, Cheng,
I. F. et al. Synthesis of graphene paper from pyrolyzed asphalt. 2852–2861, Copyright 2011, with permission from Elsevier.)
XPS reveals that GUITAR is nearly pure sp2 carbon (C 18.4.4 Summary of Physical Characteristics
1s, 284.2 eV) with minor peaks suggesting the presence
of oxygen (O 1s, 533.3 eV) and nitrogen (N 1s, 401.2 eV) GUITAR has these characteristics:
(Figure 18.8) (Cheng et al. 2011). The 284.2 eV C1 s peak
agrees well with the literature regarding the sp2 hybridized 1. Atomically flat, featureless surface; in this aspect,
carbon–carbon bond, and these include carbon nanotubes, it most resembles HOPG. SEM and AFM studies
graphite, graphene, and r-GO (Blyth et al. 2000; Estrade- determined this morphology.
Szwarckopf 2004; Jang et al. 2004; Arsat et al. 2008; Su 2. It has nanocrystalline grains of 5 nm as measured
et al. 2009; Wei et al. 2009). GUITAR’s elemental O/C ratio by Raman spectroscopy. This is closest to gra-
of 4% as calculated from XPS data is comparable to that of phene paper, and unlike HOPG- and CVD-grown
high-quality graphene (4.96%) (Brownson et al. 2011). graphene.
3. It is almost pure sp2 carbon. Both XPS and Raman
spectroscopy indicate this. The O/C elemental ratio
18.4.3 GUITAR Layered with a Flat of 4% is similar to high-quality graphene.
Surface Morphology 4. It has a layered structure consistent with graphite,
HOPG, and graphene paper, as indicated by SEM,
AFM of GUITAR is presented in Figure 18.9 and indicates a TEM, XRD, and optical microscopies.
graphitic structure with atomic flatness. Atomic-scale AFM 5. It is not a-C. The electron diffraction pattern along
indicates that GUITAR has interatomic distances of 0.148 nm with visual and SEM/TEM observations are all con-
matching that of sp2 carbon. The topographical pattern indi- sistent with this conclusion.
cates a graphitic surface. Remarkably, GUITAR is atomically
flat over several mm2. This is unlike graphene paper, which
has a mottled surface at the micron to submicron scale (Dikin 18.5 ELECTROCHEMICAL
et al. 2007).
CHARACTERISTICS OF GUITAR
Electron diffraction pattern of GUITAR is shown in
Figure 18.10. The point charges are an indication of a crystal Graphite, graphene paper, graphene, and GUITAR have simi-
lattice structure and clearly indicate that it is not amorphous lar morphologies in that they are layered sp2 carbon materi-
carbon (a–C), (since the diffraction pattern of the latter exhib- als. Furthermore, graphene, HOPG, and GUITAR are both
its diffuse ring patterns (McBride et al. 2003). The layered flat at the atomic level. As discussed above, GUITAR dif-
structure, as determined by SEM, TEM, and optical spectros- fers from HOPG and CVD graphene in that the grain size
copies, also precludes the possibility that GUITAR is a-C. is much smaller, that is, several nm versus several µm. The
Interlayer spacing (d002) deduced from Bragg’s law using data relatively small grain size and the defects between the grains
from XRD studies for GUITAR is 0.335 nm, similar to graph- in GUITAR give rise to electrochemical behaviors that distin-
ite and HOPG (McCreery 2008). guish it from all other forms of carbon electrodes.
(a)
mm/div
1.0 nm/div
0.0015
0.00050 mm/div
0.00050 mm/div
0.00 1.00 2.00 nm

(b)
Ra: 1.745 nm Height profile Rp: 0.6686 nm
Rq: 0.3387 nm Rv: –0.7733 nm
nm A A
2200 0.1485 nm
1800
1400
1000
0.00 0.40 0.80 nm
Cursor
Ad: 0.1457 nm
Ad: 0.1494 nm
Ad: 0.1496 nm
Ad: 0.1485 nm
FIGURE 18.9 (a) 2.2 nm × 2.2 nm AFM topography image of GUITAR in 2D and 3D. (b) Line analysis of topography showing the peri-
odicity of atomic structure. Interatomic distances in GUITAR are 0.148 nm, consistent with sp2 carbon. (Reprinted from Carbon, 49, Cheng,
I. F. et al. Synthesis of graphene paper from pyrolyzed asphalt. 2852–2861, Copyright 2011, with permission from Elsevier.)
18.5.1 GUITAR Has High Anodic Stability

from Lack of Electrolyte Intercalation
Voltammetric results indicate that GUITAR is as durable

as boron-doped diamond (BDD) anodes. Cyclic voltam-
metry (CV) in 1 M H2SO4(aq) in Figure 18.11 indicates that
significant anodic current onset does not begin until about
2.0 V versus Ag/AgCl for the GUITAR anode. During
the CV, we attribute the anodic current to water oxidation,
H2O → 2H+ + ½O2 + 2e− GUITAR exceeds the anodic lim-
its of HOPG and is competitive with BDD. (Gandini 2000;
Granger 2000; Iniesta 2001; Panizza et al. 2001; X. L. Cheng
2010). In Figure 18.11, anodic current onset is 1.6 V at 200 µA/
cm2 for pyrolytic graphite electrodes. Furthermore, GUITAR
exceeds the anodic limits of metal oxide and noble metal
dimensionally stable anodes (Jarzebski and Marton 1976; Hsu
and Ghandhi 1980; Comninellis and Pulgarin 1991; Pavlov
1992; Gattrell and Kirk 1993; Comninellis 1994; Pavlov and
Monahov 1996; Polcaro and Palmas 1997; Dominguez et al.
Camera length = 50 cm 1999; Pleskov 1999; Polcaro et al. 2000; Rao and Fujishima
2000; Zanta et al. 2000; Gherardini et al. 2001; Panizza and
Cerisola 2003; Fujishima et al. 2005; Panizza and Cerisola
2007).
FIGURE 18.10 An electron diffraction pattern of exfoliated Basis of anodic stability: Electrolyte does not intercalate
GUITAR. TEMs were conducted on a JEOL 1200 EX II TEM oper- into GUITAR. Murray and coworkers hypothesize that cor-
ated at 120 KV. The point charges are an indication of a crystal lat- rosion of HOPG is initiated by intercalation of aqueous elec-
tice structure. Amorphous carbon exhibits diffuse ring patterns. trolyte through the grain boundaries into subsurface layers
–100 –26,000
–0.64V (200 μA/cm2)
Basal plane pyrolytic graphite
μA/Cm2
400 (BPPG), 1M H2SO4 –11,000
1.68V (200 μA/cm2)
900 4000
–1.2 –0.8 –0.4 0 0.8 1.6 2.4
0.46 V
–100 0.22 V –26,000
Basal plane (BP) GUITAR,

μA/Cm2
400 1M H2SO4 –11,000

–1.1V (200 μA/cm2)
–1.9V (200 μA/cm2)
900 4000
–1.2 –0.8 –0.4 0 0.8 1.6 2.4
E (V) vs Ag/AgCl
FIGURE 18.11 Comparing cyclic voltammograms (cathodic and anodic) at GUITAR and pyrolytic graphite (PG). CVs were obtained
in 1 M H2SO4 at 50 mV/s with Ag/AgCl and graphite rod as reference and counter electrodes respectively. GUITAR extends ≈0.7 V total
window than PG in sulfuric acid. PG also exhibits electrolyte intercalation, indicated by the peak on the reverse scan of the anodic voltam-
mogram. Lack of electrolyte intercalation in GUITAR is a unique feature.
HOPG 5 μm GUITAR 5 nm
Micro- Inter-grain
grains defect region
Nano-crystal
grains
NO3– (300 pm)

ClO4– (350 pm)
SO4– (400 pm)
Grain—Porous gap—Grain Grain—Non-porous defect—Grain
FIGURE 18.12 Hypothesis: GUITAR resists intercalation of electrolytes from the presence of a nonporous intergrain defect region con-
sisting of nonhexagonal sp2 carbon. Left, HOPG has porous gaps between micron-size grains. The observed intercalating anions have their
hydrated radii indicated.
(Hathcock et al. 1995). Intercalation of electrolyte into HOPG together through nonhexagonal covalent links, that is, 5- and/
is detectable by a reductive current on the reverse sweep of an or 7-membered ring systems or (ii) there are interplanar bonds
anodic cyclic voltammogram (CV) where cathodic and anodic that control flow of electrolyte through its bulk. In our pre-
charge ratio (Qc/Qa) is greater than zero (Hathcock et al. liminary studies, KNO3, H2SO4, and HClO4 electrolytes were
1995). Figure 18.11 shows that the pyrolytic graphite electrode all observed to intercalate into pyrolytic graphite. This is con-
has this characteristic. On the other hand, GUITAR anodes sistent with McCreery’s studies with HOPG (Alsmeyer and
−
lack this behavior (see Figure 18.11). The Qc/Qa ratio is zero McCreery 1992). In that study, it was found that SO2− 4 , CIO 4 ,
−
in all voltammetric potential window studies of GUITAR. It and NO3 with hydrated radii of 400 pm or less were able to
is apparent that this material is unlike HOPG in that it does intercalate into HOPG and initiate corrosion. On the other
not allow electrolyte to intercalate through its basal plane. hand, H 2 PO −4 (r = 450 pm) did not intercalate.
With HOPG, electrolyte leakage is through the grain edges
(Goss et al. 1993; Hathcock et al. 1995). Despite much smaller
18.5.2 GUITAR Has High Cathodic Potential Limits
grain size, GUITAR does not allow for electrolyte intercala-
tion during the time scales of the cyclic voltammetric experi- The cathodic limits at GUITAR electrodes were examined
ments. The lack of this characteristic implies that either one in 1 M aqueous solutions of KNO3, H3PO4, H2SO4, LiClO4,
or both of the following hypotheses are applicable (Figures and (NH4)2SO4 by CV. Figure 18.14 illustrates the results.
18.12 and 18.13); (i) GUITAR’s grain boundaries are stitched Cathodic potential limits are estimated from this Figure at
GUITAR anodic limit 2.5 V

HOPG anodic limit 1.8 V
Electrolytic penetration
Micron size grains Nano-size grains w/structural defects

Pin-holes Pin-hole free?
-Fewer DOS -Higher DOS
FIGURE 18.13 Current hypotheses regarding GUITAR electrode behavior. No electrolyte intercalation increases corrosion resistance.
GUITAR resists intercalation of electrolytes from the lack of leaky gaps in the lattice. Left, initiation of corrosion in HOPG through the
intercalation model. The GUITAR electrode does not allow intercalation of electrolyte because: (i) lack of leaky gaps between nano-crystals
at the basal plane and/or (ii) inter-plane crosslinks. (Cheng, I. F et al. Highest measured anodic stability in aqueous solutions: Graphenic
electrodes from the thermolyzed asphalt reaction. RSC Adv. 3(7), 2013: 2379–2384. Reproduced by permission of The Royal Society of
Chemistry.)
(NH4)2SO4
400 μA/cm2
Li2SO4
KNO3
H3PO4
H2SO4
–2 –1.5 –1 –0.5 0
E (V) vs Ag/AgCl
FIGURE 18.14 Cyclic voltammograms at GUITAR at 50 mV/s in the indicated electrolytes (1 M). Starting potential was zero and scan
direction was toward more negative potentials. Counter and reference electrodes were graphite rod and Ag/AgCl, respectively. Potentials for
hydrogen evolution were extrapolated at 200 µA/cm2 from these voltammograms.
200 µA/cm2. Table 18.1 presents these limits over three cyclic

voltammetric runs in each electrolyte system along with stan- TABLE 18.1
dard deviations. The anodic limits are also included in that Cathodic and Anodic Potential Limits on GUITAR and
table along with the complete electrochemical potential win- Pyrolytic Graphite (PG) in Various Aqueous Electrolytes
dow which encompasses both the cathodic and anodic limits. with Graphite Rod and Ag/AgCl as Counter and
Studies in these electrolytes are important because of their Reference Electrodes, Respectively
use in regular electrochemistry, battery and fuel cell systems,
and GUITAR extends to higher potential limits than most Potential Limits at 200 μA/cm2 vs SHE (n = 3)
1 M Aqueous
materials reported in the literature (Zittel and Miller 1965; Electrolyte Anodic (V) Cathodic (V) Total Window (V)
Engstrom and Strasser 1984; Steele and Heinzel 2001; Norouzi H2SO4 2.10 ± 0.03 −0.90 ± 0.08 3.00
et al. 2009; Wessells et al. 2010). For example, a potential KNO3 1.93 ± 0.02 −1.18 ± 0.04 3.11
window of 2 V (−0.56 V to 1.54 V vs SHE) is reported for LiClO4 1.96 ± 0.05 −1.41 ± 0.16 3.37
glassy carbon in 0.1N H3PO4 (Zittel and Miller 1965). Also (NH4)2SO4 1.95 ± 0.02 −1.44 ± 0.10 3.39
noteworthy is that HOPG experiences intercalation in KNO3, H3PO4 2.03 ± 0.03 −1.24 ± 0.06 3.27
H2SO4, LiClO4, and (NH4)2SO4 at its anodic limits (between Pyrolytic graphite 1.88 ± 0.03 −0.44 ± 0.08 2.32
1.8 and 2.0 V vs SHE) (Hathcock et al. 1995), a feature not (1 M H2SO4)
observed with GUITAR anodes (Table 18.2).
TABLE 18.2
Potential Limits (Hydrogen and Oxygen Evolution Overpotentials (vs SHE)) for Various
Electrode Materials in Sulfuric Acid Medium
Cathodic Anodic Total H2SO4 Current Density
Material Limit (V) Limit (V) Window (V) Conc. (M) (μA/cm2) References
GUITAR −0.90 2.10 3.00 1.0 200 This work
Pyrolytic graphite −0.44 1.88 2.32
Platinum −0.10 1.30 1.40
HOPG −0.413 1.667 2.08 0.1 200 Tanaka et al. (2011)
GC −0.504 1.536 2.04
BDD −0.488 2.127 2.62
Diamond-like carbon −0.933 2.037 2.97
BDD −0.956 2.324 3.28 0.1 300 Teofilo et al. (2007)
BDD −1.056 2.444 3.5
BDD −0.545 1.955 2.5 0.1 200 Alves et al. (2012)
18.5.3 Electron Transfer and Density a quasi-reversible system sensitive to electrode surface micro-
of Electronic States structure and impurities on carbon electrodes (McCreery
2008). Figure 18.15 shows the CVs of both redox systems on
Another distinguishing characteristic of GUITAR electrodes GUITAR electrode at a potential sweep rate of 0.1 V/s. It was
over HOPG and graphene electrodes is in the facile hetero- found that the cathodic and anodic potential peak separation
geneous electron transfer rates. Both HOPG and graphene (ΔEp) is 71 mV between scan rates of 10 mV/s and 0.150 V/s
have barriers to this process unlike GUITAR. The assess- for Ru(NH 3 )3+/2+
6 , and 80 mV for Fe(CN)36− /4 − . Peak current
ment of these rates were conducted by cyclic voltammetric varied in a linear fashion as a function of ν1/2 over this range
characterization using Ru(NH 3 )63+/2+ and Fe(CN)36− /4 − redox indicating that both species are diffusion limited in behavior
systems (McCreery 2008). The Ru(NH 3 )63+/2+ couple under- on GUITAR.
goes an outer-sphere electron transfer, and is insensitive to GUITAR exhibits greater DOS than graphene or HOPG
surface microstructure and oxides. The Fe(CN)36− /4 − couple is electrodes. Table 18.3 is a literature comparison of the ΔEp
5.50E–05
2.5E–05
Fe(CN)63–/4– Ru(NH3)63+/2+
2.0E–05
3.5E–05
1.5E–05
1.0E–05 1.50E–05
Current (A)
5.0E–06
–5.00E–06
0.0E+00
–5.0E–06
–2.50E+05
–1.0E–05
–1.5E–05 –4.50E–05
600 500 400 300 200 100 0 100 0 –100 –200 –300 –400 –500
Potential (mV) vs Ag/AgCl
FIGURE 18.15 Cyclic voltammograms of 1 mM of indicated redox species with 1 cm2 GUITAR electrodes in 0.1 M KCl electrolyte at a
potential sweep rate of 0.1 V/s. counter electrode was a graphite rod. Analyses of ΔEp of Ru(NH 3 )63+/2+ (71 mV) and Fe(CN)36 − /4 − (80 mV) in
1 M KCl(aq) imply facile electron transfer kinetics. This differs from the basal planes of other graphitic materials where there is a larger
barrier to electron transfer due to low density of states (DOS) near the Fermi-level. (Cheng, I. F et al. Highest measured anodic stability in
aqueous solutions: Graphenic electrodes from the thermolyzed asphalt reaction. RSC Adv. 3(7), 2013: 2379–2384. Reproduced by permis-
sion of The Royal Society of Chemistry.)
TABLE 18.3
Peak Separation ΔEp (mV) Obtained from Cathodic and Anodic Waves of 1 mM Fe(CN)63− /4 −
and of 1 mM Ru(NH3 )63+ /2+ Cyclic Voltammograms at 100 mV/s on Various Carbon Electrodes
in 1 M KCl(aq)
Material ΔEp (mV) Fe(CN)63− /4− ΔEp (mV) Ru(NH3 )63+ /2+ References
GUITAR (basal plane) 80 71 This work
Pristine epitaxial graphene No peaks N/A Lim et al. (2010)
Pretreated epitaxial graphene (edge)a 81 81
BDD 282 101
Glassy carbon 68 64 Tang (2009)
Edge plane pyrolytic graphite (EPPG) 60 N/A Brownson et al. (2011)
Basal plane pyrolytic graphite (BPPG) 242 N/A
Graphene-coated EPPG 192 N/A
HOPG (basal plane) (200 mV/s) 700–1370 285 Kneten and McCreery (1992)
Source: Cheng, I. F. et al. Highest measured anodic stability in aqueous solutions: Graphenic electrodes from the thermolyzed
asphalt reaction. RSC Adv. 3(7), 2013: 2379–2384. Reproduced by permission of The Royal Society of Chemistry.
a Electrode was treated at 2.0 V (Ag/AgCl) in pH 7 phosphate buffer for 500 s. Edge defects were generated on the anodized
epitaxial graphene surface.
for the Ru(NH 3 )63+/2+ and Fe(CN)36− /4 − redox systems. These the graphitic systems. Despite good bulk electrical conductiv-
include graphene-containing materials, BDD, HOPG, and ity, the basal planes of HOPG have a very slow electron trans-
glassy carbon. The GUITAR electrodes exhibit ΔEp for fer with dissolved redox systems (Cline et al. 1994). This is
Fe(CN)36− /4 − and Ru(NH 3 )63+/2+ within the range of other car- attributed to low densities of electronic states (DOS) near the
bon materials. It is important to note that anodized epitax- Fermi level (Cline et al. 1994). Present observations describe
ial graphene of Table 18.3 is not a flat electrode (Lim et al. the basal planes of HOPG as electrochemically inert (Figure
2010). SEMs provided in this reference indicate that a sig- 18.16) (Davies et al. 2005). This is in agreement with a report
nificant portion of the surface contained exposed graphene on CVD-grown graphene that its basal planes contribute lit-
edges. The SEMs of GUITAR (see Cheng et al. 2011) dem- tle to the electrochemical detection of biological analytes in
onstrate that the exposed surface on this electrode consists comparison with graphitic islands (Brownson et al. 2011). Li
primarily of the basal plane rather than edges (Figure 18.2). In and coworkers examined monolayer graphene electrodes and
this aspect, GUITAR shows most similarity with basal plane found that corrugations were found to be 10 times faster than
HOPG electrodes. BDDs are dimensionally stable anodes flat basal planes of graphite for electron transfer with ferro-
with a high degree of sp3 hybridization. In terms of ΔEp, the cenemethanol (Li et al. 2011). We hypothesize that GUITAR
HOPG electrode stands as the exception when compared with has structural defects that enhance electron transfer with
A + e– B Basal plane : fast kinetics along defect regions B

Basal plane : slow kinetics A + e–
Basal planes
De
De
fec
fec
ts
s t
Edge plane : A + e– → B, fast kinetics

Plane edges GUITAR
HOPG
FIGURE 18.16 Schematic for the heterogeneous electron transfer on HOPG and GUITAR. Fast kinetics on HOPG is only obtained at the
edge planes, whereas GUITAR’s facile kinetics is obtained along speculated defect regions on its basal plane.
dissolved redox species (Figure 18.16). At this time, it is not absorbance (Donaldson et al. 2002; Dai et al. 2010; Huang
known if surface corrugations are present with GUITAR elec- et al. 2011a,b). A 1.2 cm2 flat GUITAR anode at 2.0 V versus
trodes, or if they are flat as might be expected with multilayer Ag/AgCl effectively degraded a 13 ppm MB solution (0.5 M
graphitic structures. When compared with HOPG, Raman Na2SO4) to near completion within 80 minutes. The decay
spectroscopy indicates that GUITAR has significant struc- curve follows pseudo-first-order kinetics with a rate constant
tural defects as discussed above. of 3.5 × 10−2 ± 0.2 × 10−2 cm−2min−1, (n = 3). For compari-
son, a BDD anode system has a pseudo-first-rate constant of
2.09 × 10−3 cm−2 min−1 in this application (Wu et al. 2009).
18.6 GUITAR AS A DIMENSIONALLY
The GUITAR anode exhibits increased DSA perfor-
STABLE ANODE mance over BDD, we attribute this feature to the high DOS in
The combination of excellent anodic stability and high DOS GUITAR as outlined in the cyclic voltammetric studies with
indicates that GUITAR can be a new form of dimension- Fe(CN)64− /3− and Ru(NH 3 )63+/2+ above. Furthermore, the costs of
ally stable anodes (DSAs) for electrochemical mineraliza- producing GUITAR and the ability to fabricate high-surface-
tion of organic xenobiotics. Two attributes are required for area anodes give it advantages over other deposition methods
the anode: (i) a high oxygen overpotential and (ii) stability at that require high temperatures. We expect that GUITAR will
highly oxidizing potentials, that is, corrosion resistance. BDD have significantly lower costs and fulfill the promise of anodic
electrodes are recognized to be the materials best suited for water purification.
DSA applications, but are expensive and difficult to fabricate
into a high-surface material (Swain et al. 1998; Boudenne and 18.7 GUITAR RELATIVE TO OTHER
Cerclier 1999; Pleskov 1999; Rao and Fujishima 2000; Fryda GRAPHITIC/GRAPHENIC AND
et al. 2003; Tröster et al. 2004; Fujishima et al. 2005). RELATED CARBON MATERIALS
By comparison with graphites, including HOPG and graphene,
18.6.1 Degradation of Methylene Blue GUITAR exhibits markedly different physical, chemical and
In order to assess its performance as a DSA, GUITAR electrochemical properties. Similarities in appearance led us
anodes were examined in the electro-oxidation of methy- to believe initially that GUITAR was a form of HOPG or gra-
lene blue (MB) (λmax = 663 nm), a common surrogate for phene. These are outlined above in the Summary of Physical
this purpose (Figure 18.17) (Cheng et al. 2013). Oxidative Characterizations; however, a prominent Raman D-band
degradation of MB is apparent through its loss of optical indicates a small grain size of 5 nm. When compared with
1–10 µm for graphite and 50–3000 nm for CVD grown gra-
phene (Brownson et al. 2012), GUITAR should be precluded
(a) 14 from consideration as a form of either (Banks et al. 2005).
12 Also noteworthy is that although GUITAR has a similar grain
Concentration (mg/L)
10
size as r-GO paper, it is dissimilar in that this material has a
wavy and textured surface from its agglomeration of graphene
8
platelets (Dikin et al. 2007). Perhaps the largest deviation
6 from HOPG/graphene/r-GO paper is in terms of electrochem-
4 ical characteristics. We hypothesize that the nature of the
2 defects between the grains are fundamentally different from
0 those with graphite, HOPG, and graphene. The lack of elec-
Time (min)
(b) 0 trolyte intercalation and relatively fast electron transfer rates
0 20 40 60 80 from its basal planes with dissolved redox species indicates
–1
that the defects may give rise to higher DOS than with HOPG
Ln(Ct/Co)
–2 or graphene. GUITAR is a much better electrode than either

material (Brownson et al. 2011). Furthermore, GUITAR has a
–3
larger aqueous electrochemical potential window. We expect
–4 that GUITAR will have superior corrosion resistance to r-GO
paper as this material has significant exposure of graphene
–5
edges to solution. We propose GUITAR to be a unique mate-
rial that has an ordered, layered, structure across planes giv-
FIGURE 18.17 GUITAR as a dimensionally stable anode. (a) ing rise to atomically flat graphene planes, but with a great
Methylene blue degradation curve with a 1.2-cm2 GUITAR anode
degree of disorder within those planes. Furthermore, although
(+2.0 V vs. Ag/AgCl) in 0.5 M Na2SO4 electrolyte. (b) Pseudo-first-
order decay assumption based on the decay curve. Over three runs
Raman spectroscopy suggests that it has nanograins of 5 nm,
this is 3.5 × 10−2 ± 0.2 × 10−2 min−1 cm−2. (Cheng, I. F. et al. Highest we suggest that these grains are bound together in nonhex-
measured anodic stability in aqueous solutions: Graphenic electrodes agonal arrangements of sp2 carbon. This is in contrast with
from the thermolyzed asphalt reaction. RSC Adv. 3(7), 2013: 2379– graphites, including HOPG and graphene, where the crystal-
2384. Reproduced by permission of The Royal Society of Chemistry.) line grains are only in physical contact. A schematic for the
Graphene paper GUITAR HOPG
Wavy Atomically flat Flat with local or grain defects
FIGURE 18.18 Schematic representation of the cross sections of graphene paper, GUITAR, and HOPG. The local or grain defects in
HOPG are responsible for electrolyte intercalation. GUITAR lacks this feature and together with SEM images, the morphology above is
proposed. (Diagram for graphene paper Reprinted by permission from Macmillan Publishers Ltd. Nature, Dikin, D. A. et al. Preparation
and characterization of graphene oxide paper, 448, 457–460, copyright 2007; and that of HOPG was adapted. Reprinted with permission
from Hathcock, K. W et al. Murray. Incipient electrochemical oxidation of highly oriented pyrolytic graphite: Correlation between surface
blistering and electrolyte anion intercalation. Anal. Chem. 67, 2201–2206. Copyright 1995 American Chemical Society.)
cross-sections of graphene paper, HOPG and GUITAR is pre- The high capacitance at GUITAR is accompanied with a
sented in Figure 18.18. capacitive window of 2.0 V (−0.6 V to 1.4 V vs SHE) which
is wider than 0.8 V for activated carbon (E Frackowiak 2007).
With the wide window along with the high capacitance,
18.8 FUTURE STUDIES AND GUITAR is expected to offer energy of at least one magnitude
APPLICATIONS OF GUITAR higher than activated carbon.
18.8.1 Scanning Tunneling Microscopy
Further research is underway in order to fully characterize and 18.8.3 Fuel Cells and Flow Batteries
apply this unique carbon material. One such study is the use of Other areas of this material’s applicability are fuel cells and
scanning tunneling microscopy to quantify the density of elec- flow batteries. Of particular interest is applications in vana-
tronic states. This investigation will give a reason for the facile dium redox flow batteries (VRB). The VRB system is based on
electron transfer and relatively higher interfacial capacitance the oxidation and reduction reactions of V5+/V4+ (E° = +1.00 V
over graphite since these properties are highly dependent on vs SHE) at the positive half-cell and V3+/V2+ (E° = − 0.26 V
the density of state at the Fermi level (Gerischer 1985). vs SHE) at the negative in sulfuric acid (Xi et al. 2008). The
VRB has various advantages over other secondary batteries
because of its independently tunable power and capacity, long
18.8.2 Ultracapacitor Development
cycle life and relatively large capacitance (Kim et al. 2010;
Electrochemical properties of GUITAR have given this mate- Chen et al. 2012). However, one disadvantage is the reduc-
rial wide applicability. Exploiting the wide potential window tion in columbic and energy efficiency due to gas evolutions
in aqueous electrolytes together with the ability to deposit up (hydrogen and oxygen) (Chen et al. 2012). Electrode mate-
to large-surface-area materials like nanosprings, develop- rials that are applicable for both redox processes above are
ment of aqueous-based ultracapacitors based on GUITAR is expected to have high overpotentials for other electrochemi-
a future application. Currently, aqueous ultracapacitors have cal processes including oxygen and hydrogen evolution in
their cell voltages limited to 1.0–2.0 V (Khomenko et al. the medium of the battery’s applicability. The ideal material
2006; Li et al. 2007; Demarconnay et al. 2010; Pasta et al. should also have a high surface area, be mechanically and
2010; Xia et al. 2012). This is based on the 1.23 V required for chemically stable, should exhibit high electron transfer rates
water degradation along with overpotentials. The impetuses for the redox processes above (Huang et al. 2008a,b; Li et al.
for aqueous electrolytes in this application are superior ionic 2011). Preliminary studies indicate that GUITAR should sur-
conductivity, nonflammability, lower costs and green solvent pass present electrode designs in this application.
considerations. On the other hand, organic solvents offer up to
5 V of cell potential, but with the danger of their flammabil-
18.8.4 Waste Water Treatment
ity and the problem of their costs (Qu and Shi 1998; Kötz and
Carlen 2000; Frackowiak and Béguin 2001; Khomenko et al. Results from the degradation of MB, indicates GUITAR’s
2008). Preliminary studies have shown that the capacitance of ability as a DSA for water purification. In addition to
GUITAR of 500 μ F/cm2 at 0.1 V in 1 M H2SO4 far exceeds GUITAR’s high oxygen overpotential, the slow kinetics for
literature values of activated carbon (10–20 μF/cm2), which hydrogen evolution will allow for a simplified electroly-
is the most common material for commercial ultracapacitors. sis reactor design that will destructively oxidize and reduce
xenobiotics and microbes (Vogel et al. 1987). Furthermore, in compacted graphite irons. J. Mater. Process. Technol. 209,
as-grown GUITAR has proven to be highly hydrophobic. 2009: 1701–1705.
Preliminary studies that utilize this property, have shown Blyth, R. I. R. et al. XPS studies of graphite electrode materials for
lithium ion batteries. Appl Surf Sci. 167, 2000: 99.
that, this material is able to adsorb organics, especially crude
Boudenne, J.-L. and O. Cerclier. Performance of carbon black-slurry
oil. In all, various properties of GUITAR make it possible electrodes for 4-chlorophenol oxidation. Water Res. 33, 1999:
to design a water treatment reactor system that incorporates 494–504.
adsorption/separation and degradation. Brownson, D. A. C., D. K. Kampouris, and C. E. Banks. Graphene
electrochemistry: Fundamental concepts through to promi-
nent applications. Chem Soc Rev. 41(21), 2012: 6944–6976.
18.9 CONCLUSION Brownson, D. A. C., Maria Gomez-Mingot, and C. E. Banks. CVD
graphene electrochemistry: Biologically relevant molecules.
A simple and cost-effective method for the synthesis of a gra- Phys. Chem. Chem. Phys. 13, 2011: 20284–20288.
phenic material has been presented. GUITAR has properties Brownson, D. A. C., L. J. Munro, D. K. Kampouris, and C. E. Banks.
that make it similar to both graphite and r-GO paper, however, Electrochemistry of graphene: Not such a beneficial electrode
GUITAR demonstrates significant differences from these material? RSC Adv. 1, 2011: 978–988.
materials, and so, has been proposed to be a unique carbon Chen, F., J. Liu, H. Chen, and C. Yan. Study on hydrogen evolution
material possibly classified between graphite and r-GO paper. reaction at a graphite electrode in the all-vanadium redox flow
battery. Int. J. Electrochem. Sci. 7, 2012: 3750–3764.
Synthesis conditions require simple laboratory apparatuses,
Chen, W., L. Yan, and P. R. Bangal. Chemical reduction of graphene
no pressure control, almost no skill, and also permits a depo- oxide to graphene by sulfur-containing compounds. J. Phys.
sition up to a vast range of substrates, since it necessitates a Chem. C 114, 2010: 19885–19890.
relatively low temperature. Large surface area and bulk syn- Cheng, I. F. et al. Synthesis of graphene paper from pyrolyzed
thesis of this material can be achieved with a wide range of asphalt. Carbon 49, 2011: 2852–2861.
starting materials including roofing tar plus sulfur. The dis- Cheng, I. F., Y. Xie, I. O. Gyan, and N. R. Nicholas. Highest mea-
covery of GUITAR solves the problems facing conventional sured anodic stability in aqueous solutions: Graphenic elec-
trodes from the thermolyzed asphalt reaction. RSC Adv. 3(7),
synthesis of carbon materials, especially graphene, since the
2013: 2379–2384.
former exists as a material with similar or better chemical, Cheng, X. L. Low-potential electrochemiluminescent sensing based
mechanical, and electrochemical properties. Preliminary on surface unpassivation of CdTe quantum dots and competition
results present GUITAR as a suitable material for wide sci- of analyte cation to stabilizer. Anal. Chem. 82, 2010: 7387–93.
entific applications including fuel cells, redox batteries, and Ci, L. et al. Double wall carbon nanotubes promoted by sulfur in
ultracapacitors, and as a DSA for waste water treatment. a floating iron catalyst CVD system. Chem. Phys. Lett. 359,
2002: 63–67.
Cline, K. K., M. T. McDermott, and R. L. McCreery. Anomalously
REFERENCES slow electron transfer at ordered graphite electrodes: Influence
of electronic factors and reactive sites. J. Phys. Chem. 98,
Alliata, D., R. Kötz, O. Haas, and H. Siegenthaler. In situ AFM study 1994: 5314–5319.
of interlayer spacing during anion intercalation into HOPG in Comninellis, C. Electrocatalysis in the electrochemical conversion/
aqueous electrolyte. Langmuir 15, 1999: 8483–8489. combustion of organic pollutants for waste water treatment.
Alsmeyer, D. C. and R L. McCreery. In Situ Raman monitoring of Electrochim. Acta 39, 1994: 1857.
electrochemical graphite intercalation and lattice damage in Comninellis, C. and C. Pulgarin. Anodic oxidation of phenol for
mild aqueous acids. Anal. Chem. 64, 1992: 1528–1533. waste water treatment. J. Appl. Electrochem. 21, 1991: 703.
Alves, S. A., F. L. Migliorini, M. R. Baldan, N. G. Ferreira, and Dai, Q., F. Liao, S. Cai, J. Wang, and J. Chen. A novel electrode for
M. R. V. Lanza. Electrochemical and morphology study of the the degradation of methylene blue in electrochemical system:
BDD/Ti electrodes with different doping levels. ECS Trans. Preparation and characterization. Adv. Mat. Res. 1407, 2010:
43(1), 2012: 191–197. 113–116.
Arsat, R. et al. Graphene-like nano-Sheets/36° LiTaO3 surface acoustic Davies, T. J., M. E. Hyde, and R. G. Compton. Nanotrench arrays
wave hydrogen gas sensor. IEEE SENSORS Conferences. 2008. reveal insight into graphite electrochemistry. Angew. Chem.
Avouris, P. and C. Dimitrakopoulos. Graphene: Synthesis and appli- 117, 2005: 5251–5256.
cations. Materials Today 15, 2012: 86–97. Demarconnay, L., E. Raymundo-Pinero, and F. Béguin. A symmet-
Bajus, M. and V. Veselý. Pyrolysis of hydrocarbons in the presence ric carbon/carbon supercapacitor operating at 1.6 V by using
of elemental sulfur. Collect. Czech. Chem. Commun. 45, a neutral aqueous solution. Electrochem. Comm. 12, 2010:
1980: 238–254. 1275–1278.
Banerjee, S., M. Sardar, N. Gayathri, A. K. Tyagi, and B. Raj. Diaz-Pinto, C., D. De, V. G. Hadjiev, and H. Peng. AB-stacked mul-
Conductivity landscape of highly oriented pyrolytic graph- tilayer graphene synthesized via chemical vapor deposition: A
ite surfaces containing ribbons and edges. Phys.Rev. B 72, characterization by hot carrier transport. ACS Nano 6, 2012:
075418, 2005: 1–7. 1142–1148.
Banks, C. E., T. J. Davies, G. G. Wildgoose, and R. G. Compton. Dikin, D. A. et al. Preparation and characterization of graphene
Electrocatalysis at graphite and carbon nanotube modified oxide paper. Nature 448, 2007: 457–460.
electrodes: Edge-plane sites and tube ends are the reactive Dominguez, C. P., J. A. Rodrigo, M. A. Villasenor, and J. Rodriguez.
sites. Chem. Commun. 2005: 829–841. Effect of the current intensity in the electrochemical oxida-
Bazdar, M., H. R. Abbasi, A. H. Yaghtin, and J. Rassizadehghani. tion of aqueous phenol wastes at an activated carbon and steel
Effect of sulfur on graphite aspect ratio and tensile properties anode. J. Ind. Eng. Chem. Res. 38, 1999: 3779.
Donaldson, J. D., S. M. Grimes, N. G. Yasri, B. Wheals, J. Parrick, Granger, M. C. Standard electrochemical behavior of high-quality,
and W. E. Errington. Anodic oxidation of the dye materials boron-doped polycrystalline diamond thin-film electrodes.
methylene blue, acid blue 25, reactive blue 2 and reactive blue Anal. Chem. 72, 2000: 3793–3804.
15 and the characterisation of novel intermediate compounds Guo, H., X. Wang, Q. Qian, F. Wang, and X. Xia. A green approach
in the anodic oxidation of methylene blue. J. Chem. Technol. to the synthesis of graphene nanosheets. ACS Nano. 9, 2009:
Biotechnol. 77, 2002: 756–760. 2653.
Double, D. D. and A. Hellawell. The nucleation and growth of Guo, P., H. Song, and X. Chen. Electrochemical performance of gra-
graphite- the modification of cast iron. Acta Metal. Mater. 43, phene nanosheets as anode material for lithium-ion batteries.
1995: 2435–2442. Electrochem. Commun. 11, 2009: 1320–1324.
Dubin, S. et al. A one-step, solvothermal reduction method for pro- Gupta, A., G. Chen, P. Joshi, S. Tadigadapa, and P. C. Eklund.
ducing reduced graphene oxide dispersions in organic sol- Raman scattering from high-frequency phonons in supported
vents. ACS Nano 4, 2010: 3845–3852. n-graphene layer films. Nano Lett. 6, 2006: 2667.
Engstrom, R. C. and V. A. Strasser. Characterization of electro- Hathcock, K. W., J. C. Brumfield, C. A. Gross, E. A. Irene, and R.
chemically pretreated glassy carbon electrodes. Anal. Chem. W. Murray. Incipient electrochemical oxidation of highly ori-
56, 1984: 136–141. ented pyrolytic graphite: Correlation between surface blister-
Estrade-Szwarckopf, H. XPS photoemission in carbonaceous mate- ing and electrolyte anion intercalation. Anal. Chem. 67, 1995:
rials: A “defect” peak beside the graphitic asymmetric peak. 2201–2206.
Carbon 42, 2004: 1713. Hsu, Y.-S. and S. K. Ghandhi. The effect of phosphorous doping on
Eswaraiah, V., S. S. J. Aravind, and S. Ramaprabhu. Top down tin oxide films made by the oxidation of phosphine and tetra-
method for synthesis of highly conducting graphene by exfoli- methyltin I: Growth and etching properties. J. Electrochem.
ation of graphite oxide using focused solar radiation. J. Mater. Soc. 127, 1980: 1592.
Chem. 21, 2011: 6800–6803. Hsu, Y.-S. and S. K. Ghandhi. The effect of phosphorous doping
Ferrari, A. and J. Robertson. Interpretation of Raman spectra of on tin oxide films made by the oxidation of phosphine and
disordered and amorphous carbon. Phys. Rev. B : Condens. tetramethyltin II: Electrical properties. J. Electrochem. Soc.
Matter. 61, 2000: 14095–14107. 127, 1980: 1595.
Frackowiak, E. Carbon materials for supercapacitor application. Huang, H., X. Liu, and J. Huang. Tubular structured hierachical
Phys. Chem. Chem. Phys. 9, 2007: 1774–1785. mesoporous titania material derived from natural cellulosic
Frackowiak, E. and F. Béguin. Carbon materials for the electro- substances and application as photocatalysts for degradation
chemical storage of energy in capacitors. Carbon 39, 2001: of methylene blue. Mater. Res. Bull.. 46, 2011a: 1814–1818.
937–950. Huang, J., Q. Zhang, F. Wei, W. Qian, D. Wang, and L. Hu. Liquefied
Fryda, M., T. Matthée, S. Mulcahy, A. Hampel, L. Schäfer, and I. petroleum gas containing sulfur as the carbon source for car-
Tröster. Fabrication and application of Diachem electrodes. bon nanotube forests. Carbon 46, 2008a: 291–296.
Diamond Relat. Mater. 12(10–11), 2003: 1950–1956. Huang, K-L., X-G. Lia, S-Q. Liu, N. Tan, and L-Q. Chen. Research
Fu, P. P. and R. Harvey. Dehydrogenation of polycyclic hydroaro- progress of vanadium redox flow battery for energy storage in
matic compounds. Chem. Rev. 78, 1978: 317–361. China. Renew. Energ. 33, 2008b: 186–192.
Fujishima, A., Y. Einaga, T. N. Rao, and D. A. Tryk. Diamond Huang, P. Y. et al. Grains and grain boundaries in single-layer gra-
Electrochemistry. Amsterdam, The Netherlands: Elsevier, phene atomic patchwork quilts. Nature 469, 2011b: 389–392.
2005. Iniesta, J. Electrochemical oxidation of 3-methylpyridine at a boron-
Gandini, D. Oxidation of carboxylic acids at boron-doped diamond doped diamond electrode: Application to electroorganic syn-
electrodes for wastewater treatment. J. Electrochem. 30, thesis and wastewater treatment. Electrochem. Comm. 3,
2000: 1345–1350. 2001: 346–351.
Gao, J., F. Liu, Y. Liu, N. Ma, Z. Wang, and X. Zhang. Environment- Jang, J. W., C. E. Lee, S. C. Lyu, T. J. Lee, and C. J. Lee. Structural
friendly method to produce graphene that employs vitamin C study of nitrogen-doping effects in bamboo-shaped mul-
and amino acid. Chem. Mater. 22, 2010: 2213–2218. tiwalled carbon nanotubes. Appl. Phys. Lett. 84, 2004:
Gattrell, M. and D. Kirk. A study of the oxidation of phenol at plati- 2877–2879.
num and preoxidized platinum surfaces. J. Electrochem. Soc. Jarzebski, Z. M. and J. P. Marton. Physical properties of SnO2 mate-
140, 1993: 1534. rials I. preparation and defect structure. J. Electrochem. Soc.
Gerischer, H. An interpretation of the double layer capacity of 123, 1976: C199–C205.
graphite electrodes in relation to the density of states at the Jiang, J., H. Tang, B. Wang, and Z. Su. Raman and infrared proper-
fermi level. J. Phys. Chem., 89, 1985: 4249–4251. ties and layer dependence of the phonon dispersions in multi-
Gherardini, L., P. A. Michaud, M. Panizza, C. Comninellis, and layered graphene. Phys. Rev. B 2008, 77, 235421, 2008: 1–8.
N. Vatistas. Electrochemical oxidation of 4-chlorophenol for Jusino, A. and H. H. Schobert. The use of sulfur to extract hydrogen
wastewater treatment: Definition of normalized current effi- from coal. Int. J. Coal Geol. 65, 2006: 223–234.
ciency. J. Electrochem. Soc. 148, 2001: D78. Khomenko, V., E. Raymundo-Pinero, and F. Béguin. High-energy
Ghosh, P. et al. Effect of sulfur concentration on the morphology density graphite/AC capacitor in organic electrolyte. J. Power
of carbon nanofibers produced from a botanical hydrocarbon. Sources 177, 2008: 643–651.
Nanoscale Res. Lett. 3, 2008: 242–248. Khomenko, V., E. Raymundo-Pinero, and F. Béguin. Optimisation
Goss, C. A., J. C. Brumfield, E. A. Irene, and R. W. Murray. of an asymmetric manganese oxide/activated carbon capaci-
Imagining the incipient electrochemical oxidation of highly tor working at 2 V in aqueous medium. J. Power Sources 153,
oriented pyrolytic graphite. Anal. Chem. 65, 1993: 1378–1389. 2006: 183–190.
Granata, S., T. Faravelli, and E. Ranzi. A wide range kinetic mod- Kim, S. et al. Cycling performance and efficiency of sulfonated
eling study of the pyrolysis and combustion of naphthenes. poly(sulfone) membranes in vanadium redox flow batteries.
Combust. Flame 132, 2003: 533–544. Electrochem. Commun. 12, 2010: 1650–1653.
Kneten, K. R. and R. L. McCreery. Effects of redox system structure Panizza, M. and G. Cerisola. Influence of anode material on the elec-
on electron-transfer kinetics at ordered graphite and glassy trochemical oxidation of 2-naphthol: Part 1. Cyclic voltam-
carbon electrodes. Anal. Chem. 64, 1992: 2518–2524. metry and potential step experiments. Electrochim. Acta 48,
Kostic, R., M. Miric, T. Radic, M. Radovic, R. Gajic, and Z. V. Popovic. 2003: 3491.
Optical characterization of graphene and highly oriented pyro- Panizza, M. and G. Cerisola. Electrocatalytic materials for the elec-
lytic graphite. Acta Physica Polonica A 116, 2009: 718. trochemical oxidation of synthetic dyes. Appl. Catal., B 75,
Kötz, R. and M. Carlen. Principles and applications of electrochemi- 2007: 95.
cal capacitors. Electrochim. Acta 45, 2000: 2483–2498. Panizza, M., P. A. Michaud, G. Cerisola, and C. Comninellis.
Kowata, T., S. H. Lee, H. Hiratsuka, and T. Hareyama. Effect of Anodic oxidation of 2-naphthol at boron-doped diamond elec-
sulfide shape on graphitization of flake graphite cast iron. trodes. J. Electroanal Chem. 507, 2001: 206–214.
J. Japan Foundry Eng. Soc. 74, 2002: 9578–9583. Pasta, M., F. La Mantia, L. Hu, H. D. Deshazer, and Y. Cui. Aqueous
Li, D., M. B. Muller, S. Gilje, R. B. Kaner, and G. G. Wallace. supercapacitors on conductive cotton. Nano Res. 3, 2010:
Processable aqueous dispersions of graphene nanosheets. Nat. 452–458.
Nanotechnol. 3, 2008: 101–105. Pavlov, D. The lead-acid battery lead dioxide active mass: A
Li, W., C. Tan, M. A. Lowe, H. D. Abruña, and D. C. Ralph. gel-crystal system with proton and electron conductivity.
Electrochemistry of individual monolayer graphene sheets. J. Electrochem. Soc. 139, 1992: 3075.
ACS Nano 5, 2011: 2264–2270. Pavlov, D. and B. Monahov. Mechanism of the elementary elec-
Li, W. et al. Nitrogen-containing carbon spheres with very large uni- trochemical processes taking place during oxygen evolution
form mesopores: The superior electrode materials for EDLC on the lead dioxide electrode. J. Electrochem. Soc. 143, 1996:
in organic electrolyte. Carbon 45, 2007: 1757–1763. 3616.
Li, X., H. Zhang, Z. Mai, H. Zhang, and I. Vankelecom. Ion exchange Pleskov, Y. V. Synthetic diamond in electrochemistry. Russ. J.
membranes for vanadium redox flow battery (VRB) applica- Electrochem. 68, 1999: 381.
tions. Energy Environ. Sci. 4, 2011: 1147. Poh, H. L., F. Sanek, A. Ambrosi, G. Zhao, Z. Sofer, and M.
Li, Y. et al. Nanographite ribbons grown from a SiC arc-discharge in Pumera. Graphenes prepared by Staudenmaier, Hofmann
a hydrogen atmosphere. Carbon 39, 2001: 626–628. and Hummers methods with consequent thermal exfoliation
Lim, C. X., Hui Ying Hoh, P. K. Ang, and K. P. Loh. Direct voltam- exhibit very different electrochemical properties. Nanoscale
metric detection of DNA and pHSensing on epitaxial gra- 4, 2012: 3515–3522.
phene: An insight into the role of oxygenated defects. Anal. Polcaro, A. M. and S. Palmas. Electrochemical oxidation of chloro-
Chem. 82, 2010: 7387–7393. phenols. Ind. Eng. Chem. Res. 36, 1997: 1791.
Luning, Z., Dušan A. P., B. Geng, and J. Marschall. Surface modi- Polcaro, A. M., S. Palmas, F. Renoldi, and M. Mascia. Three-
fication of highly oriented pyrolytic graphite by reaction with dimensional electrodes for the electrochemical combustion of
atomic nitrogen at high temperatures. Appl. Surf. Sci. 257, organic pollutants. Electrochim. Acta 46, 2000: 389.
2011: 5647–5656. Qu, D. and Hang Shi. Studies of activated carbons used in double-
Makarov, T. L. and K-H. Han. Graphite under the magnetic force layer capacitors. J. Power Sources 74, 1998: 99–107.
microscope. Phys. Stat. Sol. (b) 244, 2007: 4138–4142. Rao, T. N. and A. Fujishima. Recent advances in electrochemistry of
Matthews, M. J., M. A. Pimenta, G. Dresselhaus, and M. Endo. diamond. Diamond Relat. Mater. 9, 2000: 384–389.
Origin of dispersive effects of the Raman D band in carbon Reich, S. and C. Thomsen. Raman spectroscopy of graphite. Phil.
materials. Phys. Rev. B 59, 1999: R6585–R6588. Trans. R. Soc. Lond. A 362, 2004: 2271–2288.
McBride, W., D. J. H. Cockayne, and K. Tsuda. Deconvolution of Semmelhack, H. C. et al. Growth of highly oriented graphite films
electron diffraction patterns of amorphous materials formed at room temperature by pulsed laser deposition using carbon-
with convergent beam. Ultramicroscopy 94, 2003: 305–308. sulfur targets. Carbon 44, 2006: 3064–3072.
McCreery, R. L. Advanced carbon electrode materials for molecular Silverwood, H. A. and M. Orchin. Aromatization reactions
electrochemistry. Chem. Rev. 108, 2008: 2646–2687. with selenium and aryl diselenides. . J. Org. Chem. 27, 1962:
Morrison, N. J. The reactions of crosslink precursors in natural rub- 3401–3404.
ber. Rubber Chem. Technol. 57, 1984: 86–96. Steele, B. C. H. and A. Heinzel. Materials for fuel-cell technologies.
Morrison, N. J. and M. Porter. Temperature effects on the stability of Nature 414, 2001: 345–352.
intermediates and crosslinks in sulfur vulcanization. Rubber Strong, V. et al. Patterning and electronic tuning of laser scribed
Chem. Technol. 57, 1984: 63–85. graphene for flexible all-carbon devices. ACS Nano 6, 2012:
Norouzi, P., M. R. Ganjali, S. J. Shahtaheri, R. Dinarvand, and 1395–1403.
A. Hamzehpoor. Monitoring of methyldopa by fast Fourier Su, C. et al. Electrical and spectroscopic characterizations of ultra-
Transform continuous cyclic voltammetry at gold microelec- large reduced graphene oxide monolayers. Chem. Mater. 21,
trode. Chinese J. Chem. 27, 2009: 732–738. 2009: 5674.
Palyanov, Y. N., I. N. Kupriyanov, Y. M. Borzdov, A. G. Sokol, and Subrahmanyam, K. S., S. R. C. Vivekchand, A. Govindaraj, and C.
A. F. Khokhryakov. Diamond crystallization from a sulfur- N. R. Rao. A study of graphenes prepared by different meth-
carbon system at HPHT conditions. Cryst. Growth Des. 9, ods: Characterization, properties and solubilization. J. Mater.
2009: 2922–2926. Chem.18, 2008: 1517–1522.
Pan, D. et al. Li storage properties of disordered graphene nanosheets. Swain, G. M., A. B. Anderson, and J. C. Angus. Applications of
Chem. Mater. 21, 2009: 3136–3142. diamond thin films in electrochemistry. MRS Bull. 23, 1998:
Panigrahi, S., A. Bhattacharya, D. Bandyopadhyay, S. J. Grabowski, 56–60.
D. Bhattacharyya, and S. Banerjee. Wetting property of Tanaka, Y. et al. Electrochemical properties of N-doped hydroge-
the edges of monoatomic step on graphite: Frictional-force nated amorphous carbon films fabricated by plasma-enhanced
microscopy and ab initio quantum chemical studies. J. Phys. chemical vapor deposition methods. Electrochim. Acta 56,
Chem. C 115, 2011: 14819–14826. 2011: 1172–1181.
Tang, L. Preparation, structure, and electrochemical properties of Wei, D., Y. Liu, Y. Wang, H. Zhang, L. Huang, and G. Yu. Synthesis
reduced graphene sheet films. Adv. Funct. Mater. 19, 2009: of N-doped graphene by chemical vapor deposition and its
2782–2789. electrical properties. Nano Lett. 9, 2009: 1752.
Teofilo, R. F. et al. Improvement of the electrochemical properties Wessells, C., R. Ruff, R. A. Huggins, and Y. Cui. Investigations of
of “as-grown” boron-doped polycrystalline diamond elec- the electrochemical stability of aqueous electrolytes for lith-
trodes deposited on tungsten wires using ethanol. J Solid State ium battery applications. Electrochem. Solid State Lett. 13(5),
Electrochem. 11, 2007: 1449–1457. 2010: A59–A61.
Tröster, I., L. Schäfer, M. Fryda, and T. Matthée. Electrochemical Wu, M., G. Zhao, M. Li, L. Liu, and D. Li. Applicability of boron-
advanced oxidation process using DiaChem electrodes. Water doped diamond electrode to the degradation of chloride-medi-
Sci. Technol. 49, 2004: 207–212. ated and chloride-free wastewaters. J. Hazard. Mater. 163,
Tung, V. C., M. J. Allen, Y. Yang, and R. B. Kaner. High-throughput 2009: 26–31.
solution processing of large-scale graphene. Nat. Nanotechnol. Xi, J., Z. Wu, X. Teng, Y. Zhao, L. Chen, and X. Qiu. Self-assembled
4, 2009: 25–29. polyelectrolyte multilayer modified Nafion membrane with
Vallés, C., M. Pérez-Mendoza, P. Castell, M. T. Martínez, W. K. suppressed vanadium ion crossover for vanadium redox flow
Maser, and A. M. Benito. Towards helical and Y-shaped batteries. J. Mater. Chem. 18, 2008: 1232–1238.
carbon nanotubes: The role of sulfur in CVD processes. . Xia, H., Y. S. Meng, G. Yuan, C. Cui, and L. Lu. A symmetric
Nanotechnology 17, 2006: 4292–4299. RuO2/RuO2 supercapacitor operating at 1.6 V by using a neu-
Vogel, T. M., C. S. Criddle, and P. L. McCarty. Transformations of tral aqueous electrolyte. Electrochem. Solid State Lett. 15(4),
halogenated aliphatic coumpounds. Environ. Sci. Technol. 21, 2012: A60–A63.
1987: 722–736. Xie, Y. et al. Sulfur as an important co-factor in the formation of
Wang, G., B. Wang, J. Park, Y. Wang, B. Sun, and J. Yao. Highly multilayer graphene in the thermolyzed asphalt reaction.
efficient and large-scale synthesis of graphene by electrolytic J. Mater. Chem. 22, 2012: 5723–5729.
exfoliation. Carbon 47, 2009: 3242–3246. Zanta, C. L. P. S., A. R. De Andrade, and J. F. C. Boodts.
Wang, H., H. Tian, S. Wang, W. Zheng, and Y. Liu. Simple and eco- Electrochemical behaviour of olefins: Oxidation at ruthe-
friendly solvothermal synthesis of luminescent reduced gra- nium-titanium dioxide and iridium-titanium dioxide coated
phene oxide small sheets. Mater. Lett. 78, 2012: 170–173. electrodes. J. Appl. Electrochem. 30, 2000: 467–474.
Wang, J., M. F. Reyniers, and G. B. Martin. Influence of dimethyl Zhi, L. and K. Mullen. A bottom-up approach from molecular
disulfide on coke formation during steam cracking of hydro- nanographenes to unconventional carbon materials. J. Mater.
carbons. Ind. Eng. Chem. Res. 46, 2007: 4134–4148. Chem. 18, 2008: 1472–1484.
Wassei, J. K. and R. B. Kaner. Oh, the places you’ll go with graphene. Zittel, H. E. and F. J. Miller. A glassy-carbon electrode for voltam-
Acc. Chem. Res. Article ASAP, DOI: 10.1021/ar300184v., metry. Anal. Chem. 37, 1965: 200–203.
2012: A-J.
19 Graphene-Based Solar Cells
Tanvi Upreti, Vinay Gupta, and Suresh Chand
CONTENTS
Abstract...................................................................................................................................................................................... 361
19.1 Introduction...................................................................................................................................................................... 361
19.2 Graphene in Organic Solar Cells...................................................................................................................................... 363
19.2.1 Transparent Conducting Electrode....................................................................................................................... 363
19.2.2 Graphene as an Interfacial Layer.......................................................................................................................... 372
19.2.3 Active Layer.......................................................................................................................................................... 377
19.3 Graphene in Dye-Sensitized Solar Cells.......................................................................................................................... 379
19.4 Graphene in Solid-State Cells.......................................................................................................................................... 382
19.5 Conclusion........................................................................................................................................................................ 384
References.................................................................................................................................................................................. 384
ABSTRACT In2O3 and ~10% SnO2), is used. Commercially available ITO

has a resistance as low as 5 Ω/sq. with transparency T ≈ 80%.
Organic photovoltaic (OPV) cells have been of great interest However, it has several limitations, the most significant being
as an alternative source of energy as the conventional sources its cost and scarcity. Since it is sensitive to both acidic and
of energy are being depleted. Most of today’s solar cells are basic environments, it is brittle, due to which it cannot be used
made using silicon which is expensive because it has to be for flexible electronics; metal grids, metallic nanowires, and
purified and made into crystals that are sliced thin. Scientists metal oxides have been tried as alternatives. Recently, gra-
the world over have been exploring alternatives, such as nano- phene has emerged as a most promising material due to its
structured or hybrid solar cells. Indium tin oxide (ITO) has higher transmission over a wide wavelength range. An ideal
been used invariably as a transparent electrode in solar cells. monolayer of graphene has a transparency of 98%, sheet resis-
Being expensive, indium has found a suitable replacement in tance of 6 KΩ/sq, therefore, graphene can fulfill the role of
graphene, which is made from ubiquitous carbon. Recently, TCE in flexible electronics. Apart from TCE, graphene can
scientists have explored building semiconducting nanostruc- also be used in polymer solar cells (PSC) as an interfacial
tures directly on a pristine graphene surface without impair- layer and as an additive.
ing its electrical and structural properties due to its stable and The interfacial layer is one of the most vital components of
inert structure. In this chapter, we summarize and discuss, a solar cell as it plays a significant role in improving the fill
comprehensively, the research made so far for applications factor by minimizing the contact resistance, charge recombi-
of graphene in solar cells, including transparent electrodes, nation, and current leakage, thus enabling an efficient charge
active layers, and interface layer. extraction. These layers commonly termed as electron trans-
port layer (ETL)/hole transport layer (HTL), are incorporated
at the anode/cathode, respectively. The surface of a photoac-
19.1 INTRODUCTION
tive donor contains high densities of carrier traps and interfa-
Since its discovery by Novoselov et al.1 graphene sheet has cial trap holes that can hinder an efficient charge collection.
found its application in solar cells as a transparent conducting Incorporating interlayers that circumvent the direct contact
electrode (TCE), interfacial layer (both as an electron and hole between the photoactive donor and electrodes neutralizes this
transport layer), active layer, and in Schottky junctions.2 Table problem. However, the interlayers remain an open area for
19.1 compares the advantages and disadvantages of graphene development due to cost and compatibility with ITO.3,4
as compared with conventional materials. Optoelectronic The main functions identified with an interfacial layer are
devices, for example, touch screens, light emitting diodes described below:4,5
(LEDs), and solar cells require materials that are highly trans-
parent and have low sheet resistance. Transparent conductors 1. Decreasing the energy barrier for charge injection
currently being used are mainly indium oxide (In2O3), zinc and extraction
oxide (ZnO), or tin oxide (SnO2), as well as ternary com- 2. Forming a selective contact for single type of charge
pounds based on their combinations. In solar cells, indium tin carrier
oxide (ITO), which is a doped n-type semiconductor (~90% 3. Determining the polarity of the device
361
TABLE 19.1
Advantages and Disadvantages of Graphene as Compared to Conventional Materials for Application in Solar Cells
Used as Nongraphene Graphene
Transparent ITO: • The resistance is 18 Ω/sq. with a transparency
conducting • Resistance is 10–30 Ω/sq. while transparency is 80% at 550 nm. of 98%.
electrode • The advantages over ITO are its high
Disadvantages
transmission over a wide wavelength range, high
• High production cost.
electron mobility, easily tunable band gap, and
• Limited indium reserves and the diffusion of ions into the polymer layers.
wide application in various configuration of
• It cannot be used for flexible devices being brittle and rigid. organic solar cells.
HTL/ETL PEDOT:PSS: • It is a promising and least expensive material.
• It has a high work function and is compatible with ITO electrode. • It is a highly flexible material other than being
• Being acidic, it causes the degradation of ITO. highly conductive and transparent.
Active layer PCBM: • They exhibit high electron mobility
• They have high cost and low absorption spectrum in the visible region. (104 cm2 v−1s−1) with an easily tunable band gap,
and an enhanced solubility in organic solvents.
CNTs:
• It provides an enhanced interfacial area for the
• They have high charge mobility and a long π–π conjugation but gives low PCE.
exciton generation and charge transfer.
4. Modification in the surface property that alters the is too high for commercial applications, and its absorption
film morphology spectrum in the visible region is very low, therefore, alter-
5. Suppression of diffusion and reaction between the native acceptors were developed. Carbon nanotubes (CNTs)
electrode material and polymer were the first of such material to be tried as an acceptor due
6. Modulation of optical field as an optical spacer to its high charge mobility, long π–π conjugation, along with
large aspect ratio. Although CNT is a very cost-effective
Poly-3,4-ethylenedioxythiophene:polystyrenesulfonate alternative as an acceptor, it gives very low power conversion
(PEDOT:PSS) is the most commonly used anodic interlayer efficiency (PCE). Graphene and graphene-based materials,
or HTL due to its high work function and compatibility with in particular, graphene quantum dots (GQDs), have emerged
ITO. However, acidic PEDOT is highly hygroscopic that can in recent years as an ideal alternative for acceptor material.
cause degradation of ITO, thus reducing the device’s over- Having high electron mobility along with easily tunable
all performance considerably. Inorganic materials are being band gap, it exhibits good solubility in organic solvents after
used as an alternative to PEDOT:PSS, but the processing of functionalization. It provides a large interfacial area for the
these materials is expensive. New materials are required to exciton generation and a continuous pathway for the electron
replace the current HTL materials. In this regard, solution- transfer due to its large 2-D planar structure and atomical
processable graphene oxide (GO) sheets have emerged as the thickness property.6,7
most promising and least expensive material with high per- Apart from PSC, graphene has also found its application
formance. Similarly in the case of ETL, various metal oxides in dye-sensitized solar cells (DSSCs) (Figure 19.2)8 as well
such as calcium (Ca) and barium (Ba)3 have been used exten- as in silicon solar cells. DSSCs are attractive because they
sively. However, they are impractical for commercial use due are composed of cheap materials (which need not be of high
to their instability and high cost. Graphene, with its high con- quality), along with low printing cost. In 1991, O’Regan and
ductivity, high transparency, and flexibility has emerged as Gratzel9 discovered that a film of TiO2 nanoparticle would
one of the most promising candidates for ETL as well, not act as an n-type photo-anode in a DSSC and can increase the
only for the rigid devices but also for the flexible electronics. available surface for a dye for attachment by a factor of few
The active layer is the most significant component of a thousands. Graphene was found to facilitate a smooth transfer
solar cell. It is composed of a donor and an acceptor material of electrons in DSSC.8
that exists either as adjacent layers or in a bulk heterojunction In the case of silicon solar cells, the first carbon that was
(BHJ) formation. In BHJ, the donor and acceptor are mixed used was diamond-like amorphous film (a-C). However, due
together such that the separation between the two is at a to its unique polar nature, it was extremely difficult to pro-
nanoscale range (Figure 19.1). Light absorption of the active cess dopants by diffusion, and to remove defects by annealing
layer leads to the generation of bound electron/hole pairs because of its large bond and diffusion energies, they were
called as exciton as shown in Figure 19.1. These electrons and then replaced by CNTs. CNT film’s network exhibit a lot of
holes are collected at the two opposite electrodes contribut- porosity between bundles which decreases the conductivity
ing to the current generation. Conventional solar cell employs of the film, although it increases the transparency. Recently it
various polymers as donors and PCBM as an acceptor. The was shown that graphene films could be combined with sili-
cost of [6,6]-phenyl C71 butyric acid methyl ester (PCBM) con (Si) to form Schottky junctions for efficient solar cells.
Graphene-Based Solar Cells 363
ηED
ηCT
e Polymer L
U MO ηCC
e
ηA e PC
hv
BM LU MO
4.7 eV
h h ~Voc
5.0 eV
h h
h
Polymer H
O MO
PCBM HO
MO
ITO PEDOT:PSS Polymer/PCBM Ca/Al
AM 1.5G reference
1.6 spectrum (IEC 6090)
Irradiance (W m–2 nm–1)
P3HT:PCBM
cell response
1.2
Silicon cell
response
0.8
0.4
0.0
400 600 800 1000 1200
Wavelength (nm)
Cathode
~100 nm
Anode
~20 nm
FIGURE 19.1 (a) The operating mechanism of a PSC. (Reprinted by permission from Macmillan Publishers Ltd. Li, G., R. Zhu, and Y.
Yang. Nature Photon 6, no. 3: 153–61, copyright 2012.) (b) Comparison between solar spectrum and the photoresponse of a P3HT: PCBM
solar cell. (c) Conceptual morphology model with bicontinuous interpenetration network of the polymer and the acceptor. ((b,c) Reprinted
with permission from Yang, X. and J. Loos 2007. Toward high-performance polymer solar cells: The importance of morphology control.
Macromolecules 40, no. 5: 1353–62. Copyright 2007 American Chemical Society.)
19.2 GRAPHENE IN ORGANIC SOLAR CELLS 19.2.1 Transparent Conducting Electrode

Graphene is being used as a TCE, interfacial layer (both A TCE is a key component of the optoelectronic device
as an electron and hole transport layer), active layer through which light couples into the device. High transpar-
andDSSCs. Table 19.2 summarizes the discussion that ency, low resistance, and an appropriate work function are
follows. few properties that are taken into account while considering
(a)
Plantinum-coated FTO (back contact)
Electrolyte
Dye-covered TiO2
FTO (front contact) Sensitizing dye Titania nanoparticle
(b) –3.5 eV
LUMO (D*)
Electron injection
Electron injection
hv
Energy levels (compared with vacuum)
–4.0 eV overpotential
EF
–4.5 eV TiO2/electrolyte recombination

VOC
TiO2/dye recombination I3–/I–
–5.0 eV I2•–/I1– Dye regeneration

overpotential
HOMO (D) Dye regeneration

–5.5 eV
Mesoporous TiO2 Sensitizing dye Electrolyte
FIGURE 19.2 Dye-sensitized solar cell device schematic and operation. (a) Liquid-based DSCs are comprised of a transparent conducting
oxide (such as fluorine-doped tin oxide, FTO) on glass, a nanoparticle photoanode (such as titania) covered in a monolayer of sensitizing
dye, a hole-conducting electrolyte and a platinum-coated, FTO-coated glass back-contact70. (b) Energy level and device operation of DSCs;
the sensitizing dye absorbs a photon (energy hν); the electron is injected into the conduction band of the metal oxide (titania) and travels to
the front electrode (not shown). The oxidized dye is reduced by the electrolyte, which is regenerated at the counter-electrode (not shown) to
complete the circuit. VOC is determined by the Fermi level (EF) of titania and the redox potential (I3−/I−) of the electrolyte. (Reprinted by
permission from Macmillan Publishers Ltd. Hardin, B. E., H. J. Snaith, and M. D. McGehee. The renaissance of dye-sensitized solar cells.
Nature Photon 6, no. 3: 162–9, copyright 2012.)
a material for such a use. In conventional solar cells, ITO transparency, and flexibility. An ideal monolayer of graphene
has been used as a hole-collecting electrode with transpar- has charge carrier mobility higher than 200,000 cm2/V·s and
encies >90% at the wavelength of 550 nm, sheet resistances transparency of 98%.1 These characteristics open the possi-
of 10–30 Ω/sq, and a favorable work function, with these bility to replace ITO with graphene as the TCE, particularly
properties being the market standard for most of the transpar- for flexible and large-area device applications. Therefore, gra-
ent electrode applications.2 The scarcity of indium reserves phene can fulfill multiple functions in photovoltaic devices
and high cost led to the search for the novel electrode mate- as the transparent conductor window, photoactive material,
rial with higher stability, transparency, and conductivity. channel for charge transport, and catalyst. Graphene trans-
Recently developed high-performance ITO alternatives, parent conducting films (GTCF) can be used as window
such as s ingle-walled CNTs (SWCNTs) and metal nanowires electrodes in inorganic, organic, and DSSC devices. Initially,
are also costly.44–46 Therefore, new cost-effective and high- there were several problems associated with the large-scale
performance materials are important for the development of patterned growth of graphene, which were the main obstacles
TCE for high-performance solar cells. in using this material for device application.
Recently, graphene has emerged as a most promising TCE Kim et al.47 provided a breakthrough in 2009 by propos-
material due to its high theoretical electrical conductivity, ing a feasible method for the direct synthesis of large-scale
TABLE 19.2
Experimental Details and Performance of Graphene in Solar Cells
Application Experimental Details Conclusion References
Transparent Graphene-carbon nanotubes nanocomposites were used. Sheet resistance of graphene 240 Ω/sq. at 86% [10]
conducting Glass/graphene-CNT/ PEDOT:PSS/active layer/Ca/Al. transmittance after chemical doping. Voc = 0.583V and
electrode (TCE) PCE = 0.85%.
Graphene was synthesized via modified Hummer’s Sheet resistance of graphene 230 Ω/sq. and [11]
method. transmittance of 72%. Similar Voc for both devices.
Glass/ITO or graphene/PEDOT:PSS/CuPc/fullerene/ Enhanced capability to operate under extreme bending
BCP/Al. conditions up to 140C.
Reduced graphene oxide (rGO) film as TCE onto a PET Sustained 1000 cycles of bending PCE = 0.78%. [12]
substrate, referred to as PET/rGO, which was then used
for the OPV devices.
PET/rGO/PEDOT:PSS/P3HT:PCBM/TiO2/Al
Multilayer graphene (MLG) synthesized via CVD Film prepared at 1000°C gave the optimum sheet [13]
process. Effect of MLG synthesis temperature on the resistance of 606 Ω/sq and transmittance of 87%.
performance at 800C, 900C, 1000C was studied.
Glass/MLG/PEDOT:PSS/P3HT:PCBM/ TiOX/Al.
Chemically converted graphene (CCG) suspension Two-step reduction approach restored the sp2 carbon [14]
obtained via controlled chemical reduction of exfoliated networks of the CCG sheets. CCG-A800 films as TCE
graphene oxide (GO) in the absence of dispersant. yielded PCE = 1.01%.
Glass/Graphene/PEDOT:PSS. P3HT:PCBM/Al.
MLG with doping between graphene layers to obtain high 4-layered graphene without impurities were obtained. [15]
conductivity. Glass/4-layered graphene/ PCE = 2.5%.
MoOX + PEDOT:PSS/active layer/Al.
Solution processed rGO-SWCNT composite doped with PCE = 1.27% with rGO-SWCNT composite doped with [16]
alkali carbonates. Cs2CO3.
Glass/rGO-SWCNT doped with alkali/active
layer/V2O5/Al.
Doped graphene with Au nanoparticles and used as a top PCE = 2.7% was obtained in graphene-Au TCE. [17]
electrode in inverted configuration. Glass/ITO/ZnO/
P3HT:PCBM/PEDOT: PSS/graphene-Au.
MLG obtained via CVD was doped with SoCl2 or HNO3. Reduced sheet resistance after doping. PCE = 2.60% [18]
Doped MLG/PEDOT:PSS/ P3HT:PCBM/Ca/Al. with SoCl2 doped MLG.
Graphene mesh electrode (GME)/PEDOT:PSS/ PCE = 2.04%. [19]
P3HT:PCBM/LiF/Al.
Fabrication of package-free flexible inverted solar cell on Sheet resistance of doped graphene sheets = 68 Ω/sq. [20]
polyimide (PI) substrates with highly doped multilayer PCE = 3.2%.
graphene.
PI/Metal/ZnO/P3HT:PCBM/PEDOT:PSS(Au)/Graphene/
PMMA.
Graphene-Al-TiO2 composite prepared by self-assembly Transmittance of the modified SLG was 96.0% at [21]
method to deposit the solution-processed TiO2 on the 500 nm and sheet resistance = 1.2 kΩ/sq.
Al-covered graphene. PCE = 2.58%.
Glass/Au grid/Graphene-Al-TiO2/ PEDOT:PSS/
P3HT:PCBM/MoO3/Ag.
Doped PEDOT:PSS with perfluorinated isomer to form a PCE = 4.33%. [22]
uniform layer on graphene electrode. Graphene/modified
PEDOT:PSS/ Active layer/Al.
Few layer graphene (FLG) was electrostatically doped Sheet resistance of electrostatically doped FLG [23]
by a layer of ferroelectric polymer coating films reduced to 70 Ω/sq at 87% optical transmittance.
{poly(vinylidene fluoride-co-trifluoroethylene) Such graphene-based OSCs exhibit PCE of 2.07%
(P-(VDF-TrFE)}. with a superior stability when compared to chemically
Polarized graphene/PEDOT:PSS/ P3HT:PCBM/Ca/Al. doped graphene-based OSCs.
Interfacial layer Graphene oxide (GO) thin films as the hole transport The incorporation of GO leads to a decrease in leakage [24]
layer (HTL) in solar cells is demonstrated. ITO/GO/ currents. With 3-nm-thick layer of GO, PCE = 3.5%
P3HT: PCBM/Al. was achieved.
(Continued)
TABLE 19.2 (Continued)

Solution-processed graphene oxide GO/NiOx bi-layer as Optimum PCE = 3.48% with GO/NiOx HTL. [25]
HTL in solar cells. ITO/HTL/P3HT:PCBM/LiF/Al.
HTL = GO, NiOX, GO/NiOX and NiOX/GO.
Conventional reduced graphene oxide (rGO) and a PCE = 3.53%. [26]
newly reduced graphene oxide (pr-GO) by
using p-toluenesulfonyl hydrazide (p-TosNHNH2)
as HTL in solar cells. ITO/rGO or pr-GO/P3HT: PCBM/
Ca/Al.
Various concentration of GO:SWCNTs films were studied Optimum PCE = 4.10% with GO:SWCNT (1:0.2). [27]
as an HTL. ITO/GO:SWCNTs/P3HT: PCBM/Al.
An alternative method of refluxing exfoliated graphite Graphene sheets soluble in ethanol were synthesized. [28]
(EG) with nitric acid (HNO3) solution, followed by PCE = 3.7% with v/v ratio of .FLG/PEDOT:PSS as
ultrasonicating in ammonia–ethanol solution, and finally 2:1.
dispersing in ethanol, was used to achieve FLGs.
ITO/FLG/PEDOT:PSS/P3HT:PCBM/Ca/Al.
GO ETL by stamping transfer process. ITO/PEDOT:PSS/ PCE = 7.50% with GO/TiOx. [29]
PCDTBT:PC71BM /GO/TiOx/Al.
Solution-processable electron transport layer (ETL) based The wrapping of the graphene layers around the metal [30]
on composites of metal oxides and rGO were explored. oxide nanoparticles leads to very smooth films with
ITO/PEDOT:PSS/PTB7:PCBM/ETL/Al. ETL = ZnO, surface roughness of <3 nm. PCE = 7.50% with ZnO/
ZnO/RGO, TiO2, TiO2/RGO, BCP. RGO as ETL.
Pronounced packaging effect of multilayer graphene top Double layer graphene shows optimum PCE = 3.2%. [20]
electrodes on the solar cells in air was demonstrated. PI/
Metal/ZnO/P3HT:PCBM/PEDOT:PSS(Au)/Graphene/
PMMA. Metal: Al, Cu, Ag.
Solution-processable graphene quantum dots (GQDs) and PCE of 6.82% (DR3TBDT:PCBM) and 3.51% with [31]
GO were tested for as HTL. ITO/HTL/Active layer/LiF/ (P3HT:PCBM).
Al. HTL: GO, GQDs.
Active layer: P3HT:PCBM, DR3TBDT:PCBM.
Graphene oxide (GO)/GO-Cs (cesium neutralized GO) PCE = 3.91% with GO-Cs/Al/GO/MoO3 as an [32]
bilayer modified with ultrathin Al and MoO3 can act as interconnecting layer (ICSL).
an efficient interconnecting spacer layer (ICSL) in
tandem PSC. ITO/PEDO:PSS/PCDTBT:PCBM/ICL/
PCDTBT:PCBM/Ca/Al ICSL = GO-Cs/GO, GO-Cs/Al/
GO, GO-Cs/Al/GO/MoO3, Al/GO/MoO3, GO-Cs/Al/
MoO3, Al/MoO3.
Active layer A solution-processable functionalized graphene (SPF Postfabrication annealing at 160°C and 120°C for 10 [33]
Graphene) material as electron-acceptor material and and 20 min each, respectively, gives optimum results.
P3OT and P3HT as donor material was fabricated. ITO/ PCE = 1.4% (160°C for 20 min).
PEDOT:PSS/P3OT:SPF Graphene/LiF/Al.
Graphene-PDI hybrids were prepared by hydrothermal PCE = 1.04%. [34]
treatment of graphene oxide (GO) and PDI in DMF.
ITO/PEDOT:PSS/P3HT:PDI G hybrid::1:1/LiF/Al.
Monosubstituted C60 onto graphene nanosheets was Increase in PCE by 2.5-fold after lithiation. [35]
attached via lithiation reaction, and the resultant PCE = 1.22%.
C60-grafted graphene nanosheets were used as electron
acceptors in P3HT-based solar cell. ITO/PEDOT:PSS/
C60-G:P3HT/Al.
Graphene quantum dots (GQDs) blended with PCE = 1.14% with 1 wt% ANI-GQDs. [36]
RR-P3HT or MEH-PPV. ITO/PEDOT:PSS/P3HT:ANI-
GQDs/LiF/Al. ITO/PEDOT:PSS/MEH-PPV:MBGQDs/
LiF/Al.
Electrochemical approach for one-step preparation of PCE = 1.28%. [37]
functional GQDs with a uniform size of 3–5 nm. ITO/
PEDOT:PSS/P3HT:GQDs/Al.
(Continued)

Different types of graphene oxide quantum dots PCE = 7.60% with GQDS reduced for 5 h. [38]
(GOQDs) prepared by 5 h and 10 h reduced GQDs.
were tested in BHJ solar cells. Blended PTB7:PCBM
with GOQDS, GQDs. ITO/PEDOT:PSS/Blended
PTB7:PCBM/TiOx/Al.
DSSC Conductive, transparent, and ultrathin graphene films were PCE = 0.84%. [39]
fabricated from exfoliated graphite oxide, followed by
thermal reduction which was then employed as a window
electrode. Graphene/TiO2/dye/spiro-OMeTAD/Au.
Graphene was used as a 2-D bridge into the Faster electron transport and a lower recombination, [40]
nanocrystalline electrodes. together with a higher light scattering. PCE = 6.97%.
FTO/graphene composite electrode/TiO2/Dye/Pt.
CCG was prepared via modified Hummers method and PCE = 3.0%. [41]
MWCNTs were deposited over it. FTO/GMQNTS/
electrolyte/Pt.
GQDs were used to dope PPy as an efficient PCE = 5.27%. [42]
counter-electrode.
3-D graphene networks prepared by CVD was used as a PCE = 6.58%. [43]
photoanode.
ITO/photoanode/electrolyte/counter-electrode/ITO glass.
graphene films using chemical vapor deposition (CVD) on large-scale graphene films which gives 99% yield. This dry-
thin nickel layers for using the material in device applica- transfer process turns out to be very useful in making large-
tion. In the well-established CVD process of hydrocarbons scale graphene electrodes and devices without an additional
on reactive nickel or transition metal carbide surfaces, the lithography process.
major problem was the absorption of large amounts of car- As shown in Figure 19.5a, the transmittance of a gra-
bon sources on nickel foils, which formed thick graphite crys- phene film can vary from 76.3% to 83.7% depending upon
tals rather than graphene films. To solve this problem, Kim the number of graphene layers in the film. The transfer prop-
et al.47 used thin layers of nickel (<300 nm) deposited onto erty measurements (Figure 19.5b) showed that the electron
SiO2/ Si substrates using an electron beam evaporator, and mobility of graphene is 3750 cm2 V−1 s−1 at a carrier density
the samples were then heated to 1000°C inside a quartz tube of 5 × 1012 cm−2. In Figure 19.5b, the half integer quantum
under an argon atmosphere. After flowing reaction gas mix- hall effect observed for a high magnetic field of 8.8T cor-
tures (CH4:H2:Ar = 50:65:200 standard cm3/min), the samples responds to a monolayer graphene, indicating that the quality
were cooled down to room temperature (25°C) at the rate of of CVD-grown graphene is comparable to that of mechani-
~10°C s−1 using flowing argon. It was discovered that this fast cally cleaved graphene.11,48,49
cooling rate is critical in suppressing the formation of multiple The graphene film exhibits excellent mechanical proper-
layers and for separating graphene layers efficiently from the ties when used to make flexible and stretchable electrodes
substrate in the later process.47 (Figure 19.5c and d). The flexibility of the graphene films
Figure 19.3 shows the synthesis, etching, and transfer was measured by transferring it to a polyethylene terephthal-
process of graphene by CVD. The use of strong acid, such ate (PET) substrate (thickness, 100 mm) coated with a thin
as nitric acid (HNO3), can produce hydrogen bubbles and polydimethylsiloxane layer (thickness 200 mm; Figure 19.5c)
damage the graphene. The use of aqueous iron (III) chloride and measuring resistances with respect to the bending radii.
(FeCl3) etches the nickel layer effectively within a mild pH The resistances showed a little variation up to the bending
range without forming gaseous products or precipitates. The radius of 2.3 mm (approximate tensile strain of 6.5%), and
net ionic equation of the etching reaction can be represented are perfectly recoverable after unbending. The resistances of
as follows: graphene films with respect to uniaxial tensile strain rang-
ing from 0% to 30% (Figure 19.5d), showed similar results
2Fe3+ (aq) + Ni(s) → 2Fe 2 + (aq) + Ni 2 + (aq) (19.1) and recovery to its original resistance after stretching by 6%
(Figure 19.5d, left inset). However, further stretching often
results in mechanical failure.11
To remove the remaining nickel layers, HF solution can The first application of CVD-grown graphene as TCE
be used.47 Figure 19.4 shows the actual transfer process for in organic photovoltaic cells was shown by Gomez et al.11
(a) Patterned Ni layer (300 nm) Ni/C layer

CH4/H2/Ar Ar
~1000°C Cooling
Ni ~RT
Si SiO2 (300 nm)
(b) Graphene on a substrate

PDMS/graphene/Ni/SiO2/Si PDMS/graphene
FeCl3(aq)
or acids
Ni-layer Stamping
etching
Downside contact
(scooping up)
(c) Graphene/Ni/SiO2/Si Floating graphene/Ni Floating graphene
HF/BOE HF/BOE
SiO2-layer Ni-layer
etching etching
(short) (long)
FIGURE 19.3 Synthesis, etching, and transfer processes for the large-scale and patterned graphene films. (a) Synthesis of patterned
graphene films on thin nickel layers. (b) Etching using FeCl3 (or acids) and transfer of graphene films using a PDMS stamp. (c) Etching
using BOE or hydrogen fluoride (HF) solution and transfer of graphene films. RT, room temperature (25°C). (Reprinted by permission from
Macmillan Publishers Ltd. Kim, K. S., Y. Zhao, H. Jang et al. Large-scale pattern growth of graphene films for stretchable transparent
electrodes. Nature 457, no. 7230: 706–10, copyright 2009.)
Graphene films synthesized by CVD, were transferred to a P3HT:PCBM (poly-(3-hexylthiophene)/phenyl-C61-butyric

transparent substrate, and evaluated in organic solar cell acid methyl ester)/TiOx (titanium oxide)/Al. Devices fabri-
(OSC) heterojunctions (TCE/PEDOT:PSS/copper phthalo- cated using MLG films prepared at 1000°C (sheet resistances
cyanine (CuPc)/fullerene/bathocuproine (BCP)/aluminum) of 606 Ω/cm2 and transmittance of 87%), gave the best per-
as shown in Figure 19.6a. The device structure and current formance as an electrode in OPVs. Figure 19.7b shows the
density versus voltage (J–V) characteristics under dark and 1 incident photon to converted electron (IPCE) spectrum
sun illumination (simulated AM 1.5G irradiated at 100 mW/ and the absorption pattern for the MLG film prepared at
cm2) are shown in Figure 19.6b. The graphene and ITO-based 1000°C. Wong et al.15 performed an interface engineering of
devices have nearly identical open-circuit voltage (Voc) of layer-by-layer (LBL) stacked graphene anodes TCE for high-
0.48 V under illumination conditions which suggests similar performance OSCs. The advantage of LBL over single-layer
recombination behavior for both the cells. Besides the added graphene is that LBL has a much higher conductivity. To obtain
advantage of such films is the enhanced capability to oper- high extrinsic conductivity, appropriate doping between the
ate under extreme bending conditions up to 140°, whereas an individual graphene layers of LBL is required. Figure 19.8a
ITO shows an irreversible bending at 60° and displays cracks shows the process of obtaining LBL by the PMMA process
thereafter. However, performance of both devices was slightly as well as the doping process. The second layer of graphene
degraded upon bending (Figure 19.6c). The continuous nature can be bonded with the first layer through π–π interaction by
of CVD graphene films leads to the minimum surface rough- annealing them at 120°C for 10 min. Therefore; N-layered
ness (Figure 19.6d), which leads to a lower resistance of graphene without any organic impurity between the layers can
nearly 230 Ω/sq. at 72% transparency. Thus, the graphene be obtained. Figure 19.8b shows the LBL-stacked graphene
films obtained from such a method exhibit a great potential in from 1 to 8 layers transferred onto quartz where the trans-
flexible solar devices. mittance of graphene films with increasing number of layers
Choe et al.13 explored TCE based upon CVD-synthesized can be seen. Figure 19.8c, d shows the spectral micrographs.
multilayer graphene (MLG) films in BHJ OSCs. They stud- The effect of LBL on transmittance and sheet resistance is
ied the effect of graphene synthesis temperature on the device shown in Figure 19.8e and f, respectively. PCE for the device
performance. Figure 19.7a shows the device performance for (Figure 19.8g) with 4-layer acid-doped MLG having structure
three different MLG films prepared at different growth tem- GO/(molybdenum oxide) MoO3 + PEDOT:PSS gives the best
peratures (800°C, 900°C, and 1000°C), which were compared PCE of 2.5% as shown in Figure 19.8h. The device efficiency
for their performance as an electrode in organic photovolta- of graphene-anode-based devices (2.5%) modified by MoO3/
ics (OPV) in the configuration: glass/graphene/PEDOT:PSS/ PEDOT:PSS is very close to that of ITO-based devices (3%).
(a) (b) (c)
(d) (e)
2 cm
2 cm
(f ) (g) (h)
Stamping Patterned graphene
FIGURE 19.4 Transfer processes for large-scale graphene films. (a) A centimeter-scale graphene film grown on a Ni (300 nm)/SiO2(300 nm)/
Si substrate. (b) A floating graphene film after etching the nickel layers in 1 M FeCl3 aqueous solution. (c) Various shapes of graphene films
can be synthesized on top of patterned nickel layers. (d), (e) The dry-transfer method based on a PDMS stamp is useful in transferring the
patterned graphene films. (f) Graphene films on the PDMS substrates are transparent and flexible. (g), (h) The PDMS stamp makes confor-
mal contact with a silicon dioxide substrate. (Reprinted by permission from Macmillan Publishers Ltd. Kim, K. S., Y. Zhao, H. Jang et al.
Large-scale pattern growth of graphene films for stretchable transparent electrodes. Nature 457, no. 7230: 706–10, copyright 2009.)
Yin et al.12 examined reduced GO (rGO) films as transparent improving the overall performance of the device. Solution-
electrodes for OPV devices on highly flexible PET substrates. processed rGO-SWCNT, when doped with alkali carbonates,
The transparency of rGO determines the efficiency of light can reduce the work function of the pristine rGO-SWCNT
transmission, which dominates the device performance. It from 4.6 to 3.4 due to the formation of interfacial dipoles.
was able to sustain 1000 cycles of bending at the tensile strain Figure 19.9a shows that the presence of a thin layer of each
(approx. 2.9). The highly flexible properties of such films can of the alkali carbonates decreases the value of Φw of the
have potential applications in flexible optoelectronics. rGO-SWCNTs. In Figure 19.9b, the effect of the shift in
Another approach to enhance the electrical and mechani- the values of Φw on the J–V performance was examined
cal properties of chemically converted graphene is to form using an inverted architecture in which rGO-SWCNT films
a nanocomposite with CNTs in the form of graphene-carbon incorporating alkali carbonates were spin-cast onto the pre-
nanotubes (G-CNTs),10 which does not involve the use of sur- cleaned glass substrates. The P3HT/PCBM layer was then
factants that typically degrades the electrical performance. cast from dichlorobenzene onto the carbon-based electrode.
The G-CNT TCE-based solar cell device gives PCE of 0.85%. CsCo3-doped rGO-SWCNT (transparency 65.8% and sheet
The current density (Jsc ), open circuit voltage (Voc ), and fill resistance 331 Ω/sq) gives the optimum Jsc = 4.77 mA/cm2,
factor (FF) were 3.47 mA/cm2, 0.583 V, and 42.1%, respec- Voc = 0.60 V, FF = 44%, PCE of 1.27%.16
tively. The low Jsc and FF were likely due to poor contact at Liu et al.17 obtained an increase of more than 400% in
the interface between the G-CNT and the polymer blend. conductance of graphene by its chemical doping. They doped
Along with higher conductivity and transparency, another graphene with Au nanoparticles and PEDOT:PSS, and used
prerequisite for an efficient TCE is the appropriate work it as a top electrode for an inverted semitransparent OSC
function that can minimize the energy barriers for charge with P3HT:PCBM as an active layer and ITO as the bottom
injection.48 Modulating the work functions (Φw) to form an electrode. The device structure for solar cell fabrication is
ohmic contact improves the degree of charge injection, thus shown in Figure 19.10a. The J–V characteristics and external
(a) (b)
83.7% UV for 6 h
80.7% UV for 4 h uv 4
Resistance (Ω)
90 79.1% UV for 2 h
76.3% initial 10
2 Rxy
at 550 nm
85
Magnetoresistance (kΩ)
5 0
Transmittance (%)
–60 0 60
80 Vg (V)
1.2 Rs 84 0
Rs (kΩ per square)

75 1.0 Tr Rxx
82
0.8
Tr (%)
80 –5
70 0.6
5 μm
0.4 78
65 0.2 –10
76
0 1 2 3 4 5 6
Time (h)
60 –15
400 600 800 1000 1200 –60 –40 –20 0 20 40 60
Wavelength (nm) Vg (V)
(c) 9 (d) 108
102 104 Ry
x
8 Rx
Anistropy, Ry/Rx
107 y
Resistance (Ω)
y 103
7
101 x
6 106 102
1st 2nd 3rd
Resistance (kΩ)
Resistance (Ω)
5 105 Stretching cycles

100 101 y
0.0 0.4 0.8 1.2
4 Curvature, k (mm–1) 0 3 6 0 3 6 0 3 6
104 x
Stretching (%)
3
Ry Ry
Rx 103
2
Bending Recovery
1 102 Rx
Stable
0
101
Flat 3.5 2.7 2.3 1.0 0.8 Flat 0 5 10 15 20 25 30
Bending radius (mm)
Stretching (%)
FIGURE 19.5 Optical and electrical properties of the graphene films. (a) Transmittance of the graphene films on a quartz plate. The upper
inset shows the ultraviolet (UV)-induced thinning and the consequent enhancement of transparency. The lower inset shows the changes in
transmittance, Tr, and sheet resistance, Rs, as functions of ultraviolet illumination time. (b) Electrical properties of monolayer graphene. The
upper inset shows a four-probe electrical resistance measurement on a monolayer graphene Hall bar device (lower inset) at 1.6 K. The main
panel shows longitudinal (R xx) and transverse (R xy) magnetoresistances measured in this device for a magnetic field B58.8 T. (Here e is the
elementary charge and h is Planck’s constant.) Quantum Hall plateaux (horizontal dashed lines) are developing for higher filling factors.
(c) Variation in resistance of a graphene film transferred to a 0.3-mm-thick PDMS/PET substrate for different distances between holding
stages (i.e., for different bending radii). The left inset shows the anisotropy in four-probe resistance, measured as the ratio, Ry/R x, of the
resistances parallel and perpendicular to the bending direction, y. The right inset shows the bending process. (d) Resistance of a graphene
film transferred to a PDMS substrate isotropically stretched by 12%. The left inset shows the case in which the graphene film is transferred to
an unstretched PDMS substrate. The right inset shows the movement of holding stages and the consequent change in shape of the graphene
film. (Reprinted with permission from Gomez, D. A. et al. Continuous, highly flexible, and transparent graphene films by chemical vapor
deposition for organic photovoltaics. ACS Nano 4, no. 5: 2865–73. Copyright 2010 American Chemical Society.)
quantum efficiency (EQE) of the devices based upon second function as shown in Figure 19.10d. These kinds of OSCs
approach are shown in Figure 19.10b and c. The device with are found to be very stable after packing, as tests showed
Au and PEDOT:PSS co-doped graphene shows the maxi- very little change in efficiency even after 6 months. Most
mum PCE of 4.7% as compared to the device without the recently, a novel technique of standard industrial photoli-
Au nanoparticle, which shows the PCE of 2.02%. The best thography and O2 plasma etching process, for the synthesis
device performance was obtained with the Au nanoparti- of graphene mesh electrodes (GMEs) with good conductiv-
cles with Jsc, Voc, FF, PCE of 10.58 mA/cm 2, 0.59 V, 43.3%, ity and transparency, was proposed.19 GMEs with 50 μm
2.70%, respectively. The Voc is increased for about 20 mV period and 10 μm line width have the highest transpar-
after PEDOT:PSS doping due to an increase in the work ency with a comparable sheet resistance, and give the best
(a) 2.2 eV (b) (c)

10–2 0.6
3.0 eV
3.3 eV Graphene
(A/cm2)
4.0 eV 10–4 0.5
Power density (mW/cm2)

PEDOT 4.3 eV ITO
Graphene
4.5 eV CuPc 10–6 0.4
Current density
Al
BCP
Graphene C60 10–8
Fill factor
5.2 eV 8 Dark (mA/cm2) 0.3
5.2 eV Light (mA/cm2)
4
(mA/cm2)
6.2 eV Power (mW/cm2) 0.2
6.4 eV Al
C60
0
BCP 0.1
CuPc –4
0.0
PEDOT –8
–0.8 –0.4 0.0 0.4 0.8 0 20 40 60 80 100 120 140
Substrate
Graphene Voltage (V) Bending angle (2θ)
(d) Graphene roughness: 0.9 nm ITO roughness: 0.7 nm SWNT roughness: 8.4 nm
50 nm 50 nm 50 nm
0.8 1.0 µm 0.8 1.0 µm

0.8 1.0 µm
0.4 0.6 0.6 0.6
0.2 0.4 0.4
0.2 0.2
FIGURE 19.6 (a) Schematic representation of the energy level alignment (top) device fabrication (bottom) from the graphene films syn-
thesized by CVD, transferred to transparent substrates. (b) Logarithmic (top) and linear (bottom) current density and power density versus
voltage characteristics of graphene. (c) Fill factor dependence of the bending angle for CVD graphene and ITO devices. (d) AFM images
of the surface of CVD graphene, ITO, and SWNT films on glass. The scale bar in z-direction is 50 nm for all images. (Reprinted with
permission from Gomez, D. A. et al. Continuous, highly flexible, and transparent graphene films by chemical vapor deposition for organic
photovoltaics. ACS Nano 4, no. 5: 2865–73. Copyright 2010 American Chemical Society.)
performance as a transparent electrode in an OPV cell. OPV of the aromatic rings, thus reducing its conductivity. Also,
cells using GMEs as the transparent electrodes with a blend there exists a trade-off between the transparency and conduc-
of P3HT/PC61BM as an active layer have been fabricated: tivity along with a high sheet resistance that limits the appli-
GME/PEDOT:PSS/P3HT:PC61BM/LiF/Al, and exhibit a cation in PV cells. The cost of preparation and the possibility
PCE = 2.04%, the highest PCE for solution-processed, gra- of large-scale production (roll-to-roll process), limits the
phene, transparent, electrode-based, solar cells reported till application. Certain modifications can be adopted to tackle
date. these issues, like replacing the UV treatment with the sur-
Thus, as described in the above sections, though gra- face modification, or by developing a hydrophobic interface
phene exhibits great potential as a transparent electrode in that is compatible with the graphene layer. Tuning the work
photovoltaic cells, a few drawbacks such as the hydrophobic function sufficiently with respect to the active layer inhib-
nature of graphene makes it difficult to get a uniform HTL its the application of any interfacial layer. For improving the
(PEDOT:PSS), while its UV treatment leads to the disruption conductivity without losing on the transparency, recovering
(a) (b) 60
0 90
Absorption (%)
IPCE (%)
J (mA/Cm2)
40 60
–3
800°C 20 30
–6 900°C
1000°C
0 0
0.0 0.2 0.4 0.6 100 600 800
Voltage (V) Wave lengh (nm)
FIGURE 19.7 (a) J–V curves of photovoltaic cells with MLG films grown at different growth temperatures. (b) Incident photon-to-current
efficiency (IPCE) and absorption spectra of a photovoltaic device fabricated with a 1000°C-grown MLG film. (Reprinted from Organic
Electronics, 11(11), Choe, M. et al. Efficient bulk-heterojunction photovoltaic cells with transparent multi-layer graphene electrodes, 1864–9,
Copyright 2010, with permission from Elsevier.)
(b)
(a)
a. Spin coating PMMA

b. Etching Cu foil
1 2 3 4
N
(A) (B)
(+ doping)
5 6 7 8
Remove PMMA Remove PMMA

N in acetone in acetone (c) (d)
3 layer
1 layer
0 layer 1 layer
2 layer 2 layer
Remove PMMA in acetone
0 layer
(e) 100 600

Layer number
1
2
(f ) Wet transfer
95 3
Directly transfer by is LBL
90
4
5 Sheet resistance (Ω/sq.) 500
6 Doping by HCLintercategory+HNO3
7
85
Transparency %
8 400
80
75 100
300
Transperancy Ω 550nm
isLBL
Theoretical value
70 95
65 90 200
60 85
100
55 80
Layer number
50 0
300 400 500 600 700 800 0 1 2 3 4 5 6 7 8 9
Wavelength (nm) Layer number
0
(g) (h)
Graphene
(Anode)
3 layers
MoO1 4 layers
Current density (mA/cm2)
–2.2eV PEDOT
–2 5 layers
–3.3eV P3HT
–3.7eVPCBM
–4.3eV AM_IF
(Cathode) –4
–4.36eV
–6
e– –5.2eV –5.2eV
t
gh
–5.47eV gy
er
Li
–5.1eV
En –8
0.0 0.1 0.2 0.3 0.4 0.5 0.6

Voltage (V)
FIGURE 19.8 (a) Schematic drawing of multilayer graphene films made by normal wet transfer (A) and direct coupling LBL assembly
(B) (N = 0,1,2,3,…). (b) Optical image of multilayer graphene films (from 1 to 8 layers) on a quartz substrate. (c), (d) Typical optical micro-
scope images of 2- and 3-layer graphene films on SiO2/Si substrates. (e) UV–vis spectra of LBL graphene films with 1 to 8 layers on quartz.
The inset compares the transparency of the LBL film and its theoretical values at 550 nm. (f) Comparison of sheet resistance versus layer
number for graphene films prepared using wet transfer, LBL, and LBL acid-doped methods. (g) Schematic diagram of the OPV device. (h)
J–V characteristics of devices under light illumination for 3–5 layers of acid-doped graphene/MoO3 + PEDOT:PSS/P3HT:PCBM/LiF/Al.
(Geng, J. et al.: A simple approach for preparing transparent conductive graphene films using the controlled chemical reduction of exfoliated
graphene oxide in an aqueous suspension. Journal of Physical Chemistry C. 2010. 114. no. 34. 14433–40. Copyright Wiley-VCH Verlag
GmbH & Co. KGaA. Reproduced with permission.)
the graphene’s intrinsic structure should be considered other 19.2.2 Graphene as an Interfacial Layer
than that of doping, or the formation of a hybrid material
could prove to be an efficient tool. It can be expected that The interfaces at the acceptor’s nanostructures and donor–
OSCs with higher efficiency would be possible in the future host polymer matrix creates BHJs that are in direct electrical
by tailoring single-layer graphene and with more optimized contact with the cathode and the anode electrodes, leading
processing conditions. to the recombination of carriers and current leakage that, in
(a) (b)
rGO:Li 5
rGO:Na Li2CO3
rGO:K Na2CO3
rGO:Rb 10
K2CO3

rGO:Cs Rb2CO3
3 Cs2CO3
Intensity (a.u.)
Φ = 4.6 eV
Φ = 4.4 eV
2
Φ = 4.1 eV
Φ = 3.7 eV 1
Φ = 3.4 eV
0
18 17 16 15 –0.7 –0.6 –0.5 –0.4 –0.3 –0.2 –0.1 0.0 0.1

Binding energy (eV) Voltage (V)
FIGURE 19.9 (a) UPS spectra of the rGO doped with various alkali carbonates. (b) J–V Characteristics of the inverted P3HT/PCBM solar
cells incorporating rGO-SWCNT doped with various alkali carbonates as the cathode. (Reprinted with permission from Tung, V. C. et al.
Low-temperature solution processing of graphene–carbon nanotube hybrid materials for high-performance transparent conductors. Nano
Letters 9, no. 5: 1949–55. Copyright 2009 American Chemical Society.)
(a) (b)
10
Current density J (mA,cm2)
Graphene:
S S Dark ITO Graphene
T:P
E DO 5 Au-graphene:
Gra/P BM
Dark ITO Graphene
PC
H T/ 0
P3
Au Zn
O
O
IT –5
ass
Gl –10
0.0 0.2 0.4 0.6 0.8

(c) 80 (d)
Au-graphene 4.90
70
Graphene
Pristine
Illuminated from:
4.92
Work function (eV)
60
Au-doped
Graphene
ITO
50 Graphene
Pristine graphene
EQE (%)
4.94
Au-doped graphene
40
30 4.96
20
4.98
10
0 5.00
300 400 500 600 700 800 Transferred on Doped with
Wavelengh (nm) PEDOT:PSS PEDOT:PSS
FIGURE 19.10 (a) Schematic diagram of a semitransparent organic solar cell with the structure Glass/ITO/ZnO/P3HT:PCBM/
PEDOT:PSS/graphene. (b) J–V characteristics measured from both sides of semitransparent organic solar cells with pristine or Au-doped
graphene top electrode under solar simulator. Dark currents of the devices are presented as well. (c) EQE measured from both sides of the
organic solar cell with Au-doped graphene/PEDOT:PSS top electrode. The open and solid symbols correspond to the results characterized
from ITO and graphene sides, respectively. (d) Work functions of graphene/PEDOT: PSS electrodes prepared at different processing condi-
tions. (Reprinted with permission from Huang, J.-H. et al. Effective work function modulation of graphene/carbon nanotube composite films
as transparent cathodes for organic optoelectronics. ACS Nano 5, no. 8: 6262–71. Copyright 2011 American Chemical Society.)
turn, leads to a decrease in the PCE of solar cells. In order resistivity. Therefore, it can act as a perfect HTL with only
to minimize such effects, we require HTLs on the top of vertical conduction of charge carrier. In GO, the sp2 clusters
ITO anode, and ETL in between active layer and Al cath- are isolated laterally but are in contact with the electrodes in
ode. HTLs must be wide, band-gapped, p-type material, and a sandwich metal/GO/metal structure. Thus, carriers (both
as discussed in the introduction, several inorganic materials holes and electrons) can be injected and transported via the
such as NiO2, V2O5, MoO3 have led to efficiencies greater than isolated sp2 clusters with a relative ease in a sandwich device
5%. All these inorganic HTLs are deposited using vacuum structure. However, because the density of the isolated sp2
deposition techniques that are costly as well as incompatible clusters is sufficiently low in non-rGO, the vertical resistiv-
with solution-processable printable electronics. For this rea- ity is substantial, and the conduction and valence bands of
son, the most commonly used HTL in polymer solar cells is the sp3 sites dictate the transport. GO sheets, which are pre-
semiconducting PEDOT:PSS. Although being cost effective pared by the Hummer’s method, have the lateral size of the
and solution-processable, it is usually deposited from a highly order of 1–10 μm with an average thickness ranging from
acidic pH ~ 13 aqueous suspension that can corrode ITO over 2 to 10 nm. Li et al.24 investigated the hole-transport proper-
a period of time and induce water into the active layer leading ties of GO in OPVs. The GO was obtained from the puri-
to the degradation in the device performance. Therefore, there fied natural graphite powder using the modified Hummer’s
was a need to find a suitable HTL that can overcome the limi- method. The optical transmission spectra of the GO layers
tation of PEDOT:PSS, and is compatible with OPV materials with various thicknesses deposited on the ITO is shown in
and fabrication techniques. Figure 19.11a which shows excellent transparency of GO
Recently, GO has emerged as one of the most promis- films without any significant changes in thickness from 2 to
ing candidates for HTL in OPV. GOs are not only water- 10 nm. The optical band gap (inset in Figure 19.11a) gives
soluble but also their vertical resistivity of non-rGO is of the band gap (Eg) of 3.6 eV, which is comparable with that of
order of magnitude lower (103–104 Ω/cm 2) than the lateral the vacuum-evaporated HTLs. The device performance
(a) (b)
100
2
80 0
Transmission (%)
6
–2
60 –4
4
–6
(αhv)2
40
ITO 2 –8
GO (2 nm) GO (2 nm)
20 –10
GO (4 nm) 0 GO (4 nm)
2 3 4
GO (10 nm) hv (eV) –12 GO (10 nm)
0
300 400 500 600 700 800 0.0 0.2 0.4 0.6
(c) Wavelength (nm) Voltage (V)
(d)
2
ITO only
0
Recombination rate (s–1)
102 PEDOT:PSS
–2 GO
–4
Normalzed photevoltage (a.u.)
–6 ITO only
PEDOT:PSS
GO
–8 104
–10 GO (2 nm)
ITO only
–12 PEDOT:PSS 0 10 20 30 40 50
Time (μs)
0.0 0.2 0.4 0.6 10 100

Voltage (V) Illumination power (mWcm–2)
FIGURE 19.11 (a) Optical transmission spectra of the three GO films (2, 4, 10 nm) deposited on ITO/glass substrates. The inset shows
the Tauc plot of the 2-nm-thick GO thin film. (b) J–V characteristics of photovoltaic devices with no hole transport layer (curve labeled as
ITO), with 30-nm PEDOT:PSS layer, and 2-nm-thick GO film. (c) Current voltage characteristics of ITO/GO/P3HT: PCBM/Al devices with
different GO thickness. (d) Charge recombination rate constant k rec versus light intensities obtained using TOCVD measurements for ITO-
only, PEDOT:PSS, and 2-nm GO thin film devices. The inset shows the transient photovoltage decay curves of the corresponding devices
measured at 100 mW/cm2. (Reprinted with permission from Lee, S. et al. Flexible organic solar cells composed of P3HT: PCBM using
chemically doped graphene electrodes. Nanotechnology 23: 344013. Copyright 2012 American Chemical Society.)
parameters for P3HT:PCBM (1:0.8) with various thickness were: Jsc = 8.49 mA/cm2, Voc = 0.625 V, FF = 57.65%, and
of the interfacial layer with GO are plotted in Figure 19.11b. PCE = 3.06%.25
The J–V curves obtained from the GO are comparable with Yun et al.26 proposed a rGO (pr-GO) by reducing with
that obtained from PEDOT:PSS for the identical device for- p-toluenesulfonyl hydrazide (p-TosNHNH2) as an interfacial
mation (Figure 19.11c). The recombination rate in the case of layer in solar cells. The advantage of this process is that it
GO is much lower, as compared with PEDOT:PSS, as shown does not involve hydrazine as a reducing agent. Although the
in Figure 19.11d. The decay lifetime of ITO, PEDOT:PSS use of hydrazine reagent is highly effective for the reduction
and GO were 8.1, 9.6, and 11.6 μs (inset in Figure 19.11d), of GO, it is highly toxic and explosive. Moreover, insulating
indicating an effective charge separation and a faster hole surfactants, such as polystyrene sulfonic acid which can cause
transport in GO. a reduction in conductivity, are inevitably needed to prevent
To further enhance the hole mobility, a nickel oxide (NiOx) irreversible aggregation during the reduction process. Also,
layer can be used in combination with a GO sheet, as shown the dispersion of rGO produced using hydrazine is low, which
in Figure 19.12a.25 Transmittance of GO layer coated onto is a disadvantage, while with p-TosNHNH2 it is advantageous
ITO decreased slightly up to three layers, but at five layers, due to it being less toxic and having a high reducing capabil-
it decreased significantly down to 85% at 550 nm (Figure ity. The thickness of the pr-GO film prepared by this method
19.12b). For the purpose of optimizing the most suitable can be easily controlled. Figure 19.13a shows the J–V curve
thickness of the GO layer, various devices with the configu- performance of OSCs with pr-GO as HTL. It is clear that
ration ITO/GO/active layer/LiF/Al were fabricated. The best pr-GO dramatically enhances the performance of OSCs with
device performance has been achieved with three layers of PCE = 6.63, FF = 66.7%, Jsc = 9.33 mA/cm2, Voc = 0.59 V.
GO, namely, Jsc = 8.84 mA/cm2, Voc = 0.565 V, FF = 54.94%, These values are highly comparable to PEDOT:PSS-based
and PCE = 2.75% (Figure 19.12c). The J–V characteristics of cells. The added advantage of pr-GO is that it shows a much
the devices with and without a metal oxide layer are shown in superior lifetime as compared with that of the cell with
Figure 19.12d, while with GO/NiOx bilayer, the parameters PEDOT:PSS, as shown in Figure 19.13b.
(a) (b)
Lif:Al (cathode) 120
100
P3HT:PCBM
Transmittance (%)
80
NiOx
Graphene oxide 60 ITO
1 layer
40 2 layers
ITO (anode) 3 layers
20 5 layers
0
Substrate 300 400 500 600 700 800 900 1000
Wavelength (nm)
(c) (d) 12
12
1 layer 10
10 2 layers
3 layers
5 layers 8
8
6
6
w/o HTL
4 GO
4 NiOx
GO/NiOx
2 2 NiOx/GO
0 0
0.0 0.1 0.2 0.4 0.3 0.5 0.6 0.7 0.0 0.1 0.2 0.4 0.3 0.5 0.6 0.7
Voltage (V) Voltage (V)
FIGURE 19.12 (a) Schematic of the device using GO and NiOx as a hole transport layer by solution process. (b) Optical transmission spec-
tra of the GO films with different thickness on ITO/glass substrates. (c) J–V characteristics of devices with different GO thickness. (d) J–V
characteristics of the devices with different hole transport layers under illumination. (Reprinted from Nano Research 6, no. 7, Zhang, Q. et al.
Solution-processable graphene mesh transparent electrodes for organic solar cells, 478–84. Copyright 2013, with permission from Elsevier.)
(a) 10 (a) 2
ITO
8 0
PEDOT:PSS Bare ITO
6 GO1

GO (0.15 wt%)
–2
4 GO3 PEDOT: PSS

r-GO1 GO: CNT (1:0.2)
2 –4
r-GO3
0
pr-GO1 –6
–2 pr-GO3
–4 –8
–6
–10
–8
–10 –12
–0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6
–0.2 –0.1 0.0 0.1 0.2 0.3 0.4 0.50 0.6 0.7
Voltage (V)
Voltage (V) (b)
(b) 4.0 100

1.0

Normalized PCE
3.5 0.8
0.6
0.4 OSCs with pr-GO
3.0 OSCs with
0.2 GO (0.15 wt%)
PEDOT:PSS 95
2.5 0.0
GO: CNT (1:0.2)
0 5000 10,000 15,000 20,000
PCE (%)
Time (min.) PEDOT: PSS

2.0
1.5 OSCs with Pr-GO

90
1.0
0.5
400 500 600 700 800 900
OSCs with PEDOT:PSS
0.0 Wavelength (nm)
0 5000 10,000 15,000 20,000
Time (min.) FIGURE 19.14 (a) J–V characteristics of devices. (b) Optical
transmittance of the GO, GO: SWCNTs and PEDOT:PSS modifying
FIGURE 19.13 (a) J–V curves of OSCs based on different anode layers showing greater transparency of GO based films in the longer
interfacial layers. (b) Changes in PCE of a conventional PEDOT:PSS- wavelengths. (Zhang, D. et al.: Al-TiO2 composite modified single-
based OSC and an OSC with the pr-GO anode interfacial layer layer graphene as an efficient transparent cathode for organic solar
during exposure to air. (Liu, Z., J. Li, and F. Yan: Package-free cells. ACS Nano. 2013, 7( 2). 1740–7. Copyright Wiley-VCH Verlag
flexible organic solar cells with graphene top electrodes. Advanced GmbH & Co. KGaA. Reproduced with permission.)
Materials. 2013, 25. 4296–301. Copyright Wiley-VCH Verlag GmbH
& Co. KGaA. Reproduced with permission.)
Soluble, chemically derived, few layers of graphene sheets
For HTL, the best performance is obtained in the case of can conveniently be used as HTL.28 Ultrasonic treatment of
1–2 monolayers, which can be extremely challenging to be acid-modified, expanded, graphite sheets in the ammonia–
created via spin coating. Also, the resulting film should fully ethanol solution leads to the formation of few-layer graphene
cover the underlying surface. To overcome this problem, nanosheets (FLGs). These FLGs have an average thickness in
Kim et al.27 added a small amount of SWCNTs in GO and the range of 2.6–3 nm, that is, 8–9 layers approximately. FLG-
found out that the performance of the modified GO-based PEDOT:PSS (2:1) when used as HTL with P3HT:PCBM, gives
device can be improved. The GO:SWCNTs dispersion was Jsc = 10.68 mA/cm2 and PCE = 3.70%. The control device
obtained by direct mixing of the two components in water. having only PEDOT:PSS as HTL gives Jsc = 8.06 mA/cm2,
Incorporation of a small amount of SWCNT (1:0.05, w/w) Voc = 0.58 V, and PCE = 3.1%. It was found that the introduc-
into a GO-modifying layer (Figure 19.14a) leads to a signifi- tion of FLGs in the HTL does not alter the band alignment of
cant improvement in the device performance. For bare GO the polymer.
(0.185%), the obtained device parameters were Jsc = 9.79 mA/ It is quite evident that the GO films deposited from neutral
cm 2, FF = 62.7%, PCE = 3.66%, whereas, GO:SWCNT solutions are effective HTLs. In solar cell devices, the PCE
(1:0.2) gives enhanced parameters with Jsc = 10.82 mA/ values of the devices obtained from GO HTLS are compa-
cm 2, FF = 62.8%, PCE = 4.10%. Moreover, the transmit- rable with devices fabricated with PEDOT:PSS. Therefore,
tance values with and without SWCNT were also similar it is clear that the neutral GO can replace the acidic/basic
(Figure 19.14b). PEDOT:PSS in OPVs and LEDs.
19.2.3 Active Layer [SPF Graphene]) becomes hydrophobic and can be dissolved

in organic solvents such as 1,2 dicholorobenzene. The reac-
Unlike in a traditional inorganic solar cell where irradiation tion scheme is shown in Figure 19.15a. They fabricated hybrid
leads to the generation of electron–hole pairs, in OSCs, light solar cells using active layer comprising P3OT [poly (3-octyl-
absorption leads to the generation of excitons, or the bound thiophene)] (Figure 19.15b) as a donor and SPF Graphene as
electron and hole pairs. Most successful OPVs are based on acceptor in different content onto the ITO substrate in the
donor:acceptor configuration where the electron acceptor unit device configuration shown in Figure 19.15c. The energy band
is selected such that the electron affinity Ea is larger than that of diagram of the device configuration is shown in Figure 19.15d.
the donor polymer with ionization potential, Ip, being smaller. The best efficiency parameters are shown in Figure 19.15e. The
Conventional solar cells have employed various donor poly- device annealed at 160°C for 20 min results in a Voc = 0.92 V,
mers in conjugation with a well-known acceptor PCBM. This Jsc = 4.2 mA/cm2, FF = 0.67%, and PCE = 1.4%.
material having a low lowest unoccupied molecular orbital The incorporation of both SPF Graphene and f-MWCNTs
(LUMO) absorbs the light weakly in the visible region, which in BHJs can be more effective as the SPF Graphene helps in
is difficult to tune with high Voc in BHJ device.2 These fac- exciton dissociation, and provides a percolation path for the
tors prompted the discovery of alternative acceptor materials. electron transport, whereas, MWCNTs can provide efficient
CNT, when used as replacement for the C60-based material, as hole transportation. This can yield better carrier mobility,
it exhibits high charge mobility, long π–π conjugation along easy exciton splitting, and suppression of charge recombina-
with large aspect ratio, gives a very low PCE. In this respect tion, thereby improving the photovoltaic properties. In the
again, graphene has emerged as a potential acceptor in OSCs. photovoltaic device based on ITO/PEDOT:PSS/P3HT-F-
As discussed above, the hydrophilic nature of GO sheets MWCNTS-SPF Graphene/LiF/Al, P3HT, and f-MWCNTs
hinders their solubility in hydrophobic organic solvents. act as hole transporter and SPF Graphene as electron
Therefore, the second step of functionalization of GO/gra- transporter. When the SPF Graphene content is 0.8 wt%,
phene is to change its surface properties such that it becomes the best Jsc = 4.7 mA/cm 2, Voc = 0.76 V, FF = 0.32, and
hydrophobic. Liu et al.33 treated graphene with isocyanite. The PCE = 1.05% was achieved.50 Thereafter, several methods
resulting GO (solution-processable functionalized graphene
(a) O
NCO Graphene OH Graphene O C NH
(b) (c) LUMO

e–
e–
Work function
hv –4.3 eV
Graphene
Al
–4.7 eV
S n ITO
HOMO
h+
η=1.4% P3OT
(d)
Aluminum 2
Current density (mA cm–2)
LiF 0
P3OT:graphene
–2
PEDOT:PSS
ITO –4
5% without annealing
5% annealing at 160°C 10min
–6 5% annealing at 160°C 20min
5% annealing at 210°C 20min
hv
–0.4 –0.2 0.0 0.2 0.4 0.6 0.8 1.0
Bias (V)
FIGURE 19.15 (a) Schematic representation of the reaction of phenyl isocyanate with graphene oxide to form SPF Graphene. (b) The
idealized chemical structures of P3OT. (c) Schematic of the device with P3OT/graphene thin film as the active layer and the structure ITO
(ca.17V sq1)/PEDOT:PSS (40 nm)/P3OT:graphene (100 nm)/LiF (1 nm)/Al (70 nm). (d) Energy level diagram of P3OT and SPF Graphene.
(e) J–V curves of PV devices based on P3OT/SPF Graphene composite with SPF Graphene content of 5 wt% without annealing, after anneal-
ing at 160°C for 10 and 20 min, and at 210°C for 10 min. (Kim, K. et al.: Ultrathin organic solar cells with graphene doped by ferroelectric
polarization. ACS Applied Materials and Interfaces. 2014, 6. 3299−30. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with
permission.)
(a) (b)
PDI-G hybrid dark P3HT:PDI
PDI-G hybrid light 50
P3HT:PDI-G
0
Current density (mA/cm2) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 40
Voltage (V)
–1
30
IPCE (%)
–2
LiF/AI 20
P3HT:G composite
–3
PEDOT:PSS 10
–4 ITO
0
Glass 300 400 500 600 700
Wavelength (nm)
FIGURE 19.16 (a) Solar cell data: J–V plot of ITO/PEDOT:PSS/P3HT:PDIG hybrid (1:1)/LiF/Al device under dark (black) and light (gray)
illumination after heating at 160°C. Inset: Solar cell device structure. (b) IPCE spectra for PDI and PDIG hybrid. (Reprinted with permission
from Ryu, M. S. and J. Jang. Effect of solution processed graphene oxide/nickel oxide Bi-layer on cell performance of bulk-heterojunction
organic photovoltaic. Solar Energy Materials and Solar Cells 95, no. 10: 2893–6. Copyright 2011 American Chemical Society.)
were used to modify the graphene, one of them was to use by hydrothermal treatment of GO in DMF. The photovol-
graphene as an atomic template and structural scaffold in taic properties and IPCE of PDI and G hybrid are shown in
the nucleation and assembly of organic nanostructures. In Figure 19.16a and b, respectively. The device configuration
this process, a perylene-based PDI (perylene diimide) was is shown in the inset of Figure 19.16a. The PCE obtained
used as organic molecule, and PDI-G hybrids were prepared from PDI, PDI-G mixture, and PDI-G hybrid wires were
(a) (b)
1.0
C60:P3HT,
C60:G:P3HT,
0.8 C60-G:P3HT,
Annealing
Intensity (a.u.)
0.6 –
Al
0.4 C60-G:P3HT
0.2 PEDOT:PSS
+
ITO glass
0.0
300 400 500 600 700 800
Wavelength (nm)
(c) (d)
1
E e
0
–3.0 eV
hv
–1
–4.40 eV –4.30 eV
–4.70 eV –4.50 eV
–2 –5.00 eV
–5.20 eV
–3 –6.10 eV
C60-G:P3HT, dark
C60:P3HT, light h
–4 C60-G:3HT, light
ITO PEDOT:PSS P3HT C60 Graphene Al
C60/G mixture:P3HT, light
–5
0.0 0.2 0.6 0.4 0.8
Voltage (V)
FIGURE 19.17 (a) Absorption spectra of the C60-G:P3HT (1:1 wt/wt) film before and after annealing (130°C, 10 min) and the C60:P3HT
(1:1 wt/wt) film spin coated onto quartz plates. (b) Schematic of a hybrid photovoltaic device with the C60-G:P3HT composite as the active
layer. (c) J–V curves of the photovoltaic devices with the C60-G:P3HT (1:1 wt/wt), the C60:P3HT (1:1 wt/wt), or the C60/G mixture (12 wt%
G):P3HT (1:1 wt/wt) as the active layers after annealing treatment (130°C, 10 min). (d) Energy level diagram for the proposed photovoltaic
device using the C60-G:P3HT composite as the active layer. (Reprinted with permission from Kim, J., V. C. Tung, and J. Huang. Water pro-
cessable graphene oxide: Single walled carbon nanotube composite as anode modifier for polymer solar cells. Advanced Energy Materials
1, no. 6: 1052–7. Copyright 2011 American Chemical Society.)
0.064%, 0.002%, and 1.04%, respectively. The best device quantum confinement and edge effect, resulting in different
parameters were Voc = 0.78 V, Jsc = 3.85 mA/cm 2, FF = 35%, kinds of photoluminescence characteristics that are of signifi-
PCE = 1.4% (Figure 19.16a).34 cant importance for light harvesting. GQDs were first pro-
Since C60 molecules are well established, efficient, electron posed and synthesized by Yan et al.36 by bottom-up approach.
transport materials,51 it would be very interesting to see the The application of GQDs in OSCs were first demonstrated by
effects of hybrid materials of C60 with graphene sheets for the Gupta et al.37 and Li et al.38
photovoltaic characteristics. Yu et al.35 achieved this by using a Gupta et al.37 showed that solar cells made up of GQDs
simple lithiation reaction to covalently attach known monosub- blended with a regioregular P3HT polymer gave a significant
stituted C60 onto graphene sheets (Figure 19.17). The obtained improvement in the efficiency over those containing GSs-
C60 grafted graphene nanosheet (GSs) were used as electron blended conjugated polymers. The GQDs (<10 nm) were syn-
acceptor in P3HT-based BHJ solar cells (Figure 19.17b and d) thesized from GSs by the hydrothermal approach53 as shown
to significantly improve the electron transport, and hence, the in Figure 19.19a, which were functionalized with aniline
overall device performance (Figure 19.17c), yielding a PCE ≈ (ANI) or methylene blue (MB) (ANI-GQDs and MB-GQDs).
1.22%. Yu et al.52 further chemically grafted P3HT onto GO The solar cell characteristic of P3HT/ANI (GQD) devices is
via esterification reaction. The resultant P3HT-grafted GO shown in Figure 19.19b. Maximum values of Voc = 0.61 V,
sheets (G-P3HT) were soluble in common organic solvent. A Jsc = 3.51 mA cm−2, FF = 0.53, and PCE = 1.14% were
bilayer photovoltaic device based on solution-cast G-P3HT/C60 obtained for 10 wt% ANI-GSs in P3HT. The OPV character-
hetrerostructures were fabricated (Figure 19.18a,b). The J–V istics of reduced-controlled GQDS are shown by Kim et al.54
curves shown in Figure 19.18c indicated that there is a 200% They obtained outstandingly improved PCE of 7.6% in BHJ
increase in PCE (0.61%) of G-P3HT/C60 heterostructures with devices with reduced GQDs as compared to the reference
respect to P3HT/C60 counterpart. devices without reduced GQDs (6.7%).
The zero band gap in graphene imposes limitation for its
optoelectronic applications, since a finite band gap is favor- 19.3 GRAPHENE IN DYE-
able in achieving sharp fluorescent peaks that are key in such
applications. A band gap in graphene can be modified/altered SENSITIZED SOLAR CELLS
by chemical modifications to generate graphene quantum dots A great deal of attention has been given for the development
(GQDs).53 Zero-dimensional (0-D) GQDs possesses stronger of DSSC owing to its low cost, convenient manufacturing
(a) (b) 0
e
–3 hv –3.56 eV
Light –4
–4.40 eV –4.30 eV
Al
–4.70 eV
PEDOT:PSS
–5.00 eV
ITO glass
G-P3HT
–5 –5.39 eV
C60
–6.10 eV
–6
h
–7 ITO PEDOT:PSS G-P3HT C60 Al
(c) 3
G-P3HT/C60 light
2 P3HT/C60 light
P3HT/C60 dark
1 G-P3HT/C60 dark
–1
–2
–3
–4
0.0 0.1 0.2 0.3 0.4
Voltage (V)
FIGURE 19.18 (a) Schematic and (b) energy level diagram of a ITO/PEDOT:PSS/G P3HT/C60/Al photovoltaic device. (c) J–V characteristics
of the photovoltaic devices using P3HT/C60 or G-P3HT/C60 as the active layer. (Reprinted with permission from Nguyen, D. D. et al. Synthesis
of ethanol-soluble few-layer graphene nanosheets for flexible and transparent conducting composite films. Nanotechnology 22, no. 29: 295606.
(a) (b)
–4
5° e–
1% ANI-GQDs 3.2 eV
Current density (mAcm–2)

3.55 eV
–3 4.2 eV
4.7 eV
AI
ITO 5.1 eV
10% ANI-GSs
P3HT 5.38 eV
–2
3% ANI-GQDs
–1
5% ANI-GQDs
0
20 nm –0.1 0.1 0.3 0.5 0.7 0.9
Voltage (V)
FIGURE 19.19 (a) TEM image of the GQDs. (b) J–V characteristics of the photovoltaic devices based on ANI-GQDs with different GQDs
content and ANI-GS (under optimized condition) annealed at 160°C for 10 min at 100 mW illumination. (Reprinted with permission from
Jayawardena, K. D. G. I. et al. Solution processed reduced graphene oxide/metal oxide hybrid electron transport layers for highly efficient
polymer solar cells. Journal of Materials Chemistry A 1: 9922–7. Copyright 2013 American Chemical Society.)
processes, and comparable efficiencies (over 10%).6,9 A typi- perpendicular to the conducting substrates4,5 for introducing
cal DSSC consists of a porous, wide band gap semiconductor, charge carriers to direct the photogenerated electrons, are
such as TiO2, sensitized with a monolayer of dye and filled being adopted.39
with a hole-transporting material. The dye absorbs the inci- Graphene films can be employed in place of metal oxide
dent light and injects electrons into the semiconductor, and window electrodes in DSSC, as demonstrated by Wang et al. 40
the hole on the dye is transported out of the device via the They fabricated highly conducting (550 S/cm) and transpar-
hole transporter. The major obstruction for achieving high ent (>70%) films over a 1000–3000 nm range from exfoli-
efficiency is the transport of the photogenerated electrons ated GO. The device structure is shown in Figure 19.20a, and
across the TiO2 network that competes with the recombina- the corresponding band diagram is shown in Figure 19.20b.
tion. To improve upon the charge transport and for decreas- The I–V characteristics of the device showed Voc of 0.7 V,
ing the rate of recombination, various methods, such as using Jsc = 1.01 mA/cm2, FF = 0.36, PCE = 0.26% which is compa-
composite metal oxide as the semiconductor with different rable with the efficiency of an FTO-based cell (0.84%) (Figure
band gap,3 preparing porous structure whose direction is 19.20c).
(a) (c) 3.0
1.5
I
0.0
–1.5
–3.0
(b) –4.5
Vacuum level
–1
LUMO
–2 –0.2 0.0 0.2 0.4 0.6 0.8 1.0
Excited state S*
–3 Conduction band Voltage (V)
E (eV)
–4 HOMO
–5 Ground state S0
Graphene
–6 anode Valence band Spiro- Au
–7 Dye OMeTAD cathod
TiO2
FIGURE 19.20 (a) Illustration of dye-sensitized solar cell using graphene film as electrode, the four layers from bottom to top are Au, dye-
sensitized heterojunction, compact TiO2, and graphene film. (b) The energy level diagram of graphene/TiO2/dye/spiro-OMeTAD/Au device.
(c) J–V curve of graphene-based cell (black) and the FTO-based cell (gray). (Reprinted with permission from Chen, Y. et al. Graphene oxide-
based carbon interconnecting layer for polymer tandem solar cells. Nano Letters 14: 1467–71. Copyright 2014 American Chemical Society.)
(a) e– (b)
S* Electrode 1
Electrode 2
Electrode 3
e– Electrode 4
CB 15
Electrode CNT
e– Ba
e– TiO2 ck
I (mA/cm2)
Graphene re
FTO acti 10
Re c
on
omb
e–
inat
I3–/I– 5
n io
e–
0
S+/S 0.0 0.2 0.4 0.6 0.8
Dye V(V)
FIGURE 19.21 (a) Operational principle of the DSSC device. (b) J–V characteristics of different electrodes. The sensitizer was N3 (ruthe-
nium dye), area = 0.20 cm2, 100 mW cm2. (Reprinted with permission from Liu, Z. et al. Organic photovoltaic devices based on a novel
acceptor material: Graphene. Advanced Materials 20, no. 20: 3924–30. Copyright 2008 American Chemical Society.)
Two-dimensional graphene benefits the charge separa- 6.97%, which was an increase by 39% in comparison with
tion due to its high conductivity and good contact with TiO2 the TiO2 photoanode (Figure 19.21b). Functionalized gra-
nanoparticles. Nanocrystalline TiO2 anchors the graphene phene sheets (FGS) can also be used as a catalytic counter-
completely and forms graphene bridges, which decreases the electrode41 in DSSC because functionalized graphene has
TiO2–TiO2 contacts. Thus, a bridged graphene forms an excel- oxygen-containing sites that can perform comparably with
lent electron transfer medium and reduces the charge recom- platinum (Pt) (Figure 19.22a). The catalytic activity can be
bination rate effectively. Yang et al.55 synthesized GO from tuned by increasing the amount of oxygen-containing func-
the modified Hummer’s method and used it as a photoanode tional group (Figure 19.22b). Another application of graphene
in the fabrication of DSSC in different content. Figure 19.21a is that one can use graphene nanocomposite as a counter-
shows the working principle of DSSC. Under illumination, electrode in DSSC.42 GQDs have a much smaller size than
dye is photoexcited, and it transfers the electrons to the con- graphene while they retain the graphitic nature. Chen et al.56
duction band of TiO2. Since the TiO2 is anchored with 2D gra- doped polypyrrole (PPy) with GQD on F-doped tin oxide
phene, these excited electrons are captured by the graphene TiO2 glass as an efficient counter-electrode for a high-perfor-
and transferred to the TiO2 and to the conductive substrate mance DSSC (Figure 19.23a). The use of GQDs increases the
through graphene bridges. Due to faster electron transport porosity of the film as compared to densely packed PPy and,
and a lower recombination rate, the PCE was increased to therefore, display higher catalytic density and lower charge
(a) (b) Effect of applied bias on

Dye-sensitized solar cell performance functionalized graphene counter electrode
14
4
Platinum
12 0V
Photocurrent density (mA cm–2)
counter
Functionalized graphene
counter electrode electrode 0 .1V
10 3
0 .3V
–Z″ (Ω cm2)
0 .4V
8
2 0 .5V
6 0 .8V
4
1
2
0 0
0 0.2 0.4 0.6 3 5 7 9
Voltage (V) Z′ (Ω cm2)
FIGURE 19.22 (a) J–V curve characteristics of DSSCs using thermally decomposed chloroplatinic acid and FGS13 counter-electrodes
measured in Iodolyte AN-50 electrolyte. Active area is 0.39 cm2. (b) An expansion of the high-frequency region. (Reprinted with permission
from Liu, Z. et al. Improving photovoltaic properties by incorporating both SPFgraphene and functionalized multiwalled carbon nanotubes.
Solar Energy Materials and Solar Cells 94, no. 12: 2148–53. Copyright 2010 American Chemical Society.)
(a)
H
N
Load
I–3
Pyrrole
I– I–3 I–
GQDs
(b) 15 (c) 5.5
12 5.0
Current density (mA cm–2)
9 4.5
PCE (%)
0 3%
10% 15% PCE
6 4.0
20% 25%
30% Pt
3 4.5
X% indicate GQDs ratio (V/V)
0 3.0
0.00 0.15 0.30 0.45 0.60 0.75 0 3% 10% 15% 20% 25% 30%
Voltage (V) GQDs ratio (V/V)
FIGURE 19.23 (a) Scheme of the fabrication processes of GQD-doped PPy film on FTO glass and its application as a counter-electrode
in DSSCs. (b) J−V curves of DSSCs based on Pt, plain PPy, and various concentrations GQDs ratio (3%, 10%, 15%, 20%, 25%, and 30%)
as CEs (100 mW cm−2). (c) PCEs with different ratios of GQD-doped PPy as CEs for DSSCs. (Reprinted with permission from Wudl, F.
Fullerene materials. Journal of Materials Chemistry 12, no. 7: 1959–63. Copyright 2002 American Chemical Society.)
transfer resistance. As determined by the J–V curves, the cm2 and a FF of 45 ≈ 56%, with an energy conversion effi-
highest PCE of 5.7% could be achieved with just 10% GQDs ciency (η) of 1.0 ≈ 1.7% (Figure 19.24c). In such cells, GS
in PPy which is comparable to the Pt counter-electrode-based functions as an antireflection coating, thereby reducing the
DSSC (Figure 19.23b,c). It implies that the concept of using reflection by ≈70% in the visible region and ≈80% in the
transparent conductive films as window electrode in DSSC near-IR region.
works well. The doping of graphene with bis-(trifluoromethanesulfonyl)
amide can enhance the efficiency of a graphene/n-Si
Schottky junction cell to 8.6% which is 4–5 times higher
19.4 GRAPHENE IN SOLID-STATE CELLS
than the undoped graphene sheets (Figure 19.25a,b)58 This
Graphene has low density of electronic states that allows its is due to the doping-induced shift in the chemical potential
electrical or chemical charge transfer modulation that can that increases the graphene carrier density and increases the
improve performance of Schottky junction solar cells. Li cell’s built-in potential. This results in a decrease in the cell’s
et al.57 deposited graphene sheets on SI wafers to make solar series resistance, and an increase in the Voc (Figure 19.25c,
cells with efficiencies upto 1.5%. Figure 19.24a,b shows the d). This results in the increase in the FF of the solar cell and,
schematic illustration and the energy band diagram of the therefore, PCE.
forward bias GS/n-Si junction upon illumination. A space- Dutta et al.59 used a novel combination in solar cells,
charge region accompanied by a large built-in field is cre- which is shown in Figure 19.26a. While there was no photo-
ated near the GS/n-Si interface, thus GS film not only serves voltaic effect in the controlled sample (ZnO nanowires only),
as a transparent electrode for light illumination, but also an a clear photovoltaic effect is observed from the I–V curve of
active layer for electron-hole separation and hole transport. the devices made with the ZnNW-GQDs composites (Figure
The two-dimensional, highly conductive graphene reduces or 19.26a). This clearly confirms the electron injection from
eliminates the lateral potential drop along the GS enabling a excited state of GQDs to nanowires. The IQE of the compos-
uniform carrier separation and collection. J–V data for GS/n- ite reached 87%, whereas the IQE of the controlled sample
Si-based cell gives a Voc of 0.42–0.48 V, Jsc of ≈ 4–6.5 mA/ was much lower (Figure 19.26b).
(a) (b) Vacuum level (c)

6

χ
3 0.1 cm2, dark
ϕG ϕ(VG – Vbias) ϕn-Si 0.5 cm2, dark
eVbias 0
ϕb EC
EF
Eg –3 0.5 cm2, light
hv 0.1 cm2, light
Au GS h+ SiO2 EV –6
n-Si e–
Ti/Pd/Ag 0.0 0.2 0.4 0.6
GS n-Si
Voltage (V)
FIGURE 19.24 (a) Schematic illustration of the device configuration. Bottom-left inset: cross-sectional view, Bottom-right inset: pho-
tograph of a GS/n-Si Schottky cell with a 0.1 cm2 junction area. (b) Energy diagram of the forward-biased GS/n-Si Schottky junction
upon illumination. (c) Light J–V curves of the cells. (Yu, D. et al.: Fullerene grafted graphene for efficient bulk heterojunction polymer
photovoltaic devices. The Journal of Physical Chemistry Letters. 2011. 2(10). 1113–18. Copyright Wiley-VCH Verlag GmbH & Co. KGaA.
(a) (b) 20
5 Pristine
SBH = 0.79 eV 10 TFSA doped
J (mA/cm2)
0 Pristine 0
Doped VOC = 0.43V
In J
–5 –10
SBH = 0.89 eV –20 VOC = 0.54V

–10
–1.0 –0.5 0.0 0.5 1.0 –0.2 0.0 0.2 0.4 0.6
Bias (V) Bias (V)
(c) (d)
Evac
Wgr doped
Wgr
ϕSBH Vbi doped
Vbi
doped
ϕSBH
EF Ec
FIGURE 19.25 (a) J−V characteristics in the semilogarithmic scale, (b) J−V characteristics of graphene/n-Si diodes. (c), (d) the band
diagram at the graphene/n-Si interface before and after the doping. (Reprinted with permission from Yu, D. et al. Soluble P3HT-grafted gra-
phene for efficient bilayer heterojunction photovoltaic devices. ACS Nano 4, no. 10: 5633–40. Copyright 2010 American Chemical Society.)
(a) 100 (b) 0.4

Au 2.0
1
ZnO-GQDs TPD
Current (mA/cm–2)
0.1
0.3
Photcurrent (mA/cm–2)
1.5
Photcurrent (mA/cm–2)
75
0.01 ZnO-GQDs
1E-3
GQDs AZO 1.0 1E-4 0.2
Light
NWs
IQE(%)
1E-5 Dark
50
0.5 –0.5 0.0 0.5 ZnO 0.1
Seed Voltage(V)
layer 0.0 0.0

25
ZnO –0.5 –0.1

0
–1.0 –0.5 0.0 0.5 1.0
300 400 500 600 700 800 Voltage (V)
Wavelength (nm)
FIGURE 19.26 (a) IQE of the samples. The inset shows the schematic model of the device. (b) Solar cell I − V characteristics under
white light illumination. The inset shows the logarithmic plot of dark and photo current of the GQDs-ZnO cell. (Reprinted with permission
from Yu, D. et al. Soluble P3HT-grafted graphene for efficient bilayer heterojunction photovoltaic devices. ACS Nano 4, no. 10: 5633–40.
19.5 CONCLUSION 10. Tung, V. C., L.-M. Chen, M. J. Allen et al. 2009. Low-
temperature solution processing of grapheneâ^’carbon
Is graphene an interesting material for use in solar cells? The nanotube hybrid materials for high-performance transparent
answer to this question is promising. In flexible electron- conductors. Nano Letters 9, no. 5: 1949–55.
ics, we need a very cheap and flexible alternative as an ITO 11. Gomez, D. A., Lewis, Y. Zhang et al. 2010. Continuous,
replacement. Graphene is mechanically strong, chemically highly flexible, and transparent graphene films by chemical
vapor deposition for organic photovoltaics. ACS Nano 4, no. 5:
stable, inert, highly conducting and, above all, very cheap.
2865–73.
The beauty of graphene is that one can modulate its band gap 12. Yin, Z., S. Sun, T. Salim et al. 2010. Organic photovoltaic
in various ways. Therefore, it can be used as a transparent devices using highly flexible reduced graphene oxide films as
conductor, hole/electron transport layer and active acceptor of transparent electrodes. ACS Nano 4, no. 9: 5263–8.
electrons. In this chapter, we have clearly demonstrated the 13. Choe, M., B. H. Lee, G. Jo et al. 2010. Efficient bulk-hetero-
advantages and disadvantages of using graphene in photovol- junction photovoltaic cells with transparent multi-layer gra-
taics. The research on graphene has been recently intensified phene electrodes. Organic Electronics 11, no. 11: 1864–9.
with the development of GQDs that show unique optical prop- 14. Geng, J., L. Liu, S. B. Yang, S.-C. Youn, D. W. Kim, J.-S.
Lee, J.-K. Choi, and H.-T. Jung. 2010. A simple approach for
erties. The future of graphene is very bright, and its practical preparing transparent conductive graphene films using the
applications in electronics have begun to emerge. We hope controlled chemical reduction of exfoliated graphene oxide in
that this chapter would be useful for the readers to develop an aqueous suspension. Journal of Physical Chemistry C. 114.
an understanding of the application of graphene in solar cells, no. 34. 14433–40.
and would provide new ideas for novel applications in differ- 15. Wang, Y., S. W. Tong, X. F. Xu, B. Özyilmaz, and K. P. Loh.
ent aspects of our lives in future. 2011. Interface engineering of layer-by-layer stacked graphene
anodes for high-performance organic solar cells. Advanced
Materials 23, no. 13: 1514–18.
ACKNOWLEDGMENTS 16. Huang, J.-H., J.-H. Fang, C.-C. Liu, and C.-W. Chu. 2011.
Effective work function modulation of graphene/carbon nano-
This work was supported by the Indo-UK project “Advancing tube composite films as transparent cathodes for organic opto-
the Effectiveness and Production Potential of Excitonic Solar electronics. ACS Nano 5, no. 8: 6262–71.
Cells (APEX).” The authors would like to thank the director, 17. Liu, Z., J. Li, Z.-H. Sun, G. Tai, S.-P. Lau, and F. Yan. 2012.
NPL, for his support. Tanvi Upreti would like to thank the The application of highly doped single-layer graphene as the
Department of Science and Technology for Senior Research top electrodes of semitransparent organic solar cells. ACS
Nano 6, no. 1: 810–18.
Fellowship.
18. Lee, S., J.-S. Yeo, Y. Ji, C. Cho, D.-Y. Kim, S.-I. Na, B. H.
Lee, and T. Lee. 2012. Flexible organic solar cells composed
REFERENCES of P3HT: PCBM using chemically doped graphene electrodes.
Nanotechnology 23: 344013.
1. Novoselov, K. S., A. K. Geim, S. V. Morozov et al. 2004. 19. Zhang, Q., X. Wan, F. Xing et al. 2013. Solution-processable
Electric field effect in atomically thin carbon films. Science graphene mesh transparent electrodes for organic solar cells.
306, no. 5696: 666–9. Nano Research 6, no. 7: 478–84.
2. Wan, X., G. Long, L. Huang, and Y. Chen. 2011. Graphene—A 20. Liu, Z., J. Li, and F. Yan. 2013. Package-free flexible organic
promising material for organic photovoltaic cells. Advanced solar cells with graphene top electrodes. Advanced Materials
Materials 23, no. 45: 5342–58. 25: 4296–301.
3. Gupta, V., A. K. K. Kyaw, D. H. Wang, S. Chand, G. C. Bazan, 21. Zhang, D., F. Xie, P. Lin, and W. C. H. Choy. 2013. Al-TiO2 com-
and A. J. Heeger. 2013. Barium: An efficient cathode layer for posite modified single-layer graphene as an efficient transparent
bulk-heterojunction solar cells. Scientific Reports 3, Article num- cathode for organic solar cells. ACS Nano 7, no. 2: 1740–7.
ber: 1965. DOI: 10.1038/srep01965. 22. Kim, H., S.-H. Bae, T.-H. Han, K.-G. Lim, J.-H. Ahn, and
4. Steim, R., F. R. Kogler, and C. J. Brabec. 2010. Interface mate- T.-W. Lee. 2014. Organic solar cells using CVD-grown gra-
rials for organic solar cells. Journal of Materials Chemistry phene electrodes. Nanotechnology 25, no. 1.
20, no. 13: 2499–512. 23. Kim, K., S.-H. Bae, C. T. Toh, H. Kim, J. H. Cho, D. Whang,
5. Chen, L.-M., Z. Xu, Z. Hong, and Y. Yang. 2010. Interface T.-W. Lee, B. Özyilmaz, and J.-H. Ahn. 2014. Ultrathin
investigation and engineering—Achieving high performance organic solar cells with graphene doped by ferroelectric polar-
polymer photovoltaic devices. Journal of Materials Chemistry ization. ACS Applied Materials and Interfaces 6: 3299−304.
20, no. 13: 2575–98. 24. Li, S.-S., K.-H. Tu, C.-C. Lin, C.-W. Chen, and M. Chhowalla.
6. Yang, X. and J. Loos. 2007. Toward high-performance poly- 2010. Solution-processable graphene oxide as an efficient
mer solar cells: The importance of morphology control. hole transport layer in polymer solar cells. ACS Nano 4, no. 6:
Macromolecules 40, no. 5: 1353–62. 3169–74.
7. Li, G., R. Zhu, and Y. Yang. 2012. Polymer solar cells. Nature 25. Ryu, M. S. and J. Jang. 2011. Effect of solution processed
Photon 6, no. 3: 153–61. graphene oxide/nickel oxide Bi-layer on cell performance
8. Hardin, B. E., H. J. Snaith, and M. D. McGehee. 2012. The of bulk-heterojunction organic photovoltaic. Solar Energy
renaissance of dye-sensitized solar cells. Nature Photon 6, Materials and Solar Cells 95, no. 10: 2893–6.
no. 3: 162–9. 26. Yun, J.-M., J.-S. Yeo, J. Kim et al. 2011. Solution-processable
9. O’Regan, B. and M. Gratzel. 1991. A low-cost, high-efficiency reduced graphene oxide as a novel alternative to PEDOT: PSS
solar cell based on dye-sensitized colloidal TiO2 films. Nature hole transport layers for highly efficient and stable polymer
353, no. 6346: 737–40. solar cells. Advanced Materials 23, no. 42: 4923–8.
27. Kim, J., V. C. Tung, and J. Huang. 2011. Water processable 42. Choi, H., H. Kim, S. Hwang, W. Choi, and M. Jeon. 2011.
graphene oxide: Single walled carbon nanotube composite Dye-sensitized solar cells using graphene-based carbon nano
as anode modifier for polymer solar cells. Advanced Energy composite as counter electrode. Solar Energy Materials and
Materials 1, no. 6: 1052–7. Solar Cells 95, no. 1: 323–5.
28. Nguyen, D. D., N. H. Tai, Y. L. Chueh et al. 2011. Synthesis of 43. Chen, L., C. X. Guo, Q. Zhang et al. 2013. Graphene quan-
ethanol-soluble few-layer graphene nanosheets for flexible and tum-dot-doped polypyrrole counter electrode for high-perfor-
transparent conducting composite films. Nanotechnology 22, mance dye-sensitized solar cells. ACS Applied Materials and
no. 29: 295606. DOI: 10.1088/0957-4484/22/29/295606. Interfaces 5, no. 6: 2047–52.
29. Wang, D. H., J. K. Kim, J. H. Seo, I. Park, B. H. Hong, J. 44. Wei, J., Y. Jia, Q. Shu et al. 2007. Double-walled carbon nano-
H. Park, and A. J. Heeger. 2013. Transferable graphene tube solar cells. Nano Letters 7, no. 8: 2317–21.
oxide by stamping nanotechnology: Electron-transport layer 45. De, S., P. J. King, M. Lotya et al. 2010. Flexible, transparent,
for efficient bulk-heterojunction solar cells. Angewandte conducting films of randomly stacked graphene from surfac-
Communications 52: 2874–80. tant-stabilized, oxide-free graphene dispersions. Small 6, no.
30. Jayawardena, K. D. G. I., R. Rhodes, K. K. Gandhi, M. R. R. 3: 458–64.
Prabhath, G. D. M. R. Dabera, M. J. Beliatis, L. J. Rozanski, S. 46. Wu, Z.-S., G. Zhou, L.-C. Yin, W. Ren, F. Li, and H.-M. Cheng.
J. Henley, and S. R. P. Silva. 2013. Solution processed reduced 2013. Graphene/metal oxide composite electrode materials for
graphene oxide/metal oxide hybrid electron transport layers energy storage. Nano Energy 1, no. 1: 107–31.
for highly efficient polymer solar cells. Journal of Materials 47. Kim, K. S., Y. Zhao, H. Jang et al. 2009. Large-scale pattern
Chemistry A 1: 9922–7. growth of graphene films for stretchable transparent elec-
31. Li, M., W. Ni, B. Kan, X. Wan, L. Zhang, Q. Zhang, G. Long, trodes. Nature 457, no. 7230: 706–10.
Y. Zuo, and Y. Chen. 2013. Graphene quantum dots as the hole 48. Novoselov, K. S., A. K. Geim, S. V. Morozov et al. 2005. Two-
transport layer material for high-performance organic solar dimensional gas of massless Dirac Fermions in graphene.
cells. Physical Chemistry Chemical Physics 15: 18973–8. Nature 438, no. 7065: 197–200.
32. Chen, Y., W.-C. Lin, J. Liu, and L. Dai. 2014. Graphene oxide- 49. Zhang, Y., Y.-W. Tan, H. L. Stormer, and P. Kim. 2005.
based carbon interconnecting layer for polymer tandem solar Experimental observation of the Quantum Hall effect and
cells. Nano Letters 14: 1467–71. Berry’s phase in graphene. Nature 438, no. 7065: 201–4.
33. Liu, Z., Q. Liu, Y. Huang et al. 2008. Organic photovol- 50. Liu, Z., D. He, Y. Wang, H. Wu, J. Wang, and H. Wang. 2010.
taic devices based on a novel acceptor material: Graphene. Improving photovoltaic properties by incorporating both
Advanced Materials 20, no. 20: 3924–30. SPFgraphene and functionalized multiwalled carbon nano-
34. Wang, S., B. M. Goh, K. K. Manga, Q. Bao, P. Yang, and K. P. tubes. Solar Energy Materials and Solar Cells 94, no. 12:
Loh. 2010. Graphene as atomic template and structural scaf- 2148–53.
fold in the synthesis of graphene organic hybrid wire with 51. Wudl, F. 2002. Fullerene materials. Journal of Materials
photovoltaic properties. ACS Nano 4, no. 10: 6180–6. Chemistry 12, no. 7: 1959–63.
35. Yu, D., K. Park, M. Durstock, and L. Dai. 2011. Fullerene- 52. Yu, D., Y. Yang, M. Durstock, J.-B. Baek, and L. Dai. 2010.
grafted graphene for efficient bulk heterojunction polymer Soluble P3HT-grafted graphene for efficient bilayer hetero-
photovoltaic devices. The Journal of Physical Chemistry junction photovoltaic devices. ACS Nano 4, no. 10: 5633–40.
Letters 2, no. 10: 1113–18. 53. Pan, D., J. Zhang, Z. Li, and M. Wu. 2010. Hydrothermal route
36. Yan, X., Cui, X., and Li, L.-S. 2010. Synthesis of large, stable for cutting graphene sheets into blue-luminescent graphene
colloidal graphene quantum dots with tunable size. Journal of quantum dots. Advanced Materials 22, no. 6: 734–8.
the American Chemical Society 132, no. 17: 5944–5. 54. Kim, J. K., M. J. Park, S. J. Kim et al. 2013. Balancing light
37. Gupta, V., N. Chaudhary, R. Srivastava, G. D. Sharma, R. absorptivity and carrier conductivity of graphene quantum
Bhardwaj, and S. Chand. 2011. Luminscent graphene quantum dots for high-efficiency bulk heterojunction solar cells. ACS
dots for organic photovoltaic devices. Journal of the American Nano 7, no. 8: 7207–12.
Chemical Society 133, no. 26: 9960–3. 55. Yang, N., J. Zhai, D. Wang, Y. Chen, and L. Jiang. 2010. Two-
38. Li, Y., Y. Hu, Y. Zhao et al. 2011. An electrochemical avenue dimensional graphene bridges enhanced photoinduced charge
to green-luminescent graphene quantum dots as potential transport in dye-sensitized solar cells. ACS Nano 4, no. 2:
electron-acceptors for photovoltaics. Advanced Materials 23, 887–94.
no. 6: 776–80. 56. Tang, B., G. Hu, H. Go, and Z. Shi. 2013. Three dimensional
39. Zhu, K., N. R. Neale, A. Miedaner, and A. J. Frank. 2006. graphene network assisted high performance dye sensitized
Enhanced charge-collection efficiencies and light scattering solar cells. Journal of Power Sources 234: 60–8.
in dye-sensitized solar cells using oriented TiO2 nanotubes 57. Li, X., H. Zhu, K. Wang et al. 2010. Graphene-on-silicon Schottky
arrays. Nano Letters 7, no. 1: 69–74. junction solar cells. Advanced Materials 22, no. 25: 2743–8.
40. Wang, X., L. Zhi, and K. Mullen. 2007. Transparent, conduc- 58. Dutta, M., S. Sarkar, T. Ghosh, and D. Basak. 2012. Zno/
tive graphene electrodes for dye-sensitized solar cells. Nano graphene quantum dot solid-state solar cell. The Journal of
Letters 8, no. 1: 323–7. Physical Chemistry C 116, no. 38: 20127–31.
41. Roy-Mayhew, J. D., D. J. Bozym, C. Punckt, and I. A. Aksay. 59. Miao, X., S. Tongay, M. K. Petterson et al. 2012. High effi-
2010. Functionalized graphene as a catalytic counter electrode ciency graphene solar cells by chemical doping. Nano Letters
in dye-sensitized solar cells. ACS Nano 4, no. 10: 6203–11. 12, no. 6: 2745–50.
20 Graphene Production from
Chlorination of Metallocenes
P. González-García, E. Urones-Garrote, A. Gómez-Herrero,
D. Ávila-Brande, and L. C. Otero-Díaz
CONTENTS
Abstract...................................................................................................................................................................................... 387
20.1 Introduction...................................................................................................................................................................... 387
20.2 Organometallics as Precursors of Graphene-Based Materials......................................................................................... 389
20.3 Chlorination Method........................................................................................................................................................ 389
20.3.1 Historical Perspective........................................................................................................................................... 389
20.3.2 Chlorination Mechanism...................................................................................................................................... 390
20.3.3 Chlorination of Organometallics.......................................................................................................................... 391
20.4 TEM Study of Graphene-Like Architectures................................................................................................................... 391
20.4.1 Role of Chlorination Temperature and the Carbon Precursor.............................................................................. 391
20.4.1.1 SCNM Obtained from Ferrocene.......................................................................................................... 391
20.4.1.2 Hollow and Solid Spheres Obtained from Cobaltocene........................................................................ 393
20.4.1.3 SCNM Obtained from Chromocene...................................................................................................... 393
20.4.1.4 SCNM Obtained from Nickelocene....................................................................................................... 394
20.4.1.5 Solid Spheres Obtained from Bis(cyclopentadienyl) Titanium Dichloride........................................... 394
20.4.1.6 SCNM Obtained from Bis(cyclopentadienyl) Niobium Dichloride....................................................... 395
20.4.1.7 SCNM Obtained from Bis(cyclopentadienyl) Zirconium Dichloride.................................................... 395
20.4.1.8 Carbon Nanobags Obtained from Bis(cyclopentadienyl) Tungsten Dichloride..................................... 396
20.4.2 Influence of the Reaction Time............................................................................................................................ 397
20.4.3 Effect of the Produced Halide.............................................................................................................................. 397
20.5 Bonding Ratio and Mass–Density.................................................................................................................................... 398
20.6 In-Plane Correlated Length of Graphene-Like Layers..................................................................................................... 399
20.7 Adsorption Properties and Porous Structure.................................................................................................................... 400
20.8 Summary and Conclusions............................................................................................................................................... 402
References.................................................................................................................................................................................. 402
ABSTRACT Raman spectroscopy measurements are included. The inter-

esting structural and textural characteristics found in the par-
A variety of carbon micro–nanoparticles (tubes, bags, rib- ticles mentioned before provide a wide range of alternatives
bons, chains, filled solid and hollow spheres) have been pro- to obtain different graphene-based materials with potential
duced from chlorination of metallocenes M(C5H5)2 and their applications such as gas storage (CO2, hydrogen, methane),
derivatives M(C5H5)2Cl2. Here, the low melting point and high energy-storage systems, electrodes in electrochemical double-
reactivity of the precursors in the presence of chlorine gas layer capacitor, and catalyst supports, among others.
result in a spontaneous reaction yielding carbon architectures
formed by highly disordered graphene-like layers. Changes at
20.1 INTRODUCTION
the micrometric and nanometric scales in terms of morphol-
ogy, structure, bonding, mass–density, chemical composition, Nanoscience and nanotechnology have grown exponentially,
and porous texture as a result of the precursor (M = Fe, Co, especially in the last decade, possibly due to the increase in
Ni, Cr, Nb, W, Ti,Zr) and the synthesis conditions are ana- the development of new synthetic methods and tools for the
lyzed and described in detail mainly by transmission electron study and manipulation of different types of nanomaterials.
microscopy, in both image and diffraction modes, as well as Recently, the size dependence on electrical, optical, and mag-
using the analytical tools (energy-dispersive spectroscopy, netic properties of various nanostructures which form part of
electron energy-loss spectroscopy) associated with the micro- semiconductors, metals, and other materials is better under-
scope; in addition nitrogen adsorption, x-ray diffraction, and stood [1].
387
With respect to carbon materials, a considerable part of Owing to its success, the chlorination method has been
scientific research has been focused on the production of extended to other groups of precursors. In this sense, our
nanostructured carbons as a result of the the so-called “nano- research group has decided to work with organometallic com-
revolution.” The synthesis of fullerenes in 1985 and the land- plexes as carbon precursors, making a particular emphasis on
mark paper describing the carbon nanotubes (CNTs) in 1991 their chemical composition, because it provides the carbon
created a special interest to acquire knowledge of the structure source and the metal catalyst in the same structure. In previous
and properties in these series of carbon allotropes [2,3]. In 2005, works we have described the synthesis of diverse SCNM from
isolation of the graphite foils gave way to an unexplored two- ferrocene, cobaltocene or chromocene, bis(cyclopentadienyl)
dimensional material named graphene [4,5]. This fact allowed tungsten dichloride, and bis(cyclopentadienyl) titanium dichlo-
the discovery of numerous and unusual properties for this new ride used as carbon precursors [10–15]. In the reported data for
member of the carbon family; some of them yet unknown. the chlorination of the organometallic precursors mentioned
Nowadays, engineers require new synthetic methods to before, the nature and chemical structure of the metal halides
design carbon materials at the atomic level, yielding a bet- formed during the synthesis seem to have a more important
ter control over the structure as well as capability to produce effect than the crystal structure of the starting metallocene;
selectively novel micro–nanostructures such as nanotubes, therefore, the halide crystal structure could be responsible
carbon onions, nanocrystalline diamond, or nanoporous car- for the observed carbon nanostructures. Table 20.1 summarized
bons with a well-defined pore-size distribution (PSD) for gas the different SCNM obtained by the chlorination reaction of the
storage, batteries, supercapacitors, and other energy-related organometallics mentioned before.
applications. In this sense, the selective etching of metal Significant progresses in the knowledge of these groups
carbides is considered to be a technique for the synthesis of of novel carbon nanomaterials have been achieved with the
almost all the allotropic forms of carbon depending on the appropriate characterization methods. The use of conven-
experimental conditions [4]; thereby, disordered and nano- tional transmission electron microscopy (CTEM) and associ-
crystalline graphitic carbons, nanotubes, fullerene-like struc- ated techniques such as, x-ray energy-dispersive spectroscopy
tures, carbon onions, nanocrystalline diamond, and ordered (EDS) and electron energy-loss spectroscopy (EELS) have
graphite have been obtained. Thus, this method allows the provided important information concerning the possible for-
synthesis of almost all carbon allotropes [7,8]. mation mechanisms, their nanostructure, local chemical com-
On the contrary, a recent development in organometallic position, and carbon bonding.
chemistry research has been focused on the use of organome- On the CTEM studies some fullerene-like structures and
tallic complexes as the carbon source for the synthesis of a graphite ribbons have been found in these carbon materials, as
wide variety of carbon materials having some architecture [9]; result of an incipient graphitization degree. However, a detailed
for instance, horns, several kinds of tubes, spheres (solid and scrutiny of the high-resolution transmission electron micros-
hollow), cages, bags, encapsulates, etc. This variety of carbon copy (HRTEM) images has assisted to detect that most of the
structures with interesting properties is commonly named formed nanostructure consists of highly disordered graphene-
as shaped carbon nanomaterials (SCNMs) and all of them like layers, which are responsible of their highly porous texture
exhibit interesting properties; nevertheless, their structure and and low density. Since the storage capacity of the carbon mate-
mechanism of formation is still not totally understood. rials depends on their porosity and the size distribution of these
TABLE 20.1
Shaped Carbon Nanostructures Obtained from the Chlorination of Metallocenes and Derivatives
as Result of the Reaction Temperature
Chlorination Temperature (°C)
Precursor 100 200 300 400 700 800 900
Fe(C5H5)2 Thick CNT, elongated α-CNT, lobes, Hollow spherical Hollow carbon
hollow nanobags elongated bags particles nanospheres
Co(C5H5)2 Conglomerates of spherical particles Solid and hollow spheres
Ni(C5H5)2 Solid spheres Solid spheres with
nanocrystals
Cr(C5H5)2 Round and α-CNT, solid and hollow
rod-filled particles spheres
Ti(C5H5)2Cl2 Chains of accreted solid spheres
Zr(C5H5)2Cl2 Carbon plaques Solid and hollow spheres
Nb(C5H5)2Cl2 Carbon plaques Hollow spheres
W(C5H5)2Cl2 α-carbon nanobags Irregular and elongated
α-carbon nanobags
Graphene Production from Chlorination of Metallocenes 389
pores, nitrogen adsorption–desorption measurements provide pentacarbonyl are commonly used as a catalyst and carbon
the essential information about the surface area, the pore vol- source [43], besides other metallocenes such as cobaltocene,
ume, and the PSD development in these novel carbon materials. nickelocene, or ruthenocene [44,45] (see Table 20.2).
This work presents a brief summary concerning the most CVD is also a consolidate method for the synthesis of
notable characteristics, structure, and properties of SCNM graphene and graphene-like materials by using organome-
produced via chlorination (at different temperatures and reac- tallics. This has been demonstrated since 1990 [46]. In these
tion time) of organometallic compounds and their derivatives early studies, graphite nanoribbons via decomposition of
as carbon precursors. The evolution of the carbon nanostruc- CO/H2/Fe(CO)5 at 400–700°C were described. Recent stud-
ture as well as their final shape, related with the type of precur- ies show novel techniques to produce hybrid graphene-based
sor and the synthetic conditions, is analyzed mainly by TEM materials from organometallics as precursors. For instance,
in both image and diffraction modes as well as the analytical thermally reduced graphite oxide from methylcyclopenta-
tools (EDS, EELS) associated with the microscope. Although dienyl-trimethylplatinum(IV) or PtSn nanotubes and Pt3Sn
the results of these investigations will be described in detail reduced-graphene oxide (RGO) nanohybrid thin films via a
along the chapter of the book, a brief advance of our experience simple reduction from PtCl2 cis,cis-1,5-cyclooctadiene [47,48].
in this particular field is described in the following paragraphs.
20.3 CHLORINATION METHOD

20.2 ORGANOMETALLICS AS PRECURSORS
OF GRAPHENE-BASED MATERIALS 20.3.1 Historical Perspective
In the search of carbon precursors and more efficient syn- Nowadays, the selective etching of metal carbides by halogens
thetic routes in order to prepare novel carbon materials a pat- at high temperatures is a well-established method to synthetize
ent presented by Komatsu and Endo in 1985 appears to be the carbon nanostructures named carbide-derived carbons (CDCs)
first work describing the use of an organometallic compound [7,8,49]. However, chlorination of metal carbides for the pro-
to prepare a shaped carbon material [16]. In 1993 an appara- duction of chlorides is not a recent technology. A manufactur-
tus was described for the continuous growth of carbon fibers ing process based on the reaction of chlorine gas on silicon
from a mixture of a high vapor pressure liquid (hexane) and a carbide was proposed in 1918 according to the following reac-
suitable source of iron catalyst particles: ferrocene [17]. tion [50]:
The catalytic activity of ferrocene resulted in the devel-
opment of new synthetic alternatives for optimizing exist- SiC(s) + 2Cl 2( g ) → SiCl 4( g ) + C(s)
ing carbon materials and for the production of new carbon
nanostructures, particularly films formed by CNTs [18,19]. The main product of the reaction was silicon tetrachloride in
From this moment, there was an increase in the use of fer- gas phase. Owing to its low boiling point (57.65°C) it can be eas-
rocene as catalyst, probably due to its availability, cost, sta- ily passed through the outlet pipe into a condenser. Residual car-
bility in the presence of air and high chemical reactivity. As bon that was considered to be in the form of “porous graphite”
a consequence, ferrocene has been part of numerous works, or “carbonaceous cinder” was disposed of. During the 1960s
for instance: the synthesis of aligned multiwalled carbon the method was extended to titanium, niobium, and zirconium
nanotubes (MWCNTs) by pyrolysis of ferrocene dissolved in carbides as precursors in the production of their corresponding
benzene in atmosphere argon or via aerosol dispersion of a halides [51,52]. Consequently, the kinetics of the chlorination
solution of benzene [20,21]. High-purity single-walled carbon reaction was studied and the next general formula was proposed:
nanotubes (SWCNTs) were obtained by a simple chemical
vapor deposition (CVD) of a mixture of ferrocene in toluene MeC(s) + x/2Cl2( g ) → MeClx ( g ) + C(s)

[22], among others (see Table 20.2).
Probably during last 10 years CVD has become one of the where x = 4 when Me = Ti or Zr and x = 5 when Me = Nb. Since
most common methods used to synthesize, under appropriate the objective of the method was the production of metal halides,
conditions, SCNMs. Owing to the versatility of this method, the synthesis of CDC as a main product was not significant until
several configurations have been proposed, for instance verti- the introduction of a possible “fourth class of amorphous car-
cal CVD reactor with one heat system, horizontal CVD reac- bon” later on described as “mineral active carbon” [53].
tors couple to one or two stage furnaces, atomizers, nebulizers, The porous structures of the CDC were studied in 1975 by
injectors, among others, permitting to produce single-walled, means of nitrogen and CO2 adsorption isotherms [54]. The
double-walled CNTs, or MWCNTs, filled and hollow carbon results surprisingly showed type I isotherms shape, accord-
spheres, carbon onions, fibers and bags, etc. [23–34]. Here the ing to the IUPAC (International Union of Pure and Applied
reaction atmospheres are also diverse: argon, hydrogen, nitro- Chemistry) classification [55], indicating the presence of pores
gen, acetylene, ethylene, methane, air, and their mixtures are smaller than 2 nm. A few years later, transition metal carbides
frequently reported [35–41]. The carbon source is also varied. of the IV–VI periodic groups were used as precursors for the
In most studies, the carbon used to produce SCNM is obtained synthesis of CDC [56]. Results confirmed the highly micropo-
from an external source and this can be any carbon-contain- rous character. In addition, high surface area and micropore
ing material, typically hydrocarbons [42]. Ferrocene and iron volumes were also obtained.
TABLE 20.2
Types of Carbon Nanostructures Obtained as a Result of the Precursor, Reactor Setup, and Experimental
Conditions
Nanostructure Precursors Reactor Design Temperature (°C) Atmosphere Reference
Nanofibers Ferrocene + hexane Modified vertical CVD 1100 H2–CO–CO2 mixture [17]
Nanotube bundles Ferrocene Two-furnace CVD 1077 Argon [18]
MWCNT Ferrocene–xylene mixture Two-furnace CVD 675 Argon with hydrogen [19]
Aligned CNT Ferrocene + benzene Spray pyrolysis 850–940 Argon [20]
MWCNT Ferrocene + benzene Spray pyrolysis 800 or 900 Argon [21]
MWCNT films Ferrocene + toluene Two furnace CVD 590–850 Argon–hydrogen [22]
SWCNT Hexane, ferrocene, and thiophene Vertical CVD 1150 Hydrogen [23]
SWCNT Fe(CO)5, cobaltocene–ferrocene, Two-furnace CVD 1100 Argon and acetylene [24]
nickelocene–cobaltocene, and
ferrocene–nickelocene mixtures
SWCNT Fe(CO)5 Horizontal CVD reactor 850–1200 Carbon monoxide [25]
MWCNT Ferrocene–xylene CVD reactor with a nebulizer 900 Argon [26]
DWCNT Ferrocene/sulfur mixture in Two-furnace CVD with an 900–1180 Argon–hydrogen [27]
xylene injector
SWCNT Ferrocene, CO, CO2 Vertical CVD with a hot wire 575–600 CO2, H2O [28]
DWCNT Ferrocene, methane Two-furnace CVD 1150 Argon [29]
MWCNT Ferrocene in toluene Two-furnace CVD 700–760 Hydrogen [30]
Hollow and solid Ferrocene and tetrachloroethylene Stainless steel autoclave 550 Autogenic pressure [31]
spheres
Oxide-filled CNT Ferrocene–Fe3(CO)12 mixture Two-furnace CVD 900 Argon [32]
and bucky-onions
Spherical particles Ruthenocene–ethylene mixture Stainless steel autoclave 900–1300 Ar, H2, and acetylene [33]
Two furnace CVD
Carbon nanofibers Cobaltocene–acetylene– Two furnace CVD 1100–1150 Hydrogen [34]
thiophene mixture
SWCNT Ferrocene Horizontal CVD 1100–1150 Argon–hydrogen [35]
CNT and balls Fe(CO)5 in pentane Horizontal CVD 700–1000 Nitrogen [36]
SWCNT Iron phthalocyanine Two-furnace CVD electrical 800–920 Argon–hydrogen [37]
furnace
MWCNT Ferrocene + acetylene Pyrolysis reactor 700–1000 Argon [38]
MWCNT ferrocene–ethylene Pyrolysis reactor 1000 Argon [39]
MWNT Fe(CO)5, methane Floating vertical CVD 1050–1150 Nitrogen [40]
α-CNTs Ferrocene, sulfur, and benzene Teflon-lined autoclave 200 Autogenic pressure [41]
MWNT Fe(CO)5, iron phthalocyanine, Vertical CVD 600–950 Helium [43]
acetylene, hydrogen
MWNT Ferrocene in benzene Two-furnace CVD 800–900 Argon [44]
Nanoribbons and Tetrahydrofuran and ferrocene Spray pyrolysis reactor 950 None [47]
MWCNT
All these studies clearly showed that the textural proper- method. Several factors, other than the temperature, are now
ties of the CDC could be controlled by selecting the reaction considered: chlorine flow [58], reaction time [59], posttreat-
temperature. However, it was not until 2003 that the sub-ang- ments [60], and other process variables [6,61,62]. Nevertheless,
strom control of the CDC PSD was achieved by the chlorina- the crystal structure of the carbide precursor seems to have an
tion of Ti3SiC2 [57]. important role in the produced carbon materials. For instance,
comparing the products obtained from the layered carbon
structure in a hexagonal Ti3SiC2 with the uniform distribution
20.3.2 Chlorination Mechanism
of carbon atoms in a cubic SiC (3C polytype) under the same
Last 10 years have provided a significant progress in understand- processing conditions, two different arrangements of the carbon
ing the mechanisms involved in CDC formation [6–8]. Hereby, a atoms are clearly observed [7]. CDC obtained from Ti3SiC2 con-
better control of the CDC structure and characteristics has been sists of a disordered network of graphite ribbons whereas CDC
achieved permitting to establish the basis of the chlorination from SiC shows an amorphous structure [57,63].
(f )
Cl2 Ar
(g) (c)
(e)
(d)
(b)
(a)
NaOH
FIGURE 20.1 Scheme of the reactor for synthesis of SCNM via chlorination reaction. Gases inlet (a), flowmeter (b), quartz tube (c), resi-
tance furnance (d), quartz crucible (e), termopar (f), trap and neutralizers flasks (g).
Hence, the metal carbide lattice acts as a template. With M(Cx H y )2 Cl 2(s) + nCl 2( g ) →
the increase of the reaction temperature, metals are extracted
layer by layer from the crystalline carbide structure by the xC(s) + MCl n ( g ) + yHCl( g ) M = Ti, Nb, Zr, W
controlled flow of chlorine gas forming the respective metal
halides and the structure of the remaining carbon is defined by The chlorination reactor used for their synthesis is a com-
the crystal structure of the carbide used as carbon precursor. mon setup for the treatment of solid samples in the gas flow
at elevated temperatures. Figure 20.1 shows a version of the
20.3.3 Chlorination of Organometallics experimental apparatus. Precursors are placed in a quartz ves-
sel and heated in a tubular furnace. Chlorine atmosphere is
The refractory properties of metal carbides such as hardness, inserted while rising the reactor temperature and maintained
chemical stability and high melting points (tantalum–hafnium during the reaction time; later on, chlorine gas flow is cut and
carbide has a melting point of 4215°C) are well known [64]. If replaced by argon for the removal of the rest of reacting chlo-
chlorine gas reacts with these high melting point and chemi- rine and metal halides during the reactor cooling to room tem-
cally stable compounds, which could be the result of the chlo- perature. Chlorine excess and volatile halides, at the exit of
rination of a precursor with totally opposite characteristics, the reactor, are neutralized with an NaOH saturated solution.
that is, low melting point and high reactivity? Metallocenes
seem to have the right answer to this question since their
chemical structure provides the metal catalyst and the carbon
20.4 TEM STUDY OF GRAPHENE-
source in the same molecule providing high chemical reactiv- LIKE ARCHITECTURES
ity in a whole range of temperatures and in the presence of 20.4.1 Role of Chlorination Temperature
reactive atmospheres.
and the Carbon Precursor
From these bases in 2005, a singular variation of the CDC
preparation method has been proposed by using ferrocene 20.4.1.1 SCNM Obtained from Ferrocene
as the carbon source instead of metal carbides [10]. Results The chlorination of ferrocene was the first work describing
showed that chlorination was successfully achieved yielding the synthesis of carbon nanoparticles, with diverse shape,
several SCNM (tubes, bags, solid and hollow spheres, twisted from the reaction between chlorine gas and an organometallic
branches) as a result of the variation in the process temperatures precursor [10]. Owing to the high reactivity of the precursor
and reaction times. The morphology of the SCNM was attrib- in the presence of chlorine, the reactions were performed at
uted to the template effect of iron halides during the synthesis. relatively low temperatures, ranging from 100°C to 300°C in
Iron is extracted from the metallocene structure and eliminated a short time (30 min).
as a mixture of iron chlorides (FeCl2 and FeCl3); hence, cyclo- The SEM (scanning electron microscopy) observations of the
pentadienyl rings provide the carbon source. The final carbon sample prepared at 100°C (Figure 20.2a) showed conglomerated
nanostructures are formed by highly disordered graphene-like rounded (length ranging from 180 to 350 nm) and cylindrical
layers yielding a slit-like structure with a surface area higher particles (outer diameter ≈500 nm, wall thickness ≈260 nm)
than 900 m2/g showing potential characteristics for their appli- [65]. The TEM examinations of these groups of particles
cation as energy storage materials [11]. Chlorination of the revealed that they actually are thick CNTs and elongated hollow
organometallic precursors for the production of SCNM has nanobags (wall thickness ≈15 nm). Sometimes, these nanobags
been described according to the following possible reactions: are partially filled with crystals of FeCl2 [66] as observed in
Figure 20.2d (note the differences in the contrast of the image).
M(Cx H y )2(s) + nCl 2( g ) → At 150°C the SEM studies showed a decrease in the
amount of the spherical particles and a higher proportion
xC(s) + MCl n ( g ) + yHCl( g ) M = Fe, Co, Ni, Cr of carbon tubes (outer diameter ≈470 nm, wall thickness
(a) (b) (c)
1 μm 1 μm 1 μm
(d) (e) 0.5 μm (f )
200 nm 100 nm
1 μm (g) (h) (i)
200 nm 20 nm
FIGURE 20.2 SEM micrographs of the SCNM produced from ferrocene at 100°C (a), 150°C (b), and 200°C (c). TEM images showing
carbon nanobags obtained at 100°C (d), flower-like particles at 200°C (e) and 300°C (f), open and closed-ended α-CNT at 150°C (g) and
200°C (h), medium magnification TEM of a single α-CNT (i).
355 nm) [65]. These tubes present similar dimensions to the their inner diameter can be measured in TEM images. Figure
100°C sample; however, their inner diameter was clearly 20.2h shows typical open and closed-ended amorphous CNTs
smaller (Figure 20.2b) [67]. The respective low magnifica- observed in this sample. Their external diameter is ≈70 nm and
tion TEM images reveals groups of nanobags surrounding a the inner one is ≈10 nm, presenting thick walls of 30 nm.
nucleus of crystalline FeCl2 in a structure similar to a carbon The contrast at the medium-magnification TEM image and
flower (Figure 20.2e). On the contrary, branches of closed- individual α-CNT (Figure 20.2i) and the diffuse rings in the
ended CNTs were also found (Figure 20.2g). selected area electron diffraction (SAED) patterns, inset of
At 200°C the sample consisted mainly of amorphous carbon Figure 20.2d and h, provide a sight of the disordered nature
nanotubes (α-CNTs) with different architectures: amorphous of these SCNM.
lobes, branches, and elongated bags [10,65,67]. These α-CNTs Increasing the chlorination temperature from 200°C to
present a wide size distribution, varying from 1 to 10 µm; some 300°C is reflected in the formation of hollow spherical particles
tubes presented open ends and in other cases bulb-like endings (diameter ≈100 nm and wall thickness ≈10nm), more defined
(Figure 20.2c). During the TEM observations, occasionally and than in the samples of lower temperature (Figure 20.2f).
besides the α-CNT were detected a few elliptical lobes filled Surprisingly, the α-CNTs were not formed from this and
with FeCl2. Since the chlorination product of ferrocene nor- for higher chlorination temperatures [10]. In this sense, at
mally is FeCl3, nanobags were detected at 150°C and these lobes 900°C in the SEM micrographs are found spherical particles
presented a nucleus of crystalline FeCl2 surrounded by layers of (Figure 20.3a) which actually are hollow carbon nanospheres
disordered carbon which is unusual. The nonconventional pres- (diameter ≈50–150 nm and wall thickness ≈12–25 nm) [10]
ence of this halide was confirmed by EDS analyses and HRTEM with 80% of sp2/sp3 bonding content (Figure 20.3b). The analy-
[10]. The possible mechanism for the formation of FeCl2 has ses obtained from Raman spectroscopy revealed the presence
been previously described. The α-CNTs found at this temper- of small graphene-like layers with a size of around 3–4 nm as
ature are longer than 1 micrometer in most of the cases, and the main structure of the walls of these carbon spheres [11].
(a) (b) (c) 5 nm
500 nm 100 nm
FIGURE 20.3 SEM micrograph (a) and TEM image (b) of the hollow carbon nanospheres found at 900°C in the chlorination of ferrocene.
HRTEM image shows the characteristic contrast for highly disordered carbon materials (c).
This was in agreement with the highly disordered contrast In terms of the structure of these spheres, their SAED pat-
observed in the HRTEM images (Figure 20.3c). terns showed intense diffraction rings commonly attributed to
disordered materials (see inset in Figure 20.4c). The spacing
20.4.1.2 Hollow and Solid Spheres of these rings could be assigned to the hk0-type graphite
Obtained from Cobaltocene reflections indicating that these spheres were formed by
The SEM images of the products obtained from the chlorina- almost individual graphene-like layers, without graphite-like
tion of cobaltocene at 200°C and 300°C showed the presence of stacking (see inset in Figure 20.4c). In addition, their sp2/sp3
conglomerates formed by spherical particles as the major mor- bonding content was in the range from 86% to 90% [65].
phology, whose diameters lie in the range from 70 to 300 nm
and from 30 to 250 nm, respectively (Figure 20.4a and b). The 20.4.1.3 SCNM Obtained from Chromocene
low magnification TEM image (Figure 20.4d) indicated that Chlorination of bis(cyclopentadienyl)chromium (Cr(C5H5)2),
these spherical particles are solid (note the homogeneous con- with P21/c space group, (a = 0.5786 nm, b = 0.7815 nm,
trast in each sphere). HRTEM images of these spheres showed c = 1.0787 nm, β = 124.08°) at 400°C and 900°C has been
that their structure presents a stacking of 6–10 graphene-like recently reported [13,68]. The SEM examinations showed the
layers forming small graphitic domains [12,65]. loss of the flat faces in the powder precursor after the chlorina-
The increase of the reaction temperature of up to 900°C tion reactions. In both cases, samples consist of conglomerated
was reflected in the enlargement of the obtained spheres round (diameter from 150 to 650 nm) and rod-like particles
ranging now from 100 to 400 nm (Figure 20.4e). In addition, (length from 3 to 6 µm and thickness from 200 to 800 nm) see
from the TEM images it was possible to detect two kinds of Figure 20.5a and b. The diameters of the round particles and
spheres: solid and hollow. In most of the hollow spheres, an the length of the rods seem to be independent of the reaction
open-end protuberance was observed (Figure 20.4c). temperature. However, in micrographs of the 900°C sample
(a) (b) (c)
10 nm
100 nm 100 nm
(d) (e)
100
200
100 nm 1 μm 200 nm
210 110
FIGURE 20.4 SEM micrographs showing the spherical particles obtained at 200°C (a), 300 (b), and 900°C (e) from cobaltocene. Low-
magnification TEM images showing the solid spheres produced at 200°C (d) and all the kinds of spheres found at 900°C (c). The SAED inset
in (c) is associated with the nanostructure observed in the HRTEM also inset in (c).
310
(a) (b) (c) 210
110
100
200
300
1 μm 1 μm
110
(d) 300
200
(e)
100
110
210
100 310
200 nm 100 nm 500 nm
FIGURE 20.5 SEM micrographs showing the changes in morphology of the SCNM obtained from chromocene chlorinated at 400°C (a)
to 900°C (b). Low-magnification TEM images showing all the SCNM (c), the solid (d), and the hollow (e) spheres formed at 900°C. Their
corresponding SAED patterns (inset) confirm the presence of graphene-like layers in these materials.
all these particles are more defined, the thickness of the rods is a d-spacing of 0.22 nm corresponding to the (112) plane of the
thinner (150–500 nm) and in some cases these rods present an NiCl2 ⋅ 6H2O structure (Figure 20.6c).
open-end indicating that some of them are hollow. The TEM studies of the 900°C (Figure 20.6e) sample
The low magnification TEM images from the materials reveal the presence of solid spheres (diameter from 75 to
obtained at 400°C showed that the sample mainly consists of 150 nm) surrounded by some small hollow spheres (diameter
filled particles [13]. In contrast, the SCNM obtained at 900°C from 35 to 50 nm and wall thickness from 6 to 9 nm). From
were hollow (diameter from 80 to 180 nm and wall thickness their HRTEM image it is possible to detect some individual
from 25 to 55 nm) and solid spheres (diameter from 120 to and open graphene-like layers at the edge of the wall of these
220 nm), according to the low magnification TEM exami- spheres. However, the contrast in the image is a clear indica-
nations (Figure 20.5d and e). Occasionally, some open and tive of their disordered nature (Figure 20.6f). This structure
close-ended CNTs (diameter from 150 to 450 nm, wall thick- is also confirmed by the SAED patterns showing diffraction
ness from 35 to 65 nm, and length ranging from 1.5 to 10 µm) rings with spacing that can be assigned to the hk0-type graph-
were also found (Figure 20.5c). Here, their sp2/sp3 bonding ite reflections (see inset in Figure 20.6d). At this temperature,
ratio is ≈89%, similar to the α-CNT obtained from ferrocene no more NiCl2 ⋅ 6H2O was detected by EDS, XRD (x-ray
[65]. The open-end tubes were shorter and thinner in com- diffraction), or HRTEM. All these SCNM showed a sp2/sp3
parison with the close-ended ones. In the case of spheres, the bonding ratio higher than 95%, independent to the chlorina-
hollow ones were slightly bigger than the solid ones [68]. As tion temperature.
observed in the HRTEM images all these SCNM are formed
by highly disordered graphene-like layers without tridimen- 20.4.1.5 Solid Spheres Obtained from
sional stacking [13]. This nanostructure is confirmed with the Bis(cyclopentadienyl) Titanium Dichloride
features displayed by the SAED patterns which show the hk0 The preliminary XRD studies indicated the presence of dis-
diffraction rings corresponding to the turbostatic carbon- ordered carbon materials at 400°C [15]. Besides the XRD
type reflections. Note the insets in Figures 20.5c through e. diagrams, semiquantitative EDS and the TEM studies did
not show the presence of encapsulate particles or residual
20.4.1.4 SCNM Obtained from Nickelocene organometallic precursor. Probably, the main difference of
Architectures formed by spherical particles are the predomi- the chlorination of this precursor in comparison to the others
nant morphology obtained from the chlorination of nickelo- mentioned in this chapter is that it allows the formation of the
cene. The SEM images showed conglomerates formed by most homogeneous morphology since only solid spheres have
rounded particles (diameter from 90 to 500 nm) at 200°C been obtained.
(Figure 20.6a). However, this size distribution is narrower In this sense, the reaction temperature plays an important
(diameter from 90 to 230 nm) at 900°C (Figure 20.6d). role in the structure of the produced spheres. The spheres of
During the TEM observations, the structure of the low- the lowest temperature sample (400°C) are randomly distrib-
temperature spheres seems to be formed by two materials uted forming beds (Figure 20.7a). At 900°C these spheres
(Figure 206b). The darkest contrast in the image indicates that seems to be individually dispersed and in some cases are
some crystalline particles are imbibed into the carbon matrix accreted forming chains of up to 15 spheres (Figure 20.7d).
that forms the carbon spheres (with lighter contrast). These The size distribution decreases from 400°C (diameter from
nanometric crystals (diameter ≈20 nm) were identified by 70 to 525 nm) to 900°C (diameter from 50 to 150 nm). The
semiquantitative EDS analysis as NiCl2 ⋅ 6H2O, as later con- homogeneous contrast observed in the TEM images indicates
firmed by the HRTEM images where it is possible to measure that all these spheres are solid (Figures 20.7b and 20.7e).
(a) (b) (c)

0.22 nm
1 μm 50 nm 5 nm
(d) (e) (f )
110
1 μm 100 nm 10 nm
100
FIGURE 20.6 SEM micrographs from the SCNM produced at 200°C (a) and 900°C from nickelocene (d). Low-magnification TEM
images showing the spherical particles formed at 200°C (b) and 900°C (e). HRTEM images from a section with an NiCl2 ⋅ 6H2O nanocrystal
(c) and an individual hollow carbon nanosphere (f).
110
(a) (b) (c) 10 nm
100
1 μm 200 nm
110
(d) (e) (f ) 10 nm
100
1 μm 200 nm
FIGURE 20.7 SEM micrographs (a, d), low-magnification TEM images (b, e), and HRTEM images (c, f) from the solid spheres obtained
at 400°C and 900°C respectively by chlorination of titanocene dichloride. The corresponding SAED patterns, insets in (b) and (e) show the
disordered nature of carbon in these spheres.
While at 400°C the contrast of the HRTEM image is in Figure 20.8b). However, at 900°C the SEM micrographs
characteristic for the highly disordered carbon materials showed rounded particles (diameter ≈ 1.25 µm) with molten
(Figure 20.7c), at 900°C it is possible to detect some individ- appearance (Figure 20.8c). As observed in Figure 20.8c, some
ual and open graphene-like layers at the edge of the spheres of these spheroids are broken and inside it is possible to detect
(Figure 20.7f). The SAED patterns of the spheres formed at a conglomerate material.
400°C are weak and diffuse while at 900°C are slightly more A low-magnification TEM image made on one piece of
intense and defined as a result of the incipient graphitization these broken spheroids (Figure 20.8d) reveals that they actu-
degree which is clearly noted in the HRTEM images. Another ally are formed by agglomerated hollow carbon nanospheres
important feature is their mass–density which decreases from (diameter from 25 to 65 nm and wall thickness from 6 to 9 nm).
1.4 to 1.1 ± 0.1 g/cm3 from 400 to 900°C. This is later on From the medium magnification (Figure 20.8e) and HRTEM
reflected in their textural parameters [15]. However, with the (Figure 20.8f) images it is possible to detect that these hollow
sp2/sp3 bonding ratio, higher than 95%, is almost unaffected spheres are formed by disordered wavy graphene-like layers,
by the synthesis temperature. some of them reaching up to 6 nm. The bidimensionality of
these curved layers is validated with the SAED pattern inset
20.4.1.6 SCNM Obtained from Bis(cyclopentadienyl)
in Figure 20.8f where the spacing of the rings corresponds to
Niobium Dichloride the hk0 graphite reflections.
The SCNM obtained from this precursor probablypresent the
most extreme change in terms of morphology and structure. 20.4.1.7 SCNM Obtained from Bis(cyclopentadienyl)
At 200°C (Figure 20.8a and b), the sample consists of dense Zirconium Dichloride
carbon platelets of irregular dimensions formed by highly From the SEM micrographs an evolution in the shape of the
disordered graphene-like layers (see the HRTEM image inset produced carbon materials is clearly observed. At 200°C

(a) (b) 5 nm (c)
1 μm 50 nm 500 nm
200 210
(d) (e) (f )
100 110
200 nm 50 nm 5 nm
FIGURE 20.8 Low-magnification TEM images of the SCNM obtained at 200°C (a) and 900°C from NbCp2Cl2 (d). SEM micrograph of a
broken spheroid detected at 900°C (c). Medium magnification TEM images from the plaques (b) and the hollow spheres (e). HRTEM image
showing wavy graphene-like layers as hollow sphere structures (f). The intense diffraction rings of the SAED pattern inset in (f) confirm
the presence of these layers.
(a) (b) (c)
2 μm 10 nm
200 nm
(d) (e) (f )
5 nm
2 μm 1 μm 5 nm 200 nm
FIGURE 20.9 SEM micrographs of the SCNM obtained at 200°C (a) and 900°C by chlorination of zirconocene dichloride (d). Low-
magnification (b) and HRTEM image (c) of the 200°C sample. The enlargement of the squared zone in this image shows the interlayer spac-
ing of the precursor. Low-magnification TEM images for the hollow (e) and solid (f) spheres obtained at 900°C. The contrast in the HRTEM
images (inset) is characteristic of their disordered structure.
(Figure 20.9a) the sample consists of micrometric-sized 350 nm) were found in less proportion. The HRTEM images
plaques but at 900°C it is totally transformed into spheri- (inset) showed that the structure of these spheres is formed
cal particles with diameter ranging from 1.3 to 0.25 µm by graphene-like layers. However, the contrast of the images
(Figure 20.9d). The low-magnification TEM image of the suggest that these layers are disordered in the hollow spheres
low-temperature sample (Figure 20.9b) shows that the plaques yielding a highly porous structure whereas in the solid spheres
observed by SEM have imbibed small crystalline particles present a dense packing. On the other hand, the sp2/sp3 bond-
(identified by their darkest contrast). A closer examination ing content is close to 100%.
made at enlargement of the squared zone in the HRTEM of
Figure 20.9c allows to observe a interlayer spacing of 0.97 nm 20.4.1.8 Carbon Nanobags Obtained from
which corresponds to the (110) plane of the zirconocene Bis(cyclopentadienyl) Tungsten Dichloride
dichloride structure [69]. Therefore, it was possible to iden- Chlorination of bis(cyclopentadienyl)tungsten dichloride at
tify that these dark crystals actually are nonreacting precursor 400°C and 900°C results in hollow carbon nanobags (α-CNB)
nanoparticles. The presence of these crystalline nanoparticles having irregular dimensions and shape [14]. According to
is probably due to the low chlorination temperature which is the XRD diagrams and SEM micrographs (Figure 20.10a
below the melting point of the precursor (243°C) [70]. and d), the transformation of the precursor into the α-CNB
The low-magnification TEM images of the spherical par- is achieved at 400°C. The low-magnification TEM images
ticles synthesized at 900°C (Figure 20.9e and f) permitted showed changes in the overall morphology in the produced
to detect that most of the sample consist of hollow spheres α-CNB from 400°C (diameter from 25 to 150 nm and wall
(diameter from 625 to 825 nm and wall thickness from 180 to thickness from 3.5 to 20 nm) to 900°C (diameter from 85 to
250 nm). Besides, smaller solid spheres (diameter from 250 to 200 nm and wall thickness from 6 to 40 nm), see Figure 20.10b
(a) (b) (c)
2 μm 100 nm 200 nm
(d) (e) (f )
110
2 μm 100 nm 5 nm
100
FIGURE 20.10 SEM micrographs of the α-CNBs obtained at 400°C (a) and 900°C (d). Low-magnification TEM images of the 400°C (b)
and 900°C α-CNBs (c, e). HRTEM image (f) from the wall of an α-CNB from (c) with typical contrast of disordered carbon materials, as
noticed in the SAED pattern (inset).
and e. Occasionally, elongated α-CNB (Figure 20.10c) were In terms of the overall microstructure, the XRD diagrams did
also found at 900°C (length from 70 to 250 nm, width from 35 not show a significant change, both diagrams were typical of
to 85 nm, and wall thickness from 8 to 15 nm). disordered carbon materials.
The high gradient in the chlorination temperature is reflected
also at the nanometric scale. Contrast in the HRTEM images 20.4.3 Effect of the Produced Halide
of the 400°C bags is typical of disordered or amorphous car-
bons [14]. However, at 900°C is characteristic of carbon mate- As described in Section 20.3.3, during the synthesis procedure,
rials containing highly disordered graphene-like layers, as the respective halides (or mixture of halides) are formed in gas
observed in Figure 20.10f. However, some local and small gra- phase as secondary products. TiCl4 and WCl6 have the lowest
phitic domains formed by 2–4 stacked graphene–like layers can boiling points of these series (136°C and 346°C, respectively)
be found, as result of an incipient graphitization degree at this [70]. For this reason, these halides are easily eliminated from the
temperature. In addition, some individual graphene sheets were carbon networks in the SCNM obtained at chlorination temper-
imaged at the edge of the hollow particle. The HRTEM image of atures higher than 400°C, as confirmed by the EDS semiquan-
the elongated α-CNBs, also shows disordered curved graphene titative analyses, the XRD patterns, and the TEM examinations
sheets at the edge of the elongated particle [14]. The presence of [14,15]. In some cases, as a result of their physicochemical char-
the graphene-like layers as the main structure of these bags is acteristics and the low chlorination temperature, the produced
confirmed by the diffraction rings observed in the SAED pat- halides are trapped into the carbon structures. This phenomenon
terns which correspond to the hk0 graphite-type reflections. has been observed in the chlorination of ferrocene [10], chro-
mocene [13], and nickelocene. Ferrocene produces FeCl3, due
to its relatively low melting point (306°C); its elimination from
20.4.2 Influence of the Reaction Time
the carbon network is easily achieved by increasing the reaction
Reaction time is an important parameter in carbon materials temperature or time, as described in Section 20.4.1.1.
since it modifies their microstructure and these variations are It is well known that NiCl2 (yellow gas phase) results from
later mainly reflected in the textural parameters. The effect chlorination of nickelocene and CrCl3 (red-violet gas phase)
of the chlorination reaction on the textural characteristics has from chromocene. During the synthesis at relatively low
been observed in the SCNM produced from ferrocene [11], temperatures, in most of cases below 400°C, these halides
nickelocene, and titanocene dichloride at 900°C. However, are not totally extracted from the produced carbon particles
the effects of the chlorination time on the structure of the as a consequence of their high melting point (≈1001°C and
SCNM have been analyzed only for the SCNM produced 1152°C respectively) [70], as it was described for nickelocene.
from ferrocene at 200°C [10]. However, this fact is more evident in the products of the chlo-
When the TEM observations of the SCNM produced at rination of chromocene, where the halide remains encapsu-
30 min were performed, a great variety of carbon nanostruc- lated forming shell-like particles consisting of a crystalline
tures were found: α-CNT, empty carbon nanobags, solid nucleus identified as CrCl3 [71] surrounded by layers of disor-
spheres, elongated lobes, branches, and encapsulated materials dered carbon [13].
[10,11,67]. However, with the increase in the reaction time up to Probably, the most important feature inherent to the chlo-
180 min empty carbon nanobags were found in majority [10]. rination mechanism is the template effect provided by the
TEM analysis provided more information about the mor- formed halides. Carbon layers are successively deposited on
phology of these SCNM. In the 30 min sample some lobes the surface of the CrCl3. Later, with the increase of the reac-
filled with rhombohedral FeCl2 were observed surrounding tion temperature, the halide diffuses to the medium through
solid spheres (Figure 20.11a). These particles, as observed in the carbon coatings generating the characteristic porous struc-
Figure 20.11a, present two empty lobes situated in opposite ture of these SCNM and remaining intact the shape formed at
direction, as well as a pair of filled ones. In the 180 min sam- the beginning of the reaction.
ple, similar architectures are present but the filled lobes were Signs of these transformations are obtained from the filled
not detected (Figure 20.11b). The wall thickness increased and hollow spheres formed from chromocene. Here, the
from 10 to 20 nm and the number of solid spheres decreased. medium magnification TEM images (Figure 20.12) permits to
100 nm (a) (b)

C C
26 nm
24 nm
CrCl3
100 nm 20 nm 20 nm
(a) (b)
FIGURE 20.11 Low-magnification TEM images showing an FIGURE 20.12 Medium-magnification TEM images for a filled
intermediate step to the formation of amorphous carbon nanobags sphere at 400°C (a) and the resulting hollow carbon sphere at
at 30 min (a) and the hollow particles at 180 min (b). 900°C (b).
clearly compare the similarities in the dimensions of the full- bonding content by comparing the ELNES of a carbon sam-
empty cores, the wall thickness and the outer contrast, suggest- ple with a standard (usually graphite) where 100% of sp2 is
ing neither changes in the morphology nor in the microstructure. assumed [77,78]. This kind of comparison can be found in
Figure 20.13a. Here, all the spectra present graphite-like
ELNES. However, the presence of a round and featureless σ*
20.5 BONDING RATIO AND MASS–DENSITY
peak is a clear indication of the highly disordered nature of the
EELS as an associated technique to the transmission elec- graphene-like layers present in these materials. As observed
tron microscope involves the measurement of the lost energy in Figure 20.13a, the ELNES shape seems to be independent
transferred to a specimen by fast incident electrons [72,73]. of the carbon precursor and the reaction temperature.
One of the most important features in the energy-loss process Our results have shown that the calculated sp2/(sp2 + sp3)
is the atomic ionization, in which electrons are ejected from bonding ratio is high and very variable, ranging from 89% to
inner, or core, shells of atoms in the specimen. This character- 100%. A comparison of the bonding ratio values as a func-
istic signal termed “ionization edges” appear in a spectrum at tion of the metal precursor and the chlorination temperature
energy losses corresponding to the critical ionization energy for some of the carbon nanostructures described in Section
(which is a function of the specific atom and the specific shell 20.4.1 is depicted in Figure 20.13b. Once again, this param-
and is therefore uniquely defined) [72]. These ionization edges eter seems to be unaffected by the carbon precursor and the
have small intensity fluctuations termed as energy-loss near- reaction temperature. Hence, it is possible to find that dif-
edge structure (ELNES), their quantification permits the iden- ferent carbon particles present variable bonding ratios even
tification of specific elements in the specimen and details of when they are obtained at the same temperature.
the local atomic environment such as coordination, valence, Additional information can be extracted from the low-
and the type of bonding [72–74]. loss region (0–100 eV) of the EEL spectra. Here, the zero-
In carbon materials, atoms usually adopt variable hybrid loss peak (0 eV) is the most intense signal [72,73]. For carbon
orbitals such as sp, sp2, sp3, and intermediate states commonly materials, this zone displays more than one contribution that
named as spn [75]. In diamond, the tetrahedral disposition of can be detected as peaks arising from inelastic scattering by
the carbon atoms is a result of its four σ bonds formed by the the outer-shell electrons. The first two contributions are the
sp3 hybrid orbitals. In graphite, carbon atoms present a trigo- π – π* and the surface (π + σ) plasmons, localized around 7
nal geometric disposition having one p orbital to build the π and 15 eV, respectively. These are related to the number of
bonds and three strong σ bonds formed by the sp2-hybridized unsaturated bonds. Later, the bulk π + σ plasmon, related to
orbitals [70]. Carbon–K edge corresponds to the transition the valence electron density, appears at 27 eV in graphite and
from the 1s core level to the unoccupied part of the nonbond- 33 eV in diamond [79]. Thus, as the mass–density of the mate-
ing 2p states of the atoms. In diamond only the 1s → σ* tran- rials can be associated to the electron density, any displace-
sition is possible while in graphite the 1s → π* and 1s → σ* ment in the bulk π + σ plasmon is reflected as a change in the
transitions are present [76]. In the carbon EEL spectrum, mass–density values [80]. In Figure 20.14 this phenomenon is
these transitions are localized at 284 and 291 eV, respectively, clearly observed.
and are observed as two maxima with different intensity and All the SCNM have bulk π + σ plasmon below the graph-
shape [76]. ite and diamond taken as reference. There is a wide range
For carbon materials with intermediate sp2/sp3 bond- of energy-loss values where these plasmons are located
ing ratio, EELS provides a tool to estimate the relative sp2 and these positions do not present a trend associated to the
(a) (b)
σ∗
π∗ 100
Graphite 98
900°C
W(C5H5)2Cl2
sp2/(sp2 + sp3)
96
800°C
Ti(C5H5)2Cl2
94
600°C
Ni(C5H5)2
92
Cr(C5H5)2 400°C
90
Cr Fe Co
Zr(C5H5)2Cl2 200°C Ni Ti Zr W
88
260 280 300 320 340 200 400 600 800 1000
Energy loss (eV) Temperature (°C)
FIGURE 20.13 Comparative as a function of the metal precursor and chlorination temperature for some SCNM. Carbon-K ELNES (a)
and sp2/(sp2 + sp3) bonding ratio (b).
3.5
Ni(C5H5)2
Diamond 20.6 IN-PLANE CORRELATED LENGTH
Cr(C5H5)2 OF GRAPHENE-LIKE LAYERS
3.0 Co(C5H5)2
Zr(C5H5)2Cl2
SCNM present a structure formed by highly disordered
Mass–density (g/cm3)
Ti(C5H5)2Cl2 graphene-like layers, as described in Section 20.4.4. The

2.5 W(C5H5)2Cl2 order–disorder degree and the in-plane correlation length
Graphite
(La) have been analyzed by Raman spectroscopy. In this
2.0
sense, the first-order Raman spectra for some SCNM pre-
pared at 900°C (inset in Figure 20.15) show the two typical
bands for disordered carbon materials: the so-called graphite
15 (G) band at ≈1580–1600 cm−1 and the disorder-induced (D)
band at ≈1300–1350 cm−1 [81,82]. The G band corresponds
1.0 to the in-plane bond-stretching motion of carbon atoms in
18 20 22 24 26 28 30 32 34 sp2 configuration with E2g symmetry. The D band is a breath-
Energy loss (eV) ing mode with A1g symmetry which is forbidden in graph-
ite and becomes active in disordered graphitic structures. A
FIGURE 20.14 Variation of the mass–density with respect to the characteristic feature of the present SCNM is the absence
bulk π + σ plasmon for the SCNM produced at 900°C. of the characteristic second-order 2D (≈2700 cm−1) and 2G
(≈3250 cm−1) bands related to the ABAB… tridimensional
morphology of the particle. However, it can be detected that stacking of the graphene layers in graphite [83]. This feature
the organometallic precursor could play an important effect is also an indication of the highly disordered nature of these
in the final arrangement of the graphene-like layers since materials.
the displacements of the bulk π + σ plasmon are shifted The full width at half maximum (FWHM) values also pro-
to high energy-loss values in the SCNM obtained from vide information about the structure of the carbon materials.
Cr(C5H5)2 as the carbon precursor. In contrast, the plas- As we can observe from the inset of Figure 20.15, the FWHM
mons for those materials produced from Ni(C5H5)2 present of the D band is larger than the G one as a consequence of
the lowest values. As the corresponding plasmon position the highly disordered stacking of the graphene-like layers in
yields to the mass–density, from Figure 20.14 it is possible the carbon particles. Figure 20.15 shows a comparison of the
to observe that all these SCNM have mass–density values FWHM values with the nanostructure observed by HRTEM.
ranging from 1.05 to 1.78 g/cm 3. In this sense, the car- Here it is possible to detect that the lowest FWHM correspond
bon foils and the solid spheres obtained from Ni(C5H5)2 to the SCNM produced from Ni(C5H5)2 and Ti(C5H5)2Cl2 as
and Ti(C5H5)2Cl2, respectively, present the lowest mass– a consequence of some small graphite-like domains found
density, up to 50% lower than the density of graphite [15,70]. among the graphene matrix. In contrast, the highest FWHM
160
D G
Cr(C5H5)2
150 Fe(C5H5)2
Ni(C5H5)2
Ti(C5H5)2Cl2
140 W(C5H5)2Cl2
FWHM (cm–1)
1000 1500 2000 2500

130
120
110
100
Ni Ti Fe Cr W
Metal precursor
FIGURE 20.15 Relationship between the nanostructure of the SCNM obtained at 900°C from some organometallic precursors and their
corresponding FWHM values. The inset shows first-order Raman spectra for the selected SCNM.
values are for the SCNM obtained from Cr(C5H5)2 and and important micropore volume (W0) [11,13–15]. Since
W(C5H5)2Cl2, especially this last one which have the most both parameters have a strong dependence of the synthetic
disordered graphene-like structure, as it has been previously method a careful manipulation of the synthesis condi-
described in the HRTEM study and supported with their cor- tions provide the possibility to control the developments of
responding SAED patterns. high SBET values with well-defined pore-size distributions
On the contrary, the intensity ratio between the D and G (PSDs).
bands (ID/IG) also provides structural information about La of CDC have proved their versatility in the tuning of the
the graphene-like layers. According to the Tuinstra–Koenig textural parameters by selecting the crystal structure of the
(T–K) model, ID/IG increases inversely proportional to the size metal carbide, the temperature, reaction times, and posttreat-
of the perfect graphene units; greater disorder becomes more ments (activation, annealing) [57,58,60,85–87]. Along the
active in the D mode [84]. However, the T–K method was chlorination of organometallics, the presence of core–shell-
developed for disordered graphite-like materials with 100% like structures and some traces of metal halides trapped into
sp2-bonded atoms and with unit sizes larger than 2 nm. When the porous carbon networks (at low reaction temperatures)
the number of defects along the carbon network (pentagonal affect the nitrogen adsorption measurements [10,13,15]. For
or heptagonal rings, changes in bond angle, or hybridiza- this reason, the textural parameters for this series of SCNM
tion) increases and La decreases (below 2 nm) the number of are preferably performed on the high-temperature samples.
ordered carbon rings decreases; consequently, the intensity of In contrast, the SCNM produced from the M(C5H5)2Cl2 pre-
the D band and the ratio ID/IG start to decrease and the T–K cursors (M = Ti, Nb, Zr, W) present a complete transforma-
approximation is no longer valid. The three-stage model pro- tion into carbon particles due to low reaction temperatures.
posed by Ferrari and Robertson (F–R) [82] provides a better In addition, an incipient formation of the surface area porous
description of the disorder induced at the carbon lattice, spe- structure has been detected [14,15] as a result of the almost
cifically the stage two, which is a description of the transition complete elimination of the respective halides that were
from the topological disordering of graphene units and loss of formed.
aromatic bonding with purely sp2 network, due to the intro- According to the IUPAC classification [55], the isotherm
duction of up to 20% of sp3 bonds. plots for solid materials can be divided into six categories
Taking into account that the carbon particles described depending on several factors such as the porous structure and
along this work present an sp2 content varying from 89% to the adsorbate/adsorbent interactions. Our results have shown
98%, these values could be overestimated. In this sense, the that the produced SCNM present type I and type IV isotherm
application of the F–R model provides an La average value shapes related to the presence of micropore and mesopore
of 1.25 nm, in agreement with the size of the graphene-like (characterized by the presence of the hysteresis loop) materi-
layers detected in the HRTEM images shown in Figure 20.15. als, respectively [11,13–15].
The nitrogen adsorption–desorption isotherm plots from
20.7 ADSORPTION PROPERTIES AND the samples prepared at 900°C are depicted in Figure 20.16.
SCNM produced from metallocenes (M = Cr, Fe, Ni) present
POROUS STRUCTURE
an almost well-defined type IV isotherm shape (Figure 20.16a),
One of the characteristic features inherent to the nature except for the sample produced from Fe(C5H5)2, which can be
of the carbon materials is their porous structure; most of classified as type I. On the other hand, SCNM produced from
them exhibit an appreciable apparent surface area (SBET) metallocene derivatives (M = Ti, Zr, W) clearly describe a
(a) (b)
700 Cr(C5H5)2 Ti(C5H5)2Cl2
700
Fe(C5H5)2 Zr(C5H5)2Cl2
600 Ni(C5H5)2 600 W(C5H5)2Cl2
Amount adsorbed (cm3/g)
Amount adsorbed (cm3/g)
500 500
400 400
300 300
200 200
100 100
0 0
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Relative pressure (p/p0) Relative pressure (p/p0)
FIGURE 20.16 Nitrogen adsorption–desorption isotherms for the SCNM prepared at 900°C from metallocenes (a) and metallocene
derivatives (b).
type I isotherm shape (Figure 20.16b), except for the spheres point halides; that is, those produced from Ti(C5H5)2Cl2,
produced from Ti(C5H5)2Cl2, which can be assigned to Fe(C5H5)2, and W(C5H5)2Cl2.
type IV. In most of the carbon materials the reaction temperature
All the samples present an H4 hysteresis loop associated and time play an important role in the porous structure.
with the presence of aggregates of particles. However, this The SBET values of the present SCNM are not the exception
loop is more remarkable and wider in the samples produced since their SBET has a significant increase with the chlori-
from metallocenes than from metallocene derivatives. The nation temperature from 400°C to 900°C (Figure 20.17b),
presence of this type of hysteresis loop is also associated with more than twice by using W(C5H5)2Cl2 and up to five times
nonrigid slit-shaped pores. In all the samples, the closure of higher with Ti(C5H5)2Cl2 as a precursor. However, the chlo-
the desorption branch at p/p0 ≈ 0.42 is almost independent rination time seems to provide the opposite effect since
of the morphology of the sample but is associated with the the increase from 30 to 60 and 180 min decreases the SBET
nature of the adsorptive [55]. values.
Additionally, two important features are displayed by The highly disordered graphene-like layers that form
these isotherms. First, the high volume of nitrogen adsorbed these SCNM give rise to a very interesting PSD, in most of
at p/p0 ≤ 0.1 and the almost parallel plateau of the plot the cases below 2 nm (see Figure 20.18a). Once again it is
along a wide range of p/p0 values reveal the development possible to correlate this parameter with the metal halide
of micropores in the samples. Second, in some of the iso- formed during the synthesis. In this sense, the precursors that
therms it is possible to observe a high amount of nitrogen produce the most volatile halides (TiCl4, FeCl3, and WCl6)
adsorbed at the condensation step, at relative pressures p/ yield the wider PSD. An additional feature of these SCNM
p0 > 0.7, indicating the existence of textural porosity [13,15]. is that their diverse porous structures are clearly revealed by
This kind of porosity is ascribed to the interparticle voids the isotherm plots.
among the carbon particles as a result of the morphology The structural pores (detected at p/p0 ≤ 0.6) are respon-
of the samples detected in most of SEM micrographs (see sible for the adsorption and storage properties of carbon
Section 20.4.1) and finally reflected in the shape of the iso- materials while the importance of the textural pores is asso-
therm plots, e specially for those obtained for Cr(C5H5)2 and ciated with facilitating mass transport to the structural pores.
Ni(C5H5)2 and Ti(C5H5)2Cl2. A mixture of both kinds of iso- As observed in Figure 20.18b, the particles produced from
therms could be related to the development of hierarchical Cr(C5H5)2 have developed the highest structural porosity
porosity. having a total porosity of 72%. The spheres produced from
These series of SCNM present variable SBET values rang- Ti(C5H5)2Cl2 present the highest textural values; here this
ing from 572 to 1320 m 2/g and microporous surface (Smi) parameter represents approximately a 52%. However, the
areas as high as 1287 m 2/g. A comparative scheme of the SCNM obtained from Fe(C5H5)2 and W(C5H5)2Cl2 contain
SBET, Smi, and external surface (Sext) area values (Figure the lowest textural porosity (8% and 11%, respectively) devel-
20.17a) shows that the SCNM develop low Sext especially oping in higher proportion the structural pores. In addition,
the samples prepared from Ni(C5H5)2 and W(C5H5)2Cl2. It the W0 of the SCNM produced from W(C5H5)2Cl2 presents
is possible to correlate SBET values with the melting point almost the same value of their structural porosity showing
of the respective metallic halide formed during the synthe- their microporous contribution, as reflected in their corre-
sis since the highest SBET corresponds to the lowest melting sponding isotherm plots.
(a) (b)
1400 1400
SBET Smi Sext Fe(C5H5)2
1200 1200 Ni(C5H5)2
Ti(C5H5)2Cl2
1000
Surface area (m2/g)
1000 Zr(C5H5)2Cl2
W(C5H5)2Cl2
SBET (m2/g)
800
800
600
600
400
400
200
200 30 min 1 h 3h
0
Ti Cr Fe Ni Zr W 400 500 600 700 800 900 900 900
Metal precursor Temperature (°C)
FIGURE 20.17 Schemes showing variations in the SBET, Smi, and Sext values related to the metal precursor (a) and the SBET as a function
of chlorination temperature and time (b).
(a) (b)
0.06 1.0
0.04 Ti(C5H5)2Cl2 Textural Structural W0
0.02
Incremental pore volume (cm3/g)
Normalized pore volume (cm3/g)

0.00
0.08 0.8
Cr(C5H5)2
0.04
0.00
0.04 0.6
0.02 Fe(C5H5)2
0.00
0.12 0.4
0.08 Ni(C5H5)2
0.04
0.00
0.12 0.2
0.08 W(C5H5)2Cl2
0.04
0.00 0.0
0 1 2 3 4 5 Ti Cr Fe Ni W
Pore width (nm) Metal precursor
FIGURE 20.18 PSD plots from some SCNM obtained at 900°C (a) and a comparison of their corresponding porous structure (b).
20.8 SUMMARY AND CONCLUSIONS 2. Kroto, H.W., Heath J.R., O´Brien, S.C., Curl, R.F., Smalley,
R.E. 1985. C60: Buckminsterfullerene. Nature 318, 162–63.
Chlorination of organometallics is a novel technique to syn- 3. Iijima, S. 1991. Helical microtubules of graphitic carbon.
thesize several SCNM (encapsulates, tubes, bags, ribbons, Nature 354, 56–58.
chains, filled solids, and hollow spheres) in a wide range of 4. Novoselov, K.S., Geim, A.K., Morozov, S.V., Jiang, D., Zhang,
temperatures in a relative short time. Y., Dubonos, S.V. et al. 2004. Electric field effect in atomically
thin carbon films. Science 306, 666–669.
TEM and associated techniques played an important role
5. Geim, A.K., Novoselov, K.S. 2007. The raise of graphene.
in the observation of evolution in the microstructure of these Nat. Mater. 6, 186–191.
materials, mainly formed by highly disordered graphene-like 6. Nikitin, A., Gogotsi, Y. 2004. Nanostructured carbide-derived
layers. The HRTEM images allowed a direct examination carbon. In: Encyclopedia of Nanoscience and Nanotechnology,
of the order/disorder degree of these layers as a result of the Ed. H.S. Nalwa, pp. 553–74, American Scientific Publishers, CA.
synthesis parameters, the carbon precursor, and the formed 7. Yushin, G., Nikitin, A., Gogotsi, Y. 2006. Carbide-derived
halide, as confirmed by the EELS and Raman measurements. carbon. In: Nanomaterials Handbook, Ed. Y. Gogotsi, pp.
In this sense, these spectroscopy methods provided an evi- 24–273, CRC Press, Boca Raton, FL.
8. Presser, V., Heon, M., Gogotsi, Y. 2011. Carbide-derived car-
dence of the planarity loss in the graphene-like layers, con- bons—From porous networks to nanotubes and graphene.
firming the presence of carbon atoms with spn hybridization Adv. Funct. Mater. 21: 1–24.
forming heptagonal and pentagonal rings. Besides the struc- 9. Nyamori, V., Mhlanga, S., Coville, N. 2008. The use of
tural features, the textural parameters gave information about organometallic transition metal complexes in the synthesis
the high surface area and porous volume developed in these of shaped carbon nanomaterials. J. Organomet. Chem. 693:
SCNM suggesting their potential applications for energy- 2205–2222.
related materials. 10. Urones-Garrote, E., Ávila-Brande, D., Ayape-Katcho, N.,
Gómez-Herrero, A., Landa-Cánovas, A., Otero-Díaz, L.C.
Finally, we have proposed a possible chemical mechanism
2005. Amorphous carbon nanostructures from chlorination of
for the formation of these SCNM taking into account the ferrocene. Carbon 43: 978–985.
template effect induced by the trapped halides. However, this 11. Ayape-Katcho, N., Urones-Garrote, E., Ávila-Brande, D.,
and other thermodynamic facts are not totally understood and Gómez-Herrero, A., Urbonaite, S., Csillag, S. et al. 2007.
need to be solved. Carbon hollow nanospheres from chlorination of ferrocene.
Chem. Mater. 19: 2304–2309.
12. Urones-Garrote, E., Ávila-Brande, D., Ayape-Katcho, N.,
ACKNOWLEDGMENTS Gómez-Herrero, A., Landa-Cánovas, A., Otero-Díaz, L.C.
2007. Spherical carbon nanoparticles produced by direct chlo-
The authors thank the financial support through the projects
rination of cobaltocene. Carbon 45: 1696–1701.
with references S-2009/PPQ-1626, S2013/MIT-2753, MAT2010- 13. González-García, P., Urones-Garrote, E., Del Corro E., Ávila-
19460, and MAT2013-44964-R. P. González-Garcáa acknowl- Brande D., Otero-Díaz, L.C. 2013. The production of carbon
edges CONACYT (Mexico) for a postdoctoral grant. particles of different shapes produced by the chlorination of
Cr(C5H5)2. Carbon 52: 90–99.
14. González-García, P., Urones-Garrote, E., Ávila-Brande, D.,
REFERENCES
Gómez-Herrero, A., Otero-Díaz, L.C. 2010. Structural study
1. Rao, C.N.R., Cheethman, A.K. 2006. Materials science at the of carbon nanomaterials prepared by chlorination of tungsten
nanoscale. In: Nanomaterials Handbook, Ed. Y. Gogotsi, pp. carbide and bis(ciclopentadienyl)tungsten dichloride. Carbon
1–5, CRC Press, Boca Raton, FL. 48: 3667–3675.
15. González-García, P., Centeno, T.A., Urones-Garrote, E., 34. Singh, C., Quested, T., Boothroyd, C.B., Thomas, P., Kinloch,
Ávila-Brande, D., Gómez-Herrero, A., Otero-Díaz, L.C. 2011. I.A., Abou-Kandil, A.I., Windle, A.H. 2002. Synthesis and
Porous carbon nanospheres derived from chlorination of characterization of carbon nanofibers produced by the float-
bis(cyclopentadienyl)titanium dichloride and their electrochemi- ing catalyst method. J. Phys. Chem. B 106: 10915–10922.
cal capacitor performance. Mater. Chem. Phys. 130: 243–250. 35. Zhu, H., Wei, B. 2007. Direct fabrication of single-walled
16. Komatsu, Y., Endo, M. 1985. Japanese Patent 60-32818. carbon nanotube macro-films on flexible substrates. Chem.
17. Tibbetts, G.G., Gorkiewicz, D.W., Alig, R.L. 1993. A new Commun. 29: 3042–3044.
reactor for growing carbon fibers from liquid- and vapor- 36. Liu, X.-Y., Huang, B.-C., Coville, N.J. 2002. The Fe(CO)5
phase hydrocarbons. Carbon 31: 809–814. catalyzed pyrolysis of pentane: Carbon nanotube and carbon
18. Rao, C.N.R., Sen, R., Satishkumar, B.C., Govindaraj, A. 1998. nanoball formation. Carbon 40: 2791–2799.
Large aligned-nanotube bundles from ferrocene pyrolysis. 37. Harutyunyan, A.R., Chen, G., Eklund, P.C. 2003. Self-
Chem. Commun. 16: 1526–1526. assembled growth of single-walled carbon nanotubes by pyroly-
19. Andrews, R., Jacqes, D., Rao, A.M., Deryshire, F., Qian, D., sis of metalorganic precursor. Appl. Phys. Lett. 82: 4794–4796.
Fan, X. et al. 1999. Continuous production of aligned carbon 38. Lee, Y.T., Kim, N.S., Park, J., Han, J.B., Choi, Y.S., Ryu, H.,
nanotubes: A step closer to commercial realization. Chem. Lee, H.J. 2003. Temperature-dependent growth of carbon
Phys. Lett. 303: 467–474. nanotubes by pyrolysis of ferrocene and acetylene in the range
20. Kamalakaran, R., Terrones, M., Seeger, T., Kohler-Relich, P., between 700 and 1000°C. Chem. Phys. Lett. 372: 853–859.
Ruhle, M., Kim, Y.A. et al. 2000. Synthesis of thick and crys- 39. Awasthi, K., Singh, A.K., Srivastava, O.N. 2003. Formation
talline nanotube arrays by spray pyrolysis. Appl. Phys. Lett. of carbon nanotubes without iron inclusion and their align-
77: 3385–3387. ment through ferrocene and ferrocene–ethylene pyrolysis.
21. Mayne, M., Grobert, N., Terrones, M., Kamalakaran, R., J. Nanosci. Nanotechnol. 3: 540–454.
Ruhle, M., Kroto, H.W. et al. 2001. Pyrolytic production of 40. Kuo, C.-S., Bai, A., Huang, C.-M., Li, Y.-Y., Hu, C.-C., Chen,
aligned carbon nanotubes from homogeneously dispersed C.-C. 2005. Diameter control of multiwalled carbon nano-
benzene-based aerosols. Chem. Phys. Lett. 338: 101–107. tubes using experimental strategies. Carbon 43: 2760–2768.
22. Singh, C., Shaffer, M., Kinloch, I.A., Windle, A. 2002. 41. Luo, T., Chen, L., Bao, K., Yu, W., Qian, Y. 2006. Solvothermal
Production of aligned carbon nanotubes by the CVD injection preparation of amorphous carbon nanotubes and Fe/C coaxial
method. Physica B 323: 339–340. nanocables from sulfur, ferrocene, and benzene. Carbon 44:
23. Wei, B., Vajtai, R., Young, Y., Ajayan, P.M. 2002. Structural 2844–2848.
characterizations of long single-walled carbon nanotubes 42. Shaikjee A., Coville N.J. 2012. The role of the hydrocar-
strands. Nanoletters 2: 1105–1107. bon source on the growth of carbon materials. Carbon 50:
24. Satishkumar, B.C., Govindaraj, A., Sen, R., Rao, C.N.R. 1998. 3376–3398.
Single-walled nanotubes by the pyrolysis of acetylene-organo- 43. Seah, C.-M., Chai, S.-P., Mohamed, A.R. 2011. Synthesis of
metallic mixtures. Chem. Phys. Lett. 298: 47–52. aligned carbon nanotubes. Carbon 49: 4613–4635.
25. Nikolaev, P., Bronikowski, M.J., Bradley, R.K., Rohmund, F., 44. Marangoni, R., Serp, P., Feurer, R., Kihn, Y., Kalck, P.,
Colbert, D. T., Smith, K. A., Smalley, R. E. 1999. Gas-phase Vahlas, C. 2001. Carbon nanotubes produced by substrate free
catalytic growth of single-walled carbon nanotubes from car- metalorganic chemical vapor deposition of iron catalysts and
bon monoxide. Chem. Phys. Lett. 313: 91–97. ethylene. Carbon 39: 443–449.
26. Vivekchand, S.R.C., Cele, L.M., Deepak, F.L., Raju, A.R., 45. Rao, C.N.R., Govindaraj, A. 2002. Carbon nanotubes from
Govindaraj, A. 2004. Carbon nanotubes by nebulized spray organometallic precursors. Acc. Chem. Res. 35: 998–1007.
pyrolysis. Chem. Phys. Lett. 386: 313–318. 46. Murayama, H., Maeda, T. 1990. A novel form of filamentous
27. Wei, J.Q., Jiang, B., Wu, D.H., Wei, B.Q. 2004. Large-scale graphite. Nature 345: 791–793.
synthesis of long double-walled carbon nanotubes. J. Phys. 47. Hoseini, S.J., Barzegar, Z., Bahrami, M., Roushani, M.,
Chem. B 108: 8844–8847. Rashidi, M. 2014. Organometallic precursor route for the fab-
28. Nasibulin, A.G., Moisala, A., Brown, D.P, Jiang, H., rication of PtSn bimetallic nanotubes and Pt3Sn/reduced-gra-
Kauppinen, E.I. 2005. A novel aerosol method for single phene oxide nanohybrid thin films at oil-water interface and
walled carbon nanotube synthesis. Chem. Phys. Lett. 402: study of their electrocatalytic activity in methanol oxidation.
227–232. J. Organomet. Chem. 769: 1–6.
29. Wei, J., Ci, L., Jiang, B., Li, Y., Zhang, X., Zhu, H. et al. 2003. 48. Marquardt, D., Beckert, F., Pennetrau, F., Tolle, F., Mulhaupt,
Preparation of highly pure double-walled carbon nanotubes. J. R., Raint, O. et al. 2014. Hybrid materials of platinum nanopar-
Mater. Chem. 13: 1340–1344. ticles and thiol-functionalized graphene derivatives. Carbon.
30. Singh, C., Shaffer, M., Koziol, K.K.K., Kinloch, I.A., Windle, 66: 285–294.
A. 2003. Towards the production of large-scale aligned carbon 49. Urbonaite S. 2008. Synthesis and characterization of carbide
nanotubes. Phys. Chem. Lett. 372: 860–865. derived carbons. PhD thesis, Stockholm University, Sweden.
31. Xu, L., Zhang, W., Yang, Q., Ding, Y. Yu, W., Quian, Y. 2005. 50. Hutchins, O. 1918. Method for the production of silicon tetra-
A novel route to hollow and solid carbon spheres. Carbon 43: chloride. US Patent 1271713.
1090–1092. 51. Kirillova, G.F., Meerson, G.A., Zelikman, A.N. 1960. Kinetics
32. Schnitzler, M.C., Oliveira, M.M., Ugarte, D., Zarbin, A.J.G. of chlorination of titanium and niobium carbides. Izv. Vuzov.
2003. One-step route to iron oxide-filled carbon nanotubes Tsvetn. Metall. 3: 90–96.
and bucky-onions based on the pyrolysis of organometallic 52. Orekhov, V.P., Seryakov, G.V., Zelikman, A.N., Starobina,
precursors. Chem. Phys. Lett. 381: 541–548. T.M., Kahzanova, T.I., Petrova, K.V., Sverchkov, P.M. 1969.
33. Panchakarla, L.S., Govindaraj, A. 2007. Carbon nanostruc- Kinetics of chlorination of zirconium carbide briquets. J.
tures and graphite-coated metal nanostructures obtained by Appl. Chem. USSR 42: 230–237.
pyrolysis of ruthenocene and ruthenocene-ferrocene mix- 53. Mohun, W.A. 1959. A novel amorphous carbon. In: Proc. 4th
tures. Bull. Mater. Sci. 30: 23–29. Biennial Conf. on Carbon, pp. 443–453, Pergamon, Oxford.
54. Boehm, H.P., Warnecke, H.H. 1975. Structural parameters 70. Greenwood, N.N., Earnshaw, A. 1984. Chemistry of the
and molecular sieve properties of carbons prepared from Elements. Pergamon Press, UK, 1st Edition.
metal carbides. In: Proc. 12th Biennial Conf. on Carbon, pp. 71. Flower, K.R., Hitchock, P.B. 1996. Crystal and molecular
149–150, Pergamon, Oxford. structure of chromocene (η6-C5H5)2Cr. J. Organomet Chem.
55. Sing, K.S.W., Everett, D.H., Haul, R.A.W., Moscou, L., Pierotti, 507: 275–277.
R.A., Rouquérol, J. et al. 1985. Reporting physisorption data 72. Egerton R.F. 1996. Electron Energy-Loss Spectroscopy in the
for gas/solid systems with special reference to the determina- Electron Microscope. Plenum Press, New York.
tion of surface area and porosity. Pure Appl. Chem. 57: 603–18. 73. Williams, D.B., Carter, C.B. 2009. Transmission Electron
56. Fedorov, N.F., Ivakhnyuk, G.K., Gavrilov, D.N. 1982. Carbon Microscopy: A Textbook for Materials Science. Plenum
adsorbents from carbides of the IV–VI groups transition met- Publishing Company, New York, 2nd Edition.
als. Zh. Prikl. Khim. 55, 272–275. 74. Keast, V.J., Scott, A.J., Brydson, R., Williams, D.B., Bruley,
57. Yushin, G.N., Hoffman, E.N., Nikitin, A., Ye, H., Barsoum, J. 2001. Electron energy-loss near-edge structure—A tool for
M.W., Gogotsi, Y. 2005. Synthesis of nonporous carbide- the investigation of the electronic structure on the nanometre
derived carbon by chlorination of titanium silicon carbide. scale. J. Microsc. 203: 135–175.
Carbon 43: 2075–2082. 75. Inaganki M. 2010. Structure and texture of carbon materi-
58. Dash, R.K., Yushin, G., Gogotsi, Y. 2005. Synthesis, structure als. In: Carbons for Electrochemical Energy Storage and
and porosity analysis of microporous and mesoporous carbon Conversion Systems, Eds. Béguin F., and Frackowiak E., pp.
derived from zirconium carbide. Microporous Mesoporous 38–71, CRC Press, Boca Raton FL.
Mater. 86: 50–57. 76. Yuan J., Brown L.M. 2000. Investigation of atomic structures
59. Cambaz, Z.G., Yushin, G.N., Vyshnyakova, K.L., of diamond-like amorphous carbon by electron energy loss
Pereselentseva, L.N., Gogotsi, Y. 2006. Formation of carbide- spectroscopy. Micron 31: 515–525.
derived carbon on β-silicon carbide whiskers. J. Am. Ceram. 77. Urbonaite, S., Wachtmeister, S., Mirguet, C., Coronel, E., Zou,
Soc. 89: 509–514. W.Y., Csillag, S. et al. 2007. EELS studies of carbide derived
60. Yeon, S.-H, Osswald, S., Gogotsi, Y., Singer, J.P., Simmons, carbons. Carbon 45, 2047–2053.
J.M, Fischer, J.E. et al. 2009. Enhanced methane storage of 78. Bernier, N., Bocquet, F., Allouche, A., Saikaly, W., Brosset,
chemically and physically activated carbide-derived carbon. C., Thibault, J. et al. 2008. A methodology to optimize the
J. Power Sources 191: 560–567. quantification of sp2 carbon fraction from K edge EELS spec-
61. Gogotsi, Y., Weltz, S., Ersoy, D.A., McNallan, M.J. 2001. tra. J. Electron Spectrosc. Relat. Phenom. 164: 34–43.
Conversion of silicon carbide to crystalline diamond-struc- 79. Thomas, J.M. 2009. The renaissance and promise of electron
tured carbon at ambient pressure. Nature 411: 283–287. energy-loss spectroscopy. Angew. Chem. 48: 8824–8826.
62. Perkson, A., Leis, J., Arulepp, M., Kaarik, M., Urbonaite, S., 80. Ferrari, A.C., Libassi, A., Tanner, B.K., Stolojan, V., Yuan,
Svensson, G. 2003. Barrel-like carbon nanoparticles from car- J., Brown, L.M. et al. 2000. Density, sp3 fraction, and cross-
bide by catalyst assisted chlorination. Carbon 41: 1729–1735. sectional structure of amorphous carbon films determined by
63. Welz S., McNallan M.J., Gogotsi Y. 2006. Carbon structures x-ray reflectivity and electron energy-loss spectroscopy. Phys.
in silicon carbide derived carbon. J. Mater. Process. Technol. Rev. B. 62: 11089–11103.
179: 11–22. 81. Reich, S., Thomsen, C. 2004. Raman spectroscopy of graph-
64. Lide, D.R. 2005. CRC Handbook of Chemistry and Physics. ite. Philos. Trans. R. Soc. London A 362: 2271–2288.
CRC Press, Boca Raton FL, 86th Edition. 82. Ferrari, A.C., Robertson, J. 2000. Interpretation of Raman
65. Ávila-Brande, D., Urones-Garrote, E., Ayape-Katcho, N., spectra of disordered and amorphous carbon. Phys. Rev. B 61:
Lomba, E., Gómez-Herrero, A., Landa-Cánovas, A.R. et al. 14095–15007.
2007. Electron microscopy characterisation of nanostructur- 83. Larouche, N., Stansfield, B.L. (2010). Classifying nanostruc-
ated carbon obtained from chlorination of metallocenes and turated carbons using graphitic indices derived from Raman
metal carbides. Micron 38: 335–345. spectra. Carbon 48: 620–629.
66. Povarennykh, A.S. 1972. Crystal Chemistry Classification of 84. Tuinstra, F., Koenig, J.L. 1970. Raman spectrum of graphite.
Minerals. Plenum Press, New York. J. Chem. Phys. 53: 1126–1130.
67. Urones-Garrote, E., Ávila-Brande, D., Katcho, N.A., Gómez- 85. Gogotsi, Y., Nikitin, A., Ye, H., Zhou, W., Fischer, J.E., Yi, B.
Herrero, A., Landa-Cánovas, A.R., Otero-Díaz, L.C. 2006. et al. 2003. Nanoporous carbide-derived carbon with tunable
TEM study of amorphous carbon nanotubes produced by direct porous size. Nat. Mater. 2: 591–594.
chlorination of ferrocene. Microsc. Microanal. 12: 668–669. 86. Urbonaite, S., Juárez-Galán, J.M., Leis, J., Rodríguez-Reinoso,
68. Otero-Díaz, L., González García, P., Ávila Brande, D., Urones F., Svensson, G. 2008. Porosity development along the synthe-
Garrote, E. 2012. α-Carbon nanotubes and nanospheres pro- sis of carbons from metal carbides. Microporous Mesoporous
duced by chlorination of Cr(C5H5)2. Microsc. Microanal. 18: Mater. 113: 14–21.
1568–1569. 87. Pérez, C.R., Yeon, S.-H., Ségalini, J., Presser, V., Taberna, P.L.
69. Wilkinson, G., Birmingham, J.M. 1954. Bis-cyclopentadienyl Simon, P. et al. 2013. Structure and electrochemical perfor-
compounds of Ti, Zr, V, Nb and Ta. J. Am. Chem. Soc. 76, mance of carbide-derived carbon nanopowders. Adv. Funct.
4281–4284. Mater. 23: 1081–1089.
21 Chemical Modification of
Graphene with Polymers
Somayeh Mohamadi and Naser Sharifi-Sanjani
CONTENTS
Abstract...................................................................................................................................................................................... 405
21.1 Introduction...................................................................................................................................................................... 405
21.2 Chemistry of Graphene Surface....................................................................................................................................... 406
21.3 Chemical Modification of Graphene Surface................................................................................................................... 409
21.4 Covalent Modification of Graphene..................................................................................................................................410
21.5 Morphological Changes.................................................................................................................................................... 429
21.6 How Do the Electrical Properties of Graphene Change with Its Functionalization?....................................................... 429
21.7 Summary Section............................................................................................................................................................. 430
References...................................................................................................................................................................................431
ABSTRACT causes to form a large variety of compounds and structures

with a wide range of properties [1,2]. Carbon is known as
The discovery of graphene has opened up opportunities for the basis of life but other different forms of carbonic struc-
scientists and engineers to design new materials with special tures such as graphite, diamond, fullerenes, and carbon
properties for different applications so that it can be consid- nanotubes have also received a lot of attention, recently and
ered as a “magic discovery” of the center. It is believed that as expected, its chemistry is one of the most studied sub-
most of applications still have not been defined due to the lack ject. The term “graphene” first appeared in 1987 to describe
of complete understanding of the material properties of this single sheets of graphite and was presumed not to exist in
honeycomb two-dimensional structure. Following this discus- free state, being described as an “academic material” but
sion, identifying the chemistry of the graphene surface can be the experimental discovery of graphene is returned to Geim
considered as a chief issue in the field of graphene investiga- and Novoselov report in 2004 [3–5]. With the finding of gra-
tion. Actually, interactions of graphene with the surrounded phene, the interest of researchers shifted from other carbon
media, which are governed by the surface atoms, play an materials toward graphene and many chemists tried to play
important role in the reinforcing effect of this nanoparticle with the surface of graphene through a wide array of modi-
in different matrixes. As well, graphene sheets with a suf- fication methods.
ficiently large aspect ratio and high surface energy tend to To define, graphene is a two-dimensional (2D) single layer
restack and crumple which is totally undesirable due to the of carbon atoms with a hexagonal packed structure. The car-
dramatic decrease in surface area. Manipulation of graphene bon bonds are sp2 hybridized, where the in-plane σC–C bond is
surface properties with chemical modification is the way one of the strongest bonds in the nature and the out-of-plane π
which has been considered nowadays in different laboratories. bond, which contributes to a delocalized network of electrons,
Surface modification of graphene in order to obtain a stable is responsible for the electron conduction of graphene and
dispersion of graphene sheets would be virtually important provides the weak interaction among graphene layers or
since wrinkling and even globular structures, “lumps” depend between graphene and the substrate. It was shown that the
on the interactions and the surface energy reduction. Until electronic structure rapidly evolves with the number of lay-
now, the wide range of methods and also different molecules ers, single and, to a good approximation, bilayer graphene
including small and macromolecules have been examined. In have similar electronic spectra with one type of electrons
this chapter, an attempt is made to review the recent progress and one type of holes so are zero-gap semiconductors. For
in the applied methods for chemical modification of graphene three and more layers (>10) the conduction and valence bands
and its effect on the topological and electrical properties of start notably overlapping and the spectra become complicated
graphene. [6,7]. It is believed, graphene sheets with a thickness of more
than 10 layers should be considered as a thin film of graphite
with a 3D (three-dimensional) structure [8]. However, from
21.1 INTRODUCTION
an experimental point of view, at the moment large-scale pro-
Carbon is an attractive element due to its ability to form all duction of single- or few-layer graphene is not easy and so
sp, sp2, and sp3 hybridization of electronic orbitals which expensive.
405
As discussed above, it was believed 2D crystals were

thermodynamically unstable and could not exist. In fact, TABLE 21.1
the melting temperature of thin films rapidly diminishes Graphene Preparation Methods from Crystalline
with decreasing thickness, and they become unstable (segre- Natural Graphite
gate into islands or decompose) at a thickness of, typically,
Methods Descriptions
atomic layers [9,10]. However, in the perfecting viewpoint,
Colloidal suspension [12] Using proper solvent with graphite or
the extracted 2D crystals become intrinsically stable by mild
graphite oxide
crumpling and surface ripples. In other words, graphene exists Arc-discharge method [15] Using a high-current arc discharge between
in the 3D space and hence the atoms can oscillate out of plane a graphite anode and graphite cathode
and causes flexibility in this 2D crystal. Such 3D warping and Electrochemical exfoliation Using two different types of electrolytes
topological defect, lead to a gain in elastic energy but suppress [16–18] including ionic liquids and aqueous acids
thermal vibrations which can minimize the total free energy Detonation method [19] Using a mixture of natural graphite, nitric
[10]. The electrical property of graphene is also affected by acid, and CH3NO2
this intrinsic source of disorders. Thermal exfoliation [20] Oxidization, thermal expansion/exfoliation,
Nevertheless, the presence of a substrate or scaffolds that and ultrasonication.
hold graphene in place can stabilize certain degree of order in Chemical exfoliation [21] Reduction of a colloidal suspension of
graphene [11]. Hydrocarbon molecules that are often present exfoliated graphene oxide sheets in water
on top of free hanging graphene membranes could quench flex- with hydrazine hydrate
ural fluctuations, making them static [11]. In the field of nano- The scotch tape method [12] Mechanical cleavage with repeated peeling
Micromechanical cleaving Use of an atomic force microscope (AFM)
composites and graphene dispersion in the polymeric matrix in
method [22] tip
which graphene with sufficiently large aspect ratio can be con-
The sectioning method [23] Use of an ultrasharp wedge such as glass
sidered as a good candidate, flexural fluctuation and restacking
and diamond wedges
are also crucial. However, the degree of interactions between
graphene sheets and the substrate or polymeric matrix plays an
important factor in the topological structure, dispersion, and to be effective to stabilize graphene solution without
restacking of graphene. In this regard, the issue of manipula- reaggregation [13,14].
tion of graphene surface properties through chemistry can be • The arc-discharge method involves the use of a high-
an imperative subject in the field of graphene chemistry. current arc discharge between a graphite anode and
A precise understanding of the chemical properties and the graphite cathode in a chamber filled with hydrogen
functional groups that exist on the graphene surface and edges and helium gas [15].
is essential and the covalent functionalization of the surface, • Using the advantage of electrochemistry is another
the correlation between the surface chemistry and 3D geo- approach to exfoliate graphite by adjusting the driv-
metric shape, and the electrical and dispersion properties of ing potential in suitable electrolytes, ionic liquids,
graphene should also be discussed. and aqueous acids (e.g., H2SO4 or H3PO4) [16–19].
Graphite electrodes can be either negatively or posi-
21.2 CHEMISTRY OF GRAPHENE SURFACE tively intercalated to give graphene intercalation
compounds (GICs) and then exfoliated by solvent
Before talking about the surface chemistry of graphene, the decomposition directly during electrochemical
synthesis of graphene should be considered since different treatment or by further thermal treatment.
preparation methods lead to graphene sheets with different • In the detonation method a mixture of natural graph-
chemical properties and thickness. In this discussion, the ite, nitric acid, and CH3NO2 is exploded in a vessel
preparation methods were limited to that which crystalline and graphene detected in the soot obtained [20].
natural graphite has been used as a source material involving • Thermal exfoliation is a process which comprises
laboratory methods and large-scale manufacturing. three steps: oxidization, thermal expansion/exfolia-
As it is summarized in Table 21.1 the methods can be tion, and finally ultrasonication [21].
broadly classified as colloidal suspension, arc discharge, • In chemical exfoliation, high temperatures including
detonation technique, and thermal, chemical, or mechanical several process steps and chemicals are used [12,22].
exfoliation. In all these methods, graphene is contaminated with
impurities such as amorphous carbon and produced
• In the colloidal suspension method, a combination of oxygen/nitrogen functional groups.
aqueous or organic solvent with an initial raw mate-
rial such as graphite oxide is used [12]. Actually, The different methods which can be listed as mechanical
ultrasonic treatment in organic solvent is one of the exfoliation are as follows:
approaches to exfoliate graphite flakes into graphene
sheets. Some dipolar aprotic solvents (e.g., N,N′- • The scotch tape method is the popular method
dimethylformamide (DMF) or N-methylpyrrolidone of mechanical cleavage in which repeated peel-
(NMP)) or surfactant in aqueous solutions revealed ing is needed to achieve single-layer graphene but
Chemical Modification of Graphene with Polymers 407
prediction of the number of required peelings is armchair face [27]. XPS (x-ray photoelectron spectroscopy)
difficult [7]. analysis of the pristine graphite flakes in S. Park et al. report
• Micromechanical cleaving method reported by showed the oxygen concentration is only a small propor-
Lu et al. [23,24] involves the use of an atomic force tion of 0.87%. Consequently, it can be concluded mechani-
microscope (AFM) tip along with an array of highly cally exfoliated graphene nanosheets (see Figure 21.2) except
ordered pyrolytic graphite (HOPG) mesas made of just a small amount of oxygen functional groups, consist
from oxygen-plasma etching method. The HOPG of pure carbon atoms [28]. However, this amount should be
islands were transferred to an SiO2/Si substrate varied with the crystal size and defect percentage of pristine
using hydrofluoric acid. It is then manipulated using graphite.
an AFM tip to obtain multiple layers of HOPG. A In the chemical oxidation procedure, in general, gra-
variation of this method involves gluing a block of phene oxide (GO) is synthesized by either the Brodie [29],
prepared graphite to an AFM tip and scratching on Staudenmaier [30], or Hummers’ method [31], or some varia-
Si substrates. tion of these methods. All three methods involve oxidation
• The use of an ultrasharp wedge as a sectioning of graphite to various levels. Brodie and Staudenmaier used
method has been utilized in an ultramicrotome a combination of potassium chlorate (KClO3) with nitric
device for ultrathin sample preparation (as thin as acid (HNO3) to oxidize graphite, and the Hummers method
40 nm) is generated with either glass or diamond involves treatment of graphite with potassium permanganate
wedges [25]. Recently, Jayasena and Subbiah [26] (KMnO4) and sulfuric acid (H2SO4). Graphite salts made by
used an ultrasharp single crystal diamond wedge to intercalating graphite with strong acids such as H2SO4, HNO3,
exfoliate an HOPG sample. or HClO4 have also been used as precursors for the succes-
sive oxidation to GO [32]. The oxidation of graphite to GO
In mechanical exfoliation methods large-scale production breaks up the sp2-hybridized structure of stacked graphene
is not possible but they are the most compatible with elec- sheets, generating defects that manifest as clear wrinkles in
tronic device fabrication. Since no chemical reaction is used the stack and increase the distance between adjacent sheets
in this method, it is expected that the functional groups that from 3.35 Å in graphite powder to ≈6.8 Å for GO powder
exist on the graphene surface would be as same as pristine
graphite. The term “graphite” by itself describes an ideal
material with a perfect graphite structure including no defects
and functional groups whatsoever as illustrated in Figure 21.1,
schematically.
Graphite is known for its excellent chemical resistance,
but one major exception is poor resistance to the elements
of VI group (oxygen, sulfur, selenium, and tellurium), par-
5.708Å
ticularly oxygen and oxygen compounds. Oxidation begins
in air at 350–400°C but the low-temperature oxidation and b
the reaction rate is site preferential. Reaction with gases or 3.354Å
vapors occurs preferentially at “active sites” or defect sites,
such as dislocations, vacancies, and steps. Moreover the rate
of reaction is slow at the basal plane and rapid at the edge due a
to large differences in surface energy. This difference should 1.421Å
be accounted for the end of the basal planes of the crystal
which are the zigzag and the armchair face so that along the FIGURE 21.1 Perfect graphite structure including no defects and
zigzag face the rate of reaction is much higher than along the functional groups.
(a) OH
OH
O
O
(b) O OH
O OH OH O
O OH
O
HO
HO O OH O
O O
HO
FIGURE 21.2 Schematic model of a (a) graphene and (b) GO sheet.

[33]. During an oxidation reaction depends on the oxidation defective areas only have a limited influence on the conduc-
intensity basal plane of the sheet which is decorated with dif- tivity of an rGO sheet [37]. Owing to the low binding energy,
ferent oxygen functional groups. NMR (nuclear magnetic res- a single hydroxyl group attached to the interior aromatic
onance) study of 13C-labelled GO indicated that hydroxyl and domain is not stable; it is expected the mentioned groups dis-
epoxy (1,2-ether) groups are introduced on the basal plane of sociate or migrate to the edges of aromatic domains at low
the sheet. Carbonyl groups can present as carboxylic acids temperatures [37]. According to Gao’s calculations, the criti-
along the sheet edge and organic carbonyl defects within the cal dissociation temperature (Tc) of hydroxyl groups attached
sheet [34,35]. These functional groups provide reactive han- to the edges of GO is 650°C, and only above this temperature
dles for a variety of surface-modification reactions, which can can hydroxyl groups be fully removed [38,39]. For dehydrox-
be used to develop functionalized GO- and graphene-based ylation, a hydroxyl group is believed to more favorably leave
materials. the graphene sheet directly, producing an OH radical and a
From a chemical point of view, production of GO which graphene radical which does not result in the formation of a
possesses many reactive oxygen-containing groups is suit- lattice defect within the plane. On the basis of this calcula-
able for further functionalization and tuning properties of tion, after thermal annealing at temperatures of 700–1200°C
GO sheets; in other words, the reactivity of GO is significant. in vacuum, the hydroxyl groups can be fully eliminated,
However, during oxidation, specially at the highest oxidation while the epoxy groups are retained. As the experimental
levels graphene’s π-system is interrupted and the aqueous dis- work has discovered, a large number of functional groups can
persion changes to brown color [36]. In this state, its elec- be removed by moderate heating above 200°C with enough
trical conductivity drops off dramatically. It is also often time, but the full deoxygenation of GO solely by thermal
essential that the GO can be transformed back into a conduc- annealing is rather difficult even at temperatures as high as
tive graphitic material by rapid heating (thermal), microwave 1200°C [37].
irradiation (MWI), photoirradiation, and by photocatalyst, Similar to thermal reduction, chemical reduction can also
electrochemical, and chemical reagent reduction [37]. Though not fully remove the functional groups in GO. The reduction
numerous strategies have been proposed to reduce GO, there mechanism by hydrazine was first proposed by Stankovich
is still this question that can the functional groups of a GO et al. This reduction process starts from the ring opening of
sheet be fully eliminated? For the elimination of functional epoxy groups and form hydroxyl groups on the original sites.
groups, there are also two effects that should be considered: Their results show that the hydrazine reduction can only result
whether the oxygen-containing groups can be removed and in reducing epoxy groups, while no reaction path was found
whether the areas after removal can be restored to a long- for the reduction of the hydroxyl, carbonyl, and carboxyl
range conjugated structure, so that there are pathways for groups of GO. As talked before according to DFT simula-
carrier transport within the reduced graphene oxide (rGO) tion, the formed hydroxyl groups attached within an aromatic
sheet. domain are not stable even at moderate temperatures, and can
For conductivity restoring, it can be said, the reduction of be removed or migrate to the edges of aromatic domains and
GO must be mainly aimed at eliminating epoxy and hydroxyl restore the conjugated structure after dehydroxylation [22]. In
groups on the plane, while other groups, for example, any way depending on the field in which graphene is consid-
carboxyl, carbonyl, and ester groups, present at the edges or ered to used the additional process should be included.
(a) (b) 4
1.4 Mica
1.2
6 3
1.0 2
4
Graphene Mica
2 1
0.8
µm
µm
2
Deg
0
Deg
0
0.6
–2
–1
0.4 –4
1
–6 –2
0.2
0.0 0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 1 2 3 4
µm µm
FIGURE 21.3 AFM (phase) images of (a) graphene and (b) PMMA-modified graphene on mica surface. (From S. Mohamadi, H. Schonherr,
University of Siegene, Germany, unpublished data.)
Grafting from Grafting to
= Initiator = Polymer
= Monomer
FIGURE 21.4 Schematic presentation of “grafting-to” and “grafting-from” techniques.
21.3 CHEMICAL MODIFICATION charged aluminosilicate layers that are linked by single layers
OF GRAPHENE SURFACE of potassium ions. Studies have shown that the interactions
between graphene and mica surfaces are stronger than van der
As mentioned before, deviations from flatness in graphene Waals simple interaction and an additional interaction arises
sheets are theoretically to be expected, since Mermin– between the surfaces, significantly increasing the adhesion.
Wagner theorem [40] states that a long-range order is impos- Potassium atoms distributed at different regions of the mica
sible in two dimensions because of thermal fluctuations. surface can interact electronically with the graphene surface
These deviations from planarity are thought to represent a through the n or p molecular doping.
scattering mechanism that can reduce electronic mobility. After covalent attachment of poly(methyl methacrylate)
Based on Wang et al. [41] calculations and simulation the first (PMMA) chains on the graphene surface through atom trans-
wrinkle is not formed on the edge but in the interior of gra- fer radical polymerization (ATRP) or in other words, by the
phene and the graphene wrinkling depends on the structural modification of graphene surface chemistry, the electronic
size strongly due to the high surface energy. It seems interac- and morphological properties of the surface has been varied.
tions of graphene with the surrounded media can reduce the In this situation, the deposited modified graphene on the mica
surface energy and stabilize it to some extent, besides wrin- substrate decreases its contact with the mismatched substrate
kling; restacking of dispersed graphene sheets especially in surface by crumpling and folding. Therefore, it is plausible
the polymeric media is also undesirable due to the dramatic to suggest that there is dependence between the strength of
decrease in the surface area. The dispersion properties of chemical interaction at the interface and the degree of gra-
graphene can be affected by the number of factors such as phene crumpling.
size and surface properties. The minimum restacking and To date functionalization of graphene with different organic
more effective dispersion can be achieved more easily by the and inorganic compounds covers most of the published stud-
use of the smaller graphene sheets or in the higher edge-to- ies. Here, the manipulation is just restricted to the polymeric
plane ratio. However, the dramatic reduction in the thermal modification which also can be considered as non-covalent
and electrical properties of graphene will be occurred due and covalent. The non-covalent functionalization, which relies
to the reduction in surface area. Manipulation of graphene on the van der Waals force, electrostatic interaction or π–π
surface properties with chemical modification and interface stacking [42,43], is easier to carry out over the entire graphene
engineering is the way which has been considered nowadays surface without altering the chemical structure of graphene
in different laboratories. sheets, and provides effective means to tailor the electronic/
AFM pictures (phase image) shown in Figure 21.3 can optical property and solubility of the nanosheets in a revers-
confirm this claim that the wrinkling and crumpling in gra- ible manner. However, thermal energy and mechanical load
phene sheets, independent of their nature, can be strongly transferring from the surrounded media (polymer chains in
suppressed by interfacial interactions between graphene and the nanocomposite field) to graphene at interfaces cannot be
the surrounded media. In this experiment, graphene sheets done sufficiently and the poor coupling in vibration modes
were deposited from dilute DMF on the mica substrate by will suppress the transferring and dissipation. The same draw-
spin-casting technique. Graphene sheet on the mica surface back in the interface point can be observed in the mechani-
seems to have no wrinkle and is completely smooth. Actually, cal force transferring. Consequently, when stability and strong
micas are known as a material composed of negatively thermal and mechanical properties are expected, covalent
TABLE 21.2
Polymeric Graphene Modification by “Grafting-from” Method
Presence of Functional Groups
Grafting Methods Used on the Graphene Surface Attached Polymer
Surface-initiated ─OH and ─COOH 1. Poly(ϵ-caprolactone) [46]
ring-opening 2. Nylon 6 [47]
polymerization
Polycondensation –OH 1. Polyurethane (PU) [48]
–COOH 2. Epoxy resin [49]
Direct electrophilic –H 1. Aryl diazonium salt [50,51]
substitution 2. Ferrocene by Friedel–Crafts monoacylation [52]
Ziegler–Natta –OH 1. Polypropylene (PP) [53]
polymerization
Nucleophilic substitution Epoxy groups 1. Amine-terminated-branched polyethylene glycol [55]
reaction –OH (GO acts as a nucleophile 2. α-amino acid [57]
by using strong base such as 3. Tetraphenylporphyrin [58]
methoxide) 4. Amine-terminated ionic liquids [59]
5. Poly-l-lysine [60]
6. Attachment of aminophenol groups next to bromoisobutyrate groups and then
grafting of poly(N-isopropylacrylamide) (PNIPAM) via SI-ATRP [61]
7. Polyaniline (PANI)—the covalent functionalization of phenylene diamine and
in situ polymerization of aniline monomer [62]
8. Polyglycerol [63]
Reversible addition- 1. –OH 1. poly(N-vinylcarbazole) [64]
fragmentation reaction 2. –OH, –COOH 2. (N-isopropylacrylamide) (PNIPAM) [65]
[55–58,93] 3. poly(N-vinyl pyrrolidone) [66]
4. poly(2-hydroxyethyl methacrylate-co-acrylic acid) using bis(thiobenzoyl)disulfide
as an initiator [67]
5. Polystyrene (PS) and diblock coplymer [68–70]
Surface-initiated atom –OH, –COOH 1. Polystyrene (PS) [75,76]
transfer radical 2. Poly(methyl methacrylate) (PMMA) [77,81]
polymerization 3. poly(2-dimethylaminoethyl methacrylate) (PDMAEMA) [78,79]
4. poly (tert-butyl acrylate) (PtBA) (by using a silane coupling agent) [80]
Single-electron transfer –OH, –COOH, epoxy group 1. poly[poly(ethylene glycol) ethyl ether methacrylate] (PPEGEEMA) [82]
living radical 2. poly(tertbutyl methacrylate (tBMA) [83]
polymerization
Nitroxide-mediated radical –OH 1. polyisoprene (PI) [84]
polymerization 2. Polystyrene (PS) [84]
Conventional radical 1. poly(glycidyl methacrylate) (PGMA), poly (3-hydroxy ethyl methacrylate)
polymerization [69–72] , –OH, (PHEM), poly (2-(dimethylamino)ethyl methacrylate) (PDMA), etc. [85]
2. Poly(methyl methacrylate (PMMA)) (by using methacrylic anhydride)[86]
3. poly N-vinylpyrrolidone (PNVP) [87]
4. PS − polyacrylamide copolymer (PS − PAM) [88]
5. PS (using organophilic stearylamine-modified graphite oxide (stearyl-GO)) [89]
methods have been preferred. In this part of discussion, modi- oxygen groups of GO. The highly reactive free radicals can
fication methods have been narrowed to chemical polymeric attack the sp2 carbon atoms of graphene and form a cova-
functionalization. lent bond which has been used by Kosynkin and coworkers
[44,45] to decorate graphene with nitrophenyls. In the second
21.4 COVALENT MODIFICATION method, oxygen-containing functional groups are essential
for covalent modification despite the fact that the ideal gra-
OF GRAPHENE
phene should be short of functional groups especially on the
As mentioned above, covalent modification is more desirable surface. However, in most cases, when graphene sheets were
which can be divided into two general routes: (a) the forma- prepared from exfoliated GO, during incomplete reduction
tion of covalent bonds between free radicals or dienophiles the retained oxygen-containing functionalities can be used
and C=C bonds of pristine graphene and (b) the formation for further functionalization. Till now, chemical modification
of covalent bonds between organic functional groups and the has been divided into two main categories of “grafting-to”
(a)
O OH
OH HO NH
OH HO NH
HO O
OH OH Norepinephrine
O O OH
OH HO
Alkaline HO OH
O
condition
OH
(b) OH OH OH
SI-ROP
pNor/RG-O pNor/RG-O
ε-CL
FIGURE 21.5 Schematic presentation of poly(norepinephrine) coating on the surface of GO (a) and description of SI-ROP on the pNor/
RG-O (b).
and “grafting-from” techniques (Figure 21.4). In “grafting- biocompatible polyesters. The hydroxyl groups of norepi-
from” method the initiator is immobilized on the graphene nephrine which has been connected to the graphene surface
surface and growing of macromolecules occurs from these were used as a surface initiator and polymerization carried
points (Table 21.2). Usually, this approach is based of radi- out using ϵ-caprolactone as a monomer and tin alkoxide as a
cal polymerization including “conventional” and “living” catalyst (Figure 21.5) [46].
polymerization. The other study was reported by Xu and Gao [47] for the
The functional groups that exist on the GO and can be nylon family via ring-opening polymerization. As it has been
used for surface modification are plentiful, hydroxyl and shown in Figure 21.6, in their procedure the propagating
epoxy groups on the surface and carboxyl group on the edges. polymer chain was immobilized onto the GO laminates by
So apart from radical polymerization methods, the various amidation reaction of the active amino group in nylon-6 with
approaches that were already developed in organic synthe- the carboxylic groups in the edges of GO. In their opinion,
sizing can be also used. These methods include ring-opening at elevated temperature, caprolactam was initiated by 6-ami-
polymerization, polycondensation, direct electrophilic substi- nocaproic acid and polymeric chains of the PA6 step propa-
tution, Ziegler–Natta polymerization, atom transfer radical gated along with the consumption of monomers. They have
polymerization (ATRP), reversible addition-fragmentation suggested restoration of conjugated planes for the thermal
reaction (RAFT), and 1, 3-dipolar cycloaddition. reduction of GO to graphene by decomposition of labile oxy-
Surface-initiated ring-opening polymerization (SI-ROP) is gen-containing moieties such as epoxy and hydroxyl groups
a good example of surface polymerization of biodegradable/ (Figures 21.7 and 21.8).
O OH
NH + + H O
N 250°C H3N
+H2N (CH2)5COOH +H2N (CH2)5COO- N N COO-
–H2O O H
COOH
COOH
O O O
250°C
Condensation HO COOH
O
OH
HOOC
COOH
O
FIGURE 21.6 Grafting of nylon 6 onto graphene by in situ ring-opening polymerization of caprolactam in the presence of GO.
NCO
OCN O
HN O N
N
OH O
OH OH OH O
O HO O OH O O O HO O
MDI OH
O OH
HO OH 70°C, 2 h HO OH
O
N
H O
OCN
PTMG, EG
80°C, 8 h
FIGURE 21.7 Functionalization of GO sheets with PU.
Functionalization of GO sheets with polyurethane (PU) functionalization of graphene by a diazonium salt. The sur-
using 4,4′-diphenylmethane diisocyanate and then polycon- factant sodium dodecyl benzene sulfonate (SDBS) is wrapped
densation of poly(tetramethylene glycol) in the presence of on the surface of GO, followed by reduction at 80°C for 24 h
ethylene glycol was reported by Wang et al. [48], which is with hydrazine monohydrate. Then, the aryl diazonium salt
an example for using polycondensation polymerization in the is grafted on the surface of graphene sheets at room tempera-
surface modification of graphene (Figure 21.9). ture for 1 h. Friedel–Crafts acylation of GO with ferrocene
Liu et al. [49] have reported modification of the GO surface through the edge carboxyl groups was also demonstrated by
with epoxy resin. A three-step grafting procedure has been Avinash et al. (Figure 21.12) [52].
used to graft the epoxy monomers (DER332) and the cur- Huang et al. [53] proposed the preparation of polypro-
ing agents (diamino diphenyl methane [DDM]), onto the GO pylene/graphene oxide (PP/GO) nanocomposites via in situ
surface. The surface modification of GO has been performed Ziegler–Natta polymerization. They incorporated Mg/Ti
by grafting of Jeffamine D-2000, followed with subsequent catalyst species into GO via surface functional groups includ-
grafting of DER332 and DDM, respectively (Figure 21.10). ing ─OH and ─COOH, giving a supported catalyst system
Electrophilic substitution reactions with graphene involve and then subsequent propylene polymerization led to the
the displacement of a hydrogen atom by an electrophile. The in situ formation of the PP matrix (Figure 21.13).
preparation of organosoluble graphene by electrophilic sub- In the composite field, surface modification is done only
stitution of aryl diazonium salt on the surface of surfactant- for controlling of interfacial interaction and dispersion status
wrapped graphene was reported by Lomeda et al. and Zhu but in other fields such as optoelectronics [54], drug-delivery
et al. [50,51]. Figure 21.11 is a schematic illustration of the materials [55], biodevices [56], and others the special target is
H2N
33 COOH
COOH O
O
O NH
O
O C
H2N O NH2 O OH
HOOC OH
33 HOOC
HOOC
COOH D 2000 C
OH NH
OH O
O
O O
HOOC (GO) C 33
O NH2
NH O
H2N 33
(D-GO)
H 2N
CH2
DER 332
NH
HN
33 COOH
O HN
33 COOH
O O
NH
O O
NH
C O
O OH DDM C
HOOC OH
H2 O
H 2N C NH2 HOOC
C
OH NH
O C
O OH NH
O
O O
C n O
O NH O NH C 33
33
HN O NH NH
33 O
HN
HN
NH
CH2
H2C
NH2
H2N
H2 CH3 H2 H H2 CH3 H2 H
DER332: H2C CH C O C O C C C O C O C C CH2
O CH3 OH CH3 O
n
H2
DDM: H2N C NH2 D2000: H2N NH2
O 33
FIGURE 21.8 Reaction scheme of grafting epoxy monomer and curing agents onto GO surface.
R R
R
OH N2H4, H2O, PH = 10
OH OH O
O HO O OH O 80°C, 24 h O
O OH
HO OH OH
– +
BF4N2 R
1a-4a R R
rt, 1 h
1a,b R = Cl
2a,b R = NO2
3a,b R = OCH3
4a,b R = Br
FIGURE 21.9 Reduction and functionalization of intermediate SDBS-wrapped CCG with diazonium salts, starting with SDBS-wrapped
GO.
COOH
OH O O
OH OH
HOOC
O OH
Acidic aluminia
(CF3CO)2O
Ferrocene
OH O O
O Fe
OH OH
O OH
Fe
FIGURE 21.10 Covalent modification of GO with ferrocene by Friedel–Crafts monoacylation on an acidic alumina surface.
reserved. Recently, biological applications of graphene have Epoxy groups on the graphene surface are also versatile
also started to be concerned. For example, the fabrication of a sites for utilizing 1,3-dipolar cycloaddition and nucleophilic
novel graphene-based live-bacterial-hybrid device and DNA- substitution reaction to graphene sheets. Quintana et al. [57]
hybridization device with excellent sensitivity was reported reported functionalization of graphene by employing precur-
by Berry et al. [55]. In another work, Liu et al. [57] modi- sors of paraformaldehyde and a specifically designed NH2-
fied nanoscale GO sheets by amine-terminated-branched terminated α-amino acid. The amine (–NH2) functional
polyethylene glycol (PEG) (Figure 21.14). The unique ability group of organic modifiers bearing a lone pair of electrons
of graphene in the attachment and delivery of aromatic, water- can attack the epoxy group. The same approach was used by
insoluble drugs such as SN38, a camptothecin (CPT) analog, Zhang et al. [58] for the attachment of tetraphenylporphyrin to
was observed. the graphene surface (Figure 21.13).
OH OMgCl OMgCl
OH OH O OMgCl
O HO O OH O O ClMgO O OMgCl O O
RMgCl OMgCl
O OH O
HO OH HO OH
TiCl4
Cl Cl
Cl Cl
Cl Cl
Cl ClCl Cl Cl Ti Ti Cl Cl Cl
Cl Ti Ti Cl Cl Cl Cl Ti Ti Cl
Cl Cl Cl Cl Cl
Cl Mg Cl
Mg O Mg
C3H6 O O O O
O HO O
O OH
AIEt3 HO OH
O
Mg
Cl
Cl Cl
Cl Ti
Cl Cl Ti Cl
Cl Cl
FIGURE 21.11 Modification of graphene surface by PP through in situ Ziegler–Natta polymerization.
Yang et al. [59] have used amine-terminated ionic liquids

(IL-NH2) for the functionalization of graphene sheets, as
H2N shown in Figure 21.14 and enhanced their solubility in water,
DMF, and DMSO (dimethyl sulfoxide) due to the enhanced
O solubility and electrostatic inter-sheet repulsion provided by
the ionic liquid units.
O NH2 Shan et al. [60] prepared biocompatible poly-l-lysine-
O functionalized graphene through the active amino groups
O NH2
of amino acid and epoxy groups of graphene. This PLL-
O O functionalized nanomaterial has been applied in many fields
such as cell labeling, biofuel cells, and DNA electrochemical
O O sensors. In this group, the hydrogen peroxide (H2O2) detector
H2N O device was built based on the immobilization of horseradish
peroxidase (HRP) onto the PLL/rGO surface and the depo-
O sition of the resulting nanocomposite on a gold electrode
(Figure 21.15).
O Ren et al. [61] modified the surface of graphene as
C2H5 N O O NH
temperature responsive and water dispersible by graft-
HO N C2H5 OH O ing poly(N-isopropylacrylamide) (PNIPAM) via surface-
initiated ATRP.
First, graphene surfaces are functionalized with ami-
nophenol groups by diazonium reaction on water. Next,
bromoisobutyrate groups are covalently attached to the phe-
O
nol-functionalized graphene (G-OH) surface (Figure 21.16)
C2H5 O O
N by esterification of 2-bromoisobutyrate with the hydroxyl
HO N C2H5 OH groups, forming bromoisobutyrate-functionalized graphene
(G-Br). Finally, PNIPAM is then grafted from G-Br via ATRP.
FIGURE 21.12 Immobilization of SN38 CPT on PEG-GO. This temperature-responsive and water-dispersible modified
N
H
NH HN
H
N
N N
O H
ODCB, 160°C
+ CH3NHCH2CO2H + NH HN
H
N
CHO
FIGURE 21.13 Synthesis of graphene tetraphenylporphyrin.
Br NH2 + Br-
chain transfer as a means of converting dormant chains to
N N N N active propagating radicals. Zhang et al. [64] used RAFT
EtOH, reflux NH2
polymerization to grow poly(N-vinylcarbazole) from the
IL-NH2
surface of GO sheets which is useful for the fabrication of
various optoelectronic devices (Figure 21.19).
Yang et al. [65] decorated reduced graphene oxide
sheets by poly(N-isopropylacrylamide) (PNIPAM) polymer
brushes based on click chemistry and (RAFT) polymeriza-
tion. rGO sheets were modified by diazonium salt of propar-
HO
O gyl p-aminobenzoate and RAFT chain transfer agent (CTA)
IL-NH2 + KOH, water
+
was grafted to the surfaces of rGO sheets by click reaction.
80°C, reflex IL-HN PNIPAM decorated on the rGO sheets by RAFT polymeriza-
tion as shown in Figure 21.20.
Layek et al. [66] used RAFT polymerization for graft-
ing of polyvinylpyrrolidone (VP) onto the graphene surface
and used as reinforcement agent of poly(vinyl acetate) films.
In their method GO coupled with 2-bromo-propynyl bromide
GO P-CCG to produce the GO-BPB and then reacted with potassium
O-ethyl xanthate producing the GO-EX. The RAFT polymer-
FIGURE 21.14 Functionalization of graphene sheets by amine- ization of VP from the GO-EX surface occurred by reversible
terminated ionic liquids. addition and fragmentation of the C–S bond producing the
GO-g-PVP (Figure 21.21).
graphene can be useful in environmental devices as well as in Grafting of poly (2-hydroxyethyl methacrylate-co-acrylic
controlled release drug delivery by temperature controlling of acid) onto GO surface was performed by using RAFT polymer-
its dispersibility. ization of 2-hydroxyethyl methacrylate (HEMA) and acrylic acid
Remyamol et al. [62] decorated reduced graphene oxide by (AA) as monomer and azobisisobutyronitrile (AIBN) as initiator
polyaniline (PANI) through covalent functionalization with (Figure 21.22) [67]. The stimuli-responsive behavior of modified
phenylene diamine and then in situ polymerization of aniline graphene was examined in different pH (3–12) and temperature
monomer in the presence of GO-NH2. The synthesis strategy (25–55°C) values. The LCST (lower critical solution tempera-
for the grafting of PANI on reduced graphene oxide is given ture) of 30°C and 34°C has been observed for GO-P(HEMA)
in Figure 21.17. This PANI-modified graphene is a good can- and GO-P(HEMA-co-AA) nanosheets, respectively.
didate for optoelectronic and NLO (nonlinear optic) device Etmimi et al. [68] graft polystyrene (PS) in to GO by
applications. RAFT polymerization. Dodecyl isobutyric acid trithiocarbon-
Pham et al. [63] have reported a new modification tech- ate (DIBTC) has been immobilized into graphene surface as
nique in which GO acts as a nucleophile in the presence of a RAFT agent through an esterification process. The RAFT-
a base, such as potassium methoxide (Figure 21.18). Water- grafted GO (GO-DIBTC) was dispersed in styrene monomer,
dispersible polyglycerol-grafted GO nanostructure was syn- and the resultant mixtures form miniemulsions by using soni-
thesized and immobilized with boronic acid-functionalized cation and a surfactant (sodium dodecylbenzene sulfonate)
Fe@Au nanoparticles. This synthesized novel hybrid material and a hydrophobe (hexadecane). The stable miniemulsions
emerged as a promising material capable of meeting v arious thus obtained were polymerized using AIBN as the initiator
requirements for potential applications. to yield encapsulated PS-GO nanocomposites.
RAFT polymerization is a controlled-living radical polym- Another RAFT synthesizing strategy was used by Beckert
erization (CLRP) technique that is based on degenerative et al. [69] for attachment of PS into graphene surface. They
HO
HO O
O OH
N
O OH
NaBH4
KOH, 70°C 70°C HO N
HO O N
O
HO O
HO O
S C OH
EDC, NHS O
S C OH
O
OH OH
OH OH
N N
N O O
N
OH EDC, NHS
OH
N N
HO HO
O O
NH NH
C=O C=O
S S
= Gold electrode
H2 H2 H2
= HRP = H2N C C C CH2 H O
H
N C C
n
FIGURE 21.15 Schematic representation of the formation of the PLL/rGO−HRP and its subsequent deposition on a gold electrode.
used dithiourethane-, dithioester- and dithiocarbonate- homolytic halogen transfer between a dormant chain and a
functionalized graphenes for growing PS chains onto GO transition metal (usually Cu) complex in its lower oxidation
surface by RAFT polymerization. The styrene graft polymer- state. The generation of propagating radicals and the metal
ization was initiated either with AIBN at 65°C or by thermal complex in its higher oxidation state coordinated with the hal-
styrene self-initiation at 130°C. ogen atom has resulted because of this transfer. The transition
In Jiang et al. [70] strategy homo- and di-block copo- metals are complexed with ligands (e.g., amines and pyridine-
lymers were grafted into the graphene surface using based ligands), which lead to an increase in their solubility in
simultaneous coupling reaction and RAFT process. Two monomer/solvent as well as strongly influence the activation/
S-methoxycarbonylphenylmethyl S′-3-(trimethoxysilyl)prop- deactivation rates [74].
yltrithiocarbonate (MPTT), and S-4-(trimethoxysilyl)benzyl The use of GO nanosheets containing covalently attached
S′-propyltrithiocarbonate (TBPT)-functionalized graphene initiator moieties for ATRP has been reported by several
were used as couplable RAFT agents to prepare target nano- groups but still uncertainties remain as to what extent con-
composites (Figure 21.23). trol/livingness was actually achieved in this type of modi-
Matyjaszewski [71,72] and Sawamoto [73] developed the fication. Fang et al. [75] functionalized reduced GO sheets
CLRP (controlled living radical polymerization) technique, with lateral dimensions of 30–150 nm by PS based on the
independently. This method is based on a dynamic equilib- approach outlined in Figure 21.24. Bromopropionyl bro-
rium between the alkyl halide species (dormant chains) and mide was covalently attached to graphene–OH to anchor
the propagating radicals, which is established via reversible ATRP-initiating moieties. PS chains were grown from the
OH OH OH
OH
O
ONO Br
Br
TEA
H2N OH
HN O
HN O
O Br O O n
n
Br O HN
O O O
O O
n
O O Br O
NH O
O Br O O O n
O
CuBr/PMDETA
NIPAM
FIGURE 21.16 Synthesizing graphene-modified- poly(N-isopropylacrylamide) (PNIPAM).
O OH
O CI
O OH
O OH
HO
O O HO
O O
SOCl2,
O 80°C, 24 h
HO O
OH CI
O
O CI
DMF, 120°C, 72 h H2N NH2
H
O N NH2
O NH2 O OH
OH
HO HO
O O
O O
(NH4)2S2O8, TSA O
HN O NH
H2N
H2N
O TSA TSA
C H H H +
= HN N N N N N
-y y
n
FIGURE 21.17 Decoration of reduced GO by polyaniline (Pani).

OH COOH COOH
OH COOH COOH
O
O
O O O
OH O O
OH
O OH
HO O
O O MeO-K+ HO
O O
COOH OH
COOH OH
Au
Fe O
O
S
OH COOH COOH
HO
HO O
HO B O
O OH
HO o l
an
O O Eth OH
HO OH
OH
O
HO
O O
COOH OH
O
O
HO
HO O
O
OH
O O OH
B
Fe Au
FIGURE 21.18 Synthetic routes of water-dispersible PG-g-GO nanostructure and its immobilization with B-f-MNPs.
functionalized graphene sheets using (1, 2-dichloroben- mol and Mw/Mn = 1.6. The MWD (molecular weight distri-
zene/CuBr/N,N,N ′,N ′,N ′-pentamethyldiethylenetriamine bution) of polymer from sacrificial initiator is in most cases
(PMDETA)/110°C) in the presence of the sacrificial ATRP different from that of the grafted polymer; however, the high
initiator methyl-2-bromopropionate. The grafted polymer value of Mw/Mn suggests issues with the level of control/ liv-
originating from the sacrificial initiator had Mn = 60,000 g/ ingness also for the grafted polymer.
O O
OH OH
O O
O
HO HO S S HO
C12H25
S O
O O
O O DCC/DMAP O
HO O O S
DMF, 48 h S
C12H25
S
HO HO
O
NVC
OH
O AIBN
70°C
12 h
HO
O
O
O
O O S
n
S
N
S C12H25
HO
FIGURE 21.19 RAFT polymerization to grow poly(N-vinylcarbazole) from the surface of GO sheets.
Click chemistry
NIPAM
RAFT polymerization
O
F4B N2 C OCH2 C CH
S CH3
C12H25 S C S C COO
N3
CH3
FIGURE 21.20 Outline for the preparation of PNIPAM/reduced graphene oxide (rGO) nanocomposites based on click chemistry and
RAFT polymerization.
O
COOH C
COOH 1-Br
O Br
O
HO COOH 2-Pyridine
O 3-3 h, 0°C
OH
HOOC and 48 h, R. T. Br
COOH COOH
S S O C O
O O O
O
O S S O
S C HOOC
O +K-S O OH
O
C O OH
O HO COOH
Br
O
COOH
C
O
O S
COOH
HOOC HO O
S
O
n
O
yrrolido ne N
1-Vinyl p
2- DMF O
S
3- AIBN
0h O O C
4- 80°C, 2
S O
n
O
N
COOH
HO
COOH O
C O
O HO
O
COOH
FIGURE 21.21 Synthesis of GO-g-PVP by RAFT polymerization.
Revising the above functionalization by using the larger by reaction of GO with 1,3-diaminopropane aided with
GO sheets (0.5–1.5 μm) and a surfactant (sodium dodecy- N-hydroxysuccinimide and N-(3-(dimethylamino)propyl)-N′-
lbenzene sulfonate; SDBS) to achieve an aqueous disper- ethylcarbodiimide hydrochloride through the amide linkage.
sion for the synthesis of OH-functionalized sheets, had Immobilization of the ATRP initiator 2-bromo-2-methylpro-
different results [76]. Monomer conversions of 71%–85% pionyl bromide to GO can be carried out by both -OH on the
with Mn = 21,300–81,600 g/mol and Mw/Mn = 1.59–1.90 were surface and amine groups at edges. Grafting from ATRP of
determined but the high values of Mw/Mn indicate that the 2-(dimethylamino)ethyl methacrylate (DMAEMA) was done
control (and possibly also livingness) of the polymerization at 60°C using PMDETA/CuBr in the absence of solvent. After
was not of the standard expected in CLRP. polymerization, the PDMAEMA-modified GO displays good
Goncalves et al. [77] prepared GO sheets with cova- solubility in acidic aqueous solutions (pH = 1) as well as in
lently attached PMMA via attachment of ATRP initiator short-chain alcohols.
moieties by reaction with 2-bromo-2-methylpropionyl bro- In a control experiment, sacrificial ATRP initiator was
mide. PMMA brushes were subsequently grown using CuBr/ used, and analysis of the obtained free polymer yielded
PMDETA/65°C in DMF as solvent. The cleaved polymer Mn = 27,000 g/mol and Mw/Mn = 1.72, indicating poor control
chains from the GO sheets by using hydrolysis analyzed by over the MWD. AFM analysis revealed the polymer domains
GPC and yielded Mn = 1.170 g/mol and Mw/Mn = 1.09. Based on the sheets to be 4 nm in size.
on AFM measurements in friction mode it was concluded that Bak et al. [79] also prepared thermoresponsive graphene
the polymer chains were homogeneously distributed at the nanosheets by chemical modification with initiators followed
GO surface. by extension with PDMAEMA through surface-initiated
Yang et al. [78] used carboxylic groups to attach poly(2- ATRP. Modification was done by treatment of thermally
dimethylaminoethyl methacrylate) (PDMAEMA) into GO. reduced GO with 4-aminophenethyl alcohol resulted in dia-
Carboxylic acid groups were converted to amine groups zonium salts with hydroxyl groups (FGS-OH).
O
H
O N O S
N Br H
H S O N
N S
S S H
O
S
O O
CuBr, PMDETA, Cu
O S
O Br
O S
HO
OH
O
S
O O
O
O
S h
24
H C,
70°
n
O N N
H
HO
O O
O O
S
O
n S HO
HO S
O O
OH O S
O
O n
70°C H n
, 24 h O N N
H
HO
O O
O O
OH
O
O S
n
n S
FIGURE 21.22 RAFT polymerization of HEMA-AA.
The ATRP initiator on the surface of FGSs (functionalized attached to GO by esterification, has been used in the polym-
graphene sheets) (FGS-Br) was then introduced by esterifi- erization of methyl methacrylate [81].
cation between FGS-OH and α-bromoisobutyryl bromide. Single-electron transfer living radical polymerization
Grafting-from ATRP of 2-(dimethylamino)ethyl methacrylate (SET-LRP) also referred to as Cu(0)-mediated radical polym-
(DMAEMA) was done at 50°C using PMDETA/CuBr in the erization has also received significant attention due to its high
presence of ethyl α-bromoisobutyrate (EBiB) as sacrificial polymerization rate at low temperature and exceptionally
initiator and anisole as solvent. They found that the solubility high end-group livingness. The central feature of the process
of FGS can be easily tunable with the temperature changes is disproportionation of Cu(I) in suitable solvents resulting in
by exploiting the LCST behavior of the PDMAEMA. These the formation of Cu(0) and Cu(II). The high end-group fidel-
changes in solubility reveal the reversible changes in the sur- ity is an important advantage of this method both when syn-
face property of the graphene nanosheet by grafting of ther- thesizing polymer and when growing chains from surfaces
moresponsive PDMAEMA brushes (Figure 21.25). such as GO in that enables one to take the polymerization
As shown in Figure 21.26, the other initiator that has also to complete monomer conversion without significant loss of
been immobilized on the surface of graphene sheets is benzyl control/livingness.
chloride by using a silane coupling agent to graft poly(tert- Deng et al. [82] modified GO with poly[poly(ethylene gly-
butyl acrylate) (PtBA) brushes via ATRP [80]. In another col) ethyl ether methacrylate] (PPEGEEMA) using grafting-
study, a similar initiator, methacryloyl chloride which was from SET-LRP. Epoxide groups of the GO basal plane were
S M1, AIBN
(a) O
O Si S S O R
O O MPTT-PM1
MPTT R
GO, M1′, AIBN GO, M2, AIBN
R R
R R R R
R R
R R R
R
R R R R
R R R
R R
GO-graft-PM1′ GO-graft-PM2-b-PM1
S
S S M1′, AIBN
(b) O
O
Si TBPT-PM1
TBPT O
GO, M1′, AIBN GO, M2, AIBN
GO-graft-PM1′ GO-graft-PM2-b-PM2
= (GO)–OSi(OR′)2CH2CH2SC(=S)S-, R′ = H or CH3
= CH3CH2CH2SC(= S)S-; = (GO)-OSi(OR′)2ArCH2-)
FIGURE 21.23 Synthesis of homopolymers and diblock copolymers grafted GO by MPTT (a) and TBPT (b) mediated RAFT process and
coupling reaction. (Reprinted with permission from Jiang, K. et al. One-pot controlled synthesis of homopolymers and diblock copolymers
grafted graphene oxide using couplable RAFT agents, Macromolecules, 45, 1346−1355. Copyright 2012 American Chemical Society.)
reacted with the amine functionality of tris(hydroxymethyl) (tBMA)). The GO-OH sheets were functionalized with
aminomethane (TRIS), thus increasing the amount of bromopropionyl bromide to incorporate initiation sites.
hydroxyl groups. Alkyl halide initiating groups were sub- Polymerization of tBMA in DMSO in the presence of the sac-
sequently introduced via esterification reaction between rificial initiator 2-bromopropionate was conducted using Cu(0)
hydroxyl groups and α-bromoisobutyryl bromide. Surface- wire and tris[2-(dimethylamino)ethyl]amine (Me6TREN) as
initiated SET-LRP of PEGEEMA using CuBr/tris(2- ligand at 25°C for 1 day (Figure 21.29). AFM analysis revealed
(dimethylamino)ethyl)amine (Me6TREN) was carried out in that the thickness of single GO sheets increased from 1 nm to
water/THF at 40°C in the presence of the sacrificial initiator 8–10 nm which was related to the thickness of the covalently
methyl 2-bromopropionate (Figure 21.27). linked polymer layer.
It should be noted in the majority of cases; the molecular Modification of the surface of GO by grafting polymer
weights (distributions) of polymer originating from surface chains using nitroxide-mediated radical polymerization
initiation and sacrificial initiator are significantly different. (NMRP) was demonstrated by García-Valdez et al. [84].
One possible reason for the lack of control is that graphene Surface modification by NMRP was achieved using GO
sheets were reacting with the propagating radicals, which, in functionalized with 2,2,6,6-tetramethyl-piperidine 1-oxyl
tandem, affect the living character of the process. The graft- (TEMPO, T) to produce graphene oxide-TEMPO (GO-T).
ing of PPEGEEMA appears as dark spots on the GO sheets GO-T acts as a multifunctional macroalkoxyamine initiating
in TEM analysis, in sharp contrast to the original GO sheets and controlling polymerization in the presence of the mono-
(Figure 21.28). The size of these dark spots is ~70 nm in diam- mer. The grafting reactions were performed by dispersing
eter, which is significantly larger than what can be explained GO-T in DMF and heating at 130°C in the presence of the
by each spot corresponding to one single chain [82]. monomer to form graphene oxide-g-polystyrene-TEMPO
Chen et al. [83] also used SET-LRP to prepare reduced (GO-g-PS-T) and graphene oxide-g-polyisoprene-TEMPO
GO sheets functionalized with poly(tertbutyl methacrylate (GO-g-PI-T) (Figure 21.30).
OH N2H4, H2O OH
O OH OH OH
OH O O 100°C, 4 h O OH OH
OH O OH O
O OH OH
HO HO
OH O
OH H
OH
H2N OH
OH
80°C, Overnight
OH OH
O OH OH
ONO OH O
OH
HO OH
OH
OH
OH
Br
Br
O CuBr/PEMDTA/St
rt, 24 h MBP, 110°C, 10 h
FIGURE 21.24 PS chains growing from the functionalized graphene.
Kan et al. [85] employed the flat 2D structure of GO lami- and benzoyl peroxide (BPO) as an initiator (Figure 21.32). It
nates as platforms to grow polymer brushes by free radical seems in this method the grafting density increased but pen-
polymerization (Figure 21.31). They explored the widely etration of nonimmoblilized polymer chains into the polymer
known simple free radical polymerization of several fami- brushes and difficulty in the separation still remained.
lies of polymers and succeeded in growing polymer brushes Feng et al. have reported grafting of poly N-vinylpyrrolidone
from graphene. However, despite the success from a practical (PNVP) by the radical polymerization mechanism. They sug-
point of view, certain aspects related to the polymerization gested the breakage of weak bonds at the defects on the GO
mechanism remain still unknown and need further studies. surface initialized the radical polymerization of NVP [87].
The authors stated that during propagation macromolecular rGO was also used by Shen et al. [88] for the preparation
radicals were formed, some of which add to sp2 framework of of “amphiphilic” graphenes. Here reduced GO nanoplate-
GO producing brushes and generating new radicals capable lets were decorated with PS − polyacrylamide copolymer
of further reaction. Therefore, the 2D sheets have many reac- (PS − PAM) by in situ free radical polymerization, initiated by
tive sites onto which the propagating radicals can bind. As BPO as depicted in Figure 21.33. The modified GO is dispers-
well,from a structural point of view, side reactions such as ible in water, due to its hydrophilic PAM section, and simul-
V-shaped grafted polymers and cyclic chains may also occur. taneously in xylene due to the hydrophobic PS component.
In this technique the important amounts of nonimmobilized Moreover, the hydrophilic−hydrophobic balance of the func-
polymer chains strongly interacting with immobilized ones tionalized graphene can also be controlled by adjusting the
may also be present in the final product leading to an erro- ratio between the two polymeric components.
neous value of the grafting density. Since the probability of Beckert et al. [89] reported the solvent-free free radi-
grafting efficiency only with coupling reaction is low a high cal grafting of styrene homo- and copolymers onto organo-
value of grafting density cannot be expected in this proce- philic stearylamine-functionalized graphene nanosheets
dure. In another method, we [86] tried to introduce some vinyl without requiring initiator addition. They also used TEMPO
moieties on the graphene surface by the reaction of reduced reagent to control the further grafting yields by a controlled
graphene oxide with methacrylic anhydride. The conventional radical “grafting-from” polymerization. As illustrated in
radical polymerization was carried out in the presence of Figure 21.34, a different pathway reaction of graphene-
modified graphene, MMA (methyl methacrylate) as monomer initiated styrene polymerization has been suggested.
(a) O OH
O OH
R1
R1 O OH
O OH H2N HO
HO OH R1
O O
4-Aminophenethyl alcohol R1
O R1 O
O NO O
O
OH Isopentyl nitrite OH
OH OH
60°C
HO O HO O
R1 =
OH
Functional graphene sheet 4-Hydroxyethylphenyl FGS
(FGS) (FGS-OH)
Br O Et3N
THF
Br
O OH
O OH
(b)
R2
R2
R2 O OH
R2 O OH HO
HO R2
R2 O
O ATRP R2
R2
R2 O
R2 O
O O O OH
OH O OH
OH
N
HO O
HO O
O O Br
R3 = m
R2 =
O O O
O
2-Bromo-2-methylpropionate-
Polydimethylaminoethyl N 4-hydroxyethylphenyyl FGS
methacrylate FGS
(FGS-Br)
(FGS-PDMAEMA)
T > LCST
T < LCST
FIGURE 21.25 (a) Schematic illustration of FGS-PDMAEMA brushes growing from the surface of FGS via SI-ATRP. (b) Thermoresponsive
LCST behavior of FGS-PDMAEMA.
OH OH
O O
OH OR
CH2Cl
HO RO
OH OH
HO RO
O SiCl3 O
O O
HO OH RO OR
OH GO-Cl
O
R’: Si CH2Cl
OR’
O
R’O
O OH
R’O
Surface-initiated O
ATRP O
R’O OR’
GO-g-PtBA
H2
R’ : Si C C C Cl
O O
FIGURE 21.26 Functionalization of GO with poly(tertbutyl acrylate) (PtBA) brushes via surface-initiated ATRP.
OH
OH OH O OH
O HO O OH O 1. H2N
OH OH
HO
HO OH R.T OH
O
2. Br
O Br
HO E3N, R, T
OR1
CuBr/Me6TREN, H2O/THF, OR1 OR1 OR1 OR1
O OR1 O
40°C OR2
O HO OH
O OR2
O OR1 OR2
O n OR1
OH
TRIS-GO-Ini
Br OR1
R1 = N OR1
O R2 = OR1
FIGURE 21.27 Modification of GO with poly(poly[ethylene glycol] ethyl ether methacrylate) (PPEGEEMA) using grafting-from SET-LRP.
(a) (b)
100 nm 500 nm
FIGURE 21.28 TEM images of GO (a) and (b) TRIS-GO-PPEGEEMA. (Lang, M. D. et al.: Functionalization of graphene oxide towards
thermo-sensitive nanocomposites via moderate in situ SETLRP. Polym. Sci. Part A: Polym. Chem. 2011. 49. 4747–4755. Copyright Wiley-
VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
COOH COOH
OH OH OH OH
O
HOOC HOOC
O N2H4,H2O
OH HO 100°C OH HO
O
COOH COOH COOH COOH
NH2
ONO
80°C
OH
COOH COOH
R2 R1
OH R2 OH R1
R2 R1
Br
HOOC HOOC
R2 R1
O
OH HO R.T. OH HO
R2 R2 R2 R1 R1 R1
COOH COOH COOH

COOH COOH
G-Br P G-OH
O
Br OH P OH
O R1 =
R2 =
P
Cu((0), wire/Me6 TREN,
HOOC
P
OH HO O
O O
DMSO, 25°C
P P P O Br
P
tBMA SET-LRP
COOH COOH
G-PtBMA
FIGURE 21.29 Synthesis of PtBMA-modified graphene nanosheets.

– N
+Br O
OH N O O
O
DMF, N2
O
O N HO
HO
DM
F,
301
2
N
°C
,
°C
,N2
301
F,
DM
N
N O
O n n
O
O
O
GO-g-PS-T HO GO-g-PI-T O
HO
FIGURE 21.30 Modification of GO by PS and polyisoprene by using NMRP.
Radical
FIGURE 21.31 Synthesis of macromolecular brushes by free radical polymerization.
O
O O O O
OH O O
HO OH O HO
OH O O
O O
O
OH
OH O OH O
OH Et3N, 80°C
OH O OH O
O
O
O O
OH
OH
NMA, PBO, 80°C
n
O
O
O n
OCH3
O OCH3
O
O
HO
O O O
n
O
O
n OH O
O OCH3 OH O H CO
O 3
H3CO O O
O O
OH
n
FIGURE 21.32 Schematic illustration of graphene modification through conventional radical polymerization.
H2
2HC CH 2HC CH C C C C
BPO n R
m
O C O C NH
+ + NH2 125°C, N2 2
FIGURE 21.33 Used procedures for the preparation of graphene/PS–PAM.
PS NO PS
NO H
PS
C C
B B
H
PS
Styrene
385 K A
A
D
D PS
PS
FIGURE 21.34 Reaction pathways of the in situ styrene polymerization: “grafting-to” by addition of PS radicals to FG (A) and hydrogen
transfer from FG to PS radicals (B) increase FG radical content, whereas “grafting-from” by FG radical-initiated styrene polymerization
(C) and “grafting-to” by radical recombination (D) deplete C-centered graphene radicals. (Functional groups of stearyl-GO (hydroxyl and
stearyl moieties) are not shown due to clarity of the scheme.) (Reprinted with permission from Beckert, F. et al. Self-initiated free radical
grafting of styrene homo- and copolymers onto functionalized, Macromolecules, 46, 5488–5496. Copyright 2013 American Chemical
Society.)
21.5 MORPHOLOGICAL CHANGES properties. This type of functionalization of a carbon atom on

graphene converts the planar sp2 lattice point into a tetrahe-
As mentioned before, in order to tune physical changes and dral sp3 lattice point leading to the introduction of a scattering
the other properties of the individual layers assessing their site in graphene. Moreover, with the elimination of π electron
surface chemistry is of paramount importance. Some physi- from the carbon atom, the carrier density will be reduced and
cal properties of graphene sheets such as wrinkling and can introduce a band gap or a transport barrier.
dispersion are under adaptive control of the surface chemistry. Covalent functionalization of graphene is commonly done
The hydrophilic nature of GO based on the calculation of by producing GO, with oxygenated organic functionalities
reactive oxygen-containing groups distributed solely at the such as carboxyl, epoxy, hydroxyl, carbonyl, phenol, lactone,
periphery of single-layer GO would give a good dispersion and quinine. As a result, the formation of a large density of
in polar solvent such as water and methanol. TEM analysis sp3 hybridized carbons in the graphene network would [91]
(Figure 21.35) showed that the dispersed single GO sheets in disrupt the delocalized π cloud, converting graphene into an
polar solvent appeared typically flat yet wrinkled; however, insulator (Figure 21.36).
on addition of 1-ethyl-3-(3-dimethylaminopropyl) carbodi- Charge transport in the modified graphene depends on
imide (EDAC) GO separates immediately from the solution whether the functionalized site (a) forms domains or (b) is
and remains in an agglomerated state. It was observed under dispersed. The coexistence of two domains containing sp2
TEM that individual GO sheets condensed into star-like clus- (graphenic region with a low band gap) and sp3 (functional-
ters [90]. ized region with high band gap) has been shown by several
studies. Since these domains have a wide range of size distri-
21.6 HOW DO THE ELECTRICAL PROPERTIES bution, the carrier transport from one sp2 region to another of
OF GRAPHENE CHANGE WITH different sizes, shapes, and varying transport barriers occurs
via variable-range hopping. Quantum confinement-induced
ITS FUNCTIONALIZATION?
semiconducting behavior will be observed when the sp2

As has been discussed before, covalent functionalization is the region is small.
most studied form of graphene functionalization, and prob- On the other hand, the dispersed functionalization leads to
ably has the most significant effects on graphene’s electrical a reduction in the mean free path and Fermi velocity, and an
(a) (b) increase in elastic and inelastic scattering. As well, graphene

sheets have structural defects as electron–hole puddles which
can act as scattering centers during transport but in turn influ-
ence the functionalization of graphene [92].
These oxy groups can be further functionalized with other
molecular species as was talked about previously. More chemi-
cal modification can lead to molecular (dipole–dipole) interac-
tions with graphene to further change its e lectrical properties.
The attachment of polymeric chains onto graphene sheets
seems to have a complicated effect on the e lectric properties
of graphene. The grafted polymeric chains can have differ-
ent conformations from expanded to the coil to globule on the
graphene surface (see Figure 21.37) depending on the interac-
(c) (d) tions with the surrounded media. If the grafted chains collapse
into graphene sheets, the molecules interfacing can be consid-
ered; however, quantum coupling of the interfaced molecules
enhances the effective electric field on graphene due to the
dipole moment of the molecule. Doping graphene with elec-
tron-withdrawing groups creates a hole-rich graphene, while
electron-donating functionalities create electron-rich graphene
which can alter the electronic structure of graphene.
In Figure 21.38, the effect of functionalization (covalent
and non-covalent) and interactions on the electrical attributes
(charge density, scattering, electron–hole puddle size/density,
doping) of graphene due to a change in carbon’s hybridiza-
tion state, graphene’s quantum capacitance, lattice distortion,
FIGURE 21.35 Pure single-layer GO is stably suspended in charge transfer, dipole–dipole interfacing, and orbital hybrid-
water (a) with (near) planar morphology in TEM image (b, scale ization is shown [92].
bar = 200 nm), GO agglomeration with adding EDAC (c), observa-
tion of star-like clusters of GO sheets in TEM (d, scale bar = 100 nm).
(Reprinted from Carbon, 49, Whitby, R. et al. 2010 Morphological
21.7 SUMMARY SECTION
changes and covalent reactivity assessment of singlelayer gra- Graphene is defined as a single layer of carbon atoms
phene oxides under carboxylic group-targeted chemistry, 718–736. arranged in a hexagonal lattice. Since its discovery, a variety
Copyright 2011, with permission from Elsevier.)
of synthetic strategies have been explored to graphene pro-
duction. Some physical properties of graphene sheets such
as wrinkling and dispersion are under adaptive control of
sp2 to sp3
Y
Y
Removal of π-electron
lattice changes from planner to tetrahedron
109°C 28′
Z
X
Z X
= Carbon
= σ-Orbital
120°C
= π-Orbital
FIGURE 21.36 Schematic illustration of the effect of change in hybridization of carbon from sp2 to sp3 which leads to removal of the
π electron.
FIGURE 21.37 Grafted polymer chain conformations from expanded to coil to globule on the graphene surface.
Adsorption
π–π,
ole charge–transfer,
Dip ent Orbital
m dipole–dipole
Lattice mo rehybridization
interfacing
scattering
Quantum
capacitance
Quantum
Electron/hole Electron/hole capacitance
puddle puddle
scattering scattering
Bonding site
Doping
sp2 - sp3 Graphene
reduction π-electrons
Covalent Non-covalent
FIGURE 21.38 Schematic presentation of the effect of functionalization (covalent and non-covalent) on the electrical structures of
graphene. (T. S., Berry et al.: How do the electrical properties of graphene change with its functionalization? Small. 2013. 9(3). 341–350.
Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
the surface chemistry. Manipulation of graphene surface 3. Mouras, S. et al. 1987. Synthesis of first stage graphite intercala-
properties with chemical modification and interface engi- tion compounds with fluorides, Rev. Chim. Miner., 24, 572–582.
neering is the way which has been considered nowadays 4. Fradkin, E., 1986. Critical behavior of disordered degenerate
semiconductors, Phys. Rev. B, 33, 3263–3268.
in different laboratories. “Grafting-from” technique for
5. Geim, A. K., Novoselov, K. S., 2007. The rise of graphene,
modification of surface with polymeric chains was focused Nat. Mater., 6, 183–191.
in this chapter. The various approaches that were already 6. Meyer, J. C., Geim, A. K., Katsnelson, M. I., Novoselov,
developed in organic synthesizing have been used. These K. S., Booth, T. J., Roth, S., 2007. The structure of suspended
methods include ring-opening polymerization, polycon- graphene sheets, Nature, 446, 60–63.
densation, direct electrophilic substitution, nucleophilic 7. Nelson, D. R., Piran, T., Weinberg, S., 2004. Statistical
substitution, Ziegler–Natta polymerization, RAFT, ATRP, Mechanics of Membranes and Surfaces. World Scientific:
Singapore.
single-electron ATRP, NMRP, free radical polymerization,
8. Partoens, B., Peeters, F. M., 2006. From graphene to graph-
and 1,3-dipolar cycloaddition. ite: Electronic structure around the K point, Phys. Rev. B, 74,
Chemical modification of graphene surface leads to the 075404.
formation of a large density of sp3 with high band gaps. On 9. Venables, J. A., Spiller G. D. T., Hanbucken, M., 1984.
the other hand, dispersed functionalization leads to a reduc- Nucleation and growth of thin films, Rep. Prog. Phys., 47,
tion in the mean free path and Fermi velocity, and an increase 399–459.
in elastic and inelastic scattering. Attachment of polymeric 10. Evans, J. W., Thiel, P. A., Bartelt, M. C., 2006. Morphological
chains onto graphene surface can lead to molecular (dipole– evolution during epitaxial thin film growth: Formation of 2D
islands and 3D mounds, Surf. Sci. Rep., 61, 1–128.
dipole) interface interactions which have complicated effects 11. Castro Neto, A. H., Guinea, F., Peres, N. M. R., Novoselov,
on the electrical properties of graphene. K. S., Geim A. K., 2009. The electronic properties of gra-
phene, Rev. Mod. Phys., 81, 109–162.
12. Park, S., Ruoff, R. S., 2009. Chemical methods for the produc-
REFERENCES tion of graphenes, Nat. Nanotechnol., 4, 217–224.
1. Delhaes, P., 2011. Carbon-Based Solids and Materials. John 13. Hernandez, Y., Lotya, M., Rickard, D., Bergin, S. D.,
Wiley & Sons: Hoboken, NJ. Coleman, J. N., 2010. Measurement of multicomponent solu-
2. Gogotsi, Y., 2011. Controlling graphene properties through bility parameters for graphene facilitates solvent discovery,
chemistry, J. Phys. Chem. Lett., 2, 2509–2510. Langmuir, 26, 3208–3213.
14. Khan, U., Porwal, H., O’Neill, A., Nawaz, K., May, P., to graphite and surface modification with primary aliphatic
Coleman, J. N., 2011. Solvent-exfoliated graphene at extremely amines and amino acids, Langmuir, 19, 6050–6055.
high concentration, Langmuir, 27, 9077–9082. 34. Lerf, A., He, H., Forster, M., Klinowski, J., 1998. Structure of
15. Subrahmanyam, K. S., Panchakarla, L. S., Govindaraj, A., graphite oxide revisited, J. Phys. Chem. B, 102, 4477–4482.
Rao, C. N. R., 2009. Simple method of preparing graphene 35. He, H. Y., Klinowski, J., Forster, M., Lerf, A., 1998. A new struc-
flakes by an arc-discharge method, J. Phys. Chem. C, 113(11), tural model for graphite oxide, Chem. Phys. Lett., 287, 53–56.
4257–4259. 36. Compton, O. C., Nguyen, S. T., 2010. Graphene oxide, highly
16. Su, C. Y., Lu, A. Y., Xu, Y. P., Chen, F. R., Khlobystov, A. N., reduced graphene oxide, and graphene: Versatile building
Li, L. J., 2011. High-quality thin graphene films from fast blocks for carbon-based materials, Small, 6(6), 711–723.
electrochemical exfoliation, ACS Nano, 5, 2332–2339. 37. Pei, S., Cheng, H. M., 2012. The reduction of graphene oxide,
17. Parvez, K., Wu, Z. S., Li, R., Liu, X., Graf, R., Feng, X., Carbon, 50(9), 3210–3228.
Müllen, K., 2014. Exfoliation of graphite into graphene 38. Gao, X., Jang, J., Nagase, S., 2009. Hydrazine and thermal
in aqueous solutions of inorganic salts, J. Am. Chem. Soc., reduction of graphene oxide: Reaction mechanisms, product
136(16), 6083–6091. structures, and reaction design, J. Phys. Chem. C, 114(2),
18. Xia, Z. Y., Pezzini, S., Treossi, E., Giambastiani, G., Corticelli, 832–842.
F., Morandi, V., Zanelli, A., Bellani, V., Palermo, V., 2013. 39. Xu, S. C., Irle, S., Musaev, D. G., Lin M. C., 2006. Quantum
The exfoliation of graphene in liquids by electrochemical, chemical study of the dissociative adsorption of OH and H2O on
chemical, and sonication-assisted techniques: A nanoscale pristine and defective graphite (0001) surfaces: Reaction mech-
study, Adv. Funct. Mater., 23(37), 4684–4693. anisms and kinetics, J. Phys. Chem. C, 2, 111(3), 1355–1365.
19. Liu, J. L. et al. 2013. A green approach to the synthesis of high- 40. Mermin, N. D., Wagner, H., 1966. Absence of ferromagnetism
quality graphene oxide flakes via electrochemical exfoliation or antiferromagnetism in one- or two-dimensional isotropic
of pencil core, RSC Adv., 3, 11745–11750. Heisenberg models, Phys. Rev. Lett., 17, 1133–1136.
20. Sun, G., Li, X., Qu, Y., Wang, X., Yan, H., Zhang, Y., 2008. 41. Wang, C., Liu, Y., Lan, L., Tan, H., 2013. Graphene wrin-
Preparation and characterization of graphite nanosheets from kling: Formation, evolution and collapse, Nanoscale, 5,
detonation technique, Mater. Lett., 62(4–5), 703–706. 4454–4461.
21. McAllister, M. J. et al. 2007. Single sheet functionalized gra- 42. Shen, B., Zhai, W., Chen, C., Lu, D., Wang, J., Zheng, W.,
phene by oxidation and thermal expansion of graphite, Am. 2011. Melt blending in situ enhances the interaction between
Chem. Mater., 19(18), 4396–4404. polystyrene and graphene through π–π stacking, ACS Appl.
22. Stankovich, S., Dikin, D. A., Piner, R. D., Kohlhaas, A. K., Mater. Interfaces, 3, 3103–3109.
Kleinhammes, A., Jia, Y., Wu, Y., Nguyen, S. T., Ruoff R. 43. Stankovich, S., Piner, R. D., Chen, X., Wu, N., Nguyen, S. T.,
S., 2007. Synthesis of graphene-based nanosheets via chemi- Ruoff, R. S., 2006. Stable aqueous dispersions of graphitic
cal reduction of exfoliated graphite oxide, Carbon, 45(7), nanoplatelets via the reduction of exfoliated graphite oxide
1558–1565. in the presence of poly(sodium 4-styrenesulfonate), J. Mater.
23. Lu, X., Yu, M., Huang, H., Ruoff, R. S., 1999. Tailoring Chem., 16, 155–158.
graphite with the goal of achieving single sheets, Anglais, 10, 44. Kosynkin, D. V., Higginbotham, A. L., Sinitskii, A., Lomeda,
269–272. J. R., Dimiev, A., Price, B. K., Tour, J. M., 2009. Longitudinal
24. Torres, L., Armas, L. G., Seabra, A. C., 2014. Optimization unzipping of carbon nanotubes to form graphene nanorib-
of micromechanical cleavage technique of natural graphite by bons, Nature, 458, 872–876.
chemical treatment, Graphene, 3, 1–5. 45. Sinitskii, A., Dimiev, A., Corley, D. A., Fursina, A. A.,
25. Al-Amoudi, A., Dubochet, J., Gnaegi, H., Lüthi, W., Studer, Kosynkin, D. V., Tour, J. M., 2010. Kinetics of diazonium
D., 2003. An oscillating cryo-knife reduces cutting-induced functionalization of chemically converted graphene nanorib-
deformation of vitreous ultrathin sections, J Microsc (UK), bons, ACS Nano, 4, 1949–1954.
212(1), 26–33. 46. Kang, S. M., Park, S., Kim, D., Park, S., Y., Ruoff, R., S., Lee,
26. Jayasena, B., Subbiah, S., 2011. A novel mechanical cleavage H., 2011. Simultaneous reduction and surface functionaliza-
method for synthesizing few-layer graphenes, Nanoscale Res. tion of graphene oxide by mussel-inspired chemistry, Adv.
Lett., 6(95), 1–7. Funct. Mater., 21, 108–112.
27. Pierson, H. O., 1994. Handbook of Carbon Graphite Diamond 47. Xu, Z., Gao, C., 2010. In situ polymerization approach to
and Fullerenes Properties Processing and Applications, graphene-reinforced nylon-6 composites macromolecules,
Vol. 3, p. 43. Macromolecules, 43, 6716–6723.
28. Meng, L. Y., Park, S. J., 2012. Preparation and characteriza- 48. Wang, X., Hu, Y., Song, L., Yang, H., Xing, W., Lu, H., 2011.
tion of reduced graphene nanosheets via pre-exfoliation of In situ polymerization of graphene nanosheets and polyure-
graphite flakes, Bull. Korean Chem. Soc., 33(1), 209–214. thane with enhanced mechanical and thermal properties,
29. Brodie, B. C., 1860. Sur le poids atomique du graphite [On the J. Mater. Chem., 21, 4222–4227.
atomic weight of graphite.], Ann. Chim. Phys. 59, 466. 49. Liu, T., Zhao, Z., Weei Tjiu, W., Lv, J., Wei, C., 2014.
30. Staudenmaier, L., 1898. Verfahren zur Darstellung der Preparation and characterization of epoxy nanocomposites
Graphitsaure [Process for the representation of the graphitic containing surface-modified graphene oxide, J. Appl. Polym.
acid.], Ber. Dtsch. Chem. Ges., 31, 1481–1499. Sci., 131, 40236–40242.
31. Hummers, W. S., Jr., Offeman, R. E., 1958. Preparation of gra- 50. Lomeda, J. R., Doyle, C. D., Kosynkin, D. V., Hwang, W. F.,
phitic oxide, J. Am. Chem. Soc., 80, 1339–1339. Tour, J. M., 2008. Diazonium functionalization of surfactant-
32. Bartlett, N., Mcquillan, B. W., 1982. Graphite chemistry. In wrapped chemically converted graphene sheets, J. Am. Chem.
Intercalation Chemistry. Eds. A. J. Jacobson, Whittingham, Soc., 130, 16201–16206.
S. M., pp. 25–27. Academic Press: New York. 51. Zhu, Y., Higginbotham, A. L., Tour, J. M., 2009. Covalent
33. Bourlinos, A. B., Gournis, D., Petridis, D., Szabo, T., Szeri, functionalization of surfactant-wrapped graphene nanorib-
A., Dekany, I., 2003. Graphite oxide: Chemical reduction bons, Chem. Mater., 21, 5284–5291.
52. Avinash, M. B., Subrahmanyam, K. S., Sundarayya, Y., via surface RAFT-mediated miniemulsion polymerization,
Govindaraju, T., 2010. Covalent modification and exfoliation Polym. Sci. Part A: Polym. Chem., 49, 1621–1632.
of graphene oxide using ferrocene, Nanoscale, 2, 1762–1766. 69. Beckert, F., Friedrich, C., Thomann, R., Mülhaupt, R., 2012.
53. Huang, Y., Qin, Y., Zhou, Y., Niu, H., Yu, Z. Z., Dong, J. Y., Sulfur-functionalized graphenes as macro-chain-transfer
2010. Polypropylene/graphene oxide nanocomposites prepared and RAFT agents for producing graphene polymer brushes
by in situ Ziegler − Natta polymerization, Chem. Mater., 22, and polystyrene nanocomposites, Macromolecules, 45(17),
4096–4102. 7083–7090.
54. Xu, Y., Liu, Z., Zhang, X., Wang, Y., Tian, J., Huang, Y., 70. Jiang, K., Ye, C., Zhang, P., Wang, X., Zhao, Y., 2012. One-
Ma, Y., Zhang, X., Chen, Y., 2009. A graphene hybrid mate- pot controlled synthesis of homopolymers and diblock copo-
rial covalently functionalized with porphyrin: Synthesis and lymers grafted graphene oxide using couplable RAFT agents,
optical limiting property, Adv. Mater., 21, 1275–1279. Macromolecules, 45, 1346−1355.
55. Liu, Z., Robinson, J. T., Sun, X., Dai, H., 2008. PEGylated 71. Matyjaszewski, K., Xia, J., 2001. Atom transfer radical polym-
nanographene oxide for delivery of water-insoluble cancer erization, J. Chem. Rev., 101, 2921–2990.
drugs, J. Am. Chem. Soc., 130, 10876–10877. 72. Braunecker, W. A., Matyjaszewski, K., 2007. Controlled/
56. Mohanty, N., Berry, V., 2008. Graphene-based single-bac- living radical polymerization: Features, developments, and
terium resolution biodevice and DNA transistor: Interfacing perspectives, Prog. Polym. Sci., 32, 93–146.
graphene derivatives with nanoscale and microscale biocom- 73. Kamigaito, M., Ando, T., Sawamoto, M., 2001. Metal-catalyzed
ponents, Nano Lett., 8, 4469–4476. living radical polymerization, Chem. Rev., 101, 3689–3745.
57. Quintana, M., Spyrou, K., Grzelczak, M., Browne, W. R., 74. Badri, A., Whittaker, M. R., Zetterlund, P. B., 2012. Modification
Rudolf, P., Prato, M., 2010. Functionalization of graphene via of graphene/graphene oxide with polymer brushes using con-
1,3-dipolar cycloaddition, ACS Nano, 4, 3527–3533. trolled/living radical polymerization, J. Polym. Sci. Part A:
58. Zhang, X., Hou, L., Cnossen, A., Coleman, A. C., Ivashenko, Polymers, 50, 2981–2992.
O., Rudolf, P., van Wees, B. J., Browne, W. R., Feringa, B. L., 75. Fang, M., Wang, K. G., Lu, H. B., Yang, Y. L., Nutt, S., 2009.
2011. One-pot functionalization of graphene with porphyrin Covalent polymer functionalization of graphene nanosheets
through cycloaddition reactions, Chem. Eur. J., 17, 8957–8964. and mechanical properties of composites, J. Mater. Chem.,
59. Yang, H., Shan, C., Li, F., Han, D., Zhang, Q., Niu, L., 2009. 19, 7098–7105.
Covalent functionalization of polydisperse chemically con- 76. Fang, M., Wang, K. G., Lu, H. B., Yang, Y. L., Nutt, S., 2010.
verted graphene sheets with amine-terminated ionic liquid, Single-layer graphene nanosheets with controlled grafting of
Chem. Commun., 26, 3880–3882. polymer chains, J. Mater. Chem., 20, 1982–1992.
60. Shan, C., Yang, H., Han, D., Zhang, Q., Ivaska, A., Niu, L., 77. Goncalves, G., Marques, P., Barros-Timmons, A., Bdkin, I.,
2009. Water-soluble graphene covalently functionalized by Singh, M. K., Emami, N., Gracio, J., 2010. Graphene oxide
biocompatible poly-l-lysine, Langmuir, 25, 12030–12033. modified with PMMA via ATRP as a reinforcement filler,
61. Ren, L., Huang, S., Zhang, C., Wang, R., Tjiu, W. W., Liu, T., J. Mater. Chem., 20, 9927–9934.
2012. Functionalization of graphene and grafting of tempera- 78. Yang, Y., Wang, J., Zhang, J., Liu, J., Yang, X., Zhao, H., 2009.
ture-responsive surfaces from graphene by ATRP “on water”, Exfoliated graphite oxide decorated by PDMAEMA chains
J. Nanopart. Res., 14(6), 1–9. and polymer particles, Langmuir, 25(19), 11808–11814.
62. Remyamol, T., John, H., Gopinath, P., 2013. Synthesis and 79. Bak, J. M., Lee, T., Seo, E., Lee, Y., Jeong, H. M., Kim, B. Lee,
nonlinear optical properties of reduced graphene oxide cova- H., 2012. Thermoresponsive graphene nanosheets by func-
lently functionalized with polyaniline, Carbon, 59, 308–314. tionalization with polymer brushes, Polymer, 53, 316–323.
63. Pham, V. H., Cuong, T. V., Hur, S. H., Oh, E., Kim, E. J., Shin, 80. Li, G. L., Liu, G., Li, M., Wan, D., Neoh, K. G., Kang, E. T.,
E. W., Chung, J. S., 2011. Chemical functionalization of gra- 2010. Organo- and water-dispersible graphene oxide-polymer
phene sheets by solvothermal reduction of a graphene oxide nanosheets for organic electronic memory and gold nanocom-
suspension in N-methyl-2-pyrrolidone, J. Mater. Chem., 21, posites, J. Phys. Chem. C, 114, 12742–12748.
3371–3377. 81. Pramoda, K. P., Hussain, H., Koh, H. M., Tan, H. R., He, C. B.,
64. Zhang, B., Chen, Y., Xu, L. Q., Zeng, L. J., He, Y., Kang, 2010. Covalent bonded polymer–graphene nanocomposites,
E. T., Zhang, J., 2011. Growing poly(N-vinylcarbazole) from J. Polym. Sci., Part A: Polym. Chem., 48, 4262–4267.
the surface of graphene oxide via RAFT polymerization, 82. Deng, Y., Li, Y. J., Dai, J., Lang, M. D., Huang, X. Y. J., 2011.
J. Polym. Sci. Part A: Polym. Chem., 49, 2043–2050. Functionalization of graphene oxide towards thermo-sensitive
65. Yang, Y., Song, X., Yuan, L., Li, M., Liu, J., Ji, R., Zhao, H., nanocomposites via moderate in situ SET-LRP, Polym. Sci.
2012. Synthesis of PNIPAM polymer brushes on reduced gra- Part A: Polym. Chem., 49, 4747–4755.
phene oxide based on click chemistry and RAFT polymeriza- 83. Chen, X., Yuan, L., Yang, P., Hu, J., Yang, D., 2011. Covalent
tion, J. Polym. Sci. Part A: Polym. Chem., 50, 329–337. polymeric modification of graphene nanosheets via surface-
66. Layek, R. K., Kuila, A., Chatterjee, D. P., Nandi A. K., 2013. initiated single-electron-transfer living radical polymeriza-
Amphiphilic poly(N-vinyl pyrrolidone) grafted graphene by tion, J. Polym. Sci. Part A: Polym. Chem., 23, 4977–4986.
reversible addition and fragmentation polymerization and the 84. García-Valdez, O., Ledezma-Rodríguez, R., Saldívar-Guerra,
reinforcement of poly(vinyl acetate) films, J. Mater. Chem. A, E., Yate, L., Moya, S., Ziolo, R. F., 2014. Graphene oxide mod-
1, 10863–10887. ification with graft polymers via nitroxide mediated radical
67. Nikdel, M., Salami-Kalajahi, M., Salami Hosseini M., 2014. polymerization, Polymer, 55(10), 2347–2355.
Synthesis of poly(2-hydroxyethyl methacrylate-co-acrylic 85. Kan, L., Xu, Z., Gao, C., 2011. General avenue to individually
acid)-grafted graphene oxide nanosheets via reversible addi- dispersed graphene oxide-based two-dimensional molecular
tion–fragmentation chain transfer polymerization, RSC Adv., brushes by free radical polymerization, Macromolecules, 44,
4, 16743–16750. 444–452.
68. Etmimi, H. M., Tonge, M. P., Sandeson, R. D., 2011. Synthesis 86. Mohamadi, S., Sharifi-Sanjani, N., Mahdavi, H., 2011.
and characterization of polystyrene-graphite nanocomposites Functionalization of graphene sheets via chemically grafting
of PMMA chains through in-situ polymerization, Micromole. 90. Whitby, R., Korobeinyk, A., Glevatska, K. V., 2010.
Sci., Part A: Pure and Appl. Chem., 48, 577–582. Morphological changes and covalent reactivity assessment of
87. Feng, R., Zhou, W., Guan, G., Li, C., Zhang, D., Xiao, Y., single-layer graphene oxides under carboxylic group-targeted
Zhenga, L., Zhu, W., 2012. Surface decoration of graphene by chemistry, Carbon, 49, 718–736.
grafting polymerization using graphene oxide as the initiator, 91. Dreyer, D. R., Park, S., Bielawski, C. W., Ruoff, S. R., 2010. The
J. Mater. Chem., 22, 398. chemistry of graphene oxide, Chem. Soc. Rev., 39, 228–240.
88. Shen, J., Hu, Y., Li, C., Qin, C., Ye, M., 2009. Synthesis of 92. Sreeprasad, T. S., Berry, V., 2013. How do the electrical prop-
amphiphilic graphene nanoplatelets, Small, 5, 82–85. erties of graphene change with its functionalization? Small,
89. Beckert, F., Rostas, A. M., Thomann, R., Weber, S., Schleicher, 9(3), 341–350.
E., Friedrich, C., Mülhaupt, R., 2013. Self-initiated free radi-
cal grafting of styrene homo- and copolymers onto functional-
ized graphene, Macromolecules, 46, 5488−5496.
22 Strain and Screening Effects
Charge Carrier Mobility in Graphene
Tariq M. G. Mohiuddin, Raheel Shah, and Ram N. Singh
CONTENTS
Abstract...................................................................................................................................................................................... 435
22.1 Introduction...................................................................................................................................................................... 435
22.2 Electronic Properties........................................................................................................................................................ 436
22.3 Scattering Mechanisms and Graphene Transport Equation............................................................................................. 437
22.3.1 Boltzmann Transport Equation............................................................................................................................ 437
22.3.2 Relaxation-Time Approximation.......................................................................................................................... 438
22.4 Impact of Impurity, SR, and Acoustic Phonons on Scattering Probability...................................................................... 438
22.4.1 Remote Impurity Interaction................................................................................................................................ 438
22.4.2 SR Interaction....................................................................................................................................................... 439
22.4.3 Phonon Interaction................................................................................................................................................ 439
22.4.4 Numerical Results................................................................................................................................................. 440
22.5 Graphene Unit Cell and Effective Fermi Velocity........................................................................................................... 440
22.5.1 Unit Cell Representation....................................................................................................................................... 440
22.5.2 Fermi Velocity of Strained Graphene................................................................................................................... 441
22.6 Mobility for Different Dielectric Constants and Strained Graphene............................................................................... 442
22.7 Summary and Conclusions............................................................................................................................................... 444
References.................................................................................................................................................................................. 445
ABSTRACT 2D (two-dimensional) form and would curve up to form soot,

nanotubes, or fullerenes. However, this theoretical material
Electron mobility in a strained graphene sheet is investigated turned into reality when isolated free-standing graphene was
within the Born approximation by including three prominent found in 2004 [2,3] (Figure 22.1).
scattering mechanisms, namely, charged impurity scattering, Since its discovery, the most explored area in the study of
surface roughness (SR) scattering, and interaction with lattice the 2D graphene sheet is its electronic properties. Recently, a
phonons. The unsymmetrical hopping parameters between whole new wealth of information about graphene has emerged
the nearest neighbor atoms, which emanate from the induced following the reports on its mechanical and other nonelec-
strain, are included in the density of states description. tronic properties. Graphene is the strongest material ever
Mobility dependence on chirality along zigzag and armchair measured with a breaking strength of ~40 N/K and Young’s
orientations is also studied. The static dielectric screening for modulus ~1.0 TPa [4]. Its thermal conductivity has been
ionized impurity and SR scattering is included in the simula- reported at a record value of ~5000 W/m K [5]. Moreover,
tions. For the SR topology, we have examined graphene with graphene was shown capable of withstanding the reversible
the base substrate as dimethyl sulfoxide and SiO2. Unlike its strain of as high as 20% [4]. Additionally, the strain is a way
strained silicon counterpart, simulations reveal a significant forward toward bandgap engineering, which is essential to
drop in graphene’s mobility with increasing strain. Mobility convert graphene into semiconductor [6,7]. Collectively, these
anisotropy is confirmed along the principal orientations of characteristics increase graphene’s potential as a material
zigzag and armchair. Within the framework of this study of choice for nano-electromechanical systems (NEMS) and
the prime reason for the drop in mobility is attributed to the sensory device applications. Within these emerging novel
change in Fermi velocity due to the strain-induced distortions properties, the strain is considered very crucial as it prom-
in the graphene honeycomb lattice. ises whole new prospects for studying electronic transport
[8]. This is an idea being researched extensively theoreti-
cally [69–17]. Although fabricating an electronic device to
22.1 INTRODUCTION
extract Hall or field effect mobility in graphene while apply-
Until recently, it was widely thought that graphene is a theo- ing strain is a massive technological challenge, with efforts
retical material and considered too unstable to exist in the free such as in Reference 18, it does not seem long before it takes
state [1]. It was argued that graphene would be unstable in its shape. In the meantime, however, several new nonelectronic
435
TABLE 22.1
Summary of Previous Studies
Topic References
Graphene formalism/structure [1,3,9,19–21,30]
Electronic transport properties in graphene [2,7,13,31–55]
Strain in graphene [4,6,10–12,14–18,
22–25,27,29,56]
Thermal conductivity of graphene [5]
FIGURE 22.1 Graphene (left) is the basic building block of all Reviews articles [8,26,28]
graphitic forms; it can be stacked to form 3D graphite (right). Semiconductor physics and carrier transport [57–61]
Surface roughness in semiconductors [62,63]
characteristics of graphene have emerged when studied under

controllable applied strain [18–25].
22.2 ELECTRONIC PROPERTIES
The 2D nature of graphene means that strain laterally
propagates within a layer without diminishing. Coupling this Graphene is one of the most promising materials today. Its
with its high resistance to both elastic deformation and break- exceptional electronic properties are thought to make it
age [4,20], additionally, its high electronic quality [2,8,26] an ideal replacement of silicon in the electronics industry.
provides motivation to explore changes in its electronic prop- However, to reach this, a significant improvement in its prop-
erties when put under strain. A further encouragement to this erties compared to existing semiconductors is required. In
fact comes from theoretical reports on opening of a bandgap in particular, high charge carrier mobilities are a crucial require-
graphene for strains higher than 20% [6] and crystallographic ment for the commercial application. On the other hand, low
dependence of strain [27]. The importance of such a result for mobility is also detrimental for material research hindering
prospects of future graphene applications adds to our interest possible new important effects.
in further understanding its behavior under applied strain. Mobility “μ” of the charge carriers in graphene is expected
In this study, we involve the complete spectrum of the typi- to be extremely large compared to that of conventional semicon-
cal nearest neighbor hopping parameters in the tight-binding ductors. For example, low-temperature μ in metal oxide semi-
(TB) description of the density of states (DoS) for a graphene conductor field effect transistor (MOSFET) have been reported
lattice. The former is a set of three parameters, which are at ~1 m 2/V s, which are the same as in graphene at room
reduced to one due to symmetry considerations [28], which is temperature and, even much larger values are expected [31].
not applicable to graphene under strain. In the latter case, the Since the time of its discovery in 2004, no significant improve-
unit cell lattice vectors are strain dependent, and they carry ment in μ has been reported for graphene on substrates. This
this dependency into the nearest neighbor hopping param- fact is widely associated with scattering of free charge carriers
eter description [6]. This enables one to describe the DoS by charged impurities in a substrate. Here, we discuss the role
and ultimately the Fermi velocity as a function of strain [29]. of charged impurities scattering as the dominant source limit-
Considering three scattering mechanisms, namely, remote ing μ in graphene. μ in graphene can be determined from the
impurity, phonon, and surface roughness (SR), we determine gate electric field effect measurement [2,32,33] as
the Fermi velocity in the close vicinity of the Dirac point, and
ultimately the conductivity-dependent mobility of the charge σ(Vg )
µ= (22.1)
carriers. Our results show that for applied strain along the spe- en(Vg )
cial crystallographic orientations of zigzag (Z) and armchair
(A), the mobility of charge carriers is particularly degraded in
the interval 20% ≤ η ≥ 10%, the latter being the limit of strain where σ is the conductivity, n is the charge carrier concentra-
used in our simulations. The reduction is more prominent in tion, which is linearly dependent on the gate voltage Vg, and
the Z direction to that in the A. Temperature-dependent inves- e is the electron charge. It was also shown that 1/R H linearly
tigations revealed a significant drop in mobility midway of the depends on Vg [33] which yields n = αVg, where α is the sur-
abovementioned range of strain only in the Z direction. face charge density induced by the field effect. Its value was
This chapter is organized into seven sections as follows: found to be ≈7.3 × 1010 cm2/V, which is closely in agreement
Section 22.1 is the introduction; in Section 22.2, electronic with the theoretical estimation of 7.2 × 1010 cm2/V deter-
properties are discussed; Section 22.3 deals with scattering mined from the relation
mechanisms and the graphene transport equation; Section 22.4
discusses impact of impurity, SR, and acoustic phonon on scat- ε 0 εVg
n= (22.2)
tering probability; Section 22.5 discusses graphene unit cell te
and fermi velocity; Section 22.6 deals with mobility for dif-
ferent dielectric constants and strained graphene; and finally, where ε0 and ε are the permittivities of free space and the
followed by the summary and conclusion in Section 22.7. substrate, respectively, and t is the thickness of the substrate
Charge Carrier Mobility in Graphene 437
(i.e., distance from the gate to the graphene layer). In case Consequently, ρ in Equation 22.5 becomes
of graphene deposited on an oxidized silicon wafer, ε is the
dielectric constant of SiO2 and t is its thickness. m*  1 1 
Using Equation 22.2, one can determine μ for both elec- ρ=  +  = ρS + ρL (22.8)
ne2 〈 τ 〉 〈 τL〉

 S 
trons and holes. It was found [33] that the mobility is indepen-
dent of temperature T between 10 and 100 K.
The linear dependence of n on Vg implies a long-range scat- Here, the total resistivity is a combination of the compo-
tering potential. Further, for a one-atom thick system, which nents arising from short-range scatterers ρS and long-range
is unprotected from the immediate environment, charged scatterers ρL [31]. It was found that <τL> ~ n1/2 and <τS> ~ n−1/2
impurities seem to be the most natural candidates for being [38] and hence it can be deduced that at low n graphene is
the dominant scatterers [34–37]. a clean ballistic system for short-range scattering but a dirty
Electrons experience scattering whilst moving in an elec- system for long-range scatterers. In other words, at low n (or
tronic system. The average time between scattering events or low Vg), the short-range scattering plays no significant role but
the scattering rate can be given by the average value of the the long-range scattering (or charged impurity) completely
relaxation time <τ> and the relationship with mobility μ is dominates. A number of publications supported this fact, and
given as no evidence was put forward to prove the contrary. Graphene
was controllably doped with potassium where mobility μ was
found to decrease consistent with theoretical predictions [36].
e〈 τ 〉
µ= (22.3) Moreover, the measurement carried out on suspended gra-
m* phene yielded ultrahigh mobilities [31,39].
It has been pointed out that in addition to charged impurities
Here, m* = pF/vF, where pF and vF are the Fermi momentum there could be another possible factor playing the defining role
and velocity, respectively. in the scattering mechanisms taking place in graphene. Ripples
This relation can be extended to obtain the conductivity (microscopic corrugations) on a graphene sheet, which are an
σ as inherent feature of graphene due to its 2D nature, can also be a
dominant factor in determining the charge carrier mean free path.
ne2 〈 τ 〉 A certain type of ripples create a long-range scattering potential
σ= = µne (22.4) such as charge impurities and resulting in μ independent of n,
m* in agreement with experimental observations [64]. These results
definitely offer a clear alternative to the claim that charge impu-
The resistivity ρ is then rity scattering is the major factor limiting graphene μ.
1 m* 1 22.3 SCATTERING MECHANISMS AND

ρ= = 2 (22.5)
µne ne 〈 τ 〉 GRAPHENE TRANSPORT EQUATION
Graphene is endowed with unusual structural and electronic
In an electronic system, charge carriers may go through flexibility. One of the most interesting aspects is that its low
various types of scattering mechanisms. In the case of an energy excitations are massless chiral Dirac fermions. The
independent scattering event, the total relaxation time can be latter moves with a speed νF that is about 300 times smaller
written as than the speed of light c. Dirac fermions behave in unusual
ways when compared to ordinary electrons. There has been
∑τ
1 1 a great deal of interest in trying to understand how: (i) impu-
= (22.6)
τ i rities and defects, (ii) intrinsic deviations from the periodic-
i ity, and (iii) thermal vibrations, affect its transport properties.
Under certain conditions, it has been observed that the elec-
This assumes that the total scattering probability is the sum trons can propagate without scattering over large distances.
of individual scattering probabilities in a system. There are Therefore, the conductivity does not exhibit a saturation and
two scattering mechanisms, which often dominate; acoustic increases almost in proportion to the electron concentration
phonon scattering, which is a short-range scattering and the [32,33]. Charge impurities are believed to be the possible can-
charged impurity scattering considered to be of long range. didate for the dominant scattering, which is greatly influenced
It can be assumed, based on this fact that the total scatter- by screening effect. In the following, we concentrate on the
ing probabilities will be a combination of the long- <τL> and basic ingredients associated with electronic properties.
short- <τS> range scattering probability, which allows us to
write [57]
22.3.1 Boltzmann Transport Equation
1 1 1 (22.7) Boltzmann transport equation is quite a realistic description
≅ +
〈τ〉 〈τS 〉 〈τ L 〉 of the electron scattering in terms of the scattering probability,
W(k, k′) (that an electron with wave vector k suffers some col- wave vector k will suffer some collision. This quantity is, by
lisions and scatters into a new wave vector k′), and the distri- definition, just 1/τk. The total probability per unit time for a
bution function, f k. The representation of the nonequilibrium collision is given by summing over all final wave vector k′.
distribution function g(r, k, t) is complex. One can, however, It can be readily shown that [58]
construct, f, at a time t from its value an infinitesimal time dt
∑ W (k,k′)(1 − cos θ)
earlier. If we, first, ignore the possibility of collisions, then r 1
= (22.14)
and k coordinates of every electron would involve according τk
to the equation of motion: k′
dK It is quite clear that small angle scattering makes a very

F ( r, k ) = = −e(E + v × B /c) (22.9) small contribution to the effective collision rate. If the colli-
dt sions were only in the forward direction, then they would have
no consequence at all. For small changes in the wave vector,
Here, E and B are electric and magnetic fields, respec-
the distribution of electronic wave vectors will be only slightly
tively. Since dt is infinitesimal, we can find the explicit solu-
affected by the collisions. Forward scattering will contribute
tion to this equation to linear order in dt:
less to the effective collision rates than wide angle scattering
unless a property is being measured that depends delicately
f (r, k, t ) = f (r − v(k )dt; k − Fdt /; t − dt ) (22.10)
on the precise direction in which certain electrons move.
If the collisions occur, then the distribution function is

affected due to scattering in two ways: (i) electrons may scat- 22.4 IMPACT OF IMPURITY, SR, AND
ter out of the level k, and (ii) electrons may scatter into it from ACOUSTIC PHONONS ON SCATTERING
other levels. After expanding the above equation to linear PROBABILITY
order in dt in the limit dt → 0, and including the scattering
The quantum mechanical scattering probability W(k, k′) can
terms, one has
be computed within Born approximation as
eE c ∂f c 2π n 2
vk ⋅∇ r fkc − ⋅∇ k fkc = − k (22.11) W ( k, k ′ ) = V δ( Ek ′ − Ek ∓ ∆E ) (22.15)
∂t k′,k
scatt
which is the celebrated Boltzmann transport equation in the The Dirac delta function ensures the conservation of
absence of the magnetic field. vk is the charge carrier’s veloc- energy and ΔE is the change in energy, if any. W (k, k′) is
ity and Ec is the applied electric field. independent of the electronic distribution function gk by con-
sequence of the independent electron approximation. The
22.3.2 Relaxation-Time Approximation matrix element |Vkn′,k | is defined as
The electric field-dependent distribution function fkc can be 1

expanded around the equilibrium value f k such that
Vkn′,k =
A ∫
ψ ∗k ′ (r )Vsn (r )ψ k (r )dr

(22.16)
fkc = fk + gk (22.12) where ψk(r) is the wave-function associated with charge car-

rier and Vsn (r ) is the perturbation potential of the nth type
where gk is assumed to be a weak perturbation to the f k. responsible for scattering in the system of area A.
Hence, within the relaxation time approximation (RTA), the
scattering term −(∂fkc /∂t ) |scatt can be approximated as
22.4.1 Remote Impurity Interaction
∂fkc f k + gk gk Charge impurities present in the substrate are one of the signif-
− = ≈ (22.13) icant sources of mobility degradation in graphene [34,35,40].
∂t τk τk
scatt In k-space, the Coulomb scattering potential VsC (r ) is given by
where τk is the relaxation time. It is related to the scattering
2πe2e − dq
probability, W (k, k′), which defines the probability in an VsC (q) = (22.17)
infinitesimal time dt that an electron with wave vector k is κq
scattered into any one of the group of levels (with the same
spin) contained in the infinitesimal k-space volume element where charged impurities in the substrate are assumed to be at
dk′ about k′. a distance d away from the graphene sheet, the scattered wave
Given the quantity W (k, k′) and gk, one can explicitly vector q = |k − k′| = 2k sin θ/2 and k is the effective dielectric
construct the probability per unit time that an electron with constant of the system. The inclusion of screening effect is
vital in observing the impact of charge impurity. In particular, side of the interface is defined as Eeff = (e /ε g )(ni + ns ). z0 is
for graphene, the screened Coulomb potential results in the taken as Δ of the interface.
matrix element as The mismatch of the dielectric constants at the interface
also introduces image charges [43]. For sufficiently thick sub-
VsC (q) ∗ strate, the scattering potential associated with image charges
VkC′,k = (ψ k ′ ⋅ ψ k ) (22.18) is given by [63]
ε (q )
e2 ε q 2  K1 (qz0 ) ε 
where ε(q) is the static dielectric function for a 2D graphene Vsimg (q) = − K 0 (qz0 ) (22.24)
sheet. Starting from the Lindhard function and under ran- 16πε g  qz0 2 

dom phase approximation (RPA) [41] ε(q) is given in terms of
polarization function Π(q) as [35] where K0 and K1 are modified Bessel functions of the second
kind of order zero and one, respectively. Net impact of these
2πe2 two scattering sources is thus given by
ε (q ) = 1 + Π (q ) (22.19)
κq
VsSR (q) = Vspol (q) + Vsimg (q) (22.25)

∞ q/2 2
gv gs  q
Π(q) =
2 π( vF ) ∫ fk+ dk −
∫ fk+ 1 −   dk
 2k 
(22.20) Note that the net scattering strength not only depends on
the dielectric properties of the substrate but more importantly
0 0
on the difference between the dielectric properties of the two
materials at the interface. SR-induced scattering potential
where fk+ = f ( Ek ) + f ( Ek + 2ζ ) with f(Ek) is assumed to be thus yields the scattering rate as
the Fermi–Dirac function.
Finally, the scattering rate for the Coulomb interaction is
1 2 π  D( E ) 
the presence of ni charge carriers per area is then derived as =
τ SR  2π 
2π 2
1 2π  ni D( E )   1 − cos2 θ   VsC (q)  2π 2
τ Col
=
 2π  ∫  2   ε(q)  dθ (22.21)
∫
 1 − cos2 θ   VsSR (q) 
× 
 2   ε(q)  | S (q) | dθ
2
(22.26)
0 0
22.4.2 SR Interaction 22.4.3 Phonon Interaction

As with any surface and/or interface, graphene deposited on For ambient and higher temperature regime, carrier–pho-
the substrate has a non-smooth surface. The impact of SR on non interaction is a major cause of mobility degradation in
charge carrier mobility is extensively studied for quasi-2D electronic devices [59]. Low energy acoustic phonons are
structures [42,43,62,63] to name but a few. In this work, the treated under elastic scattering approximation. It is reported
impact due to the difference in the dielectric values at the that group symmetry forbids TA (transverse acoustic) phonon
graphene/substrate interface is taken into account. The inter- modes to exist for graphene [40,44], therefore, only LA (lon-
face randomness is, as usual, modeled by the autocovariance gitudinal acoustic) mode of the spectrum is taken into account
function between Δ—the rms (root mean square) height of with relatively stronger coupling (see discussion below).
the random interface “steps” and Λ—the average width of the The scattering potential associated with acoustic phonons
same fluctuation. For an exponential autocorrelation form, the is given by [58,45]
power spectrum density |S(q)|2 is given by [62]
Dac kBT i ( q⋅r −ωt )
|S (q)|2 = πΛ 2 ∆ 2 (1 + q 2 Λ 2 / 2)−3 / 2 (22.22) Vsac (q) = e (22.27)
v ph 2ρg A

In the presence of two different dielectric materials at
the interface, polarization charges are created. The potential where Dac is the deformation potential of the graphene lattice
induced by such polarization is given by [63] and ρg is the surface density of the system. Phonon velocity
is denoted by vph, kB is the Boltzmann constant, and T is the
temperature in Kelvin’s scale. Here, linear phonon disper-
Vspol (q) = eε Eeff e − qz0 (22.23)
sion is assumed (i.e., acoustic phonon frequency ω qac = v phq)
and equipartition approximation is applied which is valid for
where the parameter ε = (ε g − ε sub ) / (ε g + ε sub ) with εg and εsub moderate to higher temperature regimes.
denote the dielectric constants of the graphene and the sub- Impact of screening on electron–phonon interaction is long
strate, respectively. Effective electric field Eeff at the graphene debated, however, via simulations the ineffectiveness of the
screening in case of electron–phonon scattering is concluded 1014

[60], following this, dynamical screening is not included here. εsub = 47 ε0
Bulk phonon
Finally, the acoustic phonon scattering rate is derived
through the evaluation of matrix element as 1013 Surface roughness
Scattering rate (s–1)

1 2π  Dac
2
kBT  1012 Impurity charges
= D( E ) (22.28)
τ ac
 8v phρg 
 2

1011
Nonpolar optical phonons are treated inelastically in the T = 300 K
simulations. Similar to acoustic phonon spectrum, only LO ns = 1012 cm–2
phonons contribute toward a scattering mechanism. The inter- 1010
0.0 0.1 0.2 0.3 0.4 0.5
action potential, in this case, is given by [58] Energy (eV)
FIGURE 22.2 Scattering rate for lattice phonon, impurity charges,

Vsop (q) = Dop and SR interaction at room temperature. For electron energy
2ρg Aω op
q
>0.15 eV optical phonon emission process is possible, and thus a
sharp increase in phonon scattering rate is observed. Coulomb scat-
 1 i
×  N q + ±  e i ( q ⋅ r − ωt ) (22.29) tering rate initially increases with energy and then drops for suffi-
 2 2 ciently high carrier’s energy. (From Shah, R. et al. Modern Physics

Letters B, 2013. 27(3): p. 1350021.)
where Nq is the Bose–Einstein distribution function, Dop [40,46] (and references within). Ignoring the anisotropy of
the deformation field associated with vibrating lattice sites, the deformation potential, a single constant value Dac ~20 eV
i = −1 is for phonon absorption (upper sign), and i = +1 is for is used in this work. Phonon velocity vph of the LA branch
emission process (lower sign). Optical phonon frequency is is set to 2.0 × 106 cm/s and graphene density is taken as
denoted byω op q and in the dispersion region of interest it bears ρg = 7.6 × 10−8 g/cm2 [47]. The deformation field constant Do
a constant value ωo, independent of the transferred wave vec- appearing in optical phonon model is assumed to have the
tor q. strength of 2.0 × 109 eV/cm while LO phonon is associated
The interaction potential leads to the scattering rate as with energy of ω o = 152meV [40].
Figure 22.2 shows, in the absence of any strain, the relative
scattering strength of the three scattering mechanisms studied
1 2π  Dop
2
 1 i
=    No + ±  here. Both the phonon and SR scattering rates tend to increase
τ op
 4ρgω o   2 2 with carrier’s energy. At low energy, impurity scattering is
higher than the acoustic phonons and SR. For charge carriers
× D( E ± ω o )Θ( E ± ω o ) (22.30)
with sufficiently high energies, the momentum dissipation due
to ionized charge centers is less thus a decrease in the scatter-
where the Heaviside step function Θ(x) is introduced to ing rate for Coulomb interaction is observed.
account for only physically possible scattering events.
22.5 GRAPHENE UNIT CELL AND
22.4.4 Numerical Results EFFECTIVE FERMI VELOCITY
For the Coulomb interaction the location of charge centers 22.5.1 Unit Cell Representation
in the substrate away from the interface is taken as d ≈ 1 nm
[34] while the charge density ni is assumed to be around In equilibrium, the distance between two adjacent carbon
1.5 × 1011 cm−2. atoms is ao = 1.42 Å. The two primitive lattice vectors are
Unless graphene surface topology is thoroughly investi- a = a(1,0) and b = a(−1/ 2, 3 / 2) , where a = 3a0 while the
gated, the autocorrelation of the step width Λ and rms step lattice vectors joining site A to site B (see Figure 22.4) are
height Δ will be used as fine tuning parameters to simulate denoted by δi (i = 1,2,3).
observed mobility. With the exponential auto covariance Graphene’s electro-magneto properties are sensitive to the
model used here, Λ and Δ parameters are taken as 1.0 and edge effects [48,49], in particular along the two prominent
0.5 nm, respectively. The relative dielectric constant of gra- directions, namely, zigzag (Z) and armchair (A), depicted
phene is assumed to be 5.7 [40]. in Figure 22.3. The coordinate axes can always be chosen
In the phonon interaction model, the value of acoustic such that x-axis is aligned along the Z orientation. Angle
deformation potential Dac is not settled yet. In literature, ϑ represents arbitrary vector directed in between Z and A
values as small as 4.75 eV to as large as 30 eV are quoted orientations.
Armchair The overlap of wave function between initial and final

states can easily shown to be [35]
δ2 t2
t3 1 iθ
δ3 δ1 (ψ ∗± k ⋅ ψ ± k ′ ) = (e + 1) (22.35)
2
A t1
Unit
Zigzag
cell B where θ is the angle between initial wave vector k and the
final scattered wave vector k′.
y
b In equilibrium conditions the dispersion relation E(k) near
a θ the Dirac point is given by utilizing Equations 22.31 and
x
22.32 as [30]
E = ±vF |k| (22.36)

FIGURE 22.3 Part of infinite honeycomb network showing the
unit cell with two atoms A and B per cell. Hopping parameters tis
and bond lengths δis are also shown. Lattice primitive vectors are Note at equilibrium t1 = t2 = t3 = 3.03 eV [35] and
denoted by a and b. Also shown are two distinctive directions in δ1 = δ2 = δ3 = ao.
the network, namely, zigzag and armchair. aO is the C–C distance.
x-Axis is aligned along zigzag orientation.
22.5.2 Fermi Velocity of Strained Graphene
In the TB model, the energy bands in terms of nearest Electric transport is affected by the number of available vacant
neighbor π orbital hopping integrals (ti, in Figure 22.1) is states in the system. In case of graphene, DoS defined as
given by [16]
∑ δ[E(k) − E(k ′)]

1
D( E ) =
E = ± t2 + t3e−ik⋅(a +b) + t1e−ik ⋅ b (22.31)
A
k′

where upper (+) and lower (−) signs are for conduction and can easily be shown to be
valence bands, respectively. k = (k x,k y) is the 2D wave vec-
g g E
tor associated with the charge carriers with energy E. First D( E ) = v s (22.37)
Brillouin zone (BZ) of the graphene unit cell is a hexagon 2π (vF )2
itself.
The π orbitals of the valence and conduction bands cross at where gv = 2 and gS = 2 are introduced as the valley (K and K′
two corners K and K′ of the BZ also known as Dirac points. points) and spin degeneracy, respectively.
Charge carriers near the Dirac point behave like massless In terms of hopping parameter tO and C–C distance aO, the
particles, also for an intrinsic graphene Fermi energy is zero Fermi velocity is given by νF = 3tO aO / 2, which turns out to
at these K points. be around 300 times less than the speed of light. Fermi veloc-
Hamiltonian H of π bands near a Dirac point is described ity can alternatively be defined in terms of the unit cell area
by the Dirac–Weyl equation [28,50]: AC as [28]
H Ψ ± k = v F ( σ x k x + σ y k y ) Ψ ± k = E Ψ ± k (22.32) 1 Ac × ρt
vF = (22.38)
2
where vF is the Fermi velocity that is independent of charge
carriers energy, σx and σy are Pauli spinors and is the where Ac = 3 3aO2 / 2 and ρt = 3tO2 . This form is useful for
reduced Planck’s constant. Eigenstates of Dirac–Weyl [28,50] the F forthcoming discussion.
If strain is applied then the lattice vectors δi’s are modified as
1 i( k⋅r )
Ψ± k =
A
e ψ ±k (22.33) δi = (I2 × 2 + η) ⋅ δio (22.39)

where δ io’s correspond to the relaxed lattice vectors within the
where A is the area of the system and r = (x,y) is the in-plane graphene unit cell. I2×2 is identity matrix of order 2, and the
vector with strain matrix η is given by [6]
 eiθk 
1 −1  k y   cos2 ϑ − ν sin 2 ϑ (1 + ν)cos ϑ sin ϑ 
ψ ±k =   ; θ k = tan   (22.34) η = η  (22.40)
2  ±1   kx 
(1 + ν)cos ϑ sin ϑ sin 2 ϑ − ν cos2 ϑ 

where η is the applied strain and v = 0.14 is the Poisson’s ratio strain, with a prominent split between Z and A orientations at
for graphene [16]. The deformation of lattice vectors δis mod- around η = 10%. Thus, it is expected that the dc conductiv-
ify the hopping parameters tis accordingly [6] ity and the corresponding mobility will degrade with induced
strain.
ti = toe3.37(|δi | / aO −1) (22.41)

22.6 MOBILITY FOR DIFFERENT DIELECTRIC
The two distinctive directions in a graphene layer, CONSTANTS AND STRAINED GRAPHENE
namely, Z and A corresponds to ϑ = 0 and π/6, respectively, The electric current J in terms of carrier velocity and the dis-
while ϑ = π/3 is the periodicity of the system in ϑ. It is also tribution function gk is given by [51]
reported for η < 20%, no change in bandgap is observed [6].
Simulations in this work will be bound to this limit.
∑ ev g
gv gs
DoS of the system is also altered by the induced strain [29]: J= k k (22.44)
A
k
g g E
D( E ) = v s (22.42)

π ( Ac′ρt′) with gk = eτ k vk ⋅ E cδ(ζ − vF k ) is the solution of the linear-
ized Boltzmann equation. ζ is the chemical potential in the
where Ac′ is the area of the deformed unit cell and system. Additionally, the effective scattering rate,
∑τ
1 1
(t + t22 + t32 ) − 2 (t14 + t24 + t34 )
2
ρt′ = 2 (22.43) = (22.45)

1
τ eff ( E ) (n)
(E )
n
Comparing Equation 22.42 with Equation 22.37, one may
is utilized to compute electric conductivity σe within Kubo–
define the “effective” Fermi velocity as vF′ = (1/) ( AC′ × ρt′) / 2
Greenwood transport formalism as
which reduces to Equation 22.38 in the condition η = 0. The
parameter ρt′ decreases monotonically with increasing strain
as shown in Figure 22.4. The area of the deformed hexagon on e2 vF2 ∫ ED( E )τ eff ( E )(−∂f /∂E )dE
σe = (22.46)
the other hand increases linearly [56]. Additionally, the sig- 2 ∫ E (−∂f /∂E )dE
nificant difference in ρt′ appears for the induced strain >10%
for both along Z and A orientations. The relatively sharper
And finally, the carrier mobility μ in terms of conductivity
drop in ρt′ compared with Ac′ results in the low effective Fermi
and carrier density ns is calculated as
velocity with increasing strain.
Fermi velocity simulations reveal an initial linear drop in
vF′ with increasing η (slope of ~−1.2 × 108 cm/s for η ≤ 10%) σe
µ= (22.47)
[56] consistent with recent studies [10] and then a relatively ens
sharp drop in vF′ for η > 10%. This behavior mimics exactly
the trend observed for the parameter ρt′ against the induced where f(E) is the Fermi–Dirac probability and τeff (E) is the
net scattering time of the charge carriers. The scattering rate
is computed by taking into account three dominant scatter-
16 ing mechanisms, namely, Coulomb scattering with impurity
Zigzag
Armchair charges, scattering with lattice vibrations, and charge car-
14 rier interaction with rough interface. Flexural out of plane
phonons are not taken into account as their effect is mostly
12
prominent in suspended graphene samples under vacuum at
ρt′ (eV2)
room temperature [52]. These flexural phonon’s content and

10
their coupling with charge carriers are significantly reduced
8
by direct contact of samples with base substrate [53], which is
the case under consideration here.
6 Another important mechanism, which limits the mobil-
ity in finite size graphene samples grown with CVD (chemi-
4 cal vapor deposition) technique, is the scattering with grain
0 5 10 15 20 boundaries [54]. For simulations purpose we, however, have
Strain (%) considered a pristine graphene sample, free from bulk defects,
with charge neutrality at 0 V. This is only to observe the
FIGURE 22.4 Plots of the parameter ρ′t with varying strain. impact of strain on charge carrier’s mobility.
relatively higher amount of charge impurities present in the

Symbols : Measured data
Line : Simulated data
dielectric could result in low mobilities.
4 × 104 εsub = 66 ε0
Interface roughness is not necessarily the same for each
selected substrate, in addition, theoretically, the difference
Mobility (cm2/Vs)
3 × 104
between εsub and εg alters the impact of SR provided, Λ
and Δ parameters are assumed to be the same for different
dielectrics. Under identical conditions, our simulations reveal
2 × 104 higher mobilities for graphene on silicon oxide (εsub = 3.9ε0)
εsub = 47 ε0
T = 300 K as compared to dimethyl sulfoxide (εsub = 47ε0), this is due to
1 × 104 the reduced scattering potential VsSR (q) associated with SR.
εsub = 189 ε0 Simulations are further performed with strain induced in
the graphene sample. Figure 22.6 depicts the results obtained
0
0.0 0.3 0.6 0.9 1.2 1.5 for four sets of induced strain, both along Z and A orienta-
Sheet density (1012 cm–2)
tions. The mobility is found to decrease with increasing
applied strain, and this can be attributed to the decrease in
the Fermi velocity resulting from the lattice distortions. For
FIGURE 22.5 Simulations are benchmarked against the measured
mobilities as a function of sheet density. (Experimental data are strain induced around 10%, the decrease in mobility along Z
from Shishir, R.S. and D.K. Ferry, Journal of Physics: Condensed and A orientations is similar but as η is increased to around
Matter, 2009. 21(23): p. 232204.) 20%, a larger impact along Z orientation is observed. As out-
lined in Section 22.5.2, the strain dependence appears in the
To benchmark the transport models and their respective Fermi velocity due to its relation to the strain-dependent area
parameters, simulations are performed and compared with of the deformed unit cell Ac′, and the parameter ρ′t (Equation
the reported measured/extracted mobilities in Reference 55. 22.43). Although the area of the deformed unit cell Ac′
Figure 22.5 shows the comparison of simulated and the mea- increases with strain (see Figure 22.5a) it is, however, the
sured data. Simulations are performed for the substrate with same for both orientations, but the parameter ρ′t along the Z
dielectric constant of 47ε0 at room temperature. As it can be direction is more sensitive to the applied strain for η > 10%.
seen, the modeled mobility reasonably follows the reported This results in a larger decrease in mobility along Z than the
trend and its magnitude. A orientation.
The inclusion of high dielectric (εsub) constant possibly Results reveal that the phonon and Coulomb limited
has diverse effects on carrier mobility. First, the strength of mobility decrease with increasing strain, however, SR limited
impurity charge centers is certainly reduced due to screening, mobility increases with growing strain. This eccentric behav-
and thus mobility is expected to increase. On the other hand, ior of SR, contrary to other scattering mechanisms, is also
possible soft optical phonon modes in the substrate and the reported earlier for the quasi-2D system [61]. The sag in the
(a) 4.0 × 104 (b) 4.0 × 104

Strain along A orientation
Strain along Z orientation
3.5 × 104 3.5 × 104
η = 5%
η = 5%
3.0 × 104 3.0 × 104
Mobility (cm2/Vs)
2.5 × 104
Mobility (cm2/Vs)
2.5 × 104
T = 300 K T = 300 K
2.0 × 104 2.0 × 104
1.5 × 104 1.5 × 104

η η=0 η = 10% η=0
=
10
1.0 × 104 % 1.0 × 104
5.0 × 103 5.0 × 103 η = 20%

η = 20%
0.0 0.0
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0
Gate voltage (V) Gate voltage (V)
FIGURE 22.6 Impact of strain on simulated mobilities against increasing gate voltage. Mobility degradation is prominent for
20% ≤ η ≥ 10%. (a) Z direction and (b) A direction. The base substrate is dimethyl sulfoxide (εsub = 47ε0).
(a) 3.2 × 104 (b) 3.2 × 104

Strain along Z orientation Strain along A orientation
77 K 77 K
Mobility (cm2/Vs) 2.8 × 104 2.8 × 104
Mobility (cm2/Vs)
300 K 300 K
1.2 × 104 1.2 × 104
400 K 400 K
8.0 × 103 8.0 × 103
ns = 1012 cm–2 ns = 1012 cm–2

4.0 × 103 4.0 × 103
0 5 10 15 20 0 5 10 15 20
Strain (%) Strain (%)
FIGURE 22.7 Strain-dependent simulated mobilities for low to high temperature regimes. The base substrate is SiO2 and sheet density is
1012 cm−2.
mobility profile (around ns ~ 4 × 1012 cm2) and for η = 20% is 50

the net effect of growing SR while declining impurity and Zigzag
phonon limited mobilities. Armchair
The parameter ρ′t is a combination of the hopping param-
40
Percentage drop in mobility
eters (which are in turn dependent on the bond lengths) in the

graphene lattice. A larger decrease in ρ′t with applied strain
along the Z direction would suggest stronger distortions in the
lattice structure for this orientation. 30
Figure 22.7 gives the mobility profiles as a function of
induced strain for low, moderate, and elevated temperature
range. The base substrate is SiO2 in this case. At low tem-
perature (77 K), the contribution of phonons in scattering 20
mechanism is negligible, and only impurity charges and SR
play dominant parts. In this low-temperature regime, mobility
nominally increases as the induced strain is increased, below 10
13%. This is due to the reason that the net scattering rate stays 0.0 0.4 0.8 1.2 1.6
almost constant for low T and η values but the Fermi dis- Sheet density (cm–2)
tribution derivative term (−∂f/∂E), appearing in conductivity
expression, increases relatively rapidly and thus the effective FIGURE 22.8 Change in charge carrier’s mobility for a 20%
conductivity shows a positive slope in this region. strained graphene sheet as a function of sheet density. (From Shah,
Further analysis shows that in going from an unstrained R., T. Mohiuddin, and R. Singh, Modern Physics Letters B, 2013.
state to a maximum strain of 20% (see Figure 22.8) the rela- 27(3): p. 1350021.)
tive change in mobility, as a function of sheet density can be
as high as 45% for the Z direction. This feature can be effec- including the charge impurity scattering, the SR, and the
tively utilized for strain-dependent sensory devices. acoustic phonon scattering. Additionally, the strain is likely
to affect the mobility and is a way forward toward bandgap
engineering that is essential to convert graphene into semi-
22.7 SUMMARY AND CONCLUSIONS
conductor devices. In the framework of Boltzmann transport
The exceptional electronic properties of graphene make it a equation, we formulated the transport equation for graphene
formidable replacement of silicon in the electronics indus- and investigated the impact of impurity, SR, and the effec-
try. High charge carrier mobilities is a crucial requirement tive phonon scattering. Both the phonon and the SR scattering
for ensuing commercial applications. Charge carriers in rates are found to increase with carrier energy. However, for
graphene undergo various types of a scattering mechanism the charge carriers with sufficient high energies, the scattering
rate decreases because the momentum dissipation due to ion- 16. Ribeiro, R.M. et al. Strained graphene: Tight-binding and
ized centers is less in this region. density functional calculations. New Journal of Physics, 2009.
Computed values of mobilities for substrates of different 11(11): p. 115002.
17. Viana-Gomes, J., V.M. Pereira, and N.M.R. Peres, Magnetism
dielectric constant agree reasonably well in magnitude and fol-
in strained graphene dots. Physical Review B, 2009. 80(24):
low the reported trend of some other available data. The inclu- p. 245436.
sion of high dielectric values is found to have diverse effects 18. Kim, K.S. et al. Large-scale pattern growth of graphene
on carrier mobility. Simulated results reveal a higher mobility films for stretchable transparent electrodes. Nature, 2009.
for graphene on silicon oxide as compared to dimethyl sulf- 457(7230): pp. 706–710.
oxide. The mobility computed for monolayer graphene under 19. Bao, W. et al. Controlled ripple texturing of suspended
induced strain suggests that the dc conductivity gets degraded graphene and ultrathin graphite membranes. Nature
Nanotechnology, 2009. 4(9): pp. 562–566.
with increasing strain. This happens because the Fermi veloc-
20. Booth, T.J. et al. Macroscopic graphene membranes and their
ity experiences a relatively sharp drop as the applied strain extraordinary stiffness. Nano Letters, 2008. 8(8): pp. 2442–2446.
increases more than 10%. A large decrease in the mobility 21. Bunch, J.S. et al. Impermeable atomic membranes from gra-
is seen for the Z direction, which can be attributed to the phene sheets. Nano Letters, 2008. 8(8): pp. 2458–2462.
relatively stronger distortions along this orientation. Results, 22. Huang, M. et al. Phonon softening and crystallographic orien-
further, reveal that the phonons and Coulomb limited mobili- tation of strained graphene studied by Raman spectroscopy.
ties decrease with increasing strain, however, the SR limited Proceedings of the National Academy of Sciences, 2009.
106(18): pp. 7304–7308.
mobility increases with growing strain.
23. Mohiuddin, T.M.G. et al. Uniaxial strain in graphene by
Raman spectroscopy: G peak splitting, Grüneisen parame-
REFERENCES ters, and sample orientation. Physical Review B, 2009. 79(20):
p. 205433.
1. Fradkin, E., Critical-behavior of disordered degenerate semi- 24. Teague, M.L. et al. Evidence for strain-induced local conduc-
conductors. 1. Models, symmetries, and formalism. Physical tance modulations in single-layer graphene on SiO2 . Nano
Review B, 1986. 33(5): pp. 3257–3262. Letters, 2009. 9(7): pp. 2542–2546.
2. Novoselov, K.S. et al. Electric field effect in atomically thin 25. Tsoukleri, G. et al. Subjecting a graphene monolayer to ten-
carbon films. Science, 2004. 306(5696): pp. 666–669. sion and compression. Small, 2009. 5: p. 2397.
3. Novoselov, K.S. et al. Two-dimensional atomic crystals. 26. Weiss, N.O. et al. Graphene: An emerging electronic material.
Proceedings of the National Academy of Sciences of the Advanced Materials, 2012. 24(43): pp. 5782–5825.
United States of America, 2005. 102: pp. 10451–10453. 27. Qu, L.H. et al. Armchair graphene nanoribbons under
4. Lee, C. et al. Measurement of elastic properties and elastic shear strain. Physica E: Low-Dimensional Systems and
strength of monolayer graphene. Science, 2008. 321: p. 385. Nanostructures, 2014. 60: pp. 156–159.
5. Balandin, A.A. et al. Superior thermal conductivity of single- 28. Castro Neto, A.H. et al. The electronic properties of graphene.
layer graphene. Nano Letters, 2008. 8(3): pp. 902–907. Reviews of Modern Physics, 2009. 81(1): p. 109.
6. Pereira, V.M., A.H. Castro Neto, and N.M.R. Peres, Tight- 29. Pellegrino, F.M.D., G.G.N. Angilella, and R. Pucci, Strain
binding approach to uniaxial strain in graphene. Physical effect on the optical conductivity of graphene. Physical
Review B, 2009. 80(4): p. 045401. Review B, 2010. 81(3): p. 035411.
7. Ando, T., The electronic properties of graphene and carbon 30. Wallace, P.R., The band theory of graphite. Physical Review,
nanotubes. NPG Asia Materials 2009. pp. 17–21. 1947. 71(9): p. 622.
8. Geim, A.K., Graphene: Status and prospects. Science, 2009. 31. Morozov, S.V. et al. Giant intrinsic carrier mobilities in gra-
324: p. 1530. phene and its bilayer. Physical Review Letters, 2008. 100(1):
9. Boukhvalov, D.W. and M.I. Katsnelson, Enhancement of p. 016602.
chemical activity in corrugated graphene. Journal of Physical 32. Zhang, Y.B. et al. Experimental observation of the quantum
Chemistry C, 2009. 113(32): pp. 14176–14178. Hall effect and Berry’s phase in graphene. Nature, 2005.
10. Choi, S.-M., S.-H. Jhi, and Y.-W. Son, Effects of strain on elec- 438(7065): pp. 201–204.
tronic properties of graphene. Physical Review B, 2010. 81(8): 33. Novoselov, K.S. et al. Two-dimensional gas of massless Dirac
p. 081407. fermions in graphene. Nature, 2005. 438(7065): pp. 197–200.
11. de Andres, P.L. and J.A. Verges, First-principles calculation 34. Adam, S. et al. A self-consistent theory for graphene trans-
of the effect of stress on the chemical activity of graphene. port. Proceedings of the National Academy of Sciences, 2007.
Applied Physics Letters, 2008. 93: p. 171915. 104(47): pp. 18392–18397.
12. Farjam, M. and H. Rafii-Tabar, Uniaxial strain on gapped 35. Ando, T., Screening effect and impurity scattering in mono-
graphene. Physica E: Low-Dimensional Systems and layer graphene. Journal of the Physical Society of Japan,
Nanostructures, 2010. 42(8): pp. 2109–2114. 2006. 75(7).
13. Fogler, M.M., F. Guinea, and M.I. Katsnelson, Pseudomagnetic 36. Chen, J.H. et al. Charged-impurity scattering in graphene.
fields and ballistic transport in a suspended graphene sheet. Nature Physics, 2008. 4: pp. 377–381.
Physical Review Letters, 2008. 101(22): p. 226804. 37. Nomura, K. and A.H. MacDonald, Quantum Hall ferromag-
14. Mohr, M. et al. Two-dimensional electronic and vibrational netism in graphene. Physical Review Letters, 2006. 96.
band structure of uniaxially strained graphene from ab initio 38. Hwang, E.H., S. Adam, and S.D. Sarma, Carrier transport in
calculations. Physical Review B, 2009. 80(20): p. 205410. two-dimensional graphene layers. Physical Review Letters,
15. Pereira, V.M. and A.H. Castro Neto, Strain engineering of 2007. 98(18): p. 186806.
graphene’s electronic structure. Physical Review Letters, 39. Du, X. et al. Approaching ballistic transport in suspended gra-
2009. 103(4): p. 046801. phene. Nature Nanotechnology, 2008. 3(8): pp. 491–495.
40. Shishir, R.S. et al. Room temperature carrier transport in gra- 53. Shah, R. and T. Mohiuddin. Temperature dependent mobil-
phene. Journal of Computational Electronics, 2009. 8: pp. 43–50. ity in strained graphene. In Proceedings World Congress on
41. Ashcroft, N.W. and N.D. Mermin, Solid State Physics. Engineering, 2011. London, UK.
International ed. 1983: Holt-Saunders. 54. Song, H.S. et al. Origin of relatively low transport mobility
42. Ando, T., Screening effect and quantum transport in a sili- of graphene grown to chemical vapour deposition. Scientific
con inversion layer in strong magnetic fields. Journal of the Reports, 2012. 2.
Physical Society of Japan, 1977. 43(5): p. 1616. 55. Shishir, R.S. and D.K. Ferry, Intrinsic mobility in gra-
43. Ando, T., A.B. Fowler, and F. Stern, Electronic properties of phene. Journal of Physics: Condensed Matter, 2009. 21(23):
two-dimensional systems. Reviews of Modern Physics, 1982. p. 232204.
54(2): p. 437. 56. Shah, R., T. Mohiuddin, and R. Singh, Giant reduction of
44. Mañes, J.L., Symmetry-based approach to electron-phonon charge carrier mobility in strained graphene. Modern Physics
interactions in graphene. Physical Review B, 2007. 76(4): Letters B, 2013. 27(3): p. 1350021.
p. 045430. 57. Hamaguchi, C., Basic Semiconductor Physics. 2001: Springer.
45. Meyer, J.C. et al. The structure of suspended graphene sheets. 58. Lundstrom, M., Fundamentals of Carrier Transport. 2nd ed.
Nature, 2007. 446(7131): pp. 60–63. 2000: Cambridge University Press.
46. Hwang, E.H. and S. Das Sarma, Acoustic phonon scattering 59. Ridley, B.K., Electrons and Phonons in Semiconductor
limited carrier mobility in two-dimensional extrinsic gra- Multilayers. 2nd ed. 2009: Cambridge University Press.
phene. Physical Review B, 2008. 77(11): p. 115449. 60. Fischetti, M.V. and S.E. Laux, Monte Carlo study of electron
47. Falkovsky, L.A., Symmetry constraints on phonon dispersion transport in silicon inversion layers. Physical Review B, 1993.
in graphene. Physics Letters A, 2008. 372(31): pp. 5189–5192. 48(4): p. 2244.
48. Fujita, M. et al. Peculiar localized state at zigzag graphite 61. Packan, P. et al. High performance 32 nm logic technology
edge. Journal of the Physical Society of Japan, 1996. 65(7). featuring 2nd generation high-k + metal gate transistors. In
49. Wakabayashi, K. et al. Electronic and magnetic properties of Electron Devices Meeting (IEDM), 2009 IEEE International.
nanographite ribbons. Physical Review B, 1999. 59(12): p. 8271. 2009.
50. Hwang, E.H. and S. Das Sarma, Screening-induced temper- 62. Goodnick, S.M. et al. Surface roughness at the Si(100)-SiO2
ature-dependent transport in two-dimensional graphene. interface. Physical Review B, 1985. 32(12): p. 8171.
Physical Review B, 2009. 79(16): p. 165404. 63. S. Jin, M.V. Fischetti, and T.-W. Tang, Modeling of surface-
51. Peres, N.M.R., J.M.B. Lopes dos Santos, and T. Stauber, roughness scattering in ultrathin-body SOI MOSFETs. IEEE
Phenomenological study of the electronic transport coeffi- Transactions on Electron Devices, 2007. 54(9).
cients of graphene. Physical Review B, 2007. 76(7): p. 073412. 64. Katsnelson, M.I. and A.K. Geim, Electron scattering on
52. de Juan, F. et al. Charge inhomogeneities due to smooth microscopic corrugations in graphene. Philosophical
ripples in graphene sheets. Physical Review B, 2007. 76(16): Transactions of the Royal Society A: Mathematical Physical
p. 165409. and Engineering Sciences, 2008. 366: pp. 195–204.
23 Graphene-Based Antibacterial Materials
Van Hoa Nguyen and Jae-Jin Shim
CONTENTS
Abstract...................................................................................................................................................................................... 447
23.1 Introduction...................................................................................................................................................................... 447
23.2 Synthesis of Graphene-Based Antibacterial Materials..................................................................................................... 447
23.2.1 Synthesis of Graphene and Its Derivatives........................................................................................................... 447
23.2.2 Synthesis of Graphene-Based Antibacterial Composites..................................................................................... 448
23.3 Applications of Graphene-Based Antibacterial Materials............................................................................................... 450
23.3.1 Mechanisms of Antibacterial Action of Graphene-Based Materials................................................................... 450
23.3.2 Antibacterial Activity and Applications of Graphene-Based Materials............................................................... 453
23.4 Conclusions and Perspectives........................................................................................................................................... 461
Acknowledgment....................................................................................................................................................................... 461
References.................................................................................................................................................................................. 461
ABSTRACT generation of oxidative stress [4,5], and electrostatic interac-

tions [6]. In addition, graphene nanosheets (GNS) can encase
Graphene has been considered as an ideal component for bacteria and other cellular components because of their one
composite materials possessing superior carrier transport, atom thickness [7]. Graphene-based antibacterial materials
high mechanical stiffness, extremely large surface area, and include graphene, graphene oxide (GO), reduced graphene
fine thermal/chemical stability. Many composites based on oxide (RGO), and their modified forms, and composites of
graphene have been developed for potential applications in these graphene derivatives with organic compounds, poly-
lithium batteries, solar cells, electronic devices, sensors, and mers, and inorganic nanoparticles. Among them, GO and
so on. Despite the potential, there is a real concern that gra- GO-based materials have been studied extensively owing
phene-based materials may have detrimental effects on human to their high antibacterial activity, size controllability, abil-
health and natural environments. One particularly interesting ity to tune their properties by altering the oxidation level,
aspect of this subject is the toxic effects of graphene-based and high dispersibility in water. This chapter examines the
materials on the microscopic world of bacteria. This chap- antibacterial activity of graphene-based materials, including
ter provides the antibacterial properties of graphene-based GO, RGO, and graphene-based composites. The design and
materials, including graphite, graphene oxide, reduced gra- fabrication and potential applications of these antibacterial
phene oxide, and graphene-based composites. Some special materials are also described.
synthetic methods of graphene-based antibacterial compos-
ites are described. In addition, the potential applications of
graphene-based antibacterial materials are provided. 23.2 SYNTHESIS OF GRAPHENE-BASED
ANTIBACTERIAL MATERIALS
23.1 INTRODUCTION 23.2.1 Synthesis of Graphene and Its Derivatives
Graphene-based materials were reported to have antibacte- Graphene is a single atomic plane of graphite (Gt) that was first
rial properties. The interactions between graphene deriva- obtained directly from the manual mechanical cleavage from
tives and bioorganisms suggest that these materials have Gt using Scotch tape [8]. To date, several techniques for produc-
excellent antibacterial activity with minimal cytotoxicity [1]. ing graphene with appreciable quantities have been developed.
Although graphene-based materials are promising candidates Currently, there are five primary methods, including mechani-
for electronic applications, studies on the use of these mate- cal exfoliation, epitaxial growth, chemical vapor deposition,
rials in biological applications are limited. Graphene-based longitudinal “unzipping” of carbon nanotubes, and reduction
materials possess a range of desirable antibacterial proper- of graphene derivatives [9]. These techniques are described in
ties, including (i) a broad spectrum of antibacterial activity; detail in other chapters of this book. In the field of antibacterial
(ii) nontoxicity to many hosts; and (iii) insusceptibility to the materials, the most widely applied technique for the production
development of microbial resistance [2]. Several antibacte- of graphene is the reduction of GO (Figure 23.1) owing to its
rial mechanisms of graphene-based materials have been pro- ease of large-scale production and relatively low cost [10]. GO is
posed and studied such as rupture of the bacterial wall [2,3], oxygenated graphene sheets covered with carboxyl groups at its
447
(a)
Oxidation
Graphite
Graphite oxide
(GO)
Exfoliation
Reduction
Chemical converted
(b) (c) Graphene oxide
graphene (CCG)
2.00 2.00
μm μm
1.00 1 μm
1.00
3.00 0 3.00 0
nm nm
0
3.00 3.00 1 μm
0 1.00 2.00 μm 0 1.00 2.00 μm
FIGURE 23.1 (a) Preparation of graphene by the chemical reduction of GO synthesized using Hummers’ method. (Adapted from H.
Pandey et al. Nanoscale 3, 4104–4108, 2011, with permission.) (b) AFM images of GO (left) and RGO sheets (right). (c) Photographs of
freestanding and flexible (left) and SEM images showing the thickness of GO (top) and RGO paper (bottom). (Adapted from W. Hu et al.
ACS Nano 4, 4317–4323, 2010, with permission.)
edges as well as phenol hydroxyl and epoxy groups at its basal groups on GO [2,14,16,17,20,21]. The as-prepared RGO sheets
plane [9,11]. These groups not only help GO sheets disperse well are comprised of graphene domains interspersed with residual
in both polar and nonpolar solvents but also allow their interac- oxygen-containing functionalities, which improve the solubility
tions with cations and provide reactive sites for the preparation and interactions of the sheets. Moreover, the excellent scalabil-
of various graphene-based composites (Figure 23.2) [1]. GO can ity of these synthetic methods has enabled the preparation of
be exfoliated chemically from graphite oxide (GtO) obtained by many novel graphene-based materials with excellent physical
the oxidation of natural Gt powder with a range of strong oxi- properties [9]. Recently, some “green” syntheses of RGO were
dants such as HNO3, KMnO4, and H2SO4. The typical proce- reported such as the bacterial reduction of graphene [22], pro-
dures for the synthesis of GO were first developed by Hummers tein exfoliation and functionalization of graphene [23], gelatin
et al. [12], Brodie [13], and Staudenmaier [14]. However, both reduction, and stabilization of GO [24].
Brodie and Staudenmeier methods are not used widely because
the generation of chlorine dioxide (ClO2) gas, which is a highly
toxic compound that has a tendency to decompose in air and 23.2.2 Synthesis of Graphene-Based
explode [15]. Therefore, the Hummers method with its relatively
Antibacterial Composites
shorter reaction time and the absence of hazardous ClO2 is nor-
mally used to prepare GO for antibacterial applications [1–3,16– Studies of graphene-based antibacterial composites have
21]. To produce RGO-based antibacterial materials, a thermal or focused on the preparation of GO or RGO functionalized
chemical treatment is normally used to eliminate the functional with a range of antibacterial agents, including inorganic
Graphene-Based Antibacterial Materials 449
(a) of graphene and interactions with the antibacterial precur-

0h 2h 0h 2h 0h 2h 0h 2h
sors because pristine graphene has low solubility in both
polar and nonpolar solvents [29,33]. Tai et al. [26] prepared
graphene nanosheets (GNSs)-grafted PAA, followed by
inducing AgNO3 and converting Ag+ to Ag nanoparticles.
The PAA grafting not only realized dispersion in an aque-
ous solution of GNSs but also created a distribution of nega-
tive charges on the surfaces of the GNSs. Nguyen et al. [44]
Gt GtO GO rGO chose supercritical carbon dioxide as a medium for dispers-
(b) (d) ing RGO and dissolving the Ag precursor. The initial elec-
trostatic attachment of free Ag+ ions was attributed to the
presence of few oxygen functionalities at the RGO surface
(Figure 23.4).
In strategy II, a graphene-based composite was prepared
directly by a mixing or compounding method. The anti-
None SEI 50kV ×5.000 1μm WD 7.5mm
bacterial agents were normally polymers or small organic
compounds. Using this method, the structure of antibacte-
(c) 2 (e) rial agents could be controlled precisely because they were
designed and synthesized without disturbing the graphene
Height, mm
1
or its precursor. Som et al. [43] synthesized RGO/PLL com-
0 posites that functioned as antibacterial agents and promoted
the growth of human cell culture. After surface modification,
–1 RGO was mixed with PLL to form RGO/PLL composites via
0.0 0.3 0.6 0.9 1.2 1.5
Distance (μm) None SEI 50kV ×50.000 100 nm WD 7.3mm the electrostatic bonding of PLL and RGO. Lu et al. [45] syn-
thesized the graphene/chitosan/poly(vinyl alcohol) (CS/PVA)
FIGURE 23.2 (a) Photographs of Gt, GtO, GO, and rGO disper- nanofiber using an electrospinning method. The graphene was
sions at a concentration of 400 μg/mL. Images marked “0 h” show dispersed well in a CS/PVA solution by an ultrasonic stirring
the dispersions immediately after sonication, whereas images treatment, followed by electrospinning.
marked “2 h” are from dispersions after being left to stand for 2 h, In strategy III, the GO-based composite was first prepared
(b) AFM height image of the GO nanosheets dried on a mica sur- by in situ growth of the antibacterial agent on the surface
face (the scale bar shows 200 nm), (c) corresponding height profile of the GO sheets. To produce the graphene-based compos-
of the AFM image, (d) SEM images of GO nanosheets, and (e) SEM ite, GO was converted to graphene using a range of methods
images of rGO nanosheets dried on silicon wafers. (Adapted from such as thermal radiation, the photocatalytic reaction, or the
S. Liu et al. ACS Nano 5, 6971–6980, 2011, with permission.)
addition of reducing agents (hydrazine, NaBH4, etc.) [9,10,74].
Zhang et al. [24] reported the green synthesis of GO-coated
material (metal, metal oxide, and salt), small organic com- Ag nanoprisms with good antibacterial properties using gela-
pound, and polymers. The main antibacterial agents include tin as the reducing and stabilizing agent. Chook et al. [28]
inorganic materials (e.g., Ag [24–34], TiO2 [35], ZnO [36], prepared a GO/Ag composite with a low silver content and
Ag3PO4 [37]), small organic compound (e.g., tetradecyltriphe- comparable antibacterial performance using microwave irra-
nyl phosphonium bromide [38], benzylpenicillin (BP) [39], diation as a rapid method (Figure 23.5). Before the reduc-
chlorophenyl [40], lysozyme [5]), polymer (e.g., polydopa- tion, the positively charged silver ammonia complex could be
mine (PDA) [34], chitosan (CS) [41], polyethyleneimine (PEI) attracted easily to the negatively charged oxygen functional
[31], poly(acrylic acid) (PAA) [26], poly(N-vinyl carbazole) group on GO.
[42], and poly(l-lysine) (PLL) [43]). Generally, the synthetic In strategy IV, an antibacterial agent with a well-defined
techniques for graphene-based antibacterial composites can structure was first deposited on the surface of the GO sheets
be classified into five strategies according to the processes with vigorous stirring or ultrasonic irradiation. Similar to
of synthesizing graphene and antibacterial agents (Figure strategy III, to form a graphene-based composite, GO was
23.3). Table 23.1 compares the typical synthetic methods for converted to graphene using a range of techniques, as shown
graphene-based antibacterial composites and the properties above. Wang et al. [39] prepared GO/BP anion-intercalated
of as-prepared products. Both GO-based and graphene-based Mg–Al layered double hydroxide (LDH) hybrid films using
composites can be used for antibacterial applications. a solvent evaporation process. The composite was formed by
In strategy I, graphene was pre-prepared before inducing mixing GO with the BP-LDH colloidal suspensions with vig-
the antibacterial precursor. The graphene-based composites orous stirring. Carpio et al. [46] developed a highly effective
were synthesized by the in situ growth of an antibacterial antimicrobial composite by simple mixing, which contained
agent on the graphene sheets. Chemically modified gra- only 3 wt.% of GO well dispersed in a poly(vinyl-N-
phene has been used widely to enhance the processability carbazole) matrix. Similarly, Liu et al. prepared high-quality
In situ growth
I
Graphene Precursor
Deposition
II
Graphene Antibacterial agent
Deposition Reduction
III
GO Antibacterial agent GO-based composite Graphene-based

composite
In situ growth Reduction

IV
GO Precursor GO-based composite
One-pot growth
V
GO Precursor
FIGURE 23.3 Five general strategies for the preparation of graphene-based antibacterial composites. (Adapted from X. An et al. RSC Adv.
1, 1426–1434, 2011, with permission.)
GO/TiO2 films on the gram scale and GO/Ag composites 23.3 APPLICATIONS OF GRAPHENE-BASED
by a simple two-phase assembling method (Figure 23.6) to ANTIBACTERIAL MATERIALS
improve the antibacterial activity [25,35]. The nanoparticles
could interact with the GO sheets through electrostatic bind- 23.3.1 Mechanisms of Antibacterial Action
ing, physisorption, and charge-transfer interactions [25]. of Graphene-Based Materials
Akhavan et al. [47] synthesized the RGO/TiO2 films with
a high antibacterial activity using a two-step method. First, The mechanisms of antibacterial action of graphene-based
the GO/TiO2 films were prepared, followed by the conver- materials should be figured out. A range of mechanisms
sion of GO to RGO by a UV–visible-light-induced photo- have been reported, including the generation of oxidative
catalytic reduction. stress [1,4,5], rupture of the bacterial wall [2,3,17], electro-
In strategy V, the graphene and antibacterial agent were static interactions [6], and the encasement of bacteria [7,19].
synthesized simultaneously in a one-pot system, and the Liu et al. [1] compared the antibacterial activity of four
two components were blended during the composite forma- types of graphene-based materials (Gt, GtO, GO, and RGO)
tion process. A typical example is reducing GO and metal toward Escherichia coli. They attributed their antibacterial
ions simultaneously in a blended solution, leaving an RGO/ activities to their dispersibility, size, and oxidization capac-
metal particle composite. Dai et al. [48] reported that RGO/ ity. The synergy of membrane stress and oxidative stress was
Ag composites could be produced by reducing Ag+ and GO assigned to the antibacterial mechanism. Gurunathan et al.
simultaneously with l-ascorbic acid. Using the same reducing [4] reported that GO and RGO showed dose-dependent anti-
agent, Shen et al. [27] prepared an RGO/Ag composite with bacterial activity against Pseudomonas aeruginosa cells
GO and AgNO3 as the starting materials. They reported that through the generation of reactive oxygen species (ROS),
GO could be reduced under the hydrothermal conditions with leading to cell death (Figure 23.7). Hu et al. [3] reported the
l-ascorbic acid, whereas the Ag nanoparticles were grown on loss of cellular integrity of E. coli with a disruption of the
the RGO surface simultaneously. Yuan et al. [34] synthesized cell walls, which was attributed to either physical disruption
a graphene/Ag composite using a green one-step method with or oxidative stress mechanism. Akhavan and Ghaderi [17]
sodium citrate as an environmentally friendly reducing and suggested that graphene sheets serve as “cutters” and dis-
stabilizing agent for both GO and Ag+. rupt and damage the cell membranes, leading to the release
TABLE 23.1
Comparison of the Typical Synthetic Methods for Graphene-Based Antibacterial Composites and the Properties of
As-Prepared Products
Materials Synthetic Routes Properties of Products Reference
GO/Ag In situ deposition using gelatin Ag nanoprisms are well dispersed with different sizes on the [24]
surface of GO
GO/Ag Two-phase (toluene–water) process Ag nanoparticles anchored on GO are spherical in shape and [25]
highly monodispersed with size of 6 nm
Ag/RGO-g-PAA In situ grafting of PAA and Ag nanoparticles are uniformly deposited on the surfaces of [26]
deposition of Ag nanoparticles RGO-PAA with a size of 4–8 nm
RGO/Ag One-pot hydrothermal method GO is reduced well under the hydrothermal conditions; almost all [27]
RGO sheets are separated from each other and coupled by Ag
nanoparticles
GO/Ag A rapid and green microwave Ag nanoparticles with average size of 40.7 ± 7.5 nm are deposited [28]
irradiation synthesis method randomly on the GO sheets
RGO/Ag In situ chemical deposition method A homogeneous distribution of Ag nanoparticles (5–10 nm in [29]
size) on RGO sheets
RGO/Ag In situ chemical deposition method Ag nanoparticles with a size of several nanometers and about [30]
40 nm are loaded onto the surface of the RGO sheets
RGO/polyethyleneimine/Ag In situ grafting of PEI and deposition The composites have excellent water solubility [31]
of Ag
GO/Ag In situ reduction method Ag nanoparticles with about 80 nm are homogeneously assembled [32]
on the GO sheets
RGO/polydopamine/Ag Mussel-inspired method Ag nanoparticles with are uniformly and densely distributed on [33]
the surface of PDA-RGO sheets
RGO/Ag In situ chemical deposition method Well-dispersed Ag NPs with the size range of 20–25 nm are [34]
deposited on RGO surfaces
GO/TiO2 nanorod Two-phase (water–toluene) Procedure on a gram scale. Single layer and double layer GO [35]
assembling method sheets are entirely covered with TiO2 of 3–5 nm in diameter and
25–50 nm in length
Graphene/ZnO One-pot hydrothermal method Well separated, randomly distributed, spherical ZnO nanoparticles [36]
with an average size of 22 ± 6 nm on the graphene sheets
GO/Ag3PO4 Ion-exchange method Ag3PO4 particles possess regular rhombic dodecahedral [37]
morphology with an average diameter of 500 nm
Brilliant blue/RGO/tetradecyl Three-step approach (oxidation, Water-soluble composites, adsorption of BB, and TTP molecules [38]
triphenyl phosphonium ultrasonic treatment, and reduction) are adsorbed on the flake-like RGO sheets
bromide
GO/benzylpenicillin Solvent evaporation process GO/BP consist of irregular and thin hexagonal platelets with a [39]
relatively narrow particle size distribution within the range
50–100 nm
Graphene/chlorophenyl Solvent-free green diazotization 16.09 wt.% chlorophenyl grafted on graphene was found with a [40]
technique chlorine content of 5.08%
RGO/chitosan Mixing and drop-casting technique Superior thermal and tensile properties of chitosan/GO films [41]
Graphene/poly(N-vinyl Solution mixing, sonication, and Stable dispersions of graphene nanoplatelets have the average [42]
carbazole) electrodeposition method height of 1.8 nm and lengths that spans 10–20 µm
RGO/poly(l‑lysine) In situ grafting and self-assembly Electrostatic and covalent bonding of PLL to GO and RGO [43]
process of PLL derivatives
RGO/Ag Supercritical fluid-assisted Ag nanoparticles with diameters of 10–40 nm dispersed on [44]
deposition method the surfaces of RGO sheets
Graphene/chitosan/polyvinyl Electrospinning method The diameter distribution of nanofibers is comparatively narrow, [45]
alcohol centered at about 120 nm
GO/poly(vinyl-N-carbazole) High-yield process of adiabatic bulk 3 wt.% of GO well dispersed in a 97 wt.% PVK matrix [46]
polymerization
RGO/Ag One-step chemical reduction method Ag nanoparticles aggregate, the unstable RGO/Ag composites in [47]
solution settle to the bottom
(Continued)

Comparison of the Typical Synthetic Methods for Graphene-Based Antibacterial Composites and the Properties of
As-Prepared Products
Materials Synthetic Routes Properties of Products Reference
GO/TiO2 Mixing and spreading method The GO platelets deposited on the surface show a relatively [48]
smooth planar structure
RGO/naphthalenesulfonate/ Three-step approach (oxidation, Water-soluble Ag/NA/RGO hybrid. The solubility could reach up [49]
Ag ultrasonic treatment, reduction) to 1.1 mg/mL. The Ag content in hybrid is 3.6 wt.%
RGO/gentamicin Self-assembly method The efficient drug loading of 2.57 mg/mg was obtained at pH 7 [50]
RGO/Ag Chemical reduction method Ag nanoparticles with a diameter of 45–50 nm are homogeneously [51]
decorated onto graphene sheets without aggregation
GO/lysozyme Self-assembly and solution-cast A multilayered ordered structure with protein molecules is [52]
method intercalated between adjacent layers. Precise control of the
protein amount within the film was achieved
RGO/Ag Sonochemical method GO sheets were well decorated with Ag nanoparticles with size of [53]
8 nm
GO/polyethylene/2- Mixing and sonication method La3+ has bonded with carboxyl group of graphene oxide to form [54]
(methacryloyloxy)ethyl GeneO–La complexes
phosphoryl choline
GO/cotton fabrics Dip coating method The amount of GO coated onto the fabrics (3 × 4 cm) is 0.025 g [55]
RGO/Ag Ball milling method Ag nanoparticles are uniformly anchored on the surface of a [56]
graphene nanosheet and are mostly in the size of 1–5 nm
Ag/Fe2O3/GO Phase transfer method Water-soluble Ag/Fe2O3/GO. The particles become anchored in [57]
more than 90% area of the GO surface
Ag/graphene Supercritical fluid-assisted deposition Ag particles of size around 8–20 nm are coated on the GO surface [58]
method
GO/Ag In situ chemical deposition method The average size of the Ag nanoparticles anchored on the GO [59]
surface is 7.5 nm
Ag/halloysite/RGO Self-assembly and chemical Sandwich-like nanomaterial (Ag/HNTs/RGO). The size of Ag [60]
composites deposition method nanoparticles is 5–15 nm
GO/chitosan Mixing and solution-casting method The composite layer exhibits a layered structure (with thicknesses [61]
~200–500 nm)
GO paint Ball milling method A black homogenous GO nano-paint is obtained [62]
Au/Rubpy/GO In situ reduction synthesis method Au NPs has a diameter in the range of 10 − 20 nm. The sizes of [63]
the Au/Rubpy/S-GO and Au/Rubpy/L-GO are 111.6 and
192.8 nm
RGO/TiO2/Ag Two-phase (toluene–water) method Ag nanoparticles (4–6 nm) and TiO2 nanorods (3–5 nm in [64]
diameter and 20–50 nm in length) are anchored on GO sheet
GO/Ag/polymer hydrogel In situ reduction synthesis method Ag nanoparticles (11.5–39.0 nm) are homogeneously and densely [65]
attached to the surface of graphene
Ag/Fe3O4/RGO One-pot hydrothermal process The average size of Ag/Fe3O4with and without RGO sheets are [66]
20 and 13 nm, respectively
Au/TiO2/RGO Two-step hydrothermal method TiO2 particles and Au nanoparticles (40 nm) are densely anchored [67]
on graphene sheets
Graphene/quantum dots Electrochemically production method Composites have the average height of about 3 nm, corresponding [68]
to three graphene layers and the average diameter of 40.5 nm
GO/PVA/chitosan Mixing and electrospinning method GO sheets are distributed uniformly in the nanofibers. The [69]
nanofibers’ diameters are decreased with increasing the contents
of GO
RGO/Eu–vancomycin In situ mixing and grafting method Van-modified RGO is functionalized with a europium (Eu3+) [70]
complex complex
RGO/TiO2 nanosheets In situ mixing method An alternating stacking structure of graphene layer and the [71]
titanate layer
GO/ZnO One-pot hydrothermal process ZnO nanoparticles (4 nm) are uniformly anchored on the GO [72]
sheets
GO/6-aminoindazole Sonochemical waves method Covalent amide bonds are formed on GO surface [73]
Impregnation
(a)
Ag(hfa) Ag+ Ag+ Ag+

Add CO2 scCO2 Ag+
Graphene Ag+
Ag+
Ag+ Ag+ Ag+
Remove CO2
H(hfa)
scCO2 Ag Add H2
Ag Ag+ Ag+
Ag Ag+
Graphene–Ag H(hfa)
Purge Decomposition
(b) RGO RGO/Ag
100 nm 100 nm
FIGURE 23.4 (a) Schematic diagram of the preparation of RGO/Ag composite in scCO2 and (b) EM images of RGO and RGO/Ag com-
posites. (Adapted from V. H. Nguyen et al. J. Supercrit. Fluids 72, 28–35, 2012, with permission.)
of intracellular contents and eventual cell death. The sharp and organic agents [75]. Owing to their safety and stability
edge of the GNS was the main reason for membrane stress at high temperatures, inorganic antibacterial agents have
after cell deposition on GNS (Figure 23.8). Another study been used increasingly because organic antibacterial agents
reported that all the bacteria could be trapped between the have limited applications due to their low heat resistance,
aggregated graphene sheets and be isolated from the environ- high decomposability, and short life expectancy [75,76]. On
ment using a GO–melatonin suspension, leaving no active the other hand, most inorganic and organic materials, such
bacteria [19]. Zhou et al. [6] assumed that the electrostatic as antibiotics, metal ions, and quaternary ammonium com-
interactions between GO sheets and bacterial cells caused pounds, are associated with concerns regarding antibiotic
the inhibition of bacterial growth, including Staphylococcus resistance, environmental pollution, relatively complex pro-
aureus (Gram-positive bacterium) and E. coli (Gram- cessing, and high cost [3]. Recently, the use of nanostruc-
negative bacterium). The negatively charged groups of GO tured materials with antibacterial activity, owing to their size,
(e.g., CO− and COO−) can only bind to the positively charged structure, and surface properties, has been explored to meet
surface of S. aureus through an electrostatic interaction, these challenges [77]. These materials have been used both in
inhibiting bacterial growth. In contrast, electrical repulsion vivo and for exterior applications such as paints, textiles, and
will prevent interactions between E. coli and GO, which has medical devices [78,79]. More recently, graphene-based mate-
no inhibitory effect (Figure 23.9). rials are considered to be ideal antibacterial agents because of
their excellent physicochemical properties (e.g., high thermal
23.3.2 Antibacterial Activity and Applications stability and high mechanical strength), biocompatibility, and
often low cost. Considerable effort has been made to develop a
of Graphene-Based Materials
variety of graphene-based materials for antibacterial applica-
Antibacterial agents are used widely in daily life and tions. Table 23.2 presents the results of recent studies on the
have effectively addressed many public health issues [16]. use of graphene and its derivatives for antibacterial applica-
Generally, antibacterial agents can be classified into inorganic tions. Table 23.3 lists the composites made by a combination
(a) O
O
HO
HO O O
OH
O O
O
HO
HO OH HO HO O
HO
O O
O OH O
HO OH OH
OH
O HO O HO
HO
O OH O
O
OH O
OH
O O OH
OH O
e– R
Microwave irradiation Glucose
COH
(b)
AgNP O
AgNP
AgNP
HO O O
OH
AgNP O O
O AgNP
AgNP
AgNP AgNP
OH HO O
AgNP AgNP HO AgNP
O O
AgNP AgNP
O AgNP
O
HO AgNP AgNP AgNP AgNP AgNP
O HO HO
HO O
O OH
O
O AgNP
AgNP
AgNP O
OH
AgNP
O O
AgNP
OH
Ag(NH3)2–
AgNP
AgNP
Silver
nanoparticles
FIGURE 23.5 Formation mechanism of GO/Ag composite using microwave irradiation: (a) Electrostatic interactions of Ag(NH3)2+ with
oxygenated functional group on the GO sheets and (b) addition of glucose to reduce Ag ions under microwave irradiation. (Adapted from
S. W. Chook et al. Nanoscale Res. Lett. 7, 541–547, 2012, with permission.)
(a)
OH OH
OH OH COOH OH
OH OH
COOH OH OH
COOH
OH
OH O OH Ag Mix, wash COOH
H
OH
O
OH O
OH OH OH
OH OH
OH COOH
OH COOH
COOH OH
COOH
GO
GO–Ag
(b)
GO in water Ag in toluene Two-phase GO–Ag in water
Mix, wash
FIGURE 23.6 Steps for the synthesis of the GO–Ag composites: (a) schematic diagram of the two-phase synthesis of GO–Ag composites
and (b) photographs of the two-phase synthesis of GO–Ag composites. (Adapted from S. Liu et al. ACS Nano 5, 6971–6980, 2011, with
permission.)
M 1 2 3 4 (a) GO GO–C12 GO–C12-Hep
E. coli
(I)
S. aureus
(II)
(b) 120
50 μg/mL
100 Control
GO
80
GO–C12
60 GO–C12-hep
40
20
0
120
125 μg/mL
Growth inhibition (%)

100
FIGURE 23.7 Detection of GO and rGO induced apoptosis
80
by cellular DNA fragmentation. Notes: For DNA fragmenta-
tion, cells were incubated with GO and rGO (100 μg/mL) for 60
24 hours. After incubation, DNA extracted from cells and resolved 40
on agarose gel electrophoresis. Lane M, 1 kB ladder; lane 1, con- 20
trol; lane 2, GO; lane 3, rGO; lane 4, silver nanoparticles. (From 0
Gurunathan, S. et al., Int. J. Nanomedicine 7, 5901–5914, 2012. 120
with permission). 250 μg/mL
100
80
60
40
20
0
(a) E. coli S. aureus
+ FIGURE 23.9 (a) Optical images of GO, GO–C12, and

GO-heparin-C12 on culture plates inoculated with (I) E. coli and
(II) S. aureus and (b) antibacterial activity of GO, GO–C12, and
GO-heparin-C12 at different concentrations on E. coli and S. aureus
bacteria after 24 h of incubation. The control was considered to be
Graphene nanosheets Bacterial strain Antibacterial activity 100% growth. (Adapted from N. Zhou et al. Carbon 47, 1343–1350,
2009. With permission.)
(b) (c)
of graphene with antibacterial agents to improve the antibac-

terial activity and broaden their potential applications. The
two tables show that the types of as-prepared products are
mostly bare GO and RGO and composites based on them. For
practical applications, a range of materials have been devel-
oped, such as papers [3], films [17,39,42,51–53,80], textile fab-
0.2 μm 0.5 μm
ric coating [55], nanowalls [20], and fibers [45]. For example,
robust and flexible papers with antibacterial properties could
FIGURE 23.8 (a) Interactions between the GNS and bacte- be prepared from both GO and RGO nanosheets using a one-
ria. (Adapted from K. Krishnamoorthy et al. Sci. Adv. Mater. 4,
step vacuum filtration method [3]. Bao et al. [32] suggested
111–1117, 2012, with permission.) (b) Antibacterial activity of GO
nanosheets via TEM images of E. coli and (c) E. coli exposed to GO that micron-scale GO nanosheets could be deposited easily
nanosheets at 37°C for 2 h. (Adapted from W. Hu et al. ACS Nano 4, on the porous ceramic membranes for water treatment and
4317–4323, 2010, with permission.) purification processes.
TABLE 23.2
Recent Reports on the Use of Graphene and Its Derivatives as Antibacterial Materials
Materials Bacteria Highlight Results Reference
Gt E. coli For E. coli survival (10 to 10 CFU/mL) and the sample concentration of
6 7 [1]
GtO 40 μg/mL, the loss of E. coli viability were 26.1 ± 4.8% (Gt), 15.0 ± 3.7%
GO sheets (GtO), 69.3 ± 6.1% (GO), and 45.9 ± 4.8% (RGO) for 2 h
RGO sheets
Graphene sheets E. coli The lowest concentrations of graphene that allowed no growth of a microorganism [16]
Salmonella typhimurium (MIC) after overnight incubation were 1 μg/mL for E. coli and S. typhimurium,
Enterococcus faecalis 8 µg/mL for E. faecalis, and 4 µg/mL for B. subtilis
Bacillus subtilis
GO films E. coli No further growth of the bacteria was observed on the surface of GO film after the [17]
contact time of 48 h
GO sheets RGO sheets Ralstonia solanacearum Most of the bacteria could not survive after they were treated with GO in water. The [2]
RGO did not show any antibacterial activity
GO paper RGO paper E. coli After overnight incubation at the sample concentration of 85 µg/mL, no growth of [3]
E. coli on GO paper and only limited number of E. coli on RGO paper was found
GO sheets RGO sheets P. aeruginosa For P. aeruginosa survival (2 × 108 CFU/mL), the sample concentration (75 µg/ [4]
mL), the loss of P. aeruginosa viability are 87% (GO) and 86% (RGO) for 4 h
GO sheets E. coli The small GO sheets exhibited much weaker antibacterial activities than the large [8]
GO sheets. The loss of E. coli cell (106 − 107 CFU/mL) for 2 h is 99.4 ± 0.4%
(large GO) and 52.5 ± 6.7% (small GO)
GO sheets E. coli The maximum bacteria reactivation was about 16% at the GO concentration of [18]
1 µg/mL for 48 h
GO nanowalls RGO E. coli After 1 h, 41 ± 8% and 16 ± 3% of E. coli and 26 ± 5% and 5 ± 1% of S. aureus [20]
nanowalls S. aureus could survive on the surface of the GO nanowalls and the RGO nanowalls,
respectively
GO sheets RGO sheets E. coli The loss of E. coli cells (106 CFU/mL) was 87% (GO) and 81% (RGO) at the [21]
sample concentration of 150 µg/mL for 2 h
Graphene honeycomb P. aeruginosa The films have excellent broad spectrum antibacterial activity [73]
films E. coli
GO sheets E. coli It completely inhibited the growth of E. coli and S. iniae at the concentrations of [55]
S. iniae 100 and 125 µg/mL, respectively
GO sheets R. solanacearum The majority of the cells lost their cellular integrity and became elongated or [75]
flattened after being incubated with GO
GO sheets E. coli The MIC of GO nanosheets was 0.125 µg/mL for E. coli and S. typhimurium, [81]
S. typhimurium 0.25 µg/mL for B. subtilis, and 0.5 µg/mL for E. faecalis
B. subtilis
E. faecalis
Graphene sheets E. coli S. aureus E. coli and S. aureus cells cannot survive on the graphene film with Cu substrate [82]
To evaluate the antibacterial activity of graphene-based difficult than that into the solution (colony test), allowing only
materials, several methods have been reported, including a small chance of GO interacting with bacteria.
disc diffusion [6,40,50,53], colony counting [1,19,20], optical Many reports have described the strong antibacterial activ-
density [2,32], and microdilution method [16,38,44]. Owing ity of GO and RGO, as shown in Table 23.2. The antibacterial
to its convenience, efficiency and low cost, the disc diffusion activities of these materials can be affected by their morphol-
method is used most widely. A selection of the appropriate ogy and physicochemical properties such as density of func-
method would depend on the intended degree of accuracy, tional groups, size, and conductivity. For example, Liu et al.
convenience, availability of resources and technical exper- [1] reported that a GO dispersion has the highest antibacterial
tise, and cost. For example, Zhou et al. [6] reported that GtO activity, followed by RGO, Gt, and GtO toward E. coli, which
exhibited no toxicity towards bacterial species when evalu- were attributed to their dispersibility, size, and oxidization
ated using the disc diffusion method but showed remarkable capacity. On the other hand, Akhavan et al. [20] reported that
toxicity using the colony counting method (Figure 23.9). The RGO nanowalls are more toxic to E. coli and S. aureus than
diffusion of GO into the agar (solid state) was much more GO nanowalls. This might be due to the more sharpening of
TABLE 23.3
Recent Reports on the Use of Graphene-Based Antibacterial Composites
Graphene Content
Materials and Particle Size Bacteria Highlight Results Reference
Graphene/poly(N-vinyl 3 wt.% E. coli The antibacterial activity of G-PVK was approiximately 25% [42]
carbazole) (PVK) films B. subtilis higher than pure graphene
RGO/Poly(l-lysine) No data E. coli No biofilm of E. coli was formed on the surface of the GO/PLL [43]
composite and RGO/PLL films after 12 h
GO/benzylpenicillin films 47.48 wt.% Micrococcus lysodeikticus The inhibition rate (IR) was approximately 2.5 times higher [39]
than GO
RGO/chitosan composites 3 wt.% P. aeruginosa The growth of P. aeruginosa was inhibited completely by a low [41]
concentration of RGO (0.3 wt.%) in the chitosan solution
Graphene/poly(acrylic 56 wt.% E. coli The antibacterial activity of composites was much higher than [26]
acid)/Ag composites S. aureus that of pure graphene
GO/cotton fabrics No data E. coli The reduction of E. coli and S. iniae after 24 h exposed [55]
S. iniae with GO/cotton fabric was 74% and 100%, respectively
Graphene/chlorophenyl 83.91 wt.% E. coli The zone of inhibition was approximately 2 times larger than [40]
S. aureus that of graphene
GO/Ag composites 33.73 wt.% E. coli 99% of E. coli cells were killed at a composite concentration of [25]
80 µg/mL
GO/Ag3PO4 composites 17% E. coli 100% bacterial cell removal occurred in the presence of the [37]
composite at a sample concentration of 20 μg/mL
Graphene/ZnO composites No data 22 ± 6 nm E. coli After 12 h of incubation, 100% reduction of E. coli in the [36]
medium occurred
RGO/Ag composites 44 wt.% 10 nm E. coli The disinfection ability of RGO/Ag was 99.7% [27]
GO/Ag composites 60.5 wt.% S. typhi The antibacterial activity of GO/Ag samples was stronger [28]
40.7 ± 7.5 nm E. coli against Gram-negative (S. typhi and E. coli) than
S. aureus againstGram-positive (S. aureus and S. epidermidis) bacteria
Staphylococcus
epidermidis
RGO/Ag composites No data 5–10 nm E. coli The bacterial cultures were destroyed at a disinfection rate of [29]
S. aureus 99.80% (E.coli), 99.84% (S. aureus), and 99.76%
Candida albicans (C. albicans)
RGO/Ag composites 17 wt.% 40 nm E. coli Bacterial growth was inhibited completely at a composite [30]
concentration of 25 μg/mL
RGO/polyethyleneimine/ 9.5 wt.% 5–15 nm E. coli The killing rates were 93.7% (E. coli) and 96.1% (S. aureus) [31]
Ag S. aureus for 6 h
GO/Ag composites 79 wt.% 80 nm E. coli The reduction values of the E. coli and S. aureus cells were [32]
S. aureus 100% and 87.6%, respectively, which was two times higher
than that of the bare GO
Graphene/polydopamine/ 11 wt.% E. coli The growth of E. coli and B. subtilis was inhibited completely [33]
Ag nanosheets B. subtilis for 48 h at a composite concentration of 1% (v/v)
Brilliant blue/RGO/ 28.8 wt.% E. coli The minimum inhibitory concentration (MIC) were 250 and [38]
tetradecyltriphenyl S. aureus 150 mg/L against E. coli and S. aureus, respectively
phosphonium bromide
composite
GO/lysozyme films 11.75 wt% B. subtilis The antibacterial activity was enhanced approximately 6 times [53]
after coating with lysozyme.
GO/Ag nanoprisms 20 wt.% No data E. coli The growth inhibition rate against E. coli was 99.9% at a [24]
composite concentration of 10 μg/mL
Graphene/Ag composites No data 20–25 nm E. coli The disinfection rates were 99.76% (E. coli) and 99.80% [34]
C. albicans (C. albicans) at the composite concentration of 50 µg/mL
GO/TiO2 nanorod 40 wt.% 3–5 nm in E. coli No sunlight, no toxicity to the E. coli cells. Under sunlight, [35]
composites diameter 25–50 nm 90% of E. coli was inactivated after 27 min
in length
RGO/Ag composites No data 10–40 nm E. coli Minimum inhibitory concentration [44]
Listonella anguillarum (MIC) was 0.16 µg/mL against all bacteria
S. aureus Bacillus cereus
RGO/Ag composites No data 100 nm E. coli Composites clearly showed antibacterial activity against [48]
E. coli. GO alone had almost no effect
(Continued)

Graphene Content
GO/poly(vinyl-N- 3 wt.% E. coli 27% (E. coli), 14% (B. subtilis), 30% (C. metallidurans), and [46]
carbazole) Cupriavidus metallidurans 12% (R. opacus) higher loss of viability after exposure to the
B. subtilis composite than GO alone
R. opacus
Graphene/chitosan/ 4 wt.% E. coli Agrobacterium The antibacterial activity was enhanced considerably after [45]
poly(vinyl alcohol) fibers introducing graphene
RGO/naphthalene No data 3–20 nm E. coli At a concentration of 946 μg/mL, the sterilizing rate against [49]
sulfonate/Ag hybrids S. aureus E. coli and S. aureus is 7.5% and 11.2%, respectively
RGO/gentamicin 54 wt.% E. coli The viability losses of E. coli were 82.2% (composite) and [50]
composites 43.5% (graphene) at the sample concentration of 40 µg/mL
Graphene/Ag composites 99.4 wt.% E. coli The inhibition zone was 18.9 mm at a sample concentration of [51]
45–50 nm 100 μg/mL for 24 h
GO/heparin– No data E. coli At a concentration of 250 μg/mL, both inhibited more than [6]
benzalkoniumchloride S. aureus 99% of E. coli and S. aureus
(C12) composite
GO films No data 8 nm E. coli GO acts as an enhancer of life, increasing mammalian cell and [53]
GO/Ag films bacterial growth. GO/Ag showed large growth inhibition
zones with no bacteria
GO/polyethylene/ 0.2 wt.% E. coli Almost no activation of adhered E. coli and S. aureus on the [54]
2-(methacryloyloxy)ethyl S. aureus surface of the films
phosphorylcholine films
Ag/graphene composites 9.42 wt.% 1–5 nm E. coli The composites inhibited the growth of E. coli and S. aureus [56]
S. aureus completely at concentrations higher than 30 and 60 µg/cm3,
respectively
Ag/Fe2O3/GO composites No data E. coli The composites against E. coli was only 71.71 μg/mL [57]
S. aureus (143.43 μg/mL against B. subtilis)
Ag/graphene composites No data 8–20 nm E. coli B. cereus, L. anguillarum, and E. coli did not grow at the [58]
L. anguillarum composite concentration of 0.16 µg/mL, while S. aureus grew
S. aureus until the concentration reached 1.3 µg/ mL
B. cereus
GO/Ag composites No data 7.5 nm P. aeruginosa 100% inhibition rate of the adhered cells after exposure to the [59]
nanocomposite for 1 h
Ag/halloysite/RGO No data 5–15 nm E. coli MIC of all the nanomaterials followed the order: [60]
composites S. aureus RGO > Ag > Ag/HNTs > Ag/RGO > Ag/HNTs/RGO.
Ag/HNTs/RGO exhibited the lowest MIC of 2 µg/mL
GO/chitosan film 98.5 wt.% S. aureus The composite layers showed significant antibacterial activity [61]
(>77% inactivation after 3 h) against S. aureus bacteria
GO nanopaint 84 wt.% E. coli The MIC values of GO sheets against E. coli, S. aureus, and [62]
S. aureus P. aeruginosa were 0.5, 0.5, and 1 µg/mL, respectively
P. aeruginosa
Au/Rubpy/GO composite No data 10 − 20 nm E. coli Less than 8% for E. coli and 3% for S. aureus survived after [63]
S. aureus photothermal treatment
GO/TiO2/Ag composites 4–6 nm (Ag) E. coli The composites showed an excellent intrinsic antibacterial [64]
3–5 nm (TiO2) activity toward E. coli due to the Ag nanoparticles contents
Ag/graphene/polymer 33–83 wt.% E. coli No E. coli colonies remain on the agar plate treated with [65]
hydrogel 11.5–39.0 nm S. aureus hydrogels; only two S. aureas colonies remain
Ag–Fe3O4/RGO No data 20 nm E. coli The large diameter of inhibition zone indicated the [66]
composite antibacterial properties of Ag–Fe3O4/RGO composite
Au/TiO2/RGO composites 40 nm (Au) E. coli The loss of E. coli viability was 90.21% at a composite [67]
concentration of 150 µg/mL
Graphene/quantum dots No data E. coli More than 50% bacteria were killed with 100 µg/mL [68]
S. aureus GQD + 15 min photoirradiation while 200 µg/mL of
photoexcited GQD killed almost all bacteria
(Continued)

Graphene Content
GO/PVA/chitosan ~99 wt.% E. coli The addition of GO to PVA/CS nanofiber had a great ability to [69]
increase inhibition zone till 8.6 mm
GO/ Eu–vancomycin No data E. coli All E. coli were killed at a composite concentration of 10 mg/mL [70]
complex S. aureus while less than 3% of S. aureus survived in the same condition
RGO/TiO2 films No data ~7.0 Å in E. coli Complete sterilization of E. coli O157: H7 (~100%) was done [71]
thickness in 15 min
RGO/ZnO composites 35–65 wt.% 4 nm E. coli The concentration of ZnO in the composites increased to [72]
50 μg/mL and resulted in the complete loss of viability
GO/6-aminoindazole 40.9 wt.% E. coli The GO/6-AIND was found to be active only against S. aureus [73]
S. aureus strain at 512 mg/mL
the edges of RGO nanowalls and the better charge transfer GO to different polymers such as PEI [31], PAA [26], PLL
between the bacteria and RGO nanowalls, resulting in more [43], poly(N-vinyl carbazole) (PVK) [42,48], PDA [31],
damage to the cell membrane of the bacteria. In biomedical polyethylene (PE) [54], and CS/PVA [45]. For example,
applications, an ideal material should also have low mamma- combining PVK with GO, Carpio et al. [46] reported that
lian cell cytotoxicity and superior antimicrobial characteris- GO/PVK composites have very low cytotoxicity, making
tics [53]. them promising materials for human-related applications.
The modification and combination of graphene-based Moreover, it was possible to reduce the concentration of GO
materials with antibacterial agents could be ways of meeting to a level that was nontoxic to humans and animals but the
these challenges. Moreover, these processes can be tailored antimicrobial property of GO was still maintained. Lu et al.
to reduce the environmental risks and/or increase the appli- [45] examined the antibacterial activity of CS/PVA nano-
cation potential of the obtained products [1]. As presented fiber membranes containing graphene against E. coli and
above, a range of antibacterial agents, including inorganic Agrobacterium. They reported that the antibacterial activity
materials (metal, metal oxide, and salt), small organic com- of graphene/CS/PVA nanofibers was enhanced remarkably
pounds, and polymers have been used to combine with gra- compared to that of CS/PAV nanofibers. The nanofibers can
phene-based materials. Among them, Ag nanoparticles (Ag be used directly for wound healing. Figure 23.11 presents
NPs) with excellent antibacterial activity and low cytotoxic- the mechanism and images of the healing process of mice
ity to human cells were used mostly on graphene sheets, in wounds.
particular, GO and RGO [24,25,27–33]. After depositing the Graphene in combination with properly designed antibac-
graphene sheets, the stability and antibacterial activity of the terial agents can tailor the functionalities of the final products
Ag NPs were enhanced. In addition, tiny Ag NPs were loaded to fit into special applications (Table 23.3). For this purpose,
onto the surface to prevent aggregation and maintain their a range of agents have been combined directly with graphene
high surface reactivity. For example, Xu et al. [30] reported such as heparin benzalkonium chloride [6], BP [39], chloro-
the preparation of uniform and water-soluble RGO/Ag nano- phyll [40], and gentamicin [50]. Mondal et al. [40] coated gra-
composites, which displayed superior antibacterial properties phene with a chlorophyll pendant, which is an agent lethal to
to those of pure silver nanoparticles. Similarly, Zhang et al. the DNA of bacteria. The as-prepared product was found to
[24] showed that the growth inhibition rates against E. coli be more than twice as effective as unmodified graphene and
were 38% and 99.9% by testing with bare GO and GO/Ag GO. This system can be applied to a wide range of industries
composites, respectively, at a concentration of 10 ppm. Figure ranging from the food packaging to antimicrobial paper wrap-
23.10 gives a plausible antibacterial mechanism of GO/Ag ping. Wang et al. [39] developed novel films based on GO,
against E. coli. A silver salt was also reported to improve the BP (a bactericidal antibiotic), and Mg–Al LDH. Products with
antibacterial activity of GO [37]. Their GO/Ag3PO4 compos- multifunctionality, including biocompatibility, drug load-
ites showed both excellent intrinsic antibacterial and visible- ing and delivery, and antibacterial activities were suggested.
light-induced photocatalytic disinfection activities toward E. Panday et al. [50] reported graphene materials as a controlled
coli cells due to the synergistic effects of GO sheets and an release matrix for antibiotic drugs, gentamicin sulfate (Figure
Ag3PO4 photocatalyst. These composites are expected to be 23.12). This nanohybrid can be used as a single medication
applied to the removal of organic pollutants and pathogenic for the treatment of many topical bacterial infections. Wang
bacteria for water purification. et al. [2] examined the antimicrobial activity of different car-
To reduce the toxicity of GO and ensure its safe use for bon nanomaterials such as single- and multiwalled carbon
medical applications, attempts have been made to conjugate nanotubes, GO, and RGO. The antibacterial activity of GO
E. coli bacteria
++ + +
+++ O +
++ AgNP
AgNP
HO O O
++
++ + OH
O O
++++ O
AgNP AgNP
AgNP
AgNP e– AgNP
AgNP
e– OH HO e– O
AgNP HO AgNP
O O AgNP
AgNP AgNP
HO O AgNP O
AgNP AgNP AgNP
HO AgNP AgNP
O O HO
HO OH AgNP
e– O O
O AgNP
e–
++ + AgNP e– O
AgNP +
e–
++++ AgNP O AgNP
+
++ ++
+ AgNP ++ ++
++ ++
e– Electron transfer
AgNP AgNP
FIGURE 23.10 Plausible antibacterial mechanism of AgGO against E. coli. (From S.W. Chook et al., Nanoscale Res. Lett. 7, 541, 2012.
With permission.)
Prokaryotic cell
(a) DNA
Graphene Electron
Eukaryotic cell
(b)
FIGURE 23.11 (a) Possible mechanism for graphene as an antibacterial material that can be beneficial for wound healing and (b) images
of a mouse wound healing under (1) pure chitosan–PVA nanofibers, (2) chitosan–PVA nanofibers containing graphene, and (3) control.
(Adapted from B. Lu et al. Nanoscale 4, 2978–2982, 2012, with permission.)
(a) (b)
62.75%
3.0
pH
36.9%
pH 7.4
(c)
(d) 500
0.74 nm pH
400 9. 52.8%
0
300 0.34 nm
CH3
200 H2N
HO O
HO O R3
100 HN
OH O O C
H3C R2
0 10 20 30 40 50 60 R1
H2N NH2
20 degree
FIGURE 23.12 (a) SEM image of as prepared methanol derived graphene (MDG) nanosheets weakly held together; (b) digital photograph
of MDG nanosheets, not dispersed in water (left) and dispersed in water after drug conjugation (right); (c) schematic representation of the
release behavior of the gentamicin drug from MDG nanosheets as a function of pH; and (d) X-ray diffraction analysis of as prepared MDG
nanosheets. (From H. Pandey et al., Nanoscale, 3, 4104, 2011. With permission.)
can be applied to bacterial pathogens, either resistant or non- REFERENCES

resistant, and be used as an antimicrobial material to protect
plants from diseases. 1. S. Liu, T. H. Zeng, M. Hofmann, E. Burcombe, J. Wei, R.
Jiang, J. Kong, Y. Chen, Antibacterial activity of graphite,
graphite oxide, graphene oxide, and reduced graphene oxide:
Membrane and oxidative stress, ACS Nano 5, 6971–6980,
23.4 CONCLUSIONS AND PERSPECTIVES 2011.
In addition to electronic applications, a wide variety of gra- 2. X. Wang, X. Liu, H. Han, Evaluation of antibacterial effects of
carbon nanomaterials against copper-resistant Ralstonia sola-
phene-based materials have also been developed for antibac-
nacearum, Colloids Surf. B 103, 136–142, 2013.
terial applications. As shown in the first section, different 3. W. Hu, C. Peng, W. Luo, M. Lv, X. Li, D. Li, Q. Huang, C. Fan,
forms of graphene and its composites have been prepared Graphene-based antibacterial paper, ACS Nano 4, 4317–4323,
by many researchers using a range of techniques. The modi- 2010.
fication or functionalization of GO and RGO with antibac- 4. S. Gurunathan, J. W. Han, A. A. Dayem, V. Eppakayala, J.
terial agents is very important for tuning the properties of H. Kim, Oxidative stress-mediated antibacterial activity of
materials and in the expansion of the applications of gra- graphene oxide and reduced graphene oxide in Pseudomonas
aeruginosa, Int. J. Nanomed. 7, 5901–5914, 2012.
phene-based materials. In the second section, the potential
5. K. Krishnamoorthy, N. Umasuthan, R. Mohan, J. Lee, S. J.
applications of graphene-based materials for the antibacte- Kim, Antibacterial activity of graphene oxide nanosheets, Sci.
rial field have been illustrated with several proposed mecha- Adv. Mater. 4, 1111–1117, 2012.
nisms. Although some techniques have been performed to 6. N. Zhou, N. Meng, Y. Ma, X. Liao, J. Zhang, L. Li, J. Shen,
confirm these antibacterial mechanisms, it is important to Evaluation of antithrombogenic and antibacterial activities of
make continuous efforts to improve the performance. In a graphite oxide/heparin–benzalkonium chloride composite,
addition, for the use in daily life, the health risks associated Carbon 47, 1343–1350, 2009.
7. N. Mohanty, M. Fahrenholtz, A. Nagaraja, D. Boyle, V. Berry,
with graphene-based materials must be solved by evaluat-
Impermeable graphenic encasement of bacteria, Nano Lett.
ing the biocompatibility of these materials and preventing 11, 1270−1275, 2011.
potential toxicity. 8. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y.
Zhang, S. V. Dubonos, I. V. Grigorieva, A. A. Firsov, Electric
field effect in atomically thin carbon films, Science 306, 666–
ACKNOWLEDGMENT 669, 2004.
9. O. C. Compton, S. T. Nguyen, Graphene oxide, highly reduced
This study was supported by Basic Science Research graphene oxide, and graphene: Versatile building blocks for
Program (2012009529) and Priority Research Centers carbon-based materials, Small 6, 711–723, 2010.
Program (2014R1A6A1031189) through the National 10. H. Bai, C. Li, G. Shi, Functional composite materials based on
Research Foundation of Korea (NRF) funded by the Ministry chemically converted graphene, Adv. Mater. 23, 1089–1115,
of Education. 2011.
11. S. Park, R. S. Ruoff, Chemical methods for the production of 31. X. Cai, M. Lin, S. Tan, W. Mai, Y. Zhang, Z. Liang, Z. Lin, X.
graphenes. Nat. Nanotechnol. 4, 217–224, 2009. Zhang, The use of polyethyleneimine-modified reduced gra-
12. W. S. Hummers, R. E. Offeman, Preparation of graphitic phene oxide as a substrate for silver nanoparticles to produce
oxide, J. Am. Chem. Soc. 80, 1339–1339, 1958. a material with lower cytotoxicity and long-term antibacterial
13. B. C. Brodie, On the atomic weight of graphite, Philos. Trans. activity, Carbon 50, 3407–3415, 2012.
R. Soc. London 149, 249–259, 1859. 32. Q. Bao, D. Zhang, P. Qi, Synthesis and characterization of sil-
14. L. Staudenmaier, Method for the preparation of graphitic acid, ver nanoparticle and graphene oxide nanosheet composites as
Ber. Dtsch. Chem. Ges. 31, 1481–1487, 1898. a bactericidal agent for water disinfection, J. Colloid Interface
15. F. S. Hyde, Graphitic acid or oxide, J. Soc. Chem. Ind. 23, Sci. 360, 463–470, 2011.
300–302, 1904. 33. Z. Zhang, J. Zhang, B. Zhang, J. Tang, Mussel-inspired func-
16. K. Krishnamoorthy, M. Veerapandian, L. H. Zhang, K. Yun, tionalization of graphene for synthesizing Ag-polydopamine-
S. J. Kim, Antibacterial efficiency of graphene nanosheets graphene nanosheets as antibacterial materials, Nanoscale 5,
against pathogenic bacteria via lipid peroxidation, J. Phys. 118–123, 2013.
Chem. C 116, 17280−17287, 2012. 34. W. Yuana, Y. Gua, L. Li, Green synthesis of graphene/Ag
17. O. Akhavan, E. Ghaderi, Escherichia coli bacteria reduce gra- nanocomposites, Appl. Surf. Sci. 261, 753–758, 2012.
phene oxide to bactericidal graphene in a self-limiting man- 35. J. Liu, L. Liu, H. Bai, Y. Wang, D. D. Sun, Gram-scale pro-
ner, Carbon 50, 1853–1860, 2012. duction of graphene oxide–TiO2 nanorod composites: Towards
18. S. Liu, M. Hu, T. H. Zeng, R. Wu, R. Jiang, J. Wei, L. Wang, high-activity photocatalytic materials, Appl. Catal. B 106,
J. Kong, Y. Chen, Lateral dimension-dependent antibacterial 76–82, 2011.
activity of graphene oxide sheets, Langmuir 28, 12364−12372, 36. T. Kavitha, A. I. Gopalan, K. P. Lee, S. Y. Park, Glucose
2012. sensing, photocatalytic and antibacterial properties of gra-
19. O. Akhavan, E. Ghaderi, A. Esfandiar, Wrapping bacteria by phene–ZnO nanoparticle hybrids, Carbon 50, 2994–3000,
graphene nanosheets for isolation from environment, reactiva- 2012.
tion by sonication, and inactivation by near-infrared irradia- 37. L. Liu, J. Liu, D. D. Sun, Graphene oxide enwrapped Ag3PO4
tion, J. Phys. Chem. B 115, 6279–6288, 2011. composite: Towards a highly efficient and stable visible-
20. O. Akhavan, E. Ghaderi, Toxicity of graphene and graphene light-induced photocatalyst for water purification, Catal. Sci.
oxide nanowalls against bacteria, ACS Nano 4, 5731–5736, 2010. Technol. 2, 2525–2532, 2012.
21. S. Gurunathan, J. W. Han, A. A. Dayem, V. Eppakayala, M. 38. X. Cai, S. Tan, M. Lin, A. Xie, W. Mai, X. Zhang, Z. Lin,
R. Park, D. N. Kwon, J. H. Kim, Antibacterial activity of T. Wu, Y. Liu, Synergistic antibacterial brilliant blue/reduced
dithiothreitol reduced graphene oxide, J. Ind. Eng. Chem. 19, graphene oxide/quaternary phosphonium salt composite with
1280–1288, 2013. excellent water solubility and specific targeting capability,
22. D. C. Luo, G. X. Zhang, J. F. Liu, X. M. Sun, Evaluation crite- Langmuir 27, 7828–7835, 2011.
ria for reduced graphene oxide, J. Phys. Chem. C 115, 11327– 39. Y. Wang, D. Zhang, Q. Bao, J. Wuab, Y. Wan, Controlled
11335, 2011. drug release characteristics and enhanced antibacterial
23. P. Laaksonen, M. Kainlauri, T. Laaksonen, A. Shchepetov, effect of graphene oxide–drug intercalated layered double
H. Jiang, J. Ahopelto, M. B. Linder, Interfacial engineering hydroxide hybrid films, J. Mater. Chem. 22, 23106–23113,
by proteins: Exfoliation and functionalization of graphene by 2012.
hydrophobins, Angew. Chem. Int. Ed. 49, 4946–4949, 2010. 40. T. Mondal, A. K. Bhowmick, R. Krishnamoorti, Chlorophenyl
24. D. Zhang, X. Liu, X. Wang, Green synthesis of graphene oxide pendant decorated graphene sheet as a potential antimicrobial
sheets decorated by silver nanoprisms and their anti-bacterial agent: Synthesis and characterization, J. Mater. Chem. 22,
properties, J. Inorg. Biochem. 105, 1181–1186, 2011. 22481–22487, 2012.
25. L. Liu, J. Liu, Y. Wang, X. Yan, D. D. Sun, Facile synthe- 41. H. N. Lim, N. M. Huang, C. H. Loo, Facile preparation of
sis of monodispersed silver nanoparticles on graphene oxide graphene-based chitosan films: Enhanced thermal, mechani-
sheets with enhanced antibacterial activity, New J. Chem. 35, cal and antibacterial properties, J. Non-Cryst. Solids 358,
1418–1423, 2011. 525–530, 2012.
26. Z. Tai, H. Ma, B. Liu, X. Yan, Q. Xue, Facile synthesis of 42. C. M. Santos, J. Mangadlao, F. Ahmed, A. Leon, R. C.
Ag/GNS-g-PAA nanohybrids for antimicrobial applications, Advincula, D. F. Rodrigues, Graphene nanocomposite for bio-
Colloids Surf. B 89, 147–151, 2012. medical applications: Fabrication, antimicrobial and cytotoxic
27. J. Shen, M. Shi, B. Yan, H. Ma, N. Li, M. Ye, One-pot hydro- investigations, Nanotechnology 23, 395101–395110.
thermal synthesis of Ag-reduced graphene oxide composite 43. S. Some, S. M. Ho, P. Dua, E. Hwang, Y. H. Shin, H. Yoo, J. S.
with ionic liquid, J. Mater. Chem. 21, 7795–7801, 2011. Kang, D. K. Lee, H. Lee, Dual functions of highly potent gra-
28. S. W. Chook, C. H. Chia, S. Zakaria, M. K. Ayob, K. L. Chee, phene derivative poly(L-Lysine) composites to inhibit bacteria
N. M. Huang, H. M. Neoh, H. N. Lim, R. Jamal, R. M. F. R. A. and support human cells, ACS Nano 6, 7151–7161, 2012.
Rahman, Antibacterial performance of Ag nanoparticles and 44. V. H. Nguyen, B. K. Kim, Y. L. Jo, J. J. Shim, Preparation
AgGO nanocomposites prepared via rapid microwave-assisted and antibacterial activity of silver nanoparticles-decorated
synthesis method, Nanoscale Res. Lett. 7, 541–547, 2012. graphene composites, J. Supercrit. Fluids 72, 28–35, 2012.
29. J. Shen, M. Shi, N. Li, B. Yan, H. Ma, Y. Hu, M. Ye, Facile 45. B. Lu, T. Li, H. Zhao, X. Li, C. Gao, S. Zhang, E. Xie,
synthesis and application of Ag-chemically converted gra- Graphene-based composite materials beneficial to wound
phene nanocomposite, Nano Res. 3, 339–349, 2010. healing, Nanoscale 4, 2978–2982, 2012.
30. W. P. Xu, L. C. Zhang, J. P. Li, Y. Lu, H. H. Li, Y. N. Ma, W. 46. I. E. M. Carpio, C. M. Santos, X. Wei, D. F. Rodrigues,
D. Wang, S. H. Yu, Facile synthesis of silver@graphene oxide Toxicity of a polymer-graphene oxide composite against
nanocomposites and their enhanced antibacterial properties, bacterial planktonic cells, biofilms, and mammalian cells,
J. Mater. Chem. 21, 4593–4597, 2011. Nanoscale, 4, 4746–4757, 2012.
47. O. Akhavan, E. Ghaderi, Photocatalytic reduction of graphene 64. L. Liu, H. Bai, J. Liu, D. D. Sun, Multifunctional graphene
oxide nanosheets on TiO2 thin film for photoinactivation of oxide-TiO2-Ag nanocomposites for high performance water
bacteria in solar light irradiation, J. Phys. Chem. C 113, disinfection and decontamination under solar irradiation, J.
20214–20220, 2009. Hazard. Mater. 261, 214–223, 2013.
48. C. Dai, X. Yang, H. Xie, One-step synthesis of reduced graph- 65. Z. Fan, B. Liu, J. Wang, S. Zhang, Q. Lin, P. Gong, L. Ma, S.
ite oxide–silver nanocomposite, Mater. Res. Bull. 46, 2004– Yang, A novel wound dressing based on Ag/graphene polymer
2008, 2011. hydrogel: Effectively kill bacteria and accelerate wound heal-
49. X. Cai, S. Tan, A. Yu, J. Zhang, J. Liu, W. Mai, Z. Jiang, ing, Adv. Funct. Mater. 24, 3933–3943, 2014.
Sodium 1-naphthalenesulfonate-functionalized reduced gra- 66. M. K. Joshi, H. R. Pant, H. J. Kim, J. H. Kim, C. S. Kim,
phene oxide stabilizes silver nanoparticles with lower cyto- One-pot synthesis of Ag-iron oxide/reduced graphene oxide
toxicity and long-term antibacterial activity, Chem. Asian J. 7, nanocomposite via hydrothermal treatment, Colloids. Surf. A
1664–1670, 2012. 446, 102–108, 2014.
50. H. Pandey, V. Parashar, R. Parashar, R. Prakash, P. W. Ramteke, 67. W. He, H. Huang, J. Yan, J. Zhu, Photocatalytic and antibac-
A. C. Pandey, Controlled drug release characteristics and terial properties of Au-TiO2 nanocomposite on monolayer
enhanced antibacterial effect of graphene nanosheets contain- graphene: From experiment to theory, J Appl. Phys. 114,
ing gentamicin sulfate, Nanoscale 3, 4104–4108, 2011. 204701–204712, 2013.
51. B. Jiang, C. Tian, G. Song, W. Chang, G. Wang, Q. Wu, H. 68. B. Z. Ristic, M. M. Milenkovic, I. R. Dakic, B. M. Todorovic-
Fu, A novel Ag/graphene composite: Facile fabrication and Markovic, M. S. Milosavljevic, M. D. Budimir, V. G.
enhanced antibacterial properties, J. Mater. Sci. 48, 1980– Paunovic, M. D. Dramicanin, Z. M. Markovic, V. S. Trajkovic,
1985, 2013. Photodynamic antibacterial effect of graphene quantum dots,
52. B. Yuan, T. Zhu, Z. Zhang, Z. Jiang, Y. Ma, Self-assembly of Biomaterials 35, 4428–4435, 2014.
multilayered functional films based on graphene oxide sheets 69. Y. Liu, M. Park, H. K. Shin, B. Pant, J. Choi, Y. W. Park, J. Y.
for controlled release, J. Mater. Chem. 21, 3471–3476, 2011. Lee, S. J. Park, H. Y. Kim, Facile preparation and characteriza-
53. O. N. Ruiz, K. A. S. Fernando, B. Wang, N. A. Brown, P. G. tion of poly(vinyl alcohol)/chitosan/graphene oxide biocom-
Luo, N. D. McNamara, M. Vangsness, Y. P. Sun, C. E. Bunker, posite nanofibers, J. Ind. Eng. Chem. 20, 4415–4420, 2014.
Graphene oxide: A nonspecific enhancer of cellular growth, 70. X. Yang, Z. Li, E. Ju, J. Ren, X. Qu, Reduced graphene oxide
ACS Nano 5, 8100–8107, 2011. functionalized with a luminescent rare-earth complex for the
54. S. Jin, D. Xu, N. Zhou, J. Yuan, J. Shen, Antibacterial and tracking and photothermal killing of drug-resistant bacteria,
anticoagulation properties of polyethylene/GeneO-MPC Chem. Eur. J. 20, 394–398, 2014.
nanocomposites, J. Appl. Polym. Sci. 129, 884–891, 2013. 71. I. Y. Kim, S. Park, H. Kim, S. Park, R. S. Ruoff, S. J. Hwang,
55. K. Krishnamoorthy, U. Navaneethaiyer, R. Mohan, J. Lee, S. Strongly-coupled freestanding hybrid films of graphene and
J. Kim, Graphene oxide nanostructures modified multifunc- layered titanate nanosheets: An effective way to tailor the
tional cotton fabrics, Appl. Nanosci. 2, 119–126, 2012. physicochemical and antibacterial properties of graphene film,
56. G. Y. He, W. Dai, Y. T. Zhao, Q. Chen, X. Q. Sun, H. Q. Chen, Adv. Funct. Mater. 24, 2288–2294, 2014.
X. Wang, A facile synthesis of Ag@graphene-nanosheet com- 72. Y. W. Wang, A. Cao, Y. Jiang, X. Zhang, J. H. Liu, Y. Liu,
posite with enhanced antibacterial activity and acceptable H. Wang, Superior antibacterial activity of zinc oxide/gra-
environmental safety, Monatsh. Chem. 145, 3–10, 2014. phene oxide composites originating from high zinc concentra-
57. N. Gao, Y. Chen, J. Jiang, Ag@Fe2O3-GO nanocomposites tion localized around bacteria, ACS Appl. Mater. Interfaces 6,
prepared by a phase transfer method with long-term antibac- 2791−2798, 2014.
terial property, ACS Appl. Mater. Interfaces 5, 11307−11314, 73. S. S. Maktedar, S. S. Mehetre, M. Singh, R. K. Kale,
2013. Ultrasound irradiation: A robust approach for direct function-
58. Y. Haldorai, B. K. Kim, Y. L. Jo, J. J. Shim, Ag A facile green alization of graphene oxide with thermal and antimicrobial
synthetic approach, Mater. Chem. Phys. 143, 1452–1461, aspects, Ultrason. Sonochem. 21, 1407–1416, 2014.
2014. 74. X. An, J. C. Yu, Graphene-based photocatalytic composites,
59. A. F. Faria, D. S. T. Martinez, S. M. M. Meira, A. C. M. Moraes, RSC Adv. 1, 1426–1434, 2011.
A. Brandelli, A. G. S. Filho, O. L. Alves, Anti-adhesion and 75. M. Fang, J. H. Chen, X. L. Xu, P. H. Yang, H. F. Hildebrand,
antibacterial activity of silver nanoparticles supported on gra- Antibacterial activities of inorganic agents on six bacteria
phene oxide sheets, Colloids Surf. B 113, 115–124, 2014. associated with oral infections by two susceptibility tests, Int.
60. L. Yu, Y. Zhang, B. Zhang, J. Liu, Enhanced antibacterial J. Antimicrob. Agents 27, 513−517, 2006.
activity of silver nanoparticles/halloysite nanotubes/graphene 76. J. Sawai, T. Yoshikawa, Quantitative evaluation of antifun-
nanocomposites with sandwich-like structure, Sci. Rep. 11, gal activity of metallic oxide powders (MgO, CaO and ZnO)
4551–4556, 2014. by an indirect conductimetric assay, J. Appl. Microbiol. 96,
61. M. Mazaheri, O. Akhavan, A. Simchi, Flexible bactericidal 803−809, 2004.
graphene oxide–chitosan layers for stem cell proliferation, 77. K. R. Raghupathi, R. T. Koodali, A. C. Manna, Size-
Appl. Surf. Sci. 301, 456–462, 2014. dependent bacterial growth inhibition and mechanism of anti-
62. K. Krishnamoorthy, K. Jeyasubramanian, M. Premanathan, bacterial activity of zinc oxide nanoparticles, Langmuir 27,
G. Subbiah, H. S. Shin, S. J. Kim, Graphene oxide nanopaint, 4020−4028, 2011.
Carbon 72, 328–337, 2014. 78. M. Veerapandian, K. S. Yun, Functionalization of biomole-
63. D. Lin, T. Qin, Y. Wang, X. Sun, L. Chen, Graphene oxide cules on nanoparticles specialized for anti-bacterial applica-
wrapped SERS tags: Multifunctional platforms toward opti- tions, Appl. Microbiol. Biotechnol. 90, 1655−1667, 2011.
cal labeling, photothermal ablation of bacteria, and the 79. A. Kumar, P. K. Vemula, P. M. Ajayan, G. John, Silver-
monitoring of killing effect, ACS Appl. Mater. Interfaces 6, nanoparticle-embedded antimicrobial paints based on veg-
1320 − 1329, 2014. etable oil, Nat. Mater. 7, 236−241, 2008.
80. S. Yin, Y. Goldovsky, M. Herzberg, L. Liu, H. Sun, Y. Zhang, 82. J. Li, G. Wang, H. Zhu, M. Zhang, X. Zheng, Z. Di, X. Liu,
F. Meng et al. Functional Free-Standing Graphene Honeycomb X. Wang, Antibacterial activity of large-area monolayer
Films, Adv. Funct. Mater. 23, 1–7, 2013. graphene film manipulated by charge transfer, Sci. Rep. 4,
81. M. Veerapandian, L. Zhang, K. Krishnamoorthy, K. Yun, 4359–4367, 2014.
Surface activation of graphene oxide nanosheets by ultraviolet
irradiation for highly efficient anti-bacterials, Nanotechnology
24, 395706–395717, 2013.
24 Nanofluidics in Graphene-Based
Material Systems
Ling Liu and Lin Zhang
CONTENTS
Abstract...................................................................................................................................................................................... 465
24.1 Introduction...................................................................................................................................................................... 465
24.2 Uniqueness of Nanofluidics.............................................................................................................................................. 466
24.3 Nanofluid Structures and Transport in Graphene-Based Material Systems..................................................................... 468
24.3.1 Structural Properties............................................................................................................................................. 469
24.3.2 Transport Properties............................................................................................................................................. 470
24.4 Nanofluid Dynamical Behaviors in Graphene-Based Material Systems.......................................................................... 471
24.4.1 Nanofluid Morphology......................................................................................................................................... 471
24.4.2 Nanofluid Evaporation.......................................................................................................................................... 471
24.5 Nanofluid Purification in Graphene-Based Material Systems.......................................................................................... 472
24.5.1 Purification by Nanoporous Graphene................................................................................................................. 472
24.5.2 Purification by Graphene-Based Membranes....................................................................................................... 474
24.6 Conclusion........................................................................................................................................................................ 474
References.................................................................................................................................................................................. 474
ABSTRACT achievable at larger scales. Confined in nanopores sized from

sub-nanometer to a few nanometers, most ions and molecules
Nanofluidics is an emerging area that has attracted enormous of nanofluids are in close contact with the atoms of surround-
attention in recent years. Under nanoconfinement, fluids usu- ing solid materials. The ultra-large solid–fluid interfaces
ally exhibit unconventional and sometimes counter-intuitive strongly perturb nanofluid behaviors, leading to many unique
behaviors. This has led to a variety of promising applica- properties and behaviors of nanofluids, for example, low
tions of nanofluidics in energy conversion and storage, transport viscosity (Chen et al. 2008), reduced ionic solubility
phase separation, and mass transport, among others. Recent (Malani et al. 2006), hydration (Tang et al. 2002), ion-specific
advances in synthesis and fabrication techniques have made infiltration (Liu et al. 2009a), and diffusion (Lynden-Bell and
graphene monolayer, nanoribbon, nanoporous graphene, and Rasaiah 1996). Due to these superior properties and intrigu-
their derivatives and hybrids a promising material platform ing behaviors, nanofluidics has become an important source
for constructing highly customizable nanofluidic systems of for engineering innovations in energy absorption (Qiao et al.
superior properties. This chapter reviews several fundamental 2009; Surani et al. 2005), actuation (Liu et al. 2009c; Xu
aspects of nanofluidics in graphene-based material systems. et al. 2011), and energy and environmental sustainability (An
Molecular structures, transport behaviors, dynamical mor- et al. 2009; Carruthers et al. 2012; Kim et al. 2010; Yuan et al.
phologies, evaporation physics, and purification processes of 2008), among others.
nanofluids confined in such novel nanofluidic systems will be Many materials can provide nanocavities to accommodate
systematically introduced and discussed. This short review is nanofluids, including synthetic nanoporous materials such as
expected to promote the understanding of nanofluidics, and metal–organic frameworks (MOFs) (Li et al. 1999) and cova-
to provide valuable design guidelines for building graphene- lent organic frameworks (COFs) (El-Kaderi et al. 2007), natu-
based nanofluidic systems. ral materials such as zeolites and silica, biomaterials such as
various biological ion channels, and synthetic low-dimensional
materials such as carbon nanotubes (CNTs) (Hummer et al.
24.1 INTRODUCTION
2001) and boron nitride nanotubes (BNTs) (Won and Aluru
Nanofluidics is an emerging research area that studies intrigu- 2007). Among these examples (Figure 24.1), carbon nanotube
ing properties and behaviors of single/multiphase fluids that (CNT)-confined nanofluidics is arguably the most extensively
are confined in structures of nanoscale dimensions. It has investigated in recent years, owing to the fact that CNTs have
attracted enormous attention in the past decade due to the many attributes that can significantly enhance desired prop-
increasing demand for extraordinary fluid properties not erties of nanofluids, making them advantageous in many
465
(a) (b) exploring the structures, transport, electrochemistry, and ener-

(d)
Trp48
getics of graphene-confined nanofluids. Several possibilities
have been shown for graphene-based nanofluidics to under-
Arg206
pin novel applications addressing pressing healthcare, safety,
energy, and sustainability issues. These applications are envi-
sioned to continuously stimulate and benefit from fundamental
Phe200
Asn203 nanofluidics research, which will be a focus of this review.
(c)
Asn68 In this chapter, we will first elucidate several important
aspects of nanofluidics that make it distinct from classical
His66 fluid mechanics (Section 24.2). In Section 24.3, we will use
Ala65 a simplified model to illustrate the unique structures and
Gly64
transport behaviors of water confined in graphene-based
3 nm systems. In Section 24.4, several case studies will be intro-
duced to demonstrate the complex dynamics of nanofluids
FIGURE 24.1 Nanofluidics in several representative forms of in graphene-based confinement. In Section 24.5, several rep-
nanoconfinement: (a) zeolites (Liu, L. and X. Chen: Inside cover: resentative graphene-based systems will be introduced to
Fast ion transport and phase separation in a mechanically driven illustrate the permeation dynamics of fluids through various
flow of electrolytes through tortuous sub-nanometer nanochan- graphene-based porous materials and the associated phase
nels (Chem Phys Chem 11/2013). Chem Phys Chem. 2013b. 14 (11). separation applications. Insights from these theoretical and
2354–2354. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. experimental studies are expected to improve the understand-
Reproduced with permission); (b) CNTs (Reprinted with permission
ing of nanofluidics in graphene-based material systems, and
from Macmillan Publishers Ltd. Nature, Hummer, G., J. C. Rasaiah,
and J. P. Noworyta. 414 (6860):188–190, copyright 2001); (c) TEM to guide future designing efforts for novel nanofluidic applica-
micrograph of a water strand confined in a CNT with the diameter tions with extraordinary performance.
of a few nanometers (Reprinted with permission from Naguib, N. et
al. Observation of water confined in nanometer channels of closed
carbon nanotubes. Nano Letters 4 (11): 2237–2243. Copyright 2004
24.2 UNIQUENESS OF NANOFLUIDICS
American Chemical Society); and (d) a representative biological Nanofluidics is a subject exploring physics (i.e., structures,
channel (i.e., GlpF) in biological systems (From Hub, J. S. and B. transport, energetics, and electrochemistry) of fluids under
L. De Groot. Mechanism of selectivity in aquaporins and aqua- nanoconfinement in various materials including solid-
glyceroporins. Proceedings of the National Academy of Sciences of
state nanoporous materials, biological channels, and low-
the United States of America 105 (4):1198–1203, Copyright 2008
National Academy of Sciences, U.S.A.). dimensional nanomaterials (Figure 24.1). Notably, these
confining materials feature distinctive physical and chemical
properties that play key roles in determining the intrinsic and
applications. For example, the simple composition of CNT extrinsic properties of nanofluidics. In solid-state nanoporous
provides a highly stable accommodation for nanofluids; the materials such as zeolites (Polarz and Smarsly 2002), nanopo-
ultrasmooth surface permits ultrafast transport of nanofluids; rous carbon, MOFs (Li et al. 1999), and COFs (El-Kaderi et al.
and moreover, thanks to the development of advanced synthe- 2007), nanochannels are interconnected to form n etworks of
sis techniques, many aspects of CNTs including size, topology, complex topologies and rough surfaces. In biological chan-
and surface chemistry are highly tunable, which not only pro- nels such as AQP1 and ASIC1, their surfaces feature natu-
vides new opportunities for novel applications, but also enables rally derived, site-dependent residues which give rise to
fine-tuning nanofluidic systems for extraordinary performance. inherent multifunctionality for manipulating fluids (de Groot
Recently, graphene monolayer, nanoribbon (Jin et al. 2008), and Grubmuller 2001; Murata et al. 2000; Sui et al. 2001).
nanoporous graphene, and their derivatives and hybrids have In synthetic low-dimensional nanomaterials including CNTs
emerged as a new class of materials for multifunctional appli- and graphene, the nanochannels have more regular geome-
cations (Li et al. 2008; Son et al. 2006). Substantial research tries and atomic arrangements, which have led to intriguing
efforts have been devoted to the exploration and understand- nanofluidics different from that in other nanochannels (Hilder
ing of their remarkable mechanical, thermal, and electronic et al. 2009; Iijima 1991).
properties (Dutta and Pati 2010; Jia et al. 2011). Meanwhile, Despite these differences, nanocavities in these materials
nanofluidics associated with graphene-based materials has share several characteristics in common, for example, ultra-
also shown enormous promises for a wide range of applica- large surface-to-volume ratio (~102–103 m2/g) and ultrasmall
tions. Similar to CNTs, graphene-based materials have simple pore size (<1 nm) (Lu and Zhao 2004). Confined in such
composition, smooth surfaces, and highly tunable physical and small nanocavities, most fluid ions and molecules are in close
chemical properties. Moreover, with their flat surfaces, gra- contact with the confining solid atoms. As such, nanofluids
phene-based materials have more flexibility in constructing have shown several intrinsic properties radically different
nanofluidic systems. As such, they are considered advantageous from that of their bulk counterparts.
over CNTs for many nanofluidic applications, which have been First, the spatial distributions of fluid particles are
demonstrated in recent experimental and theoretical studies inhomogeneous under nanoconfinement. Take the radial

Nanofluidics in Graphene-Based Material Systems 467
(c)
(a) 4
(60,60) (40,40)
3 (30,30)
Radial density profile
Flow
(20,20)
2
(10,10)
0 a
2 3 4 5 6
Radial distance from wall (Å)
(b) b
200
Velocity (m/s)
100
P
0
0 10 20 30 40 LiC1 NaC1 KC1
Radial distance from center (Å)
FIGURE 24.2 (a) Density and (b) velocity profiles of water confined in CNTs. In the density profile, only the region close to the CNT wall
is shown; the multiple curves correspond to CNTs of different sizes. The velocity profile is shown by discrete data points. (c) MD simulation
snapshots of pressurized infiltration of three different electrolyte solutions into a molecular-sized silicon dioxide nanotube. Ions inside the
nanotube are found to form ion couples. Both the separation between two neighboring ion couples (b) and the distance between two paired
ions (a) in an ion couple are found to be dependent on ion size. (Reprinted with permission from Chen, X. et al., Nanoscale fluid transport:
Size and rate effects. Nano Letters 8 (9), 2988–2992. Copyright 2008 American Chemical Society.)
distribution of water in CNT as an example (Figure 24.2a). 1 nm for concentrated salt solutions; Prakash 2007), a num-
Due to the van der Waals (vdW) interaction between water and ber of phenomena negligible in the macroscopic world will
the confining CNT, water molecules at the interface with CNT become dominant. As a result, nanofluid flow usually exhibits
have different balance compared with those in the interior of partial slip against the solid surface, which can be charac-
the water file. As a result, the radial distribution is shown to terized by the so-called slip length (~3.4–68 μm; Majumder
be highly fluctuating close to the CNT surface, with the peak et al. 2005). With the definition of slip length, Thomas and
density almost three times higher than that of the bulk water McGaughey (2008b) modified the Hagen–Poiseuille equation
(Chen et al. 2008; Joseph and Aluru 2008; Poynor et al. 2006). as Qs = π ⋅ [(D/2)4 + 4(D/2)3Ls] ⋅ (∂P/∂z)/(8 μ), where ∂P/∂z is
The situation can become even more complicated if multiple the pressure gradient along the fluid transport direction, μ the
phases are present in the system. For instance, in nanoconfined viscosity of the fluid, D the diameter of nanoconfinement, and
electrolyte solutions (where solvent and solute are considered Ls is the slip length. Note that fluid viscosity in this equation
as different phases), all components would have different inter- can be fundamentally different at the nanoscale, compared to
actions with the solid phase and surrounding liquid particles. its counterpart in the bulk. According to a theoretical investi-
Consequently, the density profiles of each kind of particles (sol- gation (Chen et al. 2008), water confined in a (10,10) CNT has
vent molecule or solute cation/anion) are unique, and they all a viscosity four orders of magnitude lower than that of the bulk
feature inhomogeneous distributions, as shown in Figure 24.2a. water (8.5 × 10−8 Pa s), indicating that water can flow fast in
Moreover, fluids under nanoconfinement have unique trans- CNTs (Hummer et al. 2001; Majumder et al. 2005; Skoulidas
port properties. Figure 24.2b shows the radial velocity profile et al. 2002). Due to the complexity of nanofluid viscosity
of water inside a (60,60) CNT as the entire phase is pressur- (i.e., strong size and flow velocity dependencies; Chen et al.
ized to transport through the CNT at a velocity of 165 m/s. It 2008; Liu et al. 2012b) and slip length (which is not a physi-
is apparent that a plug-flow profile is formed, which is consis- cal parameter), it is difficult to employ models based on these
tent with the literature (Hanasaki and Nakatani 2006; Joseph parameters for a rigorous analysis of transport phenomena
and Aluru 2008) but contradictory to the nonslip assumption of nanofluids. As an alternative, a modified Laplace–Young
widely adopted in continuum fluid mechanics. Indeed, as the equation (Qiao et al. 2009) has been proposed for describing
dimension of nanostructures becomes close to the characteris- the infiltration and transport processes of nanofluids, with key
tic length scale of nanofluids (e.g., the Debye length is less than parameters directly extracted from MD (molecular dynamics)
simulations. The new model does not only account for the ini- 2006; Kalra et al. 2003; Majumder et al. 2005). In addition
tial energy barrier for a fluid to enter a nanopore, but also con- to the unique properties of nanofluids discussed above, the
siders the resistance against the continuous transport of the enhanced transport rate in CNTs is also attributable to the
fluid phase inside the nanopore. The equation takes the form atomic smoothness of CNT surfaces (leading to low friction
of P = Pin + ΔP = 4(Δγ + ηL)/D, where Pin = 4Δγ/D represents (Cottin-Bizonne et al. 2005; Liu et al. 2012b; Sokhan et al.
the classic capillary effect, Δγ the solid–liquid interfacial 2002)) and their hydrophobicity (inducing a depletion region
energy, ΔP = 4ηL/D quantifies the transport resistance, and L (Cottin-Bizonne et al. 2005; Hummer et al. 2001; Majumder
is length of the fluid file that has entered the nanopore, and η et al. 2005) at the liquid–solid interface).
is the resistance per unit lateral surface area. With these unique properties and intriguing behav-
In addition to the intriguing structural and transport prop- iors, nanofluidics has shown many prospective applications
erties, nanofluids also feature intrinsic size effects. Various (Garcia-Martinez 2010; Ghosh et al. 2003; Heyden et al.
fluid characteristics (e.g., structure, density, ionic hydration, 2005; Liu et al. 2007, 2008, 2009a, 2012a; Meng and Allen
hydrogen bonding, and dipole orientation) can be drastically 2008; Pennathur et al. 2007). In particular, a wide range of
varied with the size of nanoconfinement. For instance, when separation and purification applications have been conceived
the nanopore dimension is at the molecular level (<1 nm), con- based on the intrinsic sorption and size exclusion capabilities
fined water molecules tend to form a quasi-one-dimensional of nanoporous materials. Examples include dehydration of
chain and transport as a single file (Alexiadis and Kassinos ethanol to improve purity (Wang et al. 2010), capturing CO2
2008; Hummer et al. 2001); as the pressure is sufficiently from gas mixtures to contain carbon emission (An et al. 2009;
high, some more complex structures may emerge such as the Carruthers et al. 2012; Nachtigall et al. 2012), and water desal-
n-gonal ring (Koga et al. 2001) and the double helical chain ination to provide alternative water sources (Kim et al. 2010;
(Qiao et al. 2009). Inside larger nanopores, however, a multi- Liu and Chen 2013a,b; Yuan et al. 2008). In the following
layered structure is often formed (Thomas and McGaughey sections, we will review several fundamental aspects of nano-
2008a; Wang et al. 2004); as a result of the less restrictive fluidics in graphene-based materials systems, followed by its
confinement, water molecules are usually more hydrogen applications in purification and permeation technologies.
bonded (Gordillo and Marti 2000) and have more diverse
dipole orientations (Wang et al. 2004). For electrolytes, ionic 24.3 NANOFLUID STRUCTURES AND
hydration is often severely suppressed as the liquid enters a TRANSPORT IN GRAPHENE-BASED
nanopore, which reduces the coordination number consid-
MATERIAL SYSTEMS
erably (Lynden-Bell and Rasaiah 1996). Similarly, other
nanofluid characteristics such as the contact angle (Liu et al. In Section 24.2, nanofluids are shown to feature several dis-
2009c; Werder et al. 2003), infiltration pressure (Liu et al. tinctive structural and transport properties that set them apart
2009a; Qiao et al. 2009), and viscosity (Chen et al. 2008) from classic fluids. However, these conclusions are mainly
have certain size effects. In particular, given the inhomoge- drawn from the investigation of nanofluidics in CNTs and
neous molecular and ionic distributions in small confinement, various solid-state nanoporous materials. For nanofluids
the variation of electrical potential is expected to be much confined in graphene-based materials, even more intriguing
more complicated than that predicted by various mean-field behaviors are expected because of the unique physical and
theories (e.g., the Gouy–Chapman model). When the confine- chemical properties of graphene (e.g., CNT-resembling con-
ment is extremely small (close to molecular size), the electri- struction and flat surface).
cal double layer may be so tightly squeezed in some cases that To illustrate the structure and transport of nanofluids in
the diffuse layer vanishes, leaving the compact layer alone graphene-based confinement, a conceptual nanoribbon–gra-
(Chmiola et al. 2006; Liu et al. 2012a). phene hybrid fluid channel (NRC) has been recently proposed
Due to these unique intrinsic properties (structure, trans- and explored (Liu et al. 2012b). As shown in Figure 24.3a, the
port, and size effects), nanofluids have shown many intrigu- fluid phase in NRCs is sandwiched in between two layers of
ing behaviors. As electrolyte solutions are pressurized into graphene sheets; a nanoribbon is then attached to each of the
a molecular-sized silica nanotube (Figure 24.2c), water two graphene monolayers to shape a pathway for conducting
molecules, and ions pushed inside the nanotube tend to main- nanofluidic flow. Such a design creates a layered nanofluidic
tain a highly regular quasi-one-dimensional chain in which system with composite construction, namely, “nanoribbon–
each anion pairs with a cation, forming anion–cation couples graphene–fluid–graphene–nanoribbon.” Here, graphene
separated by water molecules. This unconventional phenome- sheets are introduced to contain vapors, and the nanoribbons
non suggests that salts confined in molecular-sized nanopores are incorporated to serve as a “fluid guide.” For a systematic
can hardly be fully dissolved as cations and anions stick with understanding of nanofluidics confined in NRCs, four repre-
each other to form little “crystals” (Liu et al. 2009a, 2010). On sentative systems are considered as shown in Figure 24.3b.
the other hand, according to Hummer et al. (2001), water mol- They have the same length (along the z-direction) of 213 Å
ecules can spontaneously invade empty CNTs within several and same width (along the x-direction) of 36 Å. All graphene
tens of picoseconds, much faster than the prediction of con- layers are modeled wide enough to eliminate edge effects,
tinuum hydrodynamics. Similar phenomena have also been and the distance between neighboring nanoribbon and gra-
experimentally discovered by several other groups (Holt et al. phene layers (along the y-direction) is set to be a constant of
(e)
y FSS
MID θ
x
(a) (c) Dipole angle
(f ) 0.1
NRC15, MID/FSS
NRC20, FSS
NRC20, MID
Probability
0.075 NRC30, FSS
NRC30, MID
0.05
0.025
500 1000 1500 2000 2500 3000 (kg/m3)
0 90 180 270 360

(g) Dipole angle (degree)
(b) (d) 4
Hydrogen bond per molecule

NRC15
3000 NRC20
Water density (kg/m3) NRC25
NRC30
2000
3
y 1000 NRC15
NRC20
x NRC25
NRC30
0 2
0 0.2 0.4 0.6 1 0 0.2 0.4 0.6 1
y/(h/2) y/(h/2)
FIGURE 24.3 (a) Illustration of a NRC. (b) Cross-sectional view of four models of NRC with the separation between opposite graphene
sheets, h, equal to 15, 20, 25, and 30 Å, respectively. They are denoted by NRCh (h = 15, 20, 25, and 30). Long dark gray lines indicate the
vertical positions of graphene sheets, and short light gray lines depict the positions of nanoribbons. (c) Contours of point-wise water density
within the NRCs. The charts are scaled in the y-direction by h so that the top and bottom edges (dark gray lines) all correspond to the vertical
positions of the graphene sheets. The light gray segment over the dark gray lines indicate the width of the nanoribbons. (d) Point-wise water
density along the thickness direction (y-axis). Only y > 0 is shown because of the symmetry of the system about the x-axis. The vertical
coordinate, y, is normalized by h/2, such that the right edge of the plot [y/(h/2) = 1] corresponds to the position of the top graphene sheet.
(e) In-plane orientation of the water dipole moment, θ, is defined as the angle between the y-axis and the in-plane component of the dipole
vector as illustrated by the arrow. (f) Probability distribution of the in-plane dipole moment orientation at two representative vertical posi-
tions, that is, FSS and MID as defined in (e). FSS refers to the first solvation shell, and MID (middle layer) is about the center of the NRC.
(g) Number of hydrogen bond per water molecule versus vertical position. (Liu, L. et al., 2012b. Graphene nanoribbon-guided fluid channel:
A fast transporter of nanofluids. Nanoscale 4 (20):6279–6283. Reproduced by permission of The Royal Society of Chemistry.)
3.35 Å. The only variable among the four NRCs is the separa- the most energetically favorable position for the confined fluid
tion between the two graphene sheets, denoted by h. “NRCh” phase must be centered about the perpendicular bisector of
is, therefore, employed to denote the different systems, with the nanoribbons (y-axis). This is verified by Figure 24.3b,
h = 15, 20, 25, and 30 Å, respectively. The periodical bound- in which four snapshots are illustrated to depict equilibrated
ary condition is applied along the z-direction only to ensure water phase confined in NRCs. In all cases, water is found to
continuity of the confined flow. With various h, 1190, 2940, reside in the center of the NRCs and occupies a straight flow
4900, and 6300 water molecules are confined in the model pathway shaped by the nanoribbons. The same phenomenon
NRC15, NRC20, NRC25, and NRC30, respectively. is also observed in Figure 24.3c, in which contours of water
density are plotted. Apparently, all contours are symmetric
about the y-axis, implying that the confined water is equili-
24.3.1 Structural Properties
brated in the center of the NRCs. Both graphs serve as direct
Different from CNTs and many other nanochannels in which demonstrations of the fact that, even without any geometric
nanofluids are confined by “visible” geometric boundaries constraint along the width direction (x-direction), NRCs can
composed of solid atoms, NRCs guide nanofluid flow via an still attract fluids and center them in the nanochannel.
“invisible” force field exerted by the nanoribbons. From an Figure 24.3c also suggests a layered structure for water
atomic point of view, as a fluid molecule (or ion) is confined confined in NRCs. In all cases, water is found to be signifi-
in an NRC, its balance is controlled by three factors: forces cantly densified in the first layer of water molecules from
exerted by other fluid molecules (and ions), interaction with the solid surface (also called the first solvation shell, FSS).
the graphene layers, and interaction with the nanoribbons. According to Figure 24.3d that plots the variation of water
Considering the edge effects associated with the nanoribbons, density along the y-axis, the peak density in NRC15 is found
to be about 2000 kg/m3, two times that of the bulk water; and This is partially due to the densification of water molecules
in the other cases, even higher peak densities are identified in FSS, and also partly due to the preferential alignment of
(approximately 3200 kg/m3). This uneven density distribu- dipole moment with the solid surface. After all, hydrogen
tion is similar to that in other forms of nanoconfinement. bonding is defined by both the separation and angle between
In addition to the uneven distribution of density, other two water molecules (Luzar and Chandler 1996). Therefore,
descriptors of NRC-confined water are also spatially varied both local density and dipole moment distribution can influ-
such as the orientation of dipole moment (Figure 24.3e) and ence its formation.
hydrogen bonding. Figure 24.3f plots the statistical distribu-
tion of the dipole moment orientation, θ, in three selected
24.3.2 Transport Properties
NRCs at two representative positions, FSS and MID. FSS
refers to the FSS while MID (middle layer) refers to the center Transport characteristics of nanofluids can be quanti-
of the NRC. In NRC15, FSS and MID are equivalent, as there fied by three quantities, that is, the deceleration of a fluid
exists only one layer of concentrated water due to the extreme file that freely moves inside a nanochannel, a, the shear-
confinement. According to Figure 24.3f, in the FSS of NRC15, ing stress exerted by the solid surface, τ, and the viscosity
dipole moment orientation is unevenly distributed. Water mol- of the nanoconfined fluid phase, η. These three quanti-
ecules prefer θ = 90° and 270° to other angles, implying that ties are closely correlated. Measurable by a free decelera-
dipole moments parallel to the solid surface are more ener- tion test (inset of Figure 24.4a), deceleration (a) quantifies
getically favored. By comparing NRC20 and NRC30, a simi- flow impedance per molecule. From a mechanics point of
lar distribution is noted in the FSS but the distributions in view, the deceleration is caused by the lattice resistance to
MID are found to be distinct. In NRC30, where the confine- fluid flow, quantified by a shearing stress, τ. τ is correlated
ment is larger, MID is far away from the solid surface, so the with a via τ = ρL m 0 a/2w where ρL denotes the line num-
interfacial effect is much diminished; as such, dipole moment ber density that equals to the number of water molecules
in this region has equal probability to face different direc- in a unit length of the nanochannel, m 0 the mass per water
tions, similar to that in the bulk water. In NRC20, however, molecule, and w is the width of the nanoribbons. The vis-
the interfacial effect still remains to some extent. On the other cosity, η, can be determined by η = h3wΔP/12QL, where Q
hand, Figure 24.3g plots the number of hydrogen bond per denotes the flow rate, and ΔP is the pressure drop over the
molecule, nHB, along the thickness direction. In all cases, nHB length of L. In addition, the equilibrium of the liquid file
is found to drop rapidly in FSS from a constant of about 3.6. gives whΔP/L = 2τw and Q = vwh where v denotes the flow
(a) (b) (c)

2.4 8×10–6
Deceleration (10–3 nm/ps2)
NRC15 NRC30 NRC30

NRC20 NRC25 NRC25
Shearing stress (MPa)
NRC25 1 NRC20
6×10–6
NRC20
1.6
Velocity (Pa m)
NRC30 NRC15 NRC15
0.8 4×10–6
300 0.5
Velocity
200
(m/s)
100 2×10–6
0
0
200 350 500
Time (ps)
0 0
0 50 100 150 200 0 50 100 150 200 0 50 100 150 200
Velocity (m/s) Velocity (m/s) Velocity (m/s)
(d) (e) (f )
0.75 2×10–6
Deceleration (10–3 nm/ps2)
CNT08 CNT41 CNT41

1.6
Shearing stress (MPa)
CNT14 CNT27 CNT27

CNT27 CNT14 CNT14
1.5×10–6
Velocity (Pa m)
CNT41 CNT08 CNT08

1.2 0.5
1×10–6
0.8
0.25 5×10–7
0.4
0
0 0
0 50 100 150 200 0 50 100 150 200 0 50 100 150 200
Velocity (m/s) Velocity (m/s) Velocity (m/s)
FIGURE 24.4 Velocity-dependent (a) deceleration, (b) shearing stress, and (c) nominal viscosity of water confined in the four models
NRCs. Inset of (a) plots velocity decreasing with time during a freely decelerated transport process for water confined in NRC30. (d–f)
Corresponding nanofluid descriptors in four representative CNTs. The CNTs are armchair with the diameters equal to 8.12 Å (CNT08),
13.56 Å (CNT14), 27.12 Å (CNT27), and 40.68 Å (CNT41), respectively. (Liu, L. et al., 2012b. Graphene nanoribbon-guided fluid channel:
A fast transporter of nanofluids. Nanoscale 4 (20):6279–6283. Reproduced by permission of The Royal Society of Chemistry.)
velocity. Finally, a simple relationship between τ and η is several factors including the curvature of CNTs, the additional
obtained in the form of η = τh/6v. nanoribbon layers in NRCs that are also in close interaction
Figure 24.4a through c plots the deceleration, shearing with the nanofluids, and the fact that no geometric constraint
stress, viscosity for water confined in the four model NRCs. exists in NRCs along the width direction.
All these quantities are found to strongly depend on the size
of the NRC and the velocity of the nanofluid flow. In small
24.4 NANOFLUID DYNAMICAL BEHAVIORS IN
NRCs such as NRC15, all water molecules reside in the FSS
and are in direct contact with the solid surface, so the flow GRAPHENE-BASED MATERIAL SYSTEMS
impedance caused by the lattice resistance is peaked. With In the last section, physics of nanofluids in an idealized
the size of NRC increased, more water molecules are sub- nanofluidic system was introduced to illustrate the intrinsic
jected to low or negligible interactions with the solid surface, molecular structures and transport behaviors of water under
thus decreasing the flow impedance (Figure 24.4a). Another graphene-based nanoconfinement. In realistic nanofluidic
important factor is flow velocity. As the confined nanofluid systems, however, many practical factors may take effects,
transports at a high velocity, the water molecules residing in which will significantly complicate the situation. These prac-
the boundary layers are given less time to respond to the rela- tical factors include the deformability of graphene layers,
tive motion of solid lattice, among other dynamically chang- coexistence of the liquid and vapor phases and the associated
ing local environmental variables. Hence, at high v, it is less phase transition, and ambient conditions such as temperature
likely for the interfacial water molecules to adjust to the most and humidity. In this section, several case studies will be
energetically favorable positions, thus increasing the flow introduced to illustrate dynamical behaviors of nanofluids in
impedance (Figure 24.4a). practical systems.
Different from the deceleration that quantifies the per mol-
ecule flow impedance, shearing stress quantifies the overall
24.4.1 Nanofluid Morphology
interaction between solid surfaces and confined nanofluids.
From Figure 24.3d, it is known that there are two shells of An important area for exploration is to understand the mor-
concentrated water in most of the NRCs except the smallest phology of nanofluids confined in graphene-based material
one under investigation, NRC15. The two shells, FSS and SSS systems. In the theoretical investigation shown in Section
(second solvation shell) are subjected to the strongest inter- 24.3, graphene layers are fixed so the confined fluid layer
actions with the solid phase, which determines the shearing has the same thickness across its entirety. Situation changes
stress. Due to the similar interactions of FSS and SSS with when the graphene layer is considered flexible. Figure 24.5a
solid surfaces in most NRCs that are reasonably large, the illustrates the morphology of a thin layer of water confined
shearing stress in these NRCs should be almost the same. As between a graphene monolayer and a mica substrate (He et al.
shown in Figure 24.4b, τ of NRC25 and NRC30 are found 2012). Under dynamical conditions, the water film is shown to
almost identical. As the confinement size decreases, τ of feature uneven thickness. Using a CNT height reference, the
NRC20 is slightly reduced but is still very close to that of the number of water layers and its spatial distribution are identi-
larger NRCs. τ of NRC15, however, is a lot smaller because fied in Figure 24.5b. It is readily seen that only in a small
only FSS exists and interacts with the solid surface. Same as area, the water film has three layers of water molecules; in
the velocity effect on a, shearing stress also increases with most other areas, there are only one or two layers or even no
velocity for similar reasons. water molecules. Interestingly, the few layers of water trapped
As an intrinsic property of fluids, viscosity measures the in such a system have different bonding structures. The first
resistance of fluids to the deformation caused by shearing layer is found to be highly ordered and strongly bound while
stress or tensile stress. Due to the layered structure in NRCs, the other two layers, if any, are amorphous. Using an SEM
the viscosity of nanofluids is subjected to prominent size tip, one can even manipulate the trapped water film. However,
effects. Water confined in NRC15 is shown to have the low- such operation only works for the amorphous layers; it can-
est viscosity (Figure 24.4c), implying that the deformation of not perturb or alter the bonding structure in the ordered layer.
nanofluids can be assisted by larger free surfaces in smaller Moreover, protected by the graphene coating, water keeps
NRCs. As the size of NRC increases, η keeps increasing, stable even during high-temperature processing in a vacuum.
but for all the four cases under investigation, the viscosity is This, again, validates the conceptual design introduced in
about three to four orders of magnitude lower than that of the Section 24.3, which employs graphene to contain liquids and
bulk water. Figure 24.4d through f plots a, τ, and η for water vapors. Indeed, with the dense and strong atomic structure,
in four selected CNTs with diameters equal to 8.12, 13.56, graphene is an ideal container at the nanoscale for various
27.12, and 40.68 Å, respectively. They are denoted by CNT8, nanofluidic applications.
CNT14, CNT27, and CNT41, respectively, in consistence with
the notation of the NRCs. By comparing Figure 24.4d through
24.4.2 Nanofluid Evaporation
f with Figure 24.4a through c, the computed viscosities all
show prominent size and velocity effects; moreover, the abso- Another area of immense interest is the evaporation of flu-
lute values of viscosity are found very comparable in both ids under nanoconfinement, driven by hydrophobic inter-
types of nanochannels. The differences can be attributed to action, humidity gradient, and temperature. Hydrophobic
(a) 5 nm (b) 3rd water

2.0 nm Nanofluid evaporation can also be controlled by humid-
layer
1.8 ity. By varying ambient humidity, a recent experiment
1.6
2nd water 1.4 demonstrates the feasibility of controlling wetting and
layer 1.2 dewetting of a planar slit formed by mica and graphene sur-
1.0 faces (Severin et al. 2012). Mica is employed because, at
1st water 0.8
layer 0.6 room humidity (RH), its surface is covered by a water film
0.4 due to its hydrophilicity (Hu et al. 1995). Then, by g ently
0.2 pressing a monolayer graphene onto the mica surface, a
20 nm 0 nm
Time clapsed (minutes): (d)
(c) 5 200 sandwiched structure of graphene–water–mica is formed
280
70
160 390 for the investigation of the evaporation process. Initially at
RH, water film persists in the entire region, so the graphene
layer keeps flat. As the ambient humidity is decreased,
water evaporates and small depressions can be identified in
the height of the graphene layer. Interestingly, such depres-
sions take fractal shapes and can nucleate to form larger
fractals over time (Figure 24.5d), which resembles the
growth of crystals. Figure 24.5c illustrates the growth of
dried fractal structures over time. Colored regions corre-
spond to the depressions where no water exists between the
FIGURE 24.5 (a) Morphology of a water thin film confined in graphene and the mica surface; while, in the white regions,
between a graphene monolayer and a mica surface. (b) The thickness
water film remains. It is readily seen that the “stem” of the
of the water film is position-dependent, varying between zero and
three layers of water molecules. (Reprinted with permission from fractal structure initiates from the edge of the nanofluidic
He, K. T. et al., Scanning tunneling microscopy study and nano- system and gradually proceeds toward the interior. Despite
manipulation of graphene-coated water on mica. Nano Letters 12 the intriguing phenomenon, a rational understanding of the
(6), 2665–2672. Copyright 2012 American Chemical Society.) (c drying process is still lacking, which calls for theoretical
and d) Drying a water film confined in between a graphene mono- investigations into the effects of the solid lattice, drying
layer and a mica surface features the growth of a fractal-patterned rate, and defects.
dried region. (Reprinted with permission from Severin, N. et al.,
Reversible dewetting of a molecularly thin fluid water film in a soft
graphene–mica slit pore. Nano Letters 12 (2), 774–779. Copyright 24.5 NANOFLUID PURIFICATION IN
2012 American Chemical Society.) GRAPHENE-BASED MATERIAL SYSTEMS
With appropriate design of nanocavity size and functional-
ization, many nanoporous materials and low-dimensional
evaporation plays a key role in multiple biological processes materials have been designed for purification applications
including the folding of globular proteins, the opening and including dehydration of ethanol (Wang et al. 2010), captur-
closing of ligand-gated ion channels, and ligand binding to ing CO2 from a gas mixture (An et al. 2009; Carruthers et al.
hydrophobic pockets (Sharma and Debenedetti 2012). A 2012; Nachtigall et al. 2012), desalination of sea water (Kim
recent study has investigated the drying process of a nano- et al. 2010; Yuan et al. 2008), and gas trapping (Hugon et al.
cavity formed by two hydrophobic graphene layers. Initially, 2000). Compared with the materials that have been exten-
the two graphene layers are immersed in a water bath with sively investigated (e.g., zeolites and CNTs), graphene-based
a prescribed separation. As the simulation begins, water materials not only possess desirable properties of CNTs, but
molecules inside the nanocavity are driven to evaporate they also feature several unique properties that make them
by hydrophobic interaction, which decreases the number extremely promising for phase separation applications. These
of water molecules inside the nanocavity. Apparently, this properties include their ultrathin monolayer configuration,
process is highly dynamical and depends heavily on fac- readily achievable modification (e.g., chemical functionaliza-
tors including gap size, surface size, and temperature. Using tion and structural alteration), and wide possibilities for build-
forward flux sampling, the evaporation rate is found to be ing thin films of various properties. To fulfill their promises,
drastically reduced by an order of magnitude with only a however, it is critical to rationally understand the permeation
slight increase in the gap size from 9 to 14 Å. Evaporation and screening physics of nanofluid flow through graphene-
rate per unit area is found to be increased by two orders of based materials. In this section, several investigations along
magnitude if the surface size is enlarged from 1 to 9 nm 2. this direction will be reviewed.
Moreover, as the gap between the two hydrophobic graphene
layers increases, the free energy barrier to evaporation
scales linearly, indicating that the free energy barrier is pre-
24.5.1 Purification by Nanoporous Graphene
dominantly controlled by line tension. Similar evaporation As illustrated in Figure 24.6a, nanoporous graphene is mono-
processes have been demonstrated in several other studies layer graphene embedding nanopores. An intriguing property
(Koishi et al. 2004). of nanoporous graphene is that it can pass ions in a highly
controllable manner. As such, the single-layer membrane has indicate strong functionalization effects, making rational
been designed for several electrical (Garaj et al. 2010) and design critical to achieving nanoporous graphene with desired
purification (Cohen-Tanugi and Grossman 2012) applications. functionality and properties.
Presumably, due to the small size of graphene nanopores (as Compared with other materials for similar applications,
small as several nanometers), ion passage, and the associ- nanoporous graphene has a unique characteristic, namely,
ated applications must depend heavily on the local environ- its ultrathin construction with only one atomic thickness.
ment that passes ions. Both chemical functionalization and This makes nanoporous graphene extremely promising
size tuning of the nanopores may significantly affect the for purification applications in which energy input is posi-
force field around the pore area, therefore varying ion pas- tively correlated with the separation membrane thickness.
sage/blockage rates. To fully understand ion passage through Cohen-Tanugi and Grossman carried out a study in which
nanoporous graphene, several theoretical studies have been nanoporous graphene is used for mechanically driven water
performed using molecular simulation techniques. Sint et al. desalination (Cohen-Tanugi and Grossman 2012). The sys-
compared two graphene nanopores terminated by fluorine– tem is illustrated in Figure 24.6d, where an aqueous solution
nitrogen (F–N) and hydrogen (H), respectively (Sint et al. of sodium chloride is driven by a pressure gradient to pass
2008). By electrically driving electrolyte solutions to pass through the nanoporous graphene membrane. With the con-
the two nanopores, they found that cations such as Li+, Na+, trolled size and chemical functionalization of the pore open-
and K+ could only pass the F–N-terminated nanopore shown ing, very high ion rejection rate close to 100% was achieved.
in Figure 24.6b, while anions such as Cl− and Br− ions can More strikingly, water permeability was found in the order
only pass the H-terminated nanopore shown in Figure 24.6c. of 100 L/cm2/day/MPa, about two orders of magnitude higher
Moreover, passage resistance time is found prominently than today’s most e fficient desalination techniques. This sets
dependent on the magnitude of driving forces. By increasing nanoporous graphene to the right top corner in Figure 24.6e,
the external electric field from 0.005 to 0.10 V/nm, ion pas- making it highly desirable for energy efficient and effective
sage becomes one order of magnitude faster. All these results water desalination applications. Further optimization of such
(a) (b) (c) (d)

Nanoporous graphene
4.5 nm
ΔP
(e) H2O
Na+ Cl–
100% Hydroxylated
graphene
Seawater RO Hydrogenated (f )
Brackish RO graphene
90% Nanofiltration
Salt rejection (%)
High-flux RO
80%
MFI Zeolite
70%
60%
50%
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000
Water permeability (L/cm2/day/MPa)
FIGURE 24.6 (a) Passage of ions and water molecules through functionalized graphene nanopores. (b) F–N-terminated nanopore and
(c) H-terminated nanopore. (d) Water desalination through a sheet of nanoporous graphene. (e) Salt rejection versus water permeability for the
nanoporous graphene and other separation materials. (f) Unimpeded permeation of water through membranes based on GO. (Reprinted with
permission from Sint, K., B. Wang, and P. Král. Selective ion passage through functionalized graphene nanopores. Journal of the American
Chemical Society 130 (49), 16448–16449. Copyright 2008 American Chemical Society; Cohen-Tanugi, D. and J. C. Grossman. Water desali-
nation across nanoporous graphene. Nano Letters 12 (7), 3602–3608. Copyright 2012 American Chemical Society; Boukhvalov, D. W., M. I.
Katsnelson, and Y.-W. Son. Origin of anomalous water permeation through graphene oxide membrane. Nano Letters 13:10.1021/nl4020292.
a system may be achieved by improving the local environment Boukhvalov, D. W., M. I. Katsnelson, and Y.-W. Son. 2013. Origin
around pore opening. As shown in Figure 24.6e, hydroxyl- of anomalous water permeation through graphene oxide mem-
ated g raphene is found superior to hydrogenated graphene brane. Nano Letters 13:10.1021/nl4020292.
Carruthers, J. D., M. A. Petruska, E. A. Sturm, and S. M. Wilson.
for larger water permeability. More improvements may be
2012. Molecular sieve carbons for CO2 capture. Microporous
achieved by a refined design based on a synergy of experi- and Mesoporous Materials 154:62–67.
mental and c omputational efforts. Chen, X., G. X. Cao, A. J. Han, V. K. Punyamurtula, L. Liu,
P. J. Culligan, T. Kim, and Y. Qiao. 2008. Nanoscale fluid
transport: Size and rate effects. Nano Letters 8 (9):2988–2992.
24.5.2 Purification by Graphene-Based Membranes Chmiola, J., G. Yushin, Y. Gogotsi, C. Portet, P. Simon, and P. L.
The membrane is another form of graphene-based materials. Taberna. 2006. Anomalous increase in carbon capaci-
tance at pore sizes less than 1 nanometer. Science 313
Reported by Nair et al. (2012), submicrometer-thick mem-
(5794):1760–1763.
branes made from graphene oxide (GO) can reject liquids, Cohen-Tanugi, D. and J. C. Grossman. 2012. Water desalination
vapors, and gases (including helium), but these membranes across nanoporous graphene. Nano Letters 12 (7):3602–3608.
are completely permeable to water. Figure 24.6f and g illus- Cottin-Bizonne, C., B. Cross, A. Steinberger, and E. Charlaix. 2005.
trates the macroscopic and microscopic views of the GO Boundary slip on smooth hydrophobic surfaces: Intrinsic
membranes, respectively. The leak rate of helium through GO effects and possible artifacts. Physical Review Letters
membranes is found to be consistently low for pressure up to 94:056102.
100 mbar while helium can easily pass a film of polyethylene de Groot, B. L. and H. Grubmuller. 2001. Water permeation across
biological membranes: Mechanism and dynamics of aquapo-
terephthalate (PET) that is 24 times thicker (at 1000 times rin-1 and GlpF. Science 294 (5550):2353–2357.
higher rates) (Figure 24.6). By contrast, water can permeate Dutta, S. and S. K. Pati. 2010. Novel properties of graphene
much faster through the film at a rate 10 orders of magnitude nanoribbons: A review. Journal of Materials Chemistry
higher than that of helium. The anomalous water permeation 20 (38):8207–8223.
phenomenon has been recently attributed to the formation of El-Kaderi, H. M., J. R. Hunt, J. L. Mendoza-Cortes, A. P. Cote, R.
hexagonal ice bilayer in between GO flakes and the melting E. Taylor, M. O’Keeffe, and O. M. Yaghi. 2007. Designed
transition of ice at the edges of flakes (Boukhvalov et al. 2013). synthesis of 3D covalent organic frameworks. Science 316
(5822):268–272.
Only when the interlayer distance d shown in Figure 24.6h Garaj, S., W. Hubbard, A. Reina, J. Kong, D. Branton, and J. A.
is within an optimal range can ice bilayer be formed within Golovchenko. 2010. Graphene as a subnanometre trans-elec-
the GO membranes, leading to unimpeded water filling and trode membrane. Nature 467 (7312):190–193.
permeation as reported in the experimental work. Garcia-Martinez, J. 2010. Nanotechnology for the Energy Challenge.
Weinheim: WILEY-VCH Verlag GmbH & Co. KGaA.
Ghosh, S., A. K. Sood, and N. Kumar. 2003. Carbon nanotube flow
24.6 CONCLUSION sensors. Science 299:1042–1044.
Gordillo, M. C. and J. Marti. 2000. Hydrogen bond structure of liq-
Graphene-based materials including graphene monolayers,
uid water confined in nanotubes. Chemical Physics Letters
nanoribbons, nanoporous graphene, and their derivatives 329 (5–6):341–345.
and hybrids in different forms (e.g., membranes) have been Hanasaki, I. and A. Nakatani. 2006. Flow structure of water in
demonstrated to be a promising material platform for build- carbon nanotubes: Poiseuille type or plug-like? Journal of
ing nanofluidic systems of various functionalities. For better Chemical Physics 124:144708.
understanding the underpinning nanofluidics, this review He, K. T., J. D. Wood, G. P. Doidge, E. Pop, and J. W. Lyding. 2012.
introduces several fundamental aspects of nanofluids (i.e., Scanning tunneling microscopy study and nanomanipula-
tion of graphene-coated water on mica. Nano Letters 12
molecular structures, transport behaviors, dynamical mor-
(6):2665–2672.
phologies, evaporation physics, and purification processes) Heyden, F. H. J. van der, D. Stein, and C. Dekker. 2005. Streaming
confined in graphene-based material systems. Further devel- currents in a single nanofluidic channel. Physical Review
opment of graphene-based nanofluidic systems will continue Letters 95:116104.
to drive and benefit from improved understanding of these fun- Hilder, T. A., D. Gordon, and S.-H. Chung. 2009. Salt rejection and
damental aspects. Scientific insights obtained from relevant water transport through boron nitride nanotubes. Small 5
studies are envisioned to enrich the theory of nanofluidics, (19):2183–2190.
Holt, J. K., H. G. Park, Y. M. Wang, M. Stadermann, A. B.
close gaps between mechanics and other disciplines, and
Artyukhin, C. P. Grigoropoulos, A. Noy, and O. Bakajin.
provide design principles and methodologies. 2006. Fast mass transport through sub-2-nanometer carbon
nanotubes. Science 312:1034–1037.
Hu, J., X. D. Xiao, D. F. Ogletree, and M. Salmeron. 1995.
REFERENCES Imaging the condensation and evaporation of molecularly
Alexiadis, A. and S. Kassinos. 2008. Molecular simulation of water thin films of water with nanometer resolution. Science 268
in carbon nanotubes. Chemical Reviews 108 (12):5014–5034. (5208):267–269.
An, J., S. J. Geib, and N. L. Rosi. 2009. High and selective CO2 Hub, J. S. and B. L. De Groot. 2008. Mechanism of selectivity
uptake in a cobalt adeninate metal−organic framework exhib- in aquaporins and aquaglyceroporins. Proceedings of the
iting pyrimidine- and amino-decorated pores. Journal of the National Academy of Sciences of the United States of America
American Chemical Society 132 (1):38–39. 105 (4):1198–1203.
Hugon, O., M. Sauvan, P. Benech, C. Pijolat, and F. Lefebvre. 2000. Liu, L., J. B. Zhao, C. Yin, P. J. Culligan, and X. Chen. 2009c.
Gas separation with a zeolite filter, application to the selectiv- Mechanisms of water infiltration into conical hydropho-
ity enhancement of chemical sensors. Sensors and Actuators bic nanopores. Physical Chemistry Chemical Physics
B: Chemical 67 (3):235–243. 11:6520–6524.
Hummer, G., J. C. Rasaiah, and J. P. Noworyta. 2001. Water conduc- Lu, G. Q. and X. S. Zhao. 2004. Nanoporous Materials: Science and
tion through the hydrophobic channel of a carbon nanotube. Engineering. London: Imperial College Press.
Nature 414 (6860):188–190. Luzar, A. and D. Chandler. 1996. Hydrogen-bond kinetics in liquid
Iijima, S. 1991. Helical microtubules of graphitic carbon. Nature water. Nature 379 (6560):55–57.
354 (6348):56–58. Lynden-Bell, R. M. and J. C. Rasaiah. 1996. Mobility and solva-
Jia, X., J. Campos-Delgado, M. Terrones, V. Meunier, and M. S. tion of ions in channels. Journal of Chemical Physics 105
Dresselhaus. 2011. Graphene edges: A review of their fabrica- (20):9266–9280.
tion and characterization. Nanoscale 3 (1):86–95. Majumder, M., N. Chopra, R. Andrews, and B. J. Hinds. 2005.
Jin, C., K. Suenaga, and S. Iijima. 2008. Plumbing carbon nanotubes. Enhanced flow in carbon nanotubes. Nature 438 (7064):44.
Nat Nano 3 (1):17–21. Malani, A., K. G. Ayappa, and S. Murad. 2006. Effect of con-
Joseph, S. and N. R. Aluru. 2008. Why are carbon nanotubes fast finement on the hydration and solubility of NaCl in water.
transporters of water? Nano Letters 8 (2):452–458. Chemical Physics Letters 431 (1–3):88–93.
Kalra, A., S. Garde, and G. Hummer. 2003. Osmotic water trans- Meng, D. D. and J. S. Allen. 2008. Micro and nanofluidics for energy
port through carbon nanotube membranes. Proceedings of the conversion. IEEE Nanotechnology Magazine 2:19–23.
National Academy of Sciences 100:10175–10180. Murata, K., K. Mitsuoka, T. Hirai, T. Walz, P. Agre, J. B. Heymann,
Kim, S. J., S. H. Ko, K. H. Kang, and J. Han. 2010. Direct seawa- A. Engel, and Y. Fujiyoshi. 2000. Structural determinants
ter desalination by ion concentration polarization. Nature of water permeation through aquaporin-1. Nature 407
Nanotechnology 5 (4):297–301. (6804):599–605.
Koga, K., G. T. Gao, H. Tanaka, and X. C. Zeng. 2001. Formation Nachtigall, P., M. R. Delgado, D. Nachtigallova, and C. O. Arean.
of ordered ice nanotubes inside carbon nanotubes. Nature 412 2012. The nature of cationic adsorption sites in alkaline
(6849):802–805. zeolites-single, dual and multiple cation sites. Physical

Koishi, T., S. Yoo, K. Yasuoka, X. C. Zeng, T. Narumi, R. Susukita, Chemistry Chemical Physics 14 (5):1552–1569.
A. Kawai et al. 2004. Nanoscale hydrophobic interaction and Naguib, N., H. Ye, Y. Gogotsi, A. G. Yazicioglu, C. M. Megaridis,
nanobubble nucleation. Physical Review Letters 93 (18):185701. and M. Yoshimura. 2004. Observation of water confined in
Li, H., M. Eddaoudi, M. O’Keeffe, and O. M. Yaghi. 1999. Design nanometer channels of closed carbon nanotubes. Nano Letters
and synthesis of an exceptionally stable and highly porous 4 (11): 2237–2243.
metal-organic framework. Nature 402 (6759):276–279. Nair, R. R., H. A. Wu, P. N. Jayaram, I. V. Grigorieva, and
Li, X., X. Wang, L. Zhang, S. Lee, and H. Dai. 2008. Chemically A. K. Geim. 2012. Unimpeded permeation of water through
derived, ultrasmooth graphene nanoribbon semiconductors. helium-leak–tight graphene-based membranes. Science 335
Science 319 (5867):1229–1232. (6067):442–444.
Liu, J., L. Dai, and J. W. Baur. 2007. Multiwalled carbon nano- Pennathur, S., J. C. T. Eijkel, and A. van den Berg. 2007. Energy
tubes for flow-induced voltage generation. Journal of Applied conversion in microsystems: Is there a role for micro/nanoflu-
Physics 101:064312. idics? Lab on a Chip 7:1234–1237.
Liu, L. and X. Chen. 2013a. Fast ion transport and phase separa- Polarz, S. and B. Smarsly. 2002. Nanoporous materials. Journal of
tion in a mechanically driven flow of electrolytes through Nanoscience and Nanotechnology 2 (6):581–612.
tortuous sub-nanometer nanochannels. ChemPhysChem 14 Poynor, A., L. Hong, I. K. Robinson, S. Granick, Z. Zhang, and
(11):2413–2418. P. A. Fenter. 2006. How water meets a hydrophobic surface.
Liu, L. and X. Chen. 2013b. Inside cover: Fast ion transport and phase Physical Review Letters 97 (26):266101.
separation in a mechanically driven flow of electrolytes through Prakash, S. 2007. Surface Mediated Transport and Reactions. Ph.D.
tortuous sub-nanometer nanochannels (ChemPhysChem diss., University of Illinois Urbana-Champaign, pp. 48–52.
11/2013). ChemPhysChem 14 (11):2354–2354. Qiao, Y., L. Liu, and X. Chen. 2009. Pressurized liquid in nano-
Liu, L., X. Chen, T. Kim, A. Han, and Y. Qiao. 2010. Effects of anion pores: A modified Laplace-Young equation. Nano Letters 9
size and concentration on electrolyte invasion into molecular- (3):984–988.
sized nanopores. New Journal of Physics 12 (3):033021. Severin, N., P. Lange, I. M. Sokolov, and J. P. Rabe. 2012. Reversible
Liu, L., X. Chen, W. Y. Lu, A. J. Han, and Y. Qiao. 2009a. Infiltration dewetting of a molecularly thin fluid water film in a soft
of electrolytes in molecular-sized nanopores. Physical Review graphene–mica slit pore. Nano Letters 12 (2):774–779.
Letters 102 (18):184501. Sharma, S. and P. G. Debenedetti. 2012. Evaporation rate of water
Liu, L., X. Chen, W. Lu, and Y. Qiao. 2009b. Infiltration of electro- in hydrophobic confinement. Proceedings of the National
lytes in molecular-sized nanopores. Physical Review Letters Academy of Sciences 109 (12):4365–4370.
102:184501 Sint, K., B. Wang, and P. Král. 2008. Selective ion passage through
Liu, L., H. Lim, W. Lu, Y. Qiao, and X. Chen. 2012a. Mechanical- functionalized graphene nanopores. Journal of the American
to-electric energy conversion by mechanically driven flow Chemical Society 130 (49):16448–16449.
of electrolytes confined in nanochannels. Applied Physics Skoulidas, A. I., D. M. Ackerman, J. K. Johnson, and D. S. Sholl.
Express 6 (1):015202. 2002. Rapid transport of gases in carbon nanotubes. Physical
Liu, L., Y. Qiao, and X. Chen. 2008. Pressure-driven water infil- Review Letters 89 (18):185901.
tration into carbon nanotube: The effect of applied charges. Sokhan, V. P., D. Nicholson, and N. Quirke. 2002. Fluid flow in
Applied Physics Letters 92 (10):101927. nanopores: Accurate boundary conditions for carbon nano-
Liu, L., L. Zhang, Z. Sun, and G. Xi. 2012b. Graphene nanorib- tubes Journal of Chemical Physics 117:8531
bon-guided fluid channel: A fast transporter of nanofluids. Son, Y.-W., M. L. Cohen, and S. G. Louie. 2006. Half-metallic
Nanoscale 4 (20):6279–6283. graphene nanoribbons. Nature 444 (7117):347–349.
Sui, H., B.-G. Han, J. K. Lee, P. Walian, and B. K. Jap. 2001. Wang, Y. H., C. M. Gong, J. S. Sun, H. Gao, S. A. Zheng, and S. M.
Structural basis of water-specific transport through the AQP1 Xu. 2010. Separation of ethanol/water azeotrope using com-
water channel. Nature 414 (6866):872–878. pound starch-based adsorbents. Bioresource Technology 101
Surani, F. B., X. Kong, D. B. Panchal, and Y. Qiao. 2005. Energy (15):6170–6176.
absorption of a nanoporous system subjected to dynamic Werder, T., J. H. Walther, R. L. Jaffe, T. Halicioglu, and
loadings. Applied Physics Letters 87 (16):163111. P. Koumoutsakos. 2003. On the water-carbon interaction
Tang, Y. W., K. Y. Chan, and I. Szalai. 2002. One-dimensional for use in molecular dynamics simulations of graphite and
capacitance behavior of electrolytes in a nanopore. Nano carbon nanotubes. Journal of Physical Chemistry B 107
Letters 3 (2):217–221. (6):1345–1352.
Thomas, J. A. and A. J. H. McGaughey. 2008a. Density, distribution, Won, C. Y. and N. R. Aluru. 2007. Water permeation through a sub-
and orientation of water molecules inside and outside carbon nanometer boron nitride nanotube. Journal of the American
nanotubes. Journal of Chemical Physics 128 (8):084715. Chemical Society 129 (10):2748–2749.
Thomas, J. A. and A. J. H. McGaughey. 2008b. Reassessing fast Xu, B. X., Y. Qiao, T. Park, M. Tak, Q. L. Zhou, and X. Chen. 2011.
water transport through carbon nanotubes. Nano Letters A conceptual thermal actuation system driven by interface
8:2788–2793. tension of nanofluids. Energy and Environmental Science 4
Wang, J., Y. Zhu, J. Zhou, and X. H. Lu. 2004. Diameter and helic- (9):3632–3639.
ity effects on static properties of water molecules confined in Yuan, J., X. Liu, O. Akbulut, J. Hu, S. L. Suib, J. Kong, and
carbon nanotubes. Physical Chemistry Chemical Physics 6 F. Stellacci. 2008. Superwetting nanowire membranes for
(4):829–835. selective absorption. Nature Nano 3 (6):332–336.
25 Nanoporous Graphene Sheets
for Gas Separation
Andreas W. Hauser and Peter Schwerdtfeger
CONTENTS
Abstract...................................................................................................................................................................................... 477
25.1 Introduction...................................................................................................................................................................... 477
25.1.1 Background........................................................................................................................................................... 478
25.1.2 Chapter Overview................................................................................................................................................. 478
25.2 Models for Gas Separation............................................................................................................................................... 479
25.2.1 Model Pores for Computation............................................................................................................................... 479
25.3 Classical Picture and Its Limitations: Methane Separation.............................................................................................. 479
25.3.1 Separation of Methane from Air.......................................................................................................................... 480
25.3.2 Kinetic Diameters................................................................................................................................................. 480
25.4 Quantum-Mechanical Effects in Gas Separation: Helium Isotopes................................................................................. 482
25.4.1 Tunneling Probabilities of 3He and 4He................................................................................................................ 482
25.4.2 Graphene Pores for 3He/4He Separation............................................................................................................... 484
25.4.3 Temperature-Driven Isotope Separation............................................................................................................... 486
25.5 Summary.......................................................................................................................................................................... 486
Acknowledgment....................................................................................................................................................................... 488
References.................................................................................................................................................................................. 488
ABSTRACT could be achieved by either posttreatments such as electron

beam cutting or direct synthesis of porous carbon materials
Chemical inertness and an essentially two-dimensional geom- via self-assembling approaches. Obviously, the main working
etry make porous graphene sheets a promising material for principle for such a type of membrane is its selectivity toward
gas separation applications. Selectivity toward a certain gas molecular size, with the pore diameter as the essential param-
species is mainly determined by molecular size, which makes eter for a fine-tuning of the separating behavior.
the pore diameter the essential parameter for fine-tuning of The transmission of molecules through graphene pores
the separating behavior. However, the transition of molecules in the sub-nanometer regime is an intrinsically quantum-
through graphene pores in the sub-nanometer regime is an mechanical process and requires the use of quantum chemi-
intrinsically quantum-mechanical process and needs to be cal methods for an accurate simulation. The transmission
looked at using computational chemistry. Effects of surface probability for a certain gas also depends on surface adsorp-
adsorption, geometry deformation, and even quantum tunnel- tion effects and geometry deformations of the pore during
ing play an important role in certain systems. This is illus- penetration. For sufficiently small gases such as hydrogen or
trated by the application of density functional theory to finite helium, even the phenomenon of quantum tunneling has to be
pore models and a selection of industrially relevant gases. taken into consideration.
In this chapter, we discuss all the effects mentioned above
by comparing the transmission probabilities for a selection of
25.1 INTRODUCTION
prominent gases (CH4, N2, O2, CO2, H2 3He, and 4He) through
Gas mixtures can be effectively separated by synthetic mem- a set of differently functionalized nanopores of various size.
branes, which will have an important future role to play in It is intended to give a general overview of the underlying
cost-effective environmental and energy-related processes mechanisms, which need to be considered when designing
such as the purification or capture and storage of hydrogen, pore types for a certain purpose. Our analysis is based on the
methane, or carbon dioxide [1]. The impermeability of pris- application of density functional theory (DFT) to size-reduced
tine graphene sheets for gas molecules of any size, paired with model pores. The transmission barriers obtained from elec-
their chemical inertness and the advantage of an essentially tronic structure theory are then used to estimate permeation
extended two-dimensional structure, suggest the potential rates for the gases. We show that the concept of kinetic diam-
usage of these new materials as membranes for gas separation, eters and effective pore sizes become less applicable in situa-
if pores of appropriate size are introduced artificially. This tions where attractive forces beyond van der Waals interaction
477
occur during the transmission. Finally, we also discuss quan-

tum tunneling and zero-point energy effects given the exam- TABLE 25.1
ple of helium transmission through sufficiently small pores, Suggested Applications of Nanoporous Graphene for
suggesting the potential usage of graphene sheets for the sepa- Gas Separation
ration of bosonic from fermionic helium.
Separation Process References
H from CH4 [30–33]
25.1.1 Background H from N2 [34]
H2S from CH4 [35]
Membrane-based gas separation is gaining larger acceptance He from noble gases and CH4 [36]
in the industry with possible applications in the field of natu- Noble gases among themselves [37]
ral gas purification, gas sensors, batteries, and fuel cells. New CH4 from air [38]
separation materials could also provide cheap solutions for CO2 from N2 [39,40]
the reduction of greenhouse gases such as carbon dioxide and CO2 from N2 and CH4 [31,41–43]
methane [1,2]. Currently, a broad spectrum of different mate- CO2 from N2, CH4, and O2 [44]
rials is under investigation, spanning from organic polymer- 3He from 4He [45–47]
based membranes to inorganic types made from ceramics, H from D [48]
metal, glass, or carbon, including also combined approaches
such as inorganic polymeric membranes [3]. Among these
options, graphene is outstanding in its advantage of high-tem- with a nominal area of 25 mm2 and a pore size between 1
perature and high-pressure robustness and chemical inertness and 10 nm were tested for size-selective molecular transport
[4,5]. Its discovery was equally stimulating for researchers in [70]. In general, the preparation of well-defined microporous
physics, chemistry, and the applied sciences [6–13], and trig- membranes thinner than several nanometers is problematic
gered the growth of a new branch in materials science. In the with current methods and materials due to the creation of
context of gas separation a free-standing sheet of graphene, undesired extra defects. Recently, Li et al. [71] demonstrated
discovered in 2004 [14], may be seen as the ultimate mem- the efficient separation of hydrogen from nitrogen and meth-
brane due to its single-atom thickness: The permeability of ane through structural defects in a graphene oxide membrane
a membrane is inversely proportional to its thickness, mak- of 18 Å thickness. Experiments on multilayered graphene
ing graphene the most effective gas separation device possible indicate additional tuning possibilities via different stacking
[15]. However, a perfect sheet of graphene is impermeable methods [72,73].
even to helium atoms, which are repelled by the electron den- Most of the studies on gas separation rely on the funda-
sity of the aromatic rings. A barrier height of 11.67 eV has to mental principle of molecular sieving, exploiting the size dif-
be overcome for He escaping through a non-defect graphene ferences between different gas molecules. Notable exceptions
sheet [16]. As a consequence, the porous character has to be are Reference 34, where different accumulation tendencies of
introduced artificially [17–21], either by invasive post-syn- nitrogen and hydrogen on the graphene surface are utilized,
thesis methods such as hole drilling with electron beams or and References 45–47, which take advantage of the mass
by the design of suitable precursors for the self-assembling dependence of the tunneling probability.
of these desired structures [22–24]. Other post-treatments are
UV-induced oxidative etching [25–27] or ion bombardment
25.1.2 Chapter Overview
[28,29]. None of these techniques have reached the level of
industrial implementation yet, but its possibility has led to We distinguish between pure size effects, which we refer to
several studies on pore design for certain applications: Gas as the “classical picture,” effects due to interactions between
separation (hydrogen from methane [30–33], hydrogen from pore and particles, and intrinsically quantum-mechanical
nitrogen [34], hydrogen sulfide from methane [35], helium effects, which lack any classical explanation. In the context
from other noble gases and methane [36], noble gases among of gas separation these effects may be used to classify a cer-
themselves [37], methane from air [38], carbon dioxide from tain separation technique as size sieving, chemical sieving, or
nitrogen [39,40], methane [31,41–43], and oxygen [44], fermi- quantum sieving, respectively, depending on which aspect is
onic from bosonic helium [45–47], hydrogen from deuterium mainly exploited for a certain application [74]. Section 25.2
[48]), the selective passage of ions [49–51], the characteriza- starts with a short discussion of different theoretical frame-
tion of DNA [52–55], the filtration [56], desalination [57–60] works for the simulation of gas separation through nanopo-
and remediation [61] of water, the storage of hydrogen, meth- rous materials and presents a selection of model pores that
ane, and carbon dioxide [62–64], and the design of superca- we use in the simulations of particle transmissions. In Section
pacitors [65–69] (see Table 25.1). However, the experimental 25.3, we test the applicability of the classical picture to the
realization of single-layer graphene sheets for gas separa- problem of separating methane from nitrogen, carbon diox-
tion is yet in its infancy. Experiments on micrometer-sized ide, oxygen, and hydrogen gas, discussing the limitations and
membranes proved at least their potential usage as molecular conceptual problems of this approach. We show that the inclu-
sieves by pressurized blister tests and mechanical resonance sion of interactions between particle and pore is essential for
measurements [27]. Larger graphene composite membranes a correct description of the separation mechanism. In Section
Nanoporous Graphene Sheets for Gas Separation 479
25.4, we leave the classical regime and analyze quantum- 3

mechanical aspects given the example of 3He/4He separation. 1 6
25.2 MODELS FOR GAS SEPARATION

The numerous approaches for an efficient modeling of gas 4
diffusion through nanoporous graphene can be classified into
two groups. One of them is the set of techniques based on 7
2
a real-time development of the actual diffusion process with
either classical or first-principles molecular dynamics [74].
These approaches are suitable for large systems and can be
5
applied to equilibrium molecular dynamics such as tempera-
ture-driven gas separation as well as systems with a pressure
difference across the membrane. A limiting factor is the time-
scale accessibility provided by current molecular dynamics
calculations [75,76]. The second set of approaches avoids the
direct simulation of gas diffusion processes by the applica- FIGURE 25.1 Model pores for gas separation. Single-ring holes
tion of statistical rate theory. A detailed five-step rate model are shown in the first row, the double-ring holes in the second. A
has been developed by Drahushuk and Strano [77]. In this partial replacement of C–H bonds with nitrogen atoms allows for
chapter, we focus on the event of a single particle transmis- a fine-tuning of the effective pore size. The third row contains a
sion through a pore, which can be interpreted as a reaction triangular shaped three-ring hole and an even larger four-ring hole.
described by a minimum energy reaction pathway on a cer- In the case where the hydrogen is absent the CH group has been
substituted by a nitrogen atom. (Reprinted with permission from
tain potential energy surface calculated with suitable methods
A. W. Hauser, J. Schrier, and P. Schwerdtfeger. Helium tunneling
of computational chemistry. The energy barrier occurring at through nitrogen-functionalized graphene pores: Pressure- and tem-
the microscopic level is linked to a macroscopic transmis- perature-driven approaches to isotope separation. J. Phys. Chem. C,
sion rate via transition state theory. Reaction pathways and 116(19):10819–10827, Copyright 2012 American Chemical Society.)
energy peaks corresponding to first-order saddle points can
be identified with methods such as the Nudged Elastic Band of the total energy with contributions from damped atom-pair-
[78], the Linear Synchronous Transit [79], the Growing, or the wise interaction potentials. It has been specifically designed
Freezing String Method [80], in combination with effective for noncovalently bound systems such as benzene complexes
eigenvalue-following routines [81]. By calculating partition [82], making it a reasonable choice for the given purpose. All
functions at these transition states, one can derive rate con- pore geometries are obtained from full optimizations using the
stants for particle transmissions. Gaussian09 quantum chemistry package [85]. The heights of
In this chapter we focus on the usage of the latter group the reaction barriers for the propagation of the selected gas
for the evaluation of gas separation properties of nanoporous molecules through these pores are obtained from calculations
graphene, and discuss its implementation on various levels of of the first-order saddle points for systems consisting of a finite
theory, starting with a classical interpretation of the barrier for pore plus gas molecule. Zero energy is set to having the fully
the separation of methane from air, but moving on to quantum relaxed pore and the gas molecule at infinite distance.
phenomena such as tunneling and zero-point energy effects.
25.3 CLASSICAL PICTURE AND ITS
25.2.1 Model Pores for Computation LIMITATIONS: METHANE SEPARATION
Our computational simulations are based on finite-size molec- For the first part of our study, we pick model pores with a
ular models of single graphene pores. The systems consist of diameter that lie reasonably close to the kinetic diameter [86]
the 12/14/15/16 remaining rings in the direct neighborhood of of methane (dk = 3.8 Å). According to Reference 34, pore 3
one-, two-, three-, and four-ring holes, respectively. An over- in Figure 25.1 is large enough for hydrogen (dk = 2.89 Å) but
view of the pore models is given in Figure 25.1. The pore rims too small for nitrogen (dk = 3.64 Å) to pass through, making it
are passivated by either hydrogen or by replacing the unsatu- the perfect membrane for H2/N2 separation at room tempera-
rated carbon atom with nitrogen. Due to the size of the systems, ture. In the classical picture, the only quantity that counts is
post-Hartree–Fock methods are computationally too expen- the ratio between the kinetic diameter of the gas molecules
sive. Instead, we apply DFT with an empirical correction for and the pore size. As a consequence, maximum selectivity
dispersion effects. We choose the B97D functional of Grimme is given when the pore size is only slightly larger than the
[82] together with a correlation-consistent triple-ζ basis set kinetic diameter of the largest waste gas molecule, in this
[83,84] for all atoms, which gives a total number of about 2100 case, hydrogen, while efficiently holding back the nitrogen
contracted Cartesian Gaussian basis functions (3500 primitive molecules. Interestingly, an inversion of the selectivity is
gaussians). The functional is based on the power-series ansatz observed for sufficiently large pores: nitrogen transits become
of Becke and accounts for long-range dispersion by corrections more likely than the passage of hydrogen atoms. The authors
explain this unexpected phenomenon by the tendency of gra- the barrier drops almost by a factor of five. In Reference 38,
phene to physisorb nitrogen [34]. The N2 molecules are kept we show that these molecules induce significant pore defor-
close to the pore, which effectively boosts the transmission mations during transmission. For these cases, the simplifica-
probability for this species. This illustrates a first limitation tion of a fixed ratio between molecular diameter and pore size
of the classical model. Even for graphene, seen as the highly has to be abandoned.
inert molecular sieve, the interactions between gas molecules Repeating the same analysis with pore 4 allows us to study
and the membrane play a crucial role for a correct prediction the influence of a pore-rim functionalization with nitrogen
of gas permeation properties. Still, the simple model yields at atoms. Transition state energies are listed in Table 25.2 and
least an explanation for the separation principle. transmission probabilities and ratios are shown in Figure 25.3.
The slightly larger pore shows a new feature of the smaller gas
25.3.1 Separation of Methane from Air molecules: the first-order saddle point at the transition state
has turned into a minimum for CO2, O2, and H2. If we allow
We start with the calculation of the relative barrier heights for the pore to relax during transmission, even the N2 pathway
CH4, N2, CO2, O2, and H2 through pore 3, keeping the pore becomes barrierless, and we end up with the very promising
fully frozen at the gas-free equilibrium geometry. The relative situation of a membrane that is permeable to all listed gases
barrier heights of Table 25.2 are obtained by subtraction of the but methane. Note, however, that the extent of the deforma-
absolute energies of the pore and the separated gas molecule tion of a real pore in the macroscopic membrane is expected
from the transition state energy. All molecules have to over- to be damped by the periodic environment. An interesting
come a barrier for propagating through the pore. As expected, detail is that the partial replacement of C–H groups by nitro-
methane shows the highest barrier, almost twice as high as gen has its largest effect on the interaction with CO2 during
the barrier for nitrogen. However, although this pore provides transmission. It even leads to a slight deformation of the CO2
a good selectivity toward methane, it is not suitable for indus- molecule, which is bent within the model pore, showing an
trial application due to the still very high barrier for nitrogen O–C–O angle of 174.5°, with the carbon atom pointing toward
transmission. If one assumes a Boltzmann distribution for the the N-doped pore rim. A similar system, consisting of CO2
kinetic energies of the gas molecules, the barrier height Eb can and a pyridine molecule, has been investigated by Vogiatzis
be related to the classical transmission probability via et al. [87]. Following their argumentation, the improved bind-
ing energy can be explained by a charge separation effect
1  Eb  in the bent molecule, turning the electron deficient carbon
tclass =  1 − erf   (25.1) atom into a Lewis acid acting on the N-doped rim. Hence,
2  kBT 
the replacement of C–H-groups by N atoms does not only
enlarge the pore but also introduces a Lewis-base function-
with k B as Boltzmann constant, T as temperature, and erf ality. This indicates new opportunities for the adjustment of
denoting the error function. Our results for the transmission transmission ratios based on the tuning of surface adsorption
probabilities based on the ab initio transition state energies are effects: In Reference 41, it was shown that the higher polariz-
shown in Figure 25.2. The negligible nitrogen flux (dark gray ability and quadruple moment of carbon dioxide compared to
line in the upper-left picture) undermines any industrial sepa- other gas molecules makes its adsorption very sensitive to the
ration effort, despite the high N2/CH4 separation ratio (lower choice of graphene functionalization.
left picture). The graph and the table also contain data for gas
transmissions where a full relaxation of the pores was allowed
25.3.2 Kinetic Diameters
during the particle propagation. This second approach is more
realistic and has a significant influence on the results, mainly In order to test the predictive power of the classical picture we
affecting the energies of carbon dioxide and oxygen, where compare the literature values for two common qualities, the
TABLE 25.2
Relative Barrier Heights (kcal/mol) for the Transmission of CH4, N2, CO2,
O2, and H2 through Unrelaxed and Fully Relaxed Pores 3 and 4
Pore CH4 N2 CO2 O2 H2
3 21.553 10.984 6.397 4.461 0.580
3 (rel) 8.286 5.568 1.183 0.849 0.284
4 4.804 1.118 −4.632 −0.693 −0.174
4 (rel) 0.529 −0.875 −5.072 −1.026 −1.668
Source: A. W. Hauser and P. Schwerdtfeger. Methane-selective nanoporous graphene membranes for

gas purification. Phys. Chem. Chem. Phys., 14:13292–13298, 2012. Reproduced with permis-
sion of the PCCP Owner Societies.
(a) Pore 3 unrelaxed Pore 3 relaxed

100 100
10−5 10−2
Transmission
Transmission
10−10 CH4 10−4 CH4
N2 N2
CO2 CO2
10−15 10−6 O2
O2
H2 H2
10−20 10−8
300 400 500 600 700 800 900 1000 300 400 500 600 700 800 900 1000
T (K) T (K)
(b) 1020 108

N2/CH4 N2/CH4
CO2/CH4 CO2/CH4
1015 O2/CH4
106 O2/CH4
H2/CH4 H2/CH4
Ratio
Ratio
1010 104
105 102
100 100
300 400 500 600 700 800 900 1000 300 400 500 600 700 800 900 1000
T (K) T (K)
FIGURE 25.2 (a) Thermally weighted transmission of CH4, N2, O2, CO2, and H2 through pore 3. (b) Corresponding transmission ratios
X/CH4, X = N2, O2, CO2, and H2. Note the different logarithmic scaling of the y-axis. (Adapted from A. W. Hauser and P. Schwerdtfeger.
Methane-selective nanoporous graphene membranes for gas purification. Phys. Chem. Chem. Phys., 14:13292–13298, 2012. Reproduced
with permission from the PCCP Owner Societies.)
(a) Pore 4 unrelaxed Pore 4 relaxed

100 100
Transmission
Transmission
CH4 10−1 CH4

N2 N2
CO2 CO2
O2 O2
H2 H2
10−5 10−2
300 400 500 600 700 800 900 1000 300 400 500 600 700 800 900 1000
T (K) T (K)
(b) 105 6
N2/CH4 N2/CH4
CO2/CH4 CO2/CH4
104 O2/CH4
5 O2/CH4
H2/CH4 H2/CH4
Ratio
103 4
Ratio
102 3
101 2
300 400 500 600 700 800 900 1000 300 400 500 600 700 800 900 1000
T (K) T (K)
FIGURE 25.3 (a) Thermally weighted transmission of CH4, N2, O2, CO2, and H2 through pore 4. (b) Corresponding transmission ratios X/
CH4, X = N2, O2, CO2, and H2. Note the different scaling of the y-axis, which is logarithmic in all graphs but in the lower right. (Adapted
from A. W. Hauser and P. Schwerdtfeger. Methane-selective nanoporous graphene membranes for gas purification. Phys. Chem. Chem.
Phys., 14:13292–13298, 2012. Reproduced with permission from the PCCP Owner Societies.)
kinetic diameter dk and the Lennard–Jones collision diameter diameter of 2.96 Å for pore 3 and 3.15 Å for pore 4, which is
dLJ, with the actual ranking of the transmission probabilities much closer to the effective diameters of the gas molecules.
obtained for the DFT calculations in the previous section. Interpreting it in a strict sense, the classical model predicts a
Both scales are estimations of the effective size of a penetrat- transmission probability of 1 for H2 and 0 for all other gases.
ing molecule and are derived from gas viscosity measure- Even if we assume slight uncertainties of the pore and particle
ments or second virial coefficient data. For light gases, dk diameters, one would still expect a much higher transmis-
is expected to give a better scaling of diffusion coefficients sion probability for H2 than for the remaining gases. Indeed,
with increasing size of the penetrating molecule [86,88], and for pore 3 such a drastic change in transmission probability
may be interpreted as the “waistline” diameter for linear occurs (see Figure 25.2, relaxed), but between the group H2, O2,
molecules. Literature values vary by a few hundredths of an and CO2, and the group N2 and CH4. Again, the predictive power
ångstrøm due to their experimental nature. For internal con- of the classical picture is marginal. This shows that the com-
sistency, we derive our own set of diameter values from the parison of kinetic diameters provides only a crude estimation of
electron densities calculated at the same level of theory as the separation tendencies in the case of nanoporous graphene. The
pore interactions in the previous section. This third set allows propagation of a single gas molecule through a pore has to be
a direct comparison of the pore and a particle size within the seen as a microscopic event involving physical (van der Waals
same theoretical framework. Our one-dimensional scaling forces, Pauli repulsion) or even chemical interactions (e.g., an
parameter, dDFT, corresponds to the cylindrical radius of the acid–base reaction with a functionalized pore rim) that can lead
electron density at a small contour value of 0.002 electrons to significant changes in the pore geometries as well as notice-
per cubic Bohr for linear molecules. In the case of methane, able deformations of the gas molecules themselves.
we pick the averaged spherical radius. The contour value has
been chosen to reproduce numerical results in close proximity 25.4 QUANTUM-MECHANICAL EFFECTS IN
to the experimental diameter values. In Table 25.3, we give a
GAS SEPARATION: HELIUM ISOTOPES
comparison of our scale to a popular list of kinetic diameters
[86] and Lennard–Jones collision values [89]. So far, we have extended the original concept of hard spheres
On first sight, none of the experimentally derived param- and rigid sieves in several ways: we take the actual shape of the
eters reproduces the ordering we obtain from the calculations molecules and pores into account, include particle–membrane
with respect to transmission probability while the DFT-based interactions, and allow for geometry deformations during propa-
diameters do. There is some agreement between the three dif- gation. All these effects are covered by the application of elec-
ferent types, but all of them predict a different ranking with tronic structure theory. However, although quantum mechanical
respect to molecular size. Further insights are gained by iden- in their origin, they might also be modeled by molecular mechan-
tifying a suitable quantity for the description of pore sizes: In ics calculations with carefully chosen parameters describing the
a first attempt, we use the same contour value for the electron intermolecular and intramolecular forces.
density and take the smaller diameter of the elliptically shaped In this section, we focus on weaker effects that cannot
pores 3 and 4, which yields the values 2.06 and 2.43 Å, respec- be mapped onto a classical picture, as they are direct conse-
tively. The fact that these values are at least 0.4 Å smaller than quences of the quantum nature of the propagating particles:
any effective radius of the gas molecules shows that an argu- phenomena of quantum tunneling and zero-point energy.
mentation based on electron density is highly problematic. In These effects are mass dependent and play a role only for
a second approach, we define the average radius of the pore lightweight particles and low temperatures. We discuss them
as the mean distance between the equilibrium position of the here with the example of 3He/4He separation via nitrogen-
penetrating gas molecule and the van der Waals surface of the functionalized graphene pores [45,47].
pore-rim atoms. This way we avoid a mixing of a theoretical
quantity obtained from electronic structure theory (electron
25.4.1 Tunneling Probabilities of 3He and 4He
density) with experimentally derived gas parameters. The
van der Waals radii for hydrogen and nitrogen are 1.10 and In Section 25.3, we assumed a classical dependence of the
1.55 Å, respectively. With these numbers, we obtain a pore transmission probability t on the particle energy E (replacing
TABLE 25.3
Comparison of Different Molecular Diameters in Å
Type H2 O2 CO2 N2 CH4
dk [86] 2.89 3.46 3.3 3.64 3.8
dLJ [89] 2.827 3.467 3.941 3.798 3.758
dDFT 2.80 3.15 3.26 3.31 4.02
Source: A. W. Hauser and P. Schwerdtfeger. Methane-selective nanoporous graphene membranes for gas purification. Phys.
Chem. Chem. Phys., 14:13292–13298, 2012. Reproduced with the permission of the PCCP Owner Societies.
t(E) by a Heaviside step function) for typical gases of air at Probabilities for 3He and 4He are plotted as solid and dashed
higher temperatures, which led to temperature dependence as lines, respectively. Note that they are multiplied by a factor of
described in Equation 25.1. However, t(E) is also a function 10−5 to allow a combined plot with the kinetic energy distribu-
of the mass and depends further on the shape of the actual tions p(E, T) for T = 20, 22, 24, and 26 K. At the given axis
energy barrier that has to be overcome. Hence, it needs to be scaling, the gray and black curves are hardly distinguishable,
evaluated separately for each propagating particle and gra- but the product functions of t(E) and p(E, T) show a noticeable
phene pore. We will derive the necessary information from difference: Figure 25.4c contains the products for T = 20 K.
one-dimensional reaction energy pathways of 3He and 4He At this temperature, the areas under the solid and dashed
obtained from quantum chemistry calculations. Assuming a curves are almost the same, which indicates that the transmis-
Boltzmann distribution for the kinetic energies of the helium sion ratio of 3He to 4He is already close to one. The following
atoms the energy of a single particle in one dimension can be insights can be gained from this theoretical example: (1) A
written as potential barrier surrounded by two adsorption minima leads
to a higher transmission probability than a barrier of the same
N height without minima. The central peak shows steeper shoul-
p( E, T ) = exp − E / kBT (25.2)
E ders and is effectively narrowed in the first case, which boosts
the tunneling probability. This shows that the inclusion of
attractive interactions via either post-Hartree–Fock methods
with T, k B, and N as temperature, Boltzmann constant, and
a normalization factor, respectively. This dependence, for-
mally a χ2-distribution with one degree of freedom, can be (a)
200
derived from the well-known Gaussian distribution of veloci-
Energy (cm–1)
ties vz. A Boltzmann distribution does not reflect the intrinsic
0
quantum statistical properties of fermionic or bosonic helium
at low temperatures, but it avoids the difficulty of describing
a mixed, nonideal quantum gas and provides a conservative –200
–6 –4 –2 0 2 4 6
estimate of the 3He/4He ratio as the Fermi–Dirac distribu- Distance (A)
tion is broader than the Bose–Einstein distribution. If t(E) is (b) × 10–5
1
known, the temperature-weighted transmission probability pw
p(E,T) and t(E)
is obtained by integration over the product of t(E) and p(E,T)

0.5
T = 26

pw (T ) =
∫ p(E, T )t(E)dE (25.3) 0
100
T = 20
150 200 250 300
Energy (cm–1)
(c) × 10–9
Two example reaction pathways for particles going through 4
two different pores are given in Figure 25.4a.
p(E,T) t(E)
They can be interpreted as the total energy of our model 2

system (finite pore plus propagating helium) plotted as a
function of the distance between the particle and a model 0
pore located at z = 0. The gray curve describes a single bar- 100 150 200 250 300
rier while the black curve shows a barrier surrounded by two Energy (cm–1)
minima. An arbitrary barrier height of 200 cm−1 has been
chosen for both barriers, while the minima for the black FIGURE 25.4 (a) Two possible potential functions (gray and black)
curve, accounting for adsorption effects, are set to −200 cm−1. for a barrier height of 200 cm−1. In one case, the central peak is sur-
Given these potential energy curves V(z) and the mass m of rounded by two minima, which effectively narrow the tunneling bar-
the helium isotopes, we apply a finite-difference recursion rier by causing steeper shoulders. (b) Corresponding transmission
algorithm to solve the time-independent one-dimensional probabilities t(E) (gray and black, almost identical) for 3He (solid
Schrödinger equation: lines) and 4He (dashed lines) are plotted together with the kinetic
energy distributions p(E, T) (see Equation 25.2) for T = 20, 22, 24,
and 26 K. Although the transmission functions of both potentials
d 2 ψ 2m are hardly distinguishable, the product functions p(E, T)t(E) differs.
+ 2 ( E − V ( x ))ψ = 0 (25.4)
dx 2 They are printed in (c) for a temperature of T = 20 K. Note that the
areas under the solid and dashed curves in (c) are almost equal to the
chosen temperature, indicating a 3He/4He transmission ratio that is
for helium atoms moving in the z-direction with the kinetic
already close to one. (Reprinted with permission from A. W. Hauser,
energy E, we obtain two pairs of transmission probabilities, J. Schrier, and P. Schwerdtfeger. Helium tunneling through nitrogen-
which are plotted in Figure 25.4b. The gray pair is derived functionalized graphene pores: Pressure- and temperature-driven
from the single barrier potential and the black pair stems from approaches to isotope separation. J. Phys. Chem. C, 116(19):10819–
the model potential including also attractive interactions. 10827. Copyright 2012 American Chemical Society.)
or empirical dispersion correction methods is important for 1000

1-ring holes
a correct estimation of tunneling probabilities; (2) the trans- 2-ring holes
800
mission probabilities in Figure 25.4b show certain symmetry 3-ring hole
in the sense that for energies lower than the barrier height 4-ring hole
600
Energy (cm−1)
the fermionic isotope is preferred while at energies above the
400
barrier a transmission of the bosonic isotope becomes more
likely. Barriers of equal height lead to transmission curves, 200
which intersect each other at the same value. As a conse-
quence, keeping in mind that both isotopes feel exactly the 0
same potential during propagation, the selective transmission −200
of 3He is maximized if the kinetic energy of all particles is
−6 −4 −2 0 2 4 6
kept below the barrier height. However, in a typical experi- Distance (A)
mental setup, these energies will be distributed according
to Equation 25.2, and the temperature is the only parameter
beside the form of the potential curve (determined by the pore FIGURE 25.5 Potential curves for a single helium atom passing
through graphene nanopores of different size: single-ring holes 1
type) that allows for a fine-tuning of the separation mecha-
and 2 (dashed), two-ring holes 3–5 (solid black, gray, and light
nism. Choosing low temperatures improves the selectivity gray), the triangular three-ring hole 6 (dotted), and a large four-ring
toward 3He, but also reduces the transmission probability and, pore seven (dash-dotted). The energy is plotted as a function of the
therefore, the total flux rate. Higher temperatures lead to bet- height above the pore perpendicular to the graphene plane. For the
ter flux rates but at the cost of broadened energy distributions single- and double-ring holes, a fine-tuning of the potential barrier
reaching far on the disadvantageous side of the transmission by nitrogen passivation of the pore rim is demonstrated. All poten-
curve, which reduces the 3He selectivity. tials are corrected for basis set superposition error. (Reprinted with
permission from A. W. Hauser, J. Schrier, and P. Schwerdtfeger.
Helium tunneling through nitrogen-functionalized graphene pores:
25.4.2 Graphene Pores for 3He/4He Separation Pressure- and temperature-driven approaches to isotope separation.
J. Phys. Chem. C, 116(19):10819–10827. Copyright 2012 American
In our search for suitable pore types, we go back on the model Chemical Society.)
pores given in Figure 25.1, and calculate the potential energy
for the transmission of a single He atom on a straight path
perpendicular to the membrane. The curves we obtain for the basis set superposition error, which has its maximum value
different model pores are shown in Figure 25.5. Removing (about 60 cm−1) at the transition state geometries. Hence,
a single ring from a perfect graphene sheet leads to pore 1 we applied the counterpoise correction method of Boys and
with a barrier height of 4110 cm−1, which can be reduced to Bernardi [90].
833 cm−1 via N-functionalization of the pore rim (pore 2). A Before we look at the macroscopically relevant quantities
slight enlargement of the pore by taking out a second ring of the promising two ring holes, namely the particle flux and
yields an even lower transmission barrier of 625 cm−1 (pore 3). the 3He/4He selectivity, we shall emphasize another concep-
Combining the strategy of ring removal with nitrogen pas- tual difference when comparing the pore adjustment pro-
sivation allows a further adjustment of the two-ring hole a cedure for methane to the optimization for helium isotope
single- or double-sided functionalization (pores 4 and 5) separation. In Section 25.3, it was shown that the saturation
with nitrogen leads to a barrier of 201 and 61 cm−1, respec- of the pore rim with nitrogen atoms led to significant effects
tively. Larger pores obtained by removal of three or more on the barrier energies, which suggested, at least in the case
rings do no longer show a barrier at all, making them use- of CO2 transmission, to interpret the nitrogen passivation as a
less for our purpose. All but the effectively barrierless larg- Lewis-base functionalization of the pore rim. This raises the
est model pores show the desired feature of van der Waals question whether a similar, though much weaker effect, is to
minima. These minima, which can be interpreted as adsorp- be expected for an inert gas or not. We define an average pore
tion intermediates, have a depth of more than 100 cm−1 and radius 〈r〉 as the mean distance between the He atom position
occur at a helium distance between 1.9 and 2.5 Å, following and the van der Waals surface created by the atoms of the
a trend toward shorter distance with increasing pore size. As pore rim (1.10 and 1.55 Å for hydrogen and nitrogen [91,92],
discussed before, their presence has a narrowing effect on respectively). In Figure 25.6, we plot the barrier height,
the central peak, which increases the transmission probabil- divided by the number of rim atoms, as a function of this aver-
ity. All barriers are identified as first-order saddle points. We aged pore radius for all pore models. An exponential fit of the
note that the pore geometries are optimized without helium nitrogen-free pores (circular markers) fits the data smoothly,
atoms present and kept fixed during the particle propagation. indicating the dominance of Pauli repulsion effects during
In contrast to the problems discussed earlier in Section 25.3, particle transmission. On the chosen energy scale, the attrac-
this approach is justifiable in the case of helium transmission: tive van der Waals interactions are hardly noticeable; negative
For the most relevant two-ring pores, the error introduced by barrier heights for the largest pores are the only indication.
freezing the pore geometry lies below 10 cm−1. However, the Interestingly, the same exponential curve is also in agreement
significantly smaller barrier heights require a correction of the with the energies obtained for nitrogen-functionalized pore
800 (a)
Barrier height/coor. num. (cm−1) 1, 2, 3 and 4-ring pores 10–5
Transmission
1-ring N-pore
600 2-ring 2N-pore
2-ring 4N-pore 10–10
Helium-3
400 Helium-4
Classical
10 15 20 25 30
200
(b) T (K)
105
Helium-3
transmission
QM/classical
0 Helium-4
0.5 1 1.5 2 2.5

Average pore radius (A)
100
10 15 20 25 30
FIGURE 25.6 Barrier height divided by the number of rim atoms
(c) T (K)
plotted against the averaged pore radius 〈r〉 derived from the cor-
responding pore geometries. The graph proves that the effective 20
ratio
pore size is the dominant parameter: a simple exponential fit of the 15
N-free pores reveals that there are no significant deviations for the
3He/4He
10
N-containing pores 2, 4, and 5. (Reprinted with permission from
A. W. Hauser, J. Schrier, and P. Schwerdtfeger. Helium tunneling 5
through nitrogen-functionalized graphene pores: Pressure- and tem- 0
perature-driven approaches to isotope separation. J. Phys. Chem. C, 10 15 20 25 30
116(19):10819–10827. Copyright 2012 American Chemical Society.) T (K)
FIGURE 25.7 (a) Weighted transmission of 3He (solid lines) and

types. From this finding, we derive that for helium particles, a 4He (dashed lines) through the single-side N-passivated graphene
nitrogen passivation is just a different way of pore-size modi- pore 4 as a function of temperature. (b) Ratio of quantum-mechan-
fication without additional types of interaction. ical to classical transmission probability. (c) The 3He/4He transmis-
To obtain the transmission probabilities for the selected sion ratio drops down from 19 to its asymptotic value of 1 within a
two-ring pores, we insert the corresponding potential energy temperature range of 10 K. The classical transmission probability
curves into Equation 25.4 and solve it via a finite-difference in (a) is shown as a dotted line. (Reprinted with permission from A.
W. Hauser and P. Schwerdtfeger. Nanoporous graphene membranes
approach. Our results are summarized in Figure 25.7, which
for efficient 3He/4He separation. J. Phys. Chem. Lett., 3:209–213.
contain (a) the weighted transmission of fermionic and bosonic Copyright 2012 American Chemical Society.)
helium as a function of the temperature, (b) the ratio of quan-
tum mechanical to classical transmission, and (c) the 3He/4He
ratio for the single-sided N-passivated graphene pore 4, which
is the most promising candidate for the given purpose.
1/ 2
Note the huge discrepancy between the classical and the  8k T 
〈u 〉 =  B  (25.6)
quantum-mechanical prediction at the temperature range  πm 

between 10 and 20 K, indicating that quantum tunneling is
essential for a correct description in the low-temperature
regime. This difference decreases with increasing tempera- with pressure p, particle mass m, and the Boltzmann con-
ture and is fully negligible for temperatures above 30 K. The stant k B. Combining Equations 25.5 and 25.6 and replacing
3He/4He ratio plotted in panel (c) shows an almost exponential ρ by p/(k BT) we finally obtain
increase with decreasing temperature from a moderate value
of 2 at 20 K to a promising factor of 19 at 10 K. The transmis-
sion probabilities lie in the range of 10−9, which can be con- p
zcoll = (25.7)
verted into an overall helium flux by multiplication with the 2πmkBT
frequency zcoll of particle–core collision events. For Maxwell–
Boltzmann distributed gas velocities, the latter quantity can Assuming a 100% porous graphene sheet and ideal gas
be written as conditions at 10 K and 1 bar (with a collision rate of approxi-
mately 1024 s−1 cm−2) this corresponds to a total helium flux of
ρ〈u〉 more than 10−9 mol s−1 cm−2, suggesting a multistage purifica-
zcoll = (25.5)
4 tion of terrestrial helium sources using several layers of the
proposed, yet hypothetical nanoporous graphene material: A
with ρ denoting the particle density and the mean velocity 〈u〉 membrane with a surface area of 1 m2 could yield a daily pro-
defined as duction of up to 3 g of 3He.
25.4.3 Temperature-Driven Isotope Separation tunneling: it leads to a 3He enrichment on the “cold” side of
the membrane. Hence, these two effects, both intrinsically
In the previous section, we looked at the isotope separation as quantum-mechanical phenomena, are competing, which leads
a single-directional, pressure-driven process at very low tem- to a more complex dependence of rqt,zp on the temperature of
perature, assuming a one-dimensional barrier between two the two reservoirs. The results for the single- and double-ring
reservoirs which are not in a thermodynamical equilibrium. pores in a temperature range between 10 and 70 K are shown
A possible extension of this simplified model description is to as contour plots in Figure 25.8. These landscapes show cer-
think of an equilibrium state between two reservoirs of differ- tain symmetry: For equal temperatures, the separation fac-
ent, constant temperatures (TH and TC), which are coupled via tor is one by definition; swapping temperatures TH and TC
the graphene membrane, allowing them to exchange 3He and leads to an inverted separation factor. All pores except for
4He particles [46]. The propagation of particles is now bidi-
the two-ring two-sided N-passivized pore 5 show a separa-
rectional, and we consider this by introducing a new quan- tion factor rqt,zp > 1 if TC is kept sufficiently low, indicating
tity, the separation factor rqt, which we define as the ratio of the dominance of quantum tunneling over the zero-point
3He/4He ratios, that is, the ratio between selectivities.
energy effect in this regime. However, the separation factor
is quenched due to the antagonistic effect of both contribu-
 4 p (T )   3 p (T )  tions. The two smaller pores 2 and 3 are impractical due to
rqt =  3 w H   4 w C  (25.8) the negligible helium transmission probability at the given
 pw (TH )   pw (TC ) 
temperatures. The two-ring single-sided N-passivated pore 4
yields a separation factor of 1.86 for TC = 10 and TH = 20 with
The subscript “qt” indicates that this separation is due to a still reasonable transmission probability of about 9 × 10−10.
the quantum-tunneling phenomena discussed above. The ther- At the same temperature values an even better separation fac-
mally weighted transmission probabilities pw for 3He and 4He tor (though in the other direction) of 0.34 is predicted for the
are denoted by the corresponding superscripts. Their temper- two-ring two-sided N-passivated pore 5, with a significantly
ature dependence can be read from Figure 25.7c. Assuming higher transmission probability of about 3 × 10−4.
that TH > TC, we obtain rqt > 1 since helium-3 transmission is
preferred over helium-4 and more likely at a lower tempera-
ture. Therefore, quantum tunneling leads to a 3He enrichment 25.5 SUMMARY
on the “hot” side. The same tendency is predicted by classical In this chapter, we investigated the applicability of nanopo-
thermodiffusion, which accounts for long-range interactions rous graphene sheets to the industrially highly relevant prob-
of a nonideal gas [46,93]. However, of different origin, these lem of gas separation. We introduced a hierarchy of model
two effects can be considered as additive. descriptions and analyzed their capabilities and limitations,
In a further extension of the steady-state model, we take given the example of methane from air and 3He from 4He sep-
formerly neglected degrees of freedom into consideration to aration. The very basic concept of comparing particle size and
account—at least partially—for the three dimensionality of pore diameter, definitely the most obvious approach and the
the particle propagation. Assuming a fully rigid pore, we have fundamental idea behind an experimental technique known
two more degrees of freedom in the system, corresponding to as “molecular sieving,” has no predictive power in situations
the two in-plane movements of a He atom in the pore. This where the kinetic diameters of the particles are too similar.
leads to two positive eigenmodes of the pore–particle system We showed that these experimentally derived parameters nei-
at the transition state, a saddle point of first order, and accord- ther provide a clear ranking of molecular size nor can they be
ing to Heisenberg’s uncertainty, these modes must be occu- directly related to the real dimension of a graphene nanopore.
pied, at least in their vibrational ground state. The associated The whole concept of size comparison breaks down as soon
zero-point energies have to be added to the barrier height, and as interactions between the membrane and the propagating
since the vibrational energies show an inverse dependence on particles become important.
particle mass we obtain different corrections Δ3 and Δ4 for An adsorption preference for a certain molecule due
3He and 4He, respectively, with Δ − Δ > 1. In the rigid model,
3 4 to higher attractive forces can lead to an accumulation of
these differences are in the order of only a few wave numbers, that species close to the membrane, as was observed for N2
while a more accurate description including all vibrational over H2. Therefore, the N2/H2 transmission ratio becomes
modes of the particle–pore system predicts values about 10 larger than 1.0 for sufficiently large pores, which cannot be
times larger [46,47]. Combining transition state theory with explained within the molecular sieving picture. Gas particles
the phenomenon of quantum tunneling one obtains a zero- with similar molecular dimensions, for example, N2 and CO2,
point corrected formula (hence the subscript “zp”) for the iso- which share about the same “waistline” of their electronic
tope separation factor: density, experience significantly different transmission bar-
riers when propagating through a nitrogen-functionalized
rqt,zp = rqt × e − ( ∆3 − ∆ 4 )((1/ kBTC ) − (1/ kBTH )) (25.9) pore due to different chemical interactions with the pore rim.

Furthermore, noticeable changes in the molecular geom-
The exponential term in this equation contributes to etry of the pore or the gas molecule itself can appear during
rqt,zp < 1 for TH > TC, which is the opposite effect of quantum the gas propagation. A comparison of transmission barrier
(a) (c) 2.2

1.4 1.2 1 1.6 1.2 1
2
70 70
0.8
1.8 1.4
60 60
0.8
50 50 1
10–12 0.8
40 0.6 10–14 40 10–4

10–16 0.8
30 30 0.6 10–5
1
transmission probability
0.6
10–6
20 1.6 1.4 1.2
10–23 20
1
10–8
1
10 10 1.8 1.6 1.4 1.2 1
TC (K)
10 20 30 40 50 60 70 10 20 30 40 50 60 70
(b) 1.5 1.25 1
(d)
2.5 1.5
70 0.75 70 1
3 2
60 60 5×10–2
3He
50 50
0.5 10–10
40 1 0.75 40
10–12
30 1
10–14 30 10–2
10–16 0.5
1.25
20 1.5 20
1.75 10–23 10–3
2.25 2
10 2.5 10
10 20 30 40 50 60 70 10 20 30 40 50 60 70
TH (K)
FIGURE 25.8 Thermally driven isotope separation factor (Equation 25.9), indicated by solid contours; dotted contours labeled on the
right axis indicate the minimum 3He thermally weighted transmission probability, see Equation 25.3. (a) Single-ring N-passivated pore 2;
(b) two-ring H-passivated pore 3; (c) two-ring single-sided N-passivated pore 4; and (d) two-ring two-sided N-passivated pore 5. (Reprinted
with permission from A. W. Hauser, J. Schrier, and P. Schwerdtfeger. Helium tunneling through nitrogen-functionalized graphene pores:
Pressure- and temperature-driven approaches to isotope separation. J. Phys. Chem. C, 116(19):10819–10827. Copyright 2012 American
Chemical Society.)
energies for rigid and fully relaxed finite model pores shows temperature difference across the membrane. Quantum tun-
significant differences between these two assumptions, which neling and classical thermodiffusion predict the enrichment
can be interpreted as lower and upper boundaries to the flex- of the lighter particles on the hot side of the membrane but
ibility of a real graphene pore in its periodic, two-dimensional compete with the effect of zero-point energy, which favors the
environment. Electronic structure theory offers the best tool opposite. Therefore, the separation factor, a parameter intro-
to account for all effects mentioned above and allows for a duced to describe the ratio between left-to-right and right-to-
prediction of barrier heights and transmission rates. We chose left selectivity for a pair of isotopes, shows a more complex
DFT due to the lower computational cost compared to molec- dependence on the temperatures of both reservoirs.
ular-orbital-based post-Hartree–Fock methods. The inclusion In summary, nanoporous graphene has many potential
of dispersion corrections is essential for an adequate descrip- applications in the field of gas separation. We suggested
tion of the attractive interactions between the gas molecule two possible pathways for a future usage of this promising
and membrane. material, used these examples to illustrate the limitations of
In the second part of this chapter, we discussed weaker common model descriptions, and presented a computational
effects that are only relevant for small particles such as molec- approach based on electronic structure theory and applied
ular hydrogen or helium at low temperatures. The phenom- it to finite model pores. Despite its conceptual simplicity in
enon of quantum tunneling made it necessary to calculate the macroscopic picture, the separation of gases via an effec-
the whole reaction paths as the transmission probability for tively two-dimensional membrane is a complex process on
a gas molecule is not only dependent on the barrier height, the molecular level. As a rule of thumb, such as an almost
but also on its actual shape and the mass of the propagat- linear and strictly monotonic dependence of the barrier height
ing particle. The latter dependence can be used for isotope on the effective size of a molecule, has to be dismissed in
separation, as was demonstrated for the example of fermionic cases of strong pore–particle interactions. On the other hand,
and bosonic helium. In a last extension of our pore model, this offers new strategies for a fine-tuning of future materials
we looked at bidirectional propagation processes driven by a that go beyond pore size optimization. Possible routes are the
chemical functionalization of the pore rim or the introduction 17. G. Compagnini, P. Russo, A. Hu. Synthesis, properties and
of nonporous “patches” with increased physisorption proper- potential applications of porous graphene: A review. Nano-
ties near the pore to increase the transmission probability for a Micro Lett., 5:260–273, 2013.
18. L. Jiang and Z. Fan. Design of advanced porous graphene
certain gas species. Since submission of this chapter a number
materials: From graphene nanomesh to 3d architectures.
of new articles appeared concerning graphene-based molecu- Nanoscale, 6:1922–1945, 2014.
lar separation. A selection of a few recent reviews is given in 19. G. Xie, R. Yang, P. Chen, J. Zhang, X. Tian, S. Wu, J. Zhao
References 94–98. et al. A general route towards defect and pore engineering in
graphene. Small, 10(11):2280–2284, 2014.
20. W. Yuan, J. Chen, and G. Shi. Nanoporous graphene materi-
ACKNOWLEDGMENT als. Mater. Today, 17(2):77–85, 2014.
21. V. Berry. Impermeability of graphene and its applications.
This work was supported by an SIF funding from Massey Carbon, 62(0):1–10, 2013.
University for the project “New Zealand’s answer to global 22. M. D. Fischbein and M. Drndíc. Electron beam nanosculpting
warming: nanomaterials for methane reduction. of suspended graphene sheets. Appl. Phys. Lett., 93:113107:1–
113107:3, 2008.
23. P. Kuhn, A. Forget, D. Su, A. Thomas, and M. Antonietti.
REFERENCES From microporous regular frameworks to mesoporous mate-
1. R. W. Baker. Membrane Technology and Applications. John rials with ultrahigh surface area: Dynamic reorganization of
Wiley & Sons, New York, 2012. porous polymer networks. J. Am. Chem. Soc., 130(40):13333–
2. S. A. Montzka, E. J. Dlugokencky, and J. H. Butler. Non-CO2 13337, 2008.
greenhouse gases and climate change. Nature, 746:43–50, 24. P. Xu, J. Yang, K. Wang, Z. Zhou, and P. Shen. Porous gra-
2011. phene: Properties, preparation, and potential applications.
3. P. Bernardo, E. Drioli, and G. Golemme. Membrane gas Chin. Sci. Bull., 57(23):2948–2955, 2012.
separation: A review/state of the art. Ind. Eng. Chem. Res., 25. S. Ozeki, T. Ito, K. Uozumi, and I. Nishio. Scanning tunnel-
48(10):4638–4663, 2009. ing microscopy of UV-induced gasification reaction on highly
4. N. Tanihara, H. Shimazaki, Y. Hirayama, S. Nakanishi, T. oriented pyrolytic graphite. Jpn. J. Appl. Phys., 35(Part 1, No.
Yoshinaga, and Y. Kusuki. Gas permeation properties of asym- 6B):3772–3774, 1996.
metric carbon hollow fiber membranes prepared from asym- 26. S. Huh, J. Park, Y. S. Kim, K. S. Kim, B. H. Hong, and J.-M.
metric polyimide hollow fiber. J. Membr. Sci., 160(2):179–186, Nam. UV/ozone-oxidized large-scale graphene platform with
1999. large chemical enhancement in surface-enhanced raman scat-
5. Z. Song, Z. Xu, X. Huang, J.-Y. Kim, and Q. Zheng. On the tering. ACS Nano, 5(12):9799–9806, 2011.
fracture of supported graphene under pressure. J. Appl. Mech., 27. S. P. Koenig, L. Wang, J. Pellegrino, and J. S. Bunch.
80(4):041046, 2013. Selective molecular sieving through porous graphene. Nat.
6. P. Avouris, Z. Chen, and V. Perebeinos. Carbon-based elec- Nanotechnol., 7(11):728–732, 2012.
tronics. Nat. Nanotechnol., 2(10):605–615, 2007. 28. D. C. Bell, M. C. Lemme, L. A. Stern, J. R. Williams, and C.
7. A. K. Geim and K. S. Novoselov. The rise of graphene. Nat. M. Marcus. Precision cutting and patterning of graphene with
Mater., 6(3):183–191, 2007. helium ions. Nanotechnology, 20(45):455301, 2009.
8. C. W. J. Beenakker. Colloquium: Andreev reflection and Klein 29. K. Celebi, J. Buchheim, R. M. Wyss, A. Droudian, P. Gasser,
tunneling in graphene. Rev. Mod. Phys., 80(4):1337–1354, I. Shorubalko, J.-I. Kye, C. Lee, and H. G. Park. Ultimate
2008. permeation across atomically thin porous graphene. Science,
9. A. H. Castro Neto, F. Guinea, N. M. R. Peres, K. S. Novoselov, 344(6181):289–292, 2014.
and A. K. Geim. The electronic properties of graphene. Rev. 30. D. E. Jiang, V. R. Cooper, and S. Dai. Porous graphene as the
Mod. Phys., 81(1):109–162, 2009. ultimate membrane for gas separation. Nano Lett., 9(12):4019–
10. S. Park and R. S. Ruoff. Chemical methods for the production 4024, 2009. PMID: 19995080.
of graphenes. Nat. Nanotechnol., 4(4):217–224, 2009. 31. B. J. Bucior, D.-L. Chen, J. Liu, and J. K. Johnson. Porous car-
11. F. Bonaccorso, Z. Sun, T. Hasan, and A. C. Ferrari. Graphene bon nanotube membranes for separation of H2/CH4 and CO2/
photonics and optoelectronics. Nat. Photonics, 4(9):611–622, CH4 mixtures. J. Phys. Chem. C, 116(49):25904–25910, 2012.
2010. 32. X. Qin, Q. Meng, Y. Feng, and Y. Gao. Graphene with line
12. O. V. Prezhdo, P. V. Kamat, and G. C. Schatz. Virtual issue: defect as a membrane for gas separation: Design via a first-
Graphene and functionalized graphene. J. Phys. Chem. C, principles modeling. Surf. Sci., 607(0):153–158, 2013.
115(8):3195–3197, 2011. 33. C. Sun, M. S. H. Boutilier, H. Au, P. Poesio, B. Bai, R. Karnik,
13. V. Palermo. Not a molecule, not a polymer, not a substrate… and N. G. Hadjiconstantinou. Mechanisms of molecular per-
the many faces of graphene as a chemical platform. Chem. meation through nanoporous graphene membranes. Langmuir,
Commun., 49:2848–2857, 2013. 30(2):675–682, 2014.
14. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. 34. H. Du, J. Li, J. Zhang, G. Su, X. Li, and Y. Zhao. Separation
Zhang, S. V. Dubonos, I. V. Grigorieva, and A. A. Firsov. of hydrogen and nitrogen gases with porous graphene mem-
Electric field effect in atomically thin carbon films. Science, brane. J. Phys. Chem. C, 115(47):23261–23266, 2011.
306(5696):666–669, 2004. 35. G. Lei, C. Liu, H. Xie, and F. Song. Separation of the hydrogen
15. S. T. Oyama, D. Lee, P. Hacarlioglu, and R. F. Saraf. Theory sulfide and methane mixture by the porous graphene mem-
of hydrogen permeability in nonporous silica membranes. J. brane: Effect of the charges. Chem. Phys. Lett., 599(0):127–
Membr. Sci., 244(1–2):45–53, 2004. 132, 2014.
16. O. Leenaerts, B. Partoens, and F. M. Peeters. Graphene: A 36. J. Schrier. Helium separation using porous graphene mem-
perfect nanoballoon. Appl. Phys. Lett., 93(19):193107, 2008. branes. J. Phys. Chem. Lett., 1(15):2284–2287, 2010.
37. A. M. Brockway and J. Schrier. Noble gas separation using 57. D. Cohen-Tanugi and J. C. Grossman. Water desalination
PG-ESX (X = 1, 2, 3) nanoporous two-dimensional polymers. across nanoporous graphene. Nano Lett., 12(7):3602–3608,
J. Phys. Chem. C, 117(1):393–402, 2013. 2012.
38. A. W. Hauser and P. Schwerdtfeger. Methane-selective nano- 58. G. Hu, M. Mao, and S. Ghosal. Ion transport through a gra-
porous graphene membranes for gas purification. Phys. Chem. phene nanopore. Nanotechnology, 23(39):395501, 2012.
Chem. Phys., 14:13292–13298, 2012. 59. E. N. Wang and R. Karnik. Water desalination: Graphene
39. M. Shan, Q. Xue, N. Jing, C. Ling, T. Zhang, Z. Yan, and cleans up water. Nat Nanotechnol., 7(9):552–554, 2012.
J. Zheng. Influence of chemical functionalization on the 60. D. Konatham, J. Yu, T. A. Ho, and Alberto Striolo. Simulation
CO2/N2 separation performance of porous graphene mem- insights for graphene-based water desalination membranes.
branes. Nanoscale, 4:5477–5482, 2012. Langmuir, 29(38):11884–11897, 2013.
40. T. Wu, Q. Xue, C. Ling, M. Shan, Z. Liu, Y. Tao, and X. Li. 61. K. C. Kemp, H. Seema, M. Saleh, N. H. Le, K. Mahesh, V.
Fluorine-modified porous graphene as membrane for CO2/N2 Chandra, and K. S. Kim. Environmental applications using
separation: Molecular dynamic and first-principles simula- graphene composites: Water remediation and gas adsorption.
tions. J. Phys. Chem. C, 118(14):7369–7376, 2014. Nanoscale, 5:3149–3171, 2013.
41. J. Schrier. Carbon dioxide separation with a two-dimensional 62. A. Du, Z. Zhu, and S. C. Smith. Multifunctional porous gra-
polymer membrane. ACS Appl. Mater. Interfaces, 4(7):3745– phene for nanoelectronics and hydrogen storage: New proper-
3752, 2012. ties revealed by first principle calculations. J. Am. Chem. Soc.,
42. H. Liu, V. R. Cooper, S. Dai, and D.-e. Jiang. Windowed car- 132(9):2876–2877, 2010. PMID: 20155897.
bon nanotubes for efficient CO2 removal from natural gas. 63. P. Reunchan and S.-H. Jhi. Metal-dispersed porous gra-
J. Phys. Chem. Lett., 3(22):3343–3347, 2012. phene for hydrogen storage. Appl. Phys. Lett., 98(9):093103,
43. H. Liu, Z. Chen, S. Dai, and D. e. Jiang. Selectivity trend of 2011.
gas separation through nanoporous graphene. J. Solid State 64. B. C. Wood, S. Y. Bhide, D. Dutta, V. S. Kandagal, A. D.
Chem., 224:2–6, 2015. Pathak, S. N. Punnathanam, K. G. Ayappa, and S. Narasimhan.
44. J. Lee and N. R. Aluru. Water-solubility-driven separation of Methane and carbon dioxide adsorption on edge-function-
gases using graphene membrane. J. Membr. Sci., 428(0):546– alized graphene: A comparative DFT study. J. Chem. Phys.,
553, 2013. 137(5):054702, 2012.
45. A. W. Hauser and P. Schwerdtfeger. Nanoporous graphene 65. Y. Wang, Z. Shi, Y. Huang, Y. Ma, C. Wang, M. Chen, and
membranes for efficient 3He/4He separation. J. Phys. Chem. Y. Chen. Supercapacitor devices based on graphene materials.
Lett., 3:209–213, 2012. J. Phys. Chem. C, 113(30):13103–13107, 2009.
46. J. Schrier and J. McClain. Thermally-driven isotope sepa- 66. C. Liu, Z. Yu, D. Neff, A. Zhamu, and B. Z. Jang. Graphene-
ration across nanoporous graphene. Chem. Phys. Lett., based supercapacitor with an ultrahigh energy density. Nano
521(0):118–124, 2012. Lett., 10(12):4863–4868, 2010.
47. A. W. Hauser, J. Schrier, and P. Schwerdtfeger. Helium tun- 67. L. L. Zhang, R. Zhou, and X. S. Zhao. Graphene-based mate-
neling through nitrogen-functionalized graphene pores: rials as supercapacitor electrodes. J. Mater. Chem., 20:5983–
Pressure- and temperature-driven approaches to isotope sepa- 5992, 2010.
ration. J. Phys. Chem. C, 116(19):10819–10827, 2012. 68. A. Yu, I. Roes, A. Davies, and Z. Chen. Ultrathin, transpar-
48. M. Hankel, Y. Jiao, A. Du, S. K. Gray, and S. C. Smith. ent, and flexible graphene films for supercapacitor application.
Asymmetrically decorated, doped porous graphene as an Appl. Phys. Lett., 96(25):253105, 2010.
effective membrane for hydrogen isotope separation. J. Phys. 69. H. M. Jeong, J. W. Lee, W. H. Shin, Y. J. Choi, H. J. Shin, J.
Chem. C, 116(11):6672–6676, 2012. K. Kang, and J. W. Choi. Nitrogen-doped graphene for high-
49. K. Sint, B. Wang, and P. Král. Selective ion passage through performance ultracapacitors and the importance of nitrogen-
functionalized graphene nanopores. J. Am. Chem. Soc., doped sites at basal planes. Nano Lett., 11(6):2472–2477,
130(49):16448–16449, 2008. 2011.
50. S. C. O’Hern, M. S. H. Boutilier, J.-C. Idrobo, Y. Song, J. 70. S. C. O’Hern, C. A. Stewart, M. S. H. Boutilier, J.-C. Idrobo,
Kong, T. Laoui, M. Atieh, and R. Karnik. Selective ionic S. Bhaviripudi, S. K. Das, J. Kong, T. Laoui, M. Atieh, and
transport through tunable subnanometer pores in single-layer R. Karnik. Selective molecular transport through intrin-
graphene membranes. Nano Lett., 14(3):1234–1241, 2014. sic defects in a single layer of CVD graphene. ACS Nano,
51. M. E. Suk and N. R. Aluru. Ion transport in sub-5-nm gra- 6(11):10130–10138, 2012.
phene nanopores. J. Chem. Phys., 140(8):084707, 2014. 71. H. Li, Z. Song, X. Zhang, Y. Huang, S. Li, Y. Mao, H. J.
52. S. Garaj, W. Hubbard, A. Reina, J. Kong, D. Branton, and J. Ploehn, Y. Bao, and M. Yu. Ultrathin, molecular-sieving gra-
A. Golovchenko. Graphene as a subnanometre trans-electrode phene oxide membranes for selective hydrogen separation.
membrane. Nature, 467(7312):190–193, 2010. Science, 342(6154):95–98, 2013.
53. H. W. Ch. Postma. Rapid sequencing of individual DNA mole- 72. H. W. Kim, H. W. Yoon, S.-M. Yoon, B. M. Yoo, B. K. Ahn, Y.
cules in graphene nanogaps. Nano Lett., 10(2):420–425, 2010. H. Cho, H. J. Shin et al. Selective gas transport through few-
PMID: 20044842. layered graphene and graphene oxide membranes. Science,
54. G. F. Schneider, S. W. Kowalczyk, V. E. Calado, G. Pandraud, 342(6154):91–95, 2013.
H. W. Zandbergen, L. M. K. Vandersypen, and C. Dekker. 73. M. S. H. Boutilier, C. Sun, S. C. O’Hern, H. Au, N. G.
DNA translocation through graphene nanopores. Nano Lett., Hadjiconstantinou, and R. Karnik. Implications of permeation
10(8):3163–3167, 2010. through intrinsic defects in graphene on the design of defect-
55. C. A. Merchant, K. Healy, M. Wanunu, V. Ray, N. Peterman, tolerant membranes for gas separation. ACS Nano, 8(1):841–
J. Bartel, M. D. Fischbein et al. DNA translocation through 849, 2014.
graphene nanopores. Nano Lett., 10(8):2915–2921, 2010. 74. Y. Jiao, A. Du, M. Hankel, and S. C. Smith. Modelling car-
56. M. E. Suk and N. R. Aluru. Water transport through ultrathin bon membranes for gas and isotope separation. Phys. Chem.
graphene. J. Phys. Chem. Lett., 1(10):1590–1594, 2010. Chem. Phys., 15:4832–4843, 2013.
75. D. Dubbeldam and R. Q. Snurr. Recent developments in the 88. Y. P. Yampolskii, I. Pinnau, and B. D. Freeman. Materials
molecular modeling of diffusion in nanoporous materials. Science of Membranes for Gas and Vapor Separation. John
Mol. Simul., 33(4–5):305–325, 2007. Wiley & Sons, New York, 2006.
76. A. F. Voter, F. Montalenti, and T. C. Germann. Extending the 89. B. E. Poling, J. M. Prausnitz, and J. P. O’Connell. The
time scale in atomistic simulation of materials. Annu. Rev. Properties of Gases and Liquids. McGraw-Hill, New York,
Mater. Res., 32(1):321–346, 2002. 5th edition, 2000.
77. L. W. Drahushuk and M. S. Strano. Mechanisms of gas per- 90. S. F. Boys and F. Bernardi. The calculation of small molecu-
meation through single layer graphene membranes. Langmuir, lar interactions by the differences of separate total energies.
28(48):16671–16678, 2012. Some procedures with reduced errors. Mol. Phys., 19:553–
78. G. Henkelman and H. Jónsson. Improved tangent estimate in the 566, 1970.
nudged elastic band method for finding minimum energy paths 91. A. Bondi. Van der Waals volumes and radii. J. Phys. Chem.,
and saddle points. J. Chem. Phys., 113(22):9978–9985, 2000. 68(3):441–451, 1964.
79. T. A. Halgren and W. N. Lipscomb. The synchronous-transit 92. R. Scott Rowland and R. Taylor. Intermolecular nonbonded
method for determining reaction pathways and locating molec- contact distances in organic crystal structures: Comparison
ular transition states. Chem. Phys. Lett., 49(2):225–232, 1977. with distances expected from van der Waals Radii. J. Phys.
80. A. Behn, P. M. Zimmerman, A. T. Bell, and M. Head-Gordon. Chem., 100(18):7384–7391, 1996.
Efficient exploration of reaction paths via a freezing string 93. S. C. Saxena and B. P. Mathur. Thermal diffusion in isoto-
method. J. Chem. Phys., 135(22):224108, 2011. pic gas mixtures and intermolecular forces. Rev. Mod. Phys.,
81. J. Baker. An algorithm for the location of transition states. 38:380–390, 1966.
J. Comput. Chem., 7(4):385–395, 1986. 94. Z. P. Smith and B. D. Freeman. Graphene oxide: A new
82. S. Grimme. Semiempirical GGA-type density functional con- platform for high-performance gas- and liquid-separation
structed with a long-range dispersion correction. J. Comput. membranes. Angew. Chem. Int. Ed., 53:10286–10288, 2014.
Phys., 27(15):1787–1799, 2006. doi:10.1002/anie.201404407.
83. D. E. Woon and T. H. Jr. Dunning. Gaussian basis sets for use 95. S. Gadipelli and Z. X. Guo. Graphene-based materials:
in correlated molecular calculations. IV. Calculation of static Synthesis and gas sorption, storage and separation, Progress
electrical response properties. J. Chem. Phys., 100(4):2975– in Materials Science, Volume 69, pp. 1–60, April 2015, ISSN
2988, 1994. 0079-6425, http://dx.doi.org/10.1016/j.pmatsci.2014.10.004.
84. T. H. Jr. Dunning. Gaussian basis sets for use in correlated 96. A. Aghigh, V. Alizadeh, H.Y. Wong, Md. Shabiul Islam,
molecular calculations. I. The atoms boron through neon and N. Amin, and M. Zaman. Recent advances in utilization of
hydrogen. J. Chem. Phys., 90(2):1007–1023, 1989. graphene for filtration and desalination of water: A review,
85. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. Desalination, 365:389–397, 2015, http://dx.doi.org/10.1016/j.
A. Robb, J. R. Cheeseman, G. Scalmani et al. Gaussian 09 desal.2015.03.024.
Revision A.02. Gaussian Inc., Wallingford CT, 2009. 97. L. Huang, M. Zhang, C. Li, and G. Shi. Graphene-based
86. D. W. Breck. Zeolite Molecular Sieves: Structure, Chemistry membranes for molecular separation, J. Phys. Chem. Lett.,
and Use. John Wiley & Sons, New York, 1974. 6(14):2806–2815, 2015. doi:10.1021/acs.jpclett.5b00914.
87. K. D. Vogiatzis, A. Mavrandonakis, W. Klopper, and G. E. 98. C. Sun, B. Wen, and B. Bai. Recent advances in nanoporous
Froudakis. Ab initio study of the interactions between CO2 graphene membrane for gas separation and water purifica-
and N-containing organic heterocycles. Chem. Phys. Chem., tion, Science Bulletin, Issue 2095–9273, pp. 1–17, October 14,
10(2):374–383, 2009. 2015. doi:10.1007/s11434-015-0914-9.
26 Photorefractive Properties of
Graphene-Based Organic Systems
N. V. Kamanina
CONTENTS
Abstract...................................................................................................................................................................................... 491
26.1 Recent Evidences to Use Innovative Perspective Nanoobjects in the Organic Optoelectronics...................................... 491
26.2 Material Synthesis and Technical Method Background................................................................................................... 492
26.3 Possible Explanation of the Nanoobjects Influence on the Materials’ Features............................................................... 492
26.3.1 Some Photorefractive Features of the Graphene-Doped Polyimide among Other Nanoobjects-Doped
Polyimide Systems................................................................................................................................................ 492
26.3.2 Some Ideas to Apply Graphene-Based Consideration to Explain the Decrease of the Resistivity
of the Intermediate Layer in the Liquid Crystal Cells.......................................................................................... 494
26.4 Conclusion........................................................................................................................................................................ 495
References.................................................................................................................................................................................. 496
ABSTRACT To analyze the optical and the nonlinear optical processes

of the materials, we should estimate the linear effect when
It is well known that an important direction in the study of the electric field of the laser wave is less than the intra-atomic
promising photorefractive materials for electro-optics, non- electric field correlated with the electron charge and with the
linear optics, semiconductor device technology, medicine, Bohr radius. However, when the electric field of the laser wave
display, biotechnology, and other applications is related to is larger than the intra-atomic electric field, we should draw
investigations of the optical and nonlinear optical proper- the attention on the nonlinear optical features [21]. It should
ties of nanostructured polymer, monomer, and liquid crystal- be remarked that the macroscopic polarization P of a nonlin-
line systems under the condition of their doping by effective ear optical medium [21] is described as
nanoobjects. In this chapter, the holographic technique is used
to study the photoinduced change in the refractive index of
P = χ(1) E + χ( 2 ) E 2 + χ(3) E 3 + + χ( n ) E n + (26.1)
a conjugate polyimide matrix sensitized by graphene oxides.
The obtained data are correlated with the ones received before
for fullerene-, nanotubes-, and shungite-doped polyimide. where E is the electric field of the light beam, and χ(n) is the
optical susceptibility of the nth order.
Using this aspect, the values of optical susceptibilities
26.1 RECENT EVIDENCES TO USE INNOVATIVE play an important role in nonlinear optical effect. The most
PERSPECTIVE NANOOBJECTS IN THE important optical characteristic of the organic materials with
ORGANIC OPTOELECTRONICS different symmetry is the induced dipole [22], which can be
expressed through dipole polarizabilities α(n). These are in
Recent development of optoelectronic devices, especially
turn related by the proportional dependence to the nonlinear
based on the fullerene- and graphene-based systems [1–10],
susceptibility χ(n) and to the local volume υ of the materials,
indicates an important task of studying and creating the new
which is shown via
organic electrooptical and photorefractive nanostructured
materials, which can be an alternative to inorganic ones
[11–20]. On one side, the organic structures modified with α(n)
χ( n ) = (26.2)
nanoobjects can be treated via low electric and light fields and ν
realize the effective modulating and recording optical data in
a wide range of radiation energy densities and frequencies. One can see that the values of nonlinear susceptibilities,
It can be useful for the practical application, basically in the for the first, are responsible for the polarizability of the local
optoelectronics and biomedicine. On the other side, the new volume of media and for the second, are connected with the
process observed in the nanostructured materials can extend macroscopic polarization of all system. Thus, the laser–matter
our knowledge in the photophysics of materials, especially of interaction provokes the change in polarization of media and
optical and nonlinear optical ones. predicts the change in important properties such as dynamic,
491
photoconductive, and photorefractive. It should be mentioned were recorded under the Raman–Nath diffraction conditions
that among these properties, the photorefractive characteris- when Λ−1 ≥ d; where Λ is the spatial frequency, Λ−1 indicates
tics can be dominated. The change in nonlinear refraction and the grating period, and d is the thickness of the structure
cubic nonlinearity reveal the modification of barrier-free elec- irradiated with laser beam. The thin gratings with spatial
tron pathway and dipole polarizability. It is connected with the frequencies of 100 and 150 cm−1 were recorded at an energy
change of the dipole moment and the charge carrier mobility, density varied within 0.05–0.6 J/cm2. It should be taken into
and it is regarded as the change of absorption cross section. account that we indicate a small local volume of our conju-
Thus, these features show the unique place of photorefractive gated organic nanostructured medium, substantially smaller
characteristics among others to characterize the spectral, pho- than the incident wavelength. For comparison, the optical
toconductive, photorefractive, and dynamic properties of the wavelength (λ = 532 nm in the experiment considered below)
optical materials, especially organics [23–27]. is larger than the dimension of the incorporated nanoobjects
It should be noted that at present time the promising (e.g., fullerene molecules are 0.65 ± 0.7 nm in size). In this
nanoobjects, which dramatically influence the photorefractive case, we can estimate the important optical characteristic,
parameters of the organic materials could be the following: namely, the induced dipole via estimation of laser-induced
fullerenes, shungites, quantum dots, carbon nanotubes (CNTs), refraction index and cubic nonlinearity.
graphenes, etc. They can stimulate the efficient intermolecular
charge transfer complex (CTC) formation in the nanoobjects-
doped organic conjugated materials [12,16,19,23,27]. More 26.3 POSSIBLE EXPLANATION OF THE
efficient charge transfer can be observed between the organic NANOOBJECTS INFLUENCE ON
intramolecular donor fragment (inside the monomer, polymer, THE MATERIALS’ FEATURES
or liquid crystals (LCs) structures compounds) and the inter-
molecular nanoobjects-based acceptor. It can be supported 26.3.1 Some Photorefractive Features of the
by the fact that the electron affinity energy of the nanoob- Graphene-Doped Polyimide among Other
jects, for example, for shungite structure is ~2 eV, fullerenes Nanoobjects-Doped Polyimide Systems
is ~2.65 eV, and quantum dots is 3.8–4.2 eV, which is larger
than that for most dyes and organic molecules intramolecu- The main nonlinear optical results of this study are summa-
lar acceptor fragment. Moreover, fullerenes, for example, can rized [19,23,24,27,30–36] in Table 26.1. An analysis of data
occupy more than six electrons [13]. It can organize the CTC presented in Table 26.1 for model organic polyimide system
formation in the nanoobjects-doped organic conjugated mate- shows that the introduction of nanoobjects as active accep-
rials with good advantage. tors of electrons significantly influences the charge trans-
fer under conditions where the intermolecular interaction
dominates over the intramolecular donor–acceptor contacts.
26.2 MATERIAL SYNTHESIS AND TECHNICAL Redistribution of the electron density during the recording of
METHOD BACKGROUND gratings in nanostructured materials changes the refractive
Let us consider specifically conjugated organic matrix based index by at least an order of magnitude as compared to that
on polyimide as a model to be modified with nanoobjects to in the initial matrix. This results in the formation of a clear
use the consideration mentioned above to the systems with interference pattern with a distribution of diffraction orders
registered high cubic nonlinearity estimated under holo- as shown in Figure 26.1. The diffusion of carriers from the
graphic recording scheme. The structures of molecules men- bright to dark region during the laser recording of the interfer-
tioned above have been previously shown in paper [24], where ence pattern proceeds in three (rather than two) dimensions,
triphenylamine fragment of polyimide compound is the donor which is manifested by a difference in the distribution of dif-
fragment of this complicated organic conjugated molecule. fraction orders along the horizontal and vertical axes. Thus,
Based on the experimentally observed large laser-induced the grating displacement takes place in a three-dimensional
change in the refractive index, we can consider the nanoob- (3D) medium (rather than in a 2D medium) formed as a result
jects-doped polyimide as the materials with high cubic opti- of the nanostructuration.
cal susceptibility. The thicknesses of thin film samples were The light-induced refractive index change Δni in the thin
within 2–4 μm. The organic matrices were sensitized by nanoobjects-doped films could be estimated from the experi-
doping with commercially available fullerenes C60 and C70 mental data of increase in diffraction efficiency using the fol-
(purchased from Alfa Aesar Company, Karlsruhe, Germany), lowing equation [19,21,23]:
with CNTs (received from some Russian organizations such
as Vladimir State University, Vladimir, Russia and, Boreskov I1  π∆ni d 
2
Institute of Catalysis, SB RAS, Novosibirsk, Russia), and with η= = (26.3)

I 0  2λ 
graphene oxides (received from Nanoinnova Technologies
SL, Spain). The concentration of dopants was varied within
0.05–0.5 wt.%. where η is the diffraction efficiency, I1 the intensity of the first
The experiments were performed analogous to the one diffraction order, I0 the incident laser beam, d the film thick-
shown in the papers [28,29]. The thin holographic gratings ness, and λ is the laser wavelength.
TABLE 26.1
Laser-Induced Change in the Refractive Index of the Nanoobjects-Sensitized Model Polyimide Matrixes
Nano-Objects Wavelength Energy Density Spatial Laser Pulse Laser-Induced Change in
Structure Studied Content (wt.%) (nm) (J/cm2) Frequency (mm1) Width (ns) the Refractive Index, Δni References
Pure polyimide 0 532 0.6 90 20 −4
10 –10 −5 [24,23,30]
Polyimide + malachite green 0.2 532 0.5–0.6 90–100 10–20 2.87 × 10−4 [19]
Polyimide + quantum dots based on CdSe(ZnS) 0.003 532 0.2–0.3 90–100 2.0 × 10−3 [31]
Polyimide + graphene oxides 0.1 532 0.2 100 10 3.4 × 10−3 [26]
Polyimide + graphene oxides 0.1 532 0.3 100 10 3.65 × 10−3 Present
Polyimide + shungite 0.2 532 0.063–0.1 150 10 3.8–5.3 × 10−3 [32]
Polyimide + С60 0.2 532 0.5–0.6 90 10–20 4.2 × 10−3 [24,33]
Polyimide + С70 0.2 532 0.5–0.6 90 10–20 2.15 × 10−4–4.68 × 10−3 [23,34]
Photorefractive Properties of Graphene-Based Organic Systems
Polyimide + С70 0.2 532 0.5–0.6 90 400 ps 3.71 × 10−4–5.04 × 10−4 [23,34]

Polyimide + С70 0.2 532 0.6 90 10–20 4.68 × 10−3 [24]
Polyimide + nanotubes 0.1 532 0.5–0.8 90 10–20 5.7 × 10−3 [24]
Polyimide + nanotubes 0.05 532 0.3 150 10 4.5 × 10−3 [35]
Polyimide + double-walled carbon nanotube powder 0.1 532 0.063–0.1 100 10 9.4 × 10−3 [27]
Polyimide + double-walled carbon nanotube powder 0.1 532 0.063–0.1 150 10 7.0 × 10−3 [27]
Polyimide + carbon nanofibers (type MIG) 0.1 532 0.6 90–100 10 11.7 × 10−3 [36]
Polyimide + carbon nanofibers (type MIG) 0.1 532 0.3–0.6 150 10 11.2 × 10−3 [27]
Polyimide + carbon nanofibers (type 65BR) 0.1 532 0.1–0.3 90–100 10 12.0 × 10−3 [36]
Polyimide + carbon nanofibers (type 65BR) 0.1 532 0.1 90 10 15.2 × 10−3 [27]
493
Using the obtained results, we have calculated the nonlin-

ear refraction n2 and nonlinear third order optical suscepti-
bility (cubic nonlinearity) χ(3) for all systems using a method
described above. It was found that these parameters fall
within n2 = 10−10 –10−9 cm2/ W and χ(3) = 10−10 –10−9 cm3/erg.
These values of the nonlinear parameters were significantly
larger than the ones observed, for example, traditional nonlin-
ear optical media such as quarts, CS2, CCl4, LiNbO3, etc. In
should be taken into account that the design of the organics
thin nanostructured films permits to apply these films in the
complicated optoelectronic schemes and in the spectral and
microscopy equipment used in the medicine.
26.3.2 Some Ideas to Apply Graphene-Based

FIGURE 26.1 Picture of the visualization of the diffraction Consideration to Explain the Decrease
response in the organic thin film doped with nanoobjects. (This of the Resistivity of the Intermediate
figure of diffraction orders has been received from Natalia V.
Layer in the Liquid Crystal Cells
Kamanina Lab for Photophysics of media with nanoobjects.)
Let us consider the possible application of the graphene-based
Based on the laser-induced change in the refractive index, structures in display technique to improve the quality of the
the nonlinear refraction, n2, and coefficients and cubic non- intermediate conducting layers and to increase the speed of
linearity, χ(3), have been calculated using the following equa- liquid crystal (LC) display element. As has been shown earlier
tions [21]: in the publications [39,40] that fullerenes and nanotubes can
be used to modify the interface between the solid substrate
∆ni and LC mesophase, in that case, higher transparency of LC
n2 = (26.4) display element owing to the absence of additional alignment
I
high conductive layers can be found. The surface electromag-
netic wave (SEW) was applied to form the orienting relief.
n2 n0c SEW source was a quasi-CW (continuous waveform) gap CO2
χ(3) = (26.5)
16π 2 laser generating p-polarized radiation with a wavelength of
10.6 μm and a power of 30 W. The skin layer thickness for
where Δni is the laser-induced change in the refractive index, this radiation was ~0.05 μm. Moreover, it has been demon-
I the intensity of the laser beam, n0 the linear refractive index strated later [41], the possibility of obtaining a homeotropic
of the system, and c is the light velocity. orientation of LC molecules using nanoobjects deposited on
It should be noticed that the thermal part of delta n in a substrate in an electric field with strength of ~100 V/cm
the materials studied is close to the value of 10−5. Thus, the and then treated by SEWs. The nanoobjects, namely, CNTs
increase in the diffraction efficiency in the current experi- were deposited by a contactless method of laser ablation. The
ments and hence in the light-induced refractive index change results of this investigation have been proposed to develop
could be explained by the high-frequency Kerr photorefrac- optical elements for displays with the vertical orientation of
tive effect (please see Equation 26.4 stimulated by intermo- nematic LC molecules as an alternative method for the clas-
lecular charge transfer processes in these compounds.) sical MVA (multidomain vertical alignment)-display technol-
One can observe that the data in Table 26.1 shows that the ogy, traditionally used by some display companies.
introduction of the graphene oxides nanoparticles reveals the It should be noticed that graphenes or graphene oxides
same nonlinear optical parameters as have been shown for materials can be used for these aims too. The relief activated
fullerene-based systems under the conditions of the same at the conducting indium-tin-oxide (ITO) layer via deposition
nanosensitizer concentration and the same spatial frequency. the grapheme oxides are presented in Figure 26.2. The area of
Adding the CNTs produces the change in the refractive prop- microscopic observation (see Figure 26.2a) is regarded to the
erties, as do fullerenes, at a much lower percentage content of dimension of 1.3 × 1.8 mm. The AFM (atomic force micros-
the CNTs as compared to that of C60 and C70 or of grapheme copy) image of the grapheme layer treated with SEW (see
oxides. Moreover, the large value of the laser-induced refrac- Figure 26.2b) is corresponded to the dimension correlated
tive index for CNTs-based materials has been found not only with a wavelength of λ = 10.6 μm. Indeed, these results show
at 90–100 mm−1 but also at the higher spatial frequencies, the situation when the homogeneity of the structure should be
namely at 150 mm−1. This implies that the possibility of vari- improved, but the relief regularity predicts the orientation of
ous charge transfer mechanisms in the systems with CNTs is the LC dipole with good advantage.
quite acceptable and may correspond to that predicted in the Moreover, attention should be drawn to the important prac-
publication [37,38]. tical fact. It has been found and statistically supported that
Photorefractive Properties of Graphene-Based Organic Systems 495
nM
nM 78
40,000 77
76
30,000
75
20,000
74
73
10,000
72
0
0 10,000 20,000 30,000 40,000 nM
FIGURE 26.2 Microscopy (a) and AFM (b) images of the graphene-based intermediate-conducting ITO contacts used in the LC cells.
the resistivity of the ITO conducting layer can be decreased Professor A.I. Plekhanov (Institute of Automation and
drastically using an additional deposition of the graphene- Electrometry SB RAS, Novosibirsk, Russia), Professor N.
based materials. The value of the resistivity can be changed M. Shmidt (Ioffe Physical-Technical Institute, St.-Petersburg,
from 300 to 400 Ω up to the values of 70–90 Ω. It permits to Russia), Professor E.F. Sheka (University of Peoples’
decrease twice the applied bias voltage to activate the modula- Friendship, Moscow, Russia), Professor L.A. Chernozatonskii
tion of the light beam through the LC elements. Thus, taken (Emanuel Institute of Biochemical Physics, Russian Academy
into account this consideration, the grapheme-based thin film of Sciences, Moscow, Russia), and foreign colleagues: Professor
elements can be predicted for the wide area of scientific and Francois Kajzar (Université d’Angers, Angers, France),
practical applications. Professor D.P. Uskokovic (Institute of Technical Sciences of
the Serbian Academy of Sciences and Arts, Belgrade, Serbia),
and Professor Iwan Kityk (Electrical Engineering Department,
26.4 CONCLUSION
Czestochowa University Technology, Czestochowa, Poland)
It should be noticed and confirmed that the nanoobjects for their helpful scientific discussions. The author would also
based on fullerenes, shungites, CNTs, carbon fibers, quantum like to acknowledge her Lab colleagues Dr. V.I. Studeonov, Dr.
dots, and graphenes, influence drastically the photorefractive P.Ya. Vasilyev, Dr. S.V. Serov, and PhD students N.A. Shurpo,
parameters of the polyimide materials. The increase of elec- S.V. Likhomanova, P.V. Kuzhakov, and A.A. Kukharchik
tron affinity energy and the specific surface area energy can (Vavilov State Optical Institute, St. Petersburg, Russia) for
change the electron pathways that predict the increase of the their participation and possible help in this study at its differ-
dipole moment and of the local volume polarizability, as well ent steps. The author would like to thank Dr. V.E. Vaganov
as of the charge carrier mobility. It has been considered that (Vladimir State University) and Dr. I.V. Mishakov (Boreskov
the distance between the donor part of the organic conjugated Institute of Catalysis, Siberian Branch of Russian Academy
molecule and nanoobjects and their dimensions influenced of Sciences, Novosibirsk) for presenting the MIG and 65BR
the migration of the charge carrier under the light irradiation. CNTs mixtures.
It is possible that these intermolecular complexes, as quasi- The basic results have been presented at different international
walls inside the organic structures can be applied to orient scientific conference, for example, International Symposium on
the human erythrocytes in order to testify their configuration Materials and Devices for Nonlinear Optics, ISOPL’5 (2009,
with good advantage under the condition when the charge at France); International Workshop on Nano and Bio-Photonics
the erythrocytes shell can be changed. The nonlinear optical (IWNBP), St Germain au Mont d’Or (2011, France), Laser
parameters, such as nonlinear refraction and cubic nonlin- Optics (2010, 2012, Saint Petersburg, Russia); YUCOMAT-2011,
earity, for nanoobjects-doped organic systems, indicate their 2012, 2013 (Serbia-Montenegro); and Seminar in Akademii
place between traditional nonlinear materials. The advanced Jana Dlugosza w Czestochowa (2012, Poland).
approach and information described in this chapter can be In this chapter, the data used in all figures and tables are
fruitful for the readers to find a sustainable solution in a fun- original or have been placed with prior permission from dif-
damental study and in the industry approach of the nanoob- ferent journals in case Dr.Sci., PhD Natalia V. Kamanina has
jects-modified organic materials. published them elsewhere in different forms.
The presented results are correlated with the work partially
supported by Russian Foundation for Basic Research, grants
ACKNOWLEDGMENTS
No.10-03-00916 (2010–2012) and No. 13-03-00044 (2013–
The author thank her Russian colleagues, Professor V.I. 2015) and the Project “BIOMOLEC” (via FP7 Program,
Berendyaev (Karpov Research Institute, Moscow, Russia), Marie Curie action, 2011–2014).
REFERENCES 17. Khoo, I.C., Williams, Y.Z., Lewis, B., Mallouk, T. 2006.
Photorefractive CdSe and gold nanowire-doped liquid
1. Yu, D., Park, K., Durstock, M., Dai, L. 2011. Fullerene- crystals and polymer-dispersed-liquid-crystal photonic
grafted graphene for efficient bulk heterojunction polymer crystals. Molecular Crystals and Liquid Crystals 446:
photovoltaic devices. Journal of Physical Chemistry Letters, 233–44.
2:1113–18. 18. Grishina, A.D., Licea-Jimenez, L., Pereshivko, L. Ya, Krivenko,
2. Basu, J., Basu, J.K., Bhattacharyya, T.K. 2010. The evolution T.V., Savel’ev, V.V., Rychwalski, R.W., Vannikov, A.V. 2006.
of graphene-based electronic devices. International Journal Infrared photorefractive composites based on polyvinylcarba-
of Smart and Nano Materials, 1(3):201–23. zole and carbon nanotubes. High Energy Chemistry, 40(5):341–7.
3. Saito, R., Hofmann, M., Dresselhaus, G., Joriod, A., Dresselhaus, 19. Kamanina, N.V., Emandi, A., Kajzar, F., Attias, A.-J. 2008.
M.S. 2011. Raman spectroscopy of graphene and carbon nano- Laser-induced change in the refractive index in the systems
tubes. Review. Advances in Physics, 60(3):413–550. based on nanostructured polyimide: Comparative study with
4. Late, D.J., Ghosha, A., Chakraborty, B., Sood, A.K., other photosensitive structures. Molecular Crystals and
Waghmare, U.V., Rao, C.N.R. 2011. Molecular charge- Liquid Crystals 486:1–11.
transfer interaction with single-layer grapheme. Journal of 20. Denisov, A. and de Bougrenet de la Tocnaye, J.-L. 2009.
Experimental Nanoscience, 6(6):641–51. Soluble fullerene derivative in liquid crystal: Polymer com-
5. Novoselov, K. S., Fal’ko, V.I., Colombo, L., Gellert, P.R., posites and their impact on photorefractive grating efficiency
Schwab, M.G., Kim, K. 2012. A roadmap for grapheme. and resolution. Applied Optics, 48(10):1926–31.
Review. Nature, 490:192–200. 21. Akhmanov, S.A. and Nikitin, S.Y. 1997. Physical Optics.
6. Li, Zh. Zheng L., Saini, V., Bourdo, S., Dervishi, E., Biris, Moscow University Press, Oxford, 657pp.
A.S. 2012. Solar cells with graphene and carbon nanotubes 22. Chemla, D.S. and Zyss, J. (Ed.) 1989. Nonlinear Optical
on silicon. Journal of Experimental Nanoscience, iFirst. 1–8. Properties of Organic Molecules and Crystals. Vol. 2,
http://dx.doi.org/10.1080/17458080.2011.572191. doi:10.1080/ Academic Press, Inc. Harcourt Brace Jovanovich, Publisher.
17458080.2011.572191. Translated in Russia, Mir Press, Moscow.
7. Choi, K.H., Ali A., Jo, J. 2013. Randomly oriented graphene 23. Kamanina, N.V. 2001. Light-induced variation in the refrac-
flakes film fabrication from graphite dispersed in N-methyl- tive index in a polyimide–fullerene system. Optics and
pyrrolidone by using electrohydrodynamic atomization tech- Spectroscopy, 90(6):867–71.
nique. Journal of Materials Science: Materials in Electronics, 24. Kamanina, N.V. and Uskokovic, D.P. 2008. Refractive index
24:4893–900. of organic systems doped with Nano-objects. Materials and
8. Liu, S., Wu, J., Zhou, Z., Gao, L., Luo, S., Xu, X., Wang, Z.M. Manufacturing Processes. 23:552–6.
2013. Influence of graphite oxide drying temperature on ultra- 25. Kamanina, N.V., Serov, S.V., Savinov, V.P., Uskoković, D.P.
fast microwave synthesis of grapheme. Journal of Materials 2010. Photorefractive and photoconductive features of the
Science: Materials in Electronics, 24:1298–302. nanostructured materials. International Journal of Modern
9. Umar, M.I.A., Yap, C.C., Awang, R., Jumali, M.H.H., Salleh, Physics B (IJMPB), 24(6–7):695– 702.
M.M., Yahaya, M. 2013. Characterization of multilayer gra- 26. Kamanina. N.V., Rozhkova, N.N., Chernozatonskii, L.A.,
phene prepared from short-time processed graphite oxide Shmidt, N.M., Ferritto, R., Kajzar, F. 2012. Influence of nano-
flake. Journal of Materials Science: Materials in Electronics, structuration process on the properties of materials. Nonlinear
24:1282–6. Optics and Quantum Optics. 45:153–60.
10. Kamanina, N.V. 2014. Features of Optical Materials Modified 27. Kamanina, N.V., Serov, S.V., Shurpo, N.A., Likhomanova,
with Effective Nanoobjects: Bulk Properties and Interface. S.V., Timonin, D.N., Kuzhakov, P.V., Rozhkova, N.N., Kityk,
Nova Science Publishers, Inc., New York, 116pp. I.V., Plucinski, K.J., Uskokovic, D.P. 2012. Polyimide–fuller-
11. Wang, Y., Herron, N., Casper, J. 1993. Bucky ball and quan- ene nanostructured materials for nonlinear optics and solar
tum dot doped polymers: A new class of optoelectronic mate- energy applications. Journal of Materials Science: Materials
rials. Materials Science and Enginering, B 19:61–6. in Electronics, 23(8):1538–42.
12. Itaya, A., Sizzuki, I., Tsuboi, Y., Miyasaaka, H. 1997. 28. Kamanina, N.V. and Vasilenko, N.A. 1995. High-speed SLM
Photoinduced electron transfer processes of C60-doped with a photosensitive polymer layer. Electronics Letters,
poly(N-vinylcarbazole) films as revealed by picosecond laser 31(5):394–5.
photolysis. Journal of Physical Chemisry B, 101(26):5118–23. 29. Kamanina, N.V. and Vasilenko, N.A. 1997. Influence of oper-
13. Brabec, C.J., Padinger, F., Sariciftci, N.S., Hummelen, J.C. ating conditions and of interface properties on dynamic char-
1999. Photovoltaic properties of conjugated polymer/metha- acteristics of liquid-crystal spatial light modulators. Optical
nofullerene composites embedded in a polystyrene matrix. and Quantum Electronics, 29(1):1–9.
Journal of Applied Physics, 85(91):6866–72. 30. Kamanina, N.V. 2003. Nonlinear optical study of fullerene-
14. Even, M., Heinrich, B., Guillon, D., Guldi, D.M., Prato, M., doped conjugated systems: New materials for nanophotonics
Deschenaux, R. 2001. A mixed fullerene—ferrocene thermo- applications. Proceedings of the NATO Advanced Research
tropic liquid crystal: Synthesis, liquid-crystalline properties, Workshop on Organic Nanophotonics, II/100, pp. 177–92.
supramolecular organization and photoinduced electron trans- 31. Kamanina, N.V., Plekhanov, A.I., Serov, S.V., Savinov, V.P.,
fer. Chemistry—A European Journal, 7(12):2595–604. Shalin, P.A., Kajzar, F. 2010. Correlation between photocon-
15. Kamanina, N.V., Sizov, V.N., Stasel’ko, D.I. 2001. Recording ductive and nonlinear optical characteristics of fullerene- and
of thin phase holograms in polymer-dispersed liquid-crystal nanotubes-doped organic composites. Nonlinear Optics and
composites based on fullerene-containing π-conjugated Quantum Optics 40:307–17.
organic systems. Optics and Spectroscopy, 90(1):1–3. 32. Kamanina, N.V., Serov, S.V., Shurpo, N.A., Rozhkova, N.N.
16. Kamanina, N.V. 2005. Fullerene-dispersed liquid crystal 2011. Photoinduced changes in refractive index of nano-
structure: Dynamic characteristics and self-organization pro- structured shungite containing polyimide systems. Technical
cesses. Physics-Uspekhi, 48(4):419–27. Physics Letters, 37(10):949–51.
Photorefractive Properties of Graphene-Based Organic Systems 497
33. Kamanina, N.V. and Sheka E.F. 2004. Optical limiters and 37. Kamanina, N.V., Serov, S.V., Savinov, V.P. 2010. Photorefractive
diffraction elements based on a COANP-fullerene system: properties of nanostructured organic materials doped with
Nonlinear optical properties and quantum-chemical simula- fullerenes and carbon nanotubes. Technical Physics Letters,
tion. Optics and Spectroscopy, 96(4):599–612. 36(1):40–2.
34. Kamanina, N.V. 2002. Mechanisms of optical limiting in 38. Kamanina, N.V. 2012. The effect of the charge transfer path-
π-conjugated organic system: Fullerene-doped polyimide. way during intermolecular complex formation on nonlinear
Synthetic Metals, 127(1–3):121–8. optical and photoconducting properties of nanocomposites.
35. Kamanina, N.V., Vasilyev, P.Y., Serov, S.V., Savinov, V.P., Technical Physics Letters, 38(2):114–17.
Bogdanov, K.Y., Uskokovic, D.P. 2010. Nanostructured mate- 39. Vasilyev, P.Y. and Kamanina, N.V. 2007. Fullerene-containing
rials for optoelectronic applications. Acta Physica Polonica A, liquid crystal spatiotemporal light modulators with surface-
117(5):786–90. electromagnetic-wave-treated conducting layers. Technical
36. Kamanina, N.V., Shurpo, N.A., Likhomanova, S.V., Timonin, Physics Letters, 33(1):8–10.
D.N., Serov, S.V., Barinov, O.V., Vasilyev, P.Ya. et al. 2011. 40. Vasilyev, P.Y. and Kamanina, N.V. 2007. Prospects for
Features of the nanostructured composites. Proceedings of using transparent conducting coatings containing fullerenes
the tenth Israeli-Russian Bi-National Workshop 2011 The and nanotubes in display elements of the new generation.
Optimization of the Composition, Structure and Properties Technical Physics Letter, 33(9):764–6.
of Metals, Oxides, Composites, Nano- and Amorphous 41. Kamanina, N.V. and Vasilyev, P.Y. 2009. On the possibility of a
Materials, Israel Academy of Science and Humanities and the homeotropic alignment in nematic liquid crystal elements using
Russian Academy of Science; June 20–23, 2011, pp. 77–85. carbon nanotubes. Technical Physics Letters, 35(6):501–3.
27 Applications of Graphene and Its
Derivatives in Electrochemical Sensors
and Comparison Study of Graphene
as a Modifier with Other Modifiers
Majid Arvand, Navid Ghodsi, and Tahereh M. Gholizadeh
CONTENTS
Abstract...................................................................................................................................................................................... 499
27.1 Introduction...................................................................................................................................................................... 499
27.1.1 GR Properties....................................................................................................................................................... 500
27.1.2 Preparation Methods............................................................................................................................................. 500
27.2 Chemically Modified GR Nanocomposites...................................................................................................................... 500
27.2.1 Preparation of CMG............................................................................................................................................. 501
27.3 Composite Materials Based on CMG............................................................................................................................... 501
27.3.1 Preparation Methods............................................................................................................................................. 501
27.3.2 Composites with Insulating Polymers.................................................................................................................. 501
27.3.3 Composites with Conducting Polymers................................................................................................................ 502
27.3.4 Composites with Other Carbon Nanomaterials.................................................................................................... 502
27.3.5 Composites with Small Organic Molecules......................................................................................................... 502
27.3.6 Composites with Metal Nanoparticles.................................................................................................................. 502
27.3.7 Composites with Nanoparticles of Metal Compounds......................................................................................... 502
27.4 Electroanalytical Applications of GR-Based Sensors...................................................................................................... 502
27.4.1 Unmodified GR for Electroanalysis..................................................................................................................... 502
27.4.2 Sensors Fabricated by Modifying GR.................................................................................................................. 504
27.5 Comparison Study of GR as a Modifier with Other Modifiers........................................................................................ 512
27.6 Conclusion........................................................................................................................................................................ 512
References.................................................................................................................................................................................. 512
ABSTRACT chapter, the effects of GR as an electrode modifier will be

compared with other modifiers.
Graphene (GR), a single-atom-thick nanostructured sheet
arranged in a two-dimensional honeycomb lattice, has
attracted tremendous research interest due to its distinctive
27.1 INTRODUCTION
properties, such as superior chemical and mechanical stabil-
ity, large specific surface area, and extraordinary electronic Given the prominence of graphene (GR) in today’s scientific
transport properties. Owing to these unique properties, GR literature, it can come as some surprise that this material
can greatly improve the performance of electrochemical was shown to exist only in 2004. That was when Kostya S.
responses and have attracted considerable attention in the field Novoselov and Andre K. Geim, the 2010 Nobel Laureates in
of electrochemistry and bioelectrochemistry. The application Physics, demonstrated that single layers of GR could be iso-
of GR as a kind of novel electrochemical sensing material is a lated from graphite and definitively identified by microscopy
most promising research hotspot. In this chapter , we examine [1]. It is the thinnest known material in the universe and the
the use of GR and GR-derived materials in electrochemical strongest ever measured till now [2].
sensors. In the first part, we will overview those aspects of GR is a new form of carbon, as shown in Figure 27.1.
GR’s properties most relevant to electrochemical applications While zero-dimensional carbons, such as fullerenes, act as
and then examine the latest advances in electrochemical sen- insulating materials [3], one-dimensional carbon structures,
sors made from GR derivatives and in the second part of the such as single-walled carbon nanotubes (SWCNTs), which are
499
600 is useful for the investigation of physical properties and for

b
500 selected applications (such as single-layer GR in field-effect
400 transistors [FETs]) [8], but for electrochemical applications,
300 bulk quantities of GR sheets are needed. (ii) Several methods
200 of producing bulk quantities of GR sheets and GR nanoplate-
100 a lets were developed based on chemical exfoliation of graph-
I/μA
0 ite. This involves intercalation of small molecules (typically

–100 sulfuric and nitric acid [9] or small organic molecules such as
–200 tetrabutyl ammonium hydroxide [10]) between the GR layers
–300
in graphite and consequent separation of the GR sheets (i.e.,
–400
by either thermal evaporation of the intercalated molecules
–500
–0.5 –0.3 –0.1 0.1 0.3 0.5 0.7
or sonication). It is remarkable that this method is based on
E/V vs. Ag/AgCl the 150-year-old work by Brodie [11]. (iii) Unzipping CNTs
either electrochemically, chemically, or by physical meth-
FIGURE 27.1 CVs of a mixture of 5.0 mM [Fe(CN)6]3−/4− and ods. (a) Electrochemical intercalation of Li+ and ammonia
0.1 M KCl on the surface of (a) GCE and (b) GR/GCE with a scan within the lattice of graphite [12,13] or multiwalled carbon
rate of 0.1 V s−1. nanotubes (MWCNTs) [14] increases the interlayer distance
from ca. 3.35 to 6.62 Å, and consequently delaminates graph-
basically wrapped-up GR sheets, possess excellent uniaxial ite or CNT to form a GR nanoribbon (GRN). (b) A physical
electron conductivity and fast heterogeneous electron trans- approach to the preparation of GRNs was demonstrated by
fer rates at their ends [4]. Two-dimensional GR sheets offer Dai and coworkers [15]. The CNTs were opened by plasma
a unique two-dimensional environment for electron transport etching of nanotubes partly embedded in a polymer film. The
and offer fast heterogeneous electron transfer at their edges resulting GNR exhibited smooth edges and a narrow width
[5]. The three-dimensional structure based on sp2-hybridized distribution (10–20 nm). (c) Tour and coworkers [16] recently
carbon is, of course, graphite. presented a solution-based oxidative process for producing
a nearly 100% yield of nanoribbon structures by lengthwise
cutting and unraveling of MWCNT side walls. The CNT wall
27.1.1 GR Properties
is subjected to permanganate oxidation to give the manganate
GR shows many advantages for electrochemical applications ester which can oxidize further to the dione in the dehydrating
when compared with graphite or carbon nanotubes (CNTs). medium. The next attack by permanganate eventually results
GR exhibits a surface area of 2630 m2 g−e, which is much in the fully cut nanotube. (d) Epitaxial growth of GR, which
greater than that of graphite (~10 m2 g−0) and even that of is useful for applications in nanoelectronics [17]. (e) Chemical
CNTs (1315 m2 g−1) [6]. The electrical conductivity of GR is vapor deposition (CVD) for the synthesis of GR nanoplatelets
excellent. When comparing GR to graphite, one should note and nanoribbons [18]. We emphasize that the CVD method
that GR sheets are flexible as opposite to brittle graphite, should be catalyst free to preclude similar problems that are
which is beneficial for use in flexible electronic and energy associated with CNTs relative to intercalated metallic impuri-
storage devices. Electrodes made from GR have significantly ties, as discussed above. Given the significant amount of inter-
more uniform distribution of electrochemically active sites est in GR, the number of methods of preparation is expected
than do those made from graphite. Their electrochemical to increase steadily. There is a problem associated with the
properties are similar to those of CNTs. However, GR has stability of individual GR sheets because they tend to restack
one huge advantage over CNTs: it is usually prepared from to form graphite nanoplatelets (<100 layers) or even thicker
graphite and does not contain any heterogeneous materials. structures (>100 layers). Solutions to this problem include
This is in contrast to the preparation of CNTs, which are chemical functionalization or the use of nanoparticles (metal-
typically grown from carbon-containing gas with the use of lic, C60, or CNTs) to prevent such restacking [19].
metallic catalytic nanoparticles. It is well documented that
such nanoparticles remain in the CNTs even after extensive 27.2 CHEMICALLY MODIFIED GR
purification procedures [7].
NANOCOMPOSITES
GR has been regarded as an important component for mak-
27.1.2 Preparation Methods
ing various functional composite electrode modifiers. GR can
The preparation of GR is a crucial issue and this area of be prepared through physical, chemical, and electrochemical
research is very active since none of the existing methods is approaches, as mentioned in the previous section. Among
suitable for all applications. The methods of preparation can them, the chemical methods were tested to be effective for
be classified into the following five categories: (i) Historically, producing chemically modified graphene (CMG) from vari-
the first method for the production of GR is the “cellophane ous precursors (such as graphite, CNTs, and polymers) in
method,” which is capable of separating a single GR sheet large scale and at low costs. Thus, CMG is more suitable for
from crystalline graphite [1]. Such dry mechanical exfoliation synthesizing high-performance GR-based composites. In this
Applications of Graphene and Its Derivatives 501
chapter, we review the recent developments in studies of the polar and apolar solvents. Therefore, special care should be
composites of CMG and small molecules, polymers, inor- taken in the process of inducing the second component to
ganic nanoparticles, or other carbon nanomaterials. avoid aggregation of CMG sheets. Usually, chemical modifi-
cation is an effective technique to increase the processability
of CMG as well as the compatibility and interfacial bonding
27.2.1 Preparation of CMG force in the composite. The main advantage of pregrapheni-
The most widely applied technique used for preparing CMG zation strategy is that the structure and properties of CMG
is chemical reduction of GR oxide. GR oxide is usually syn- can be fine-tuned. In postgraphenization strategy, a mixture
thesized by exfoliation of graphite oxide (GO) obtained by of CMG precursor (e.g., GR oxide) and the second compo-
oxidation of natural graphite powder with various oxidants nent is prepared in the first step, and then the precursor is
in acidic media. The typical method of synthesizing GO converted into CMG. The chemical conversion process can
was developed by Hummers and Offeman [20]. In this case, be carried out in solution or in solid state. In solution, the sec-
graphite is oxidized by treatment with KMnO4 and NaNO3 in ond component of the composite possibly acts as a stabilizer
concentrated H2SO4 to form GO, then GO is exfoliated into for improving the dispensability of the resulting composite.
GR oxide sheets in water by ultrasonication to form a stable GR oxide is a suitable precursor when a polar solvent is uti-
aqueous dispersion. The GR oxide product can be purified by lized and on the contrary, when an apolar solvent is applied,
centrifugation and dialysis to remove aggregates and various GR oxide modified with organic moieties is a better choice.
inorganic impurities such as metal ions and acids. However, in a solid matrix, the experimental conditions need
Recently, an improved method was reported by Marcano to be optimized for guaranteeing an efficient graphenization
et al. [21]. They used KMnO4 as the only oxidant and an acid and effective control on the structure of the final product.
mixture of concentrated H2SO4 and H3PO4 (9:1) as the acidic In syn-graphenization process, CMG and the second compo-
medium. This technique greatly increased the efficiency of nent are synthesized simultaneously [27]. This is a one-step
oxidizing graphite to GO and also prevented the formation of approach. A typical example is reducing GR oxide and metal
toxic gases, such as NO2 and N2O4. The GR oxide prepared by ions simultaneously in a blend solution, giving a CMG/metal
this method is more oxidized than that prepared by Hummers’ particle composite [28].
method and also possesses a more regular structure. Two methods have been widely applied for introducing
GR oxide can be chemically reduced to CMG. Various the second component to a CMG-based composite: mixing
inorganic and organic reducing agents such as hydrazine and in situ synthesis. Mixing includes solution mixing and
monohydrate [22], sodium borohydride [23], hydroquinone melt compounding, and the latter is usually used for prepar-
[24], and strong alkaline solutions [25] have been explored ing CMG/polymer composites. This method allows us to pre-
for the chemical reduction of GR oxide. Among the reducing cisely control the structure of the second component, because
agents described above , hydrazine monohydrate was the most the second component can be designed and synthesized with-
widely used, mainly due to its strong reduction activity and out the disturbance of CMG or its precursor. In situ synthesis
the stability in aqueous media. On reduction with hydrazine, involves preparing the second component in the system con-
most of the oxygen-containing functional groups of GR oxide taining CMG. Therefore, although in situ synthesis is a more
are eliminated and the π-electron conjugation within the aro- convenient procedure, the control on the structure and content
matic system of graphite is partially restored [26]. of the second component is more complicate.
27.3 COMPOSITE MATERIALS BASED ON CMG 27.3.2 Composites with Insulating Polymers

CMG/polymer composites have attracted a great deal of inter-
27.3.1 Preparation Methods
est because of their wide applications in high-strength and con-
A variety of methods have been developed for preparing ductive materials, catalysis, and fire retardants [29–31]. The
CMG-based composites, which can also be simply classified insulating polymers that have been incorporated into CMGs
into three groups according to the processes of synthesizing or GR oxide to form composites mainly are: poly(ethylene
CMG: (1) pregraphenization method: CMG was synthesized oxide) (PEO) [32], poly(vinyl alcohol) (PVA) [33], poly(diallyl
before the second component was induced; (2) postgrapheni- dimethylammoniumchloride) (PDDA) [34], poly(sodium
zation method: a composite composed of a CMG precursor 4-styrenesulfonate) (PSSNa) [35], poly(acrylic acid) (PAA)
(usually GR oxide) and the second component is prepre- [36], chitosan [37], Nafion [38], poly(N-isopropylacrylamide)
pared, followed by converting the precursor into CMG; and (PNIPAAm) [39], DNA [40], poly(vinyl acetate) (PVAc) [41],
(3) syn-graphenization method: CMG and the second com- polyfurfuryl alcohol (PFA) [42], poly(acrylamide) (PAM)
ponent are synthesized synchronistically in a one-pot system [43], polystyrene (PS) [44], poly(arylene disulfide) (PAS) [45],
and the two components are blended during the composite poly(methy methacrylate) (PMMA) [46], poly(acrylic ester)
forming process. In the pregraphenization process, CMG has (PAE) [47], poly(vinylidene fluoride) (PVDF) [48], polycar-
been presynthesized; thus, the incorporation of the second bonate (PC) [49], poly(allylamine) (PAAm) [50], polyure-
component is the key procedure for preparing the composite. thane (PU) [51], poly(lactic acid) (PLA) [52], polypropylene
However, pristine CMG usually has poor solubility in both (PP) [53], nylon [54], and graphitic carbon nitride (GCN) [55].
27.3.3 Composites with Conducting Polymers molecules induced by CMG. Composites of CMG and non-
aromatic molecules such as β-cyclodextrin (β-CD) can also
Conducting polymers (CP) are quite different from classical be prepared by simply mixing their solutions [69].
insulating polymers. They have conjugated backbones, which
provide them with unique electrical and optical properties.
These polymers are conductive in their doped states while 27.3.6 Composites with Metal Nanoparticles
insulating in their neutral states. Furthermore, they are usu-
CMG sheets are a useful substrate for immobilizing metal
ally brittle and weak in mechanical strength. On the contrary,
nanoparticles (MNPs) because of its high conductivity, large
CMG can be recognized as a two-dimensional conjugated
specific area, and excellent thermal stability. CMG usually
polymer according to its structure. It is mechanically strong,
can enhance the catalytic activities and reactivities of MNPs.
electrically conductive, and thermally stable. Therefore, the
Hydrophobic and static interactions are the main driving
incorporation of CMG into conducting polymers is attractive
forces of adsorbing MNPs on CMG surfaces. Most of this
for combining the properties of both components or improv-
kind of composites reported in the literature consist of noble
ing the properties of resulting composites based on synergis-
MNPs, including Au [70], Pt [71], Pd [72], and Ag [73]. In
tic effects. Since the conjugated polymer chains are very stiff,
addition, MNPs of Fe [74], Cu [75], and Co [76] were also
they are typically insoluble, intractable, and are decomposed
used for fabricating the composites with CMG.
before melting. Thus, CMG/CP composites were mostly
prepared by in situ polymerization. Up-to-date, polyaniline
(PANi) [56], poly(3-hexylthiophene) (P3HT) [57], poly(3,4- 27.3.7 Composites with Nanoparticles
ethylenedioxythiophene) (PEDOT) [58], and poly(phenylene of Metal Compounds
ethynylene) (PPE) [59] have been hybridized with CMG to
form composites. A variety of metal compounds have been used for synthesiz-
ing their composites with CMGs. They include oxides (e.g.,
Fe3O4 [77], Al2O3 [78], SnO2 [79], NiO [80], MnO2 [81], TiO2
27.3.4 Composites with Other [82], ZnO [83], and Co3O4 [84]), hydroxides (e.g., Co(OH)2
Carbon Nanomaterials [85] and Ni(OH)2 [86]) and salts (e.g., CdS [87], CdSe [88],
and LiFePO4 [89]).
Apart from GR, there are several types of carbon nanoma-
terials including fullerenes, CNTs, and so on [60]. Recently,
composites of CMG and other carbon nanomaterials such as 27.4 ELECTROANALYTICAL APPLICATIONS
CMG/CNTs [61] and CMG/fullerenes [62] have been reported. OF GR-BASED SENSORS
The applications of GR-modified electrodes can be subdivided
27.3.5 Composites with Small Organic Molecules into applications where the favorable electrochemical proper-
The first examples of these kinds of composites are GR ties are exploited to detect analytes in solution with no further
oxide intercalated by small organic molecules [63]. The pro- modification of the GR and where the GR is further modified
cedures used for fabricating these composites are similar to to give selectivity in the detection of analytes. Thus, this sec-
those applied for preparing CMG/polymer composites. Small tion of the chapter will outline examples of each subdivision.
organic molecules can be used as stabilizers for dispersing
GR oxide. Following a postgraphenization process, GR oxide
27.4.1 Unmodified GR for Electroanalysis
can be converted to CMG [64]. Besides, surfactants were used
to exfoliate graphite in water with the assistance of ultra- The GR-modified electrodes have been applied for the
sonication [65]. However, the exfoliated graphite flakes are detection of analytes at lower potentials than at some other
of different thickness; only ~3% of flakes consists of mono- electrode materials. Analytes that have been monitored at
layers. Exfoliation of natural or thermally expanded graphite GR-modified electrodes include caffeine [90], ascorbic acid
by perylene bisimide-based bolaamphiphile or coronene tet- (AA) [91], paracetamol (PC) [92], and H2O2 [93].
racarboxylic acid in water was also reported [66]. The com- Arvand et al. [23] developed a voltammetric sensor based on
posites of CMG and porphyrins or metalloporphyrins were GR-modified electrode for the determination of trace amounts
prepared by solution mixing [67]. They assembled positively of l-dopa in mouse brain extract and pharmaceuticals. They
charged 5,10,15,20-tetrakis (1-methyl-4-pyridinio)porphyrin synthesized GR from pure graphite powder via the modified
(TMPyP) molecules on CMG sheets in an aqueous medium Brodie method [94]. The prepared GR was characterized by
[68]. The driving forces for self-assembly were studied to be transmission electron microscopy (TEM), Fourier transform
π–π interaction as well as strong static attraction of negatively infrared spectroscopy (FTIR), and cyclic voltammetry (CV).
charged CMG and positively charged porphyrin. When CMG The electrochemistry characterization of GR was performed
was added into TMPyP solution, a large bathochromic shift by CV in a mixture of [Fe(CN)6]3−/4− and potassium chloride
in the absorption spectrum of TMPyP was observed, indicat- (KCl). As shown in Figure 27.1, a pair of well-defined redox
ing a strong interaction between these two components. This peaks of [Fe(CN)6]3−/4− was observed on bare glassy carbon
spectral change can be explained as the flattening of TMPyP electrode (GCE) (Figure 27.1a). The oxidation and reduction
peak currents were obtained to be 26.3 and 25.5 μA, respec- Zhu et al. [98] developed a GR-modified carbon fiber
tively. When the same solution was measured with GR/GCE microelectrode (GCFME) to determine dopamine (DA) in
under the same conditions (Figure 27.1b), it was found that mouse hippocampal tissue in the presence of ascorbic acid
the oxidation and reduction peaks were sharply increased to (AA) and uric acid (UA). Owing to the excellent electrochem-
311.7 and 314.2 μA, respectively, suggesting that the electron- ical properties of GR, they demonstrated an electrochemical
transfer kinetic of [Fe(CN)6]3−/4− was significantly accelerated sensor for in vitro measurement of DA in the mouse hippo-
by the modified film of GR. Additionally, the background cur- campal tissue. The electron transfer process for the oxidation
rents were also increased obviously, indicating the increase of DA, AA, and UA on the proposed electrode was favorable
of the capacity due to the enlargement of the specific surface because of the two-dimensional honeycomb lattices, maxi-
area of the GR/GCE. Voltammetric peak currents showed a mum surface area to volume, and high density of edge-plane-
linear response for l-dopa concentration in the range of 0.04– like defective sites of GR. In addition, the determination of
79 μM with a detection limit of 22 nM. DA using GCFME was not observed to be interfered in the
Arvand et al. [95] fabricated a GR-based electrochemical presence of UA and AA under physiological conditions.
sensor for sensitive detection of quercetin in foods, as well. Wang et al. [99] investigated the electrochemistry of dou-
They synthesized GR from graphite powder via Staudenmaier ble-stranded DNA (dsDNA) on a GR/GCE. They prepared
procedure [96] and characterized the prepared GR by TEM GR nanosheets according to a modified Hummers’ method.
and Fourier transform infrared (FT-IR). According to Figure The morphologic and electrochemical characterizations of
27.2, the electrochemical behavior of quercetin showed that GR/GCE were studied, as well. The results showed that GR
at bare GCE, quercetin was reversibly oxidized with a for- nanosheets had been successfully cast on the surface of GCE
mal potential at 0.35 V (Figure 27.2b). This redox peak has without aggregation. The electrochemistry of GR/GCE was
been known as the two electrons and two protons process of characterized by CV in a mixture of [Fe(CN)6]3–/4– and KCl.
the catechol (CC) moiety [97]. However, at GR/GCE (Figure A pair of well-defined redox peaks was obtained. The same
27.2c), a couple of redox peaks can be achieved with separa- solution was analyzed with GR/GCE and the results showed
tion (ΔEp) close to 0.083 V. Although no obvious difference that the oxidation and reduction peaks increased, suggest-
in peak potential of quercetin could be observed at GCE and ing that the electron-transfer kinetic of [Fe(CN)6]3–/4– was
GR/GCE, the redox peak current at GR/GCE (14.46 µA) was significantly accelerated by the modified electrode [100].
almost sixfold larger than at bare GCE (2.43 µA). Thus, GR Electrochemical studies of dsDNA showed that, compared
must have a great effect on the kinetics of electrode reaction with the bare GCE, the electrochemistry of dsDNA was pro-
for the quercetin and must provide a suitable environment moted by the excellent surface area effect and electron con-
for the quercetin to transfer electrons with underlying GCE. ductivity of GR.
The reason for the better performance of the GR/GCE is not Du et al. [101] developed a voltammetric sensor for simul-
only attributable to the increment of electrode surface area, taneous determination of hydroquinone (HQ) and CC at
but also is attributable to the more feasible electron transfer a GR/GCE. GR was synthesized through chemical oxida-
between quercetin and GR through π–π interactions. tion–reduction treatment of graphite. Characterization of GR,
using TEM image, indicated transparent and wrinkled GR
nanosheets. This wrinkled nature of GR helped maintain a
60
high surface area on the electrode [25]. The authors studied
50 c the electrochemical behaviors of HQ and CC at the bare and
40 GR-modified GCE in acetate buffer solution (pH 4.5) using
30 CV. According to the experimental data, at bare GCE, there
20
are a broad anodic peak at 406 mV and two cathodic peaks
at 186 and 73 mV, corresponding to the reduction of CC and
10 b
HQ, respectively. This shows that the oxidation of HQ and
I/μA
a
0 CC at the bare GCE is irreversible and undergoes a sluggish
–10 electron- transfer kinetic. However, at the GR/GCE, a pair
–20 of redox peak is observed at 207 and 181 mV with a ΔEp of
–30
26 mV. Moreover, a pair of redox peaks appeared at 319 and
286 mV with a ΔEp of 33 mV. These showed that the HQ and
–40
CC oxidation overpotential was significantly lowered and fast
–50 electron-transfer kinetics occurred at the GR/GCE.
–0.2 –0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
E/V vs. Ag/AgCl Kim et al. [102] investigated the electrochemical detection of
DA in the presence of AA using GR-modified electrode. They
FIGURE 27.2 Cyclic voltammograms of (a) GR/GCE in the
synthesized GO through a modified Hummers’ method [103].
absence of quercetin in 0.1 M citrate buffer and ethanol mixture The authors mentioned that GR could be used as advanced
(85:15 v/v), (b) bare GCE, and (c) GR/GCE in the presence of carbon-based electrode material for the selective determination
50 µM quercetin in 0.1 M citrate buffer and ethanol mixture (85:15 of DA and the electrochemical behavior of DA in an excess
v/v) (pH 4.0). amount of AA due to its unique electronic property [104].
TABLE 27.1
Using Unmodified GR as Modifier for Electrochemical Sensors
Sensitivity
Electrode Analyte Method of Synthesis Method pH Real Sample LOD (nM) (μA μM−1) References
GCE l-Dopa Modified Brodie DPV 6.2 Mouse brain extract and 22 0.043 [23]
method pharmaceuticals
GCE Quercetin Staudenmaier method DPV 4.0 Green apple, yellow apple, 3.6 0.333 [95]
white onion, yellow onion
Carbon fiber Dopamine Modified Brodie DPV 7.4 Mouse hippocampal tissue 10 – [98]
microelectrode method
GCE ds-DNA Modified Hummers’ DPV 8 – 0.35 (μg/mL) 0.0364 [99]
method
GCE Acetaminophen – DPV 3.0 Pharmaceutical 600 0.132 [101]
GCE Dopamine Modified Hummers’ DPV 7.4 – 2.64 μM 0.0659 [102]
method
GPE Chlorpromazine Hummers’ method DPV 4.0 Pharmaceutical 6.0 2.18 [105]
A GR paste electrode (GPE) for the electrochemical detec- CS has been used as protector and disperser to ensure GR to
tion of chlorpromazine has been demonstrated by Parvin be stably and uniformly dispersed in water. The GR nano-
[105]. GR was synthesized from natural flake graphite flakes were successfully immobilized on GCE to construct
through modified Hummers’ method [20,106]. In a compari- a GR-modified electrode as a biosensor to sense nitric oxide.
son between GPE and carbon paste electrode, the GPE showed Cytochrome c (Cytc) was adsorbed tightly on the surface
well-defined peaks with a small peak potential separation of the modified electrode and the direct electron transfer of
(ΔEp) of 60 mV, a substantial negative shift of the anodic peak Cytc was achieved. The adsorbed Cytc on the surface of CS–
potential, and a dramatic increase of current. These exhibited GR-modified electrode exhibited an enzyme-like catalytic
that the presence of GR, which enhanced the conductivity, activity for the reduction of NO.
surface area, electron transfer rate of the electrode surface, Kang et al. [110] studied direct electrochemistry of a glu-
and facilitated electron transfer between the biomolecules and cose oxidase (GOD)–GR–CS nanocomposite. GR-based
the electrode surface contributing to the high electrocatalytic polymer nanocomposites show an extraordinarily small elec-
activity of GPE. The results of the described literatures are trical percolation threshold because of the large conductiv-
summarized in Table 27.1. ity and aspect ratio of the GR nanosheets [111]. The authors
prepared a hybrid nanocomposite of GR–CS and then modi-
fied the surface of GCE. Afterward, GOD was absorbed on
27.4.2 Sensors Fabricated by Modifying GR
the nanocomposite film. According to the results, GR could
There have been a number of examples of immobilizing bio- provide a favorable microenvironment for the enzyme and
molecules onto GR-modified electrodes for the purpose of boosted direct electron transfer at the surface of the modified
fabricating electrochemical sensors. Su et al. [107] have dem- electrode.
onstrated a GR immunosensing platform based on a new elec- Zhang et al. [112] have demonstrated fabrication of GR
trochemical immunosensor designed for sensitive screening oxide (GO)/Prussian blue (PB) hybrid film by electropo-
of the carcinoembryonic antigen (CEA). Experimental results lymerizing PB onto the GO-modified GCE, and studied its
have shown that the electrochemical properties of the immu- electrochemical behaviors. They utilized the GO/PB hybrid
nosensing interface were improved and the sensitivity of the film-modified electrode for the sensitive detection of hydrogen
electrochemical immunosensor was enhanced, applying GR peroxide. Compared with the electrochemical performance
nanosheets. Moreover, conjugation of the three-dimensional of PB/GCE, the authors mentioned that GO/PB hybrid film
single-stranded DNA (ssDNA) network increased the surface exhibited much better electrochemical and electrocatalytic
coverage of the electrode, and amplified the immobilized activities toward H2O2, which was ascribed to the synergistic
amount of biomolecules. effects of GO and PB.
Xiong et al. [108] investigated the electrochemical behav- Yang et al. [113] investigated the effects of introducing gold
ior of AA and DA on GR oxide-modified electrodes contain- nanoparticles (AuNPs) on the electrochemical properties of
ing various contents of oxygen functional group (OxFG). GR functionalized with ionic liquid (IL). GR sheet decorated
They found that the determination of DA at the coexistence with metal NPs has been applied as an effective electrochemi-
of AA was realized just by adjusting the reduction conditions cal platform for electrochemical detection of biomolecules,
without another supplementary material. such as glucose, UA, hydrogen peroxide, AA, DA, and oth-
Wu et al. [109] prepared chitosan (CS)-dispersed GR nano- ers [114–117]. Yang et al. deposited AuNPs on the surface
flakes by a chemical route to reduce GR oxide, and dispersed of reduced GR oxide (RGO) functionalized with ILs by the
them fully in water to form a stable black aqueous solution. sonochemical synthesis procedure. The authors fabricated a
biosensor based on GOx/AuNPs/RGO-IL which was utilized which would benefit Hb to adsorb on negatively charged GR.
for the detection of glucose. The proposed electrode showed No voltammetric peak was observed at CS-DMF/GR/GCE,
enhanced electron transfer rate, reduced overpotential, and which demonstrated that the peaks of Hb–CS-DMF/GCE,
increased electrocatalytic surface area. Hb–CS/GR/GCE, and Hb–CS-DMF/GR/GCE were ascribed
Bao et al. [118] prepared GR oxide-templated polyaniline to the presence of Hb. A couple of stable, well-defined, and
(GO-PAN) microsheets via oxidation polymerization of ani- nearly reversible redox peaks at −0.257 and −0.399 V were
line with GO as the template, which was used for the modifi- observed on Hb–CS-DMF/GR/GCE. For Hb–CS-DMF/
cation of GCE to simultaneously discriminate AA, DA, and GCE, a couple of redox peaks of Hb were inconspicuous;
together with UA [20,103]. According to the results obtained, especially the peak of oxidation, and the background cur-
the DPVs of the mixture solution of AA, DA, and UA at bare rent was lower than that of the other three modified GCEs.
GCE, GO/GCE, graphene sheet (GS)/GCE, PAN/GCE, and According to these results, the existence of GR could greatly
GO-PAN/GCE have shown that the oxidation peaks of AA enhance the direct electron transfer between Hb molecules
and DA overlapped and a broad overlapped peak at 180 mV at and the underlying electrode. Hb–CS/GR/GCE also led to a
the bare GCE was observed. However, the DPV peaks were pair of redox peaks of Hb which was ascribed to the addition
well resolved at GO-PAN/GCE with the peak potentials at of DMF. The DMF played an important role in the direct elec-
about –55, 195, and 385 mV for AA, DA, and UA, respec- tron transfer of Hb in the Hb–CS film.
tively. Besides, the separations of the oxidation peak poten- Jiang et al. [124] prepared GR sheets by mild chemical
tials for AA–DA, DA–UA, and AA–UA were 250, 190, and reduction of GO, and PB was electrodeposited onto the GR
440 mV, respectively. The large separations of the peak poten- matrix by CV scanning. Owing to the remarkable synergistic
tials contributed to the capability of selective determination of effect of GR and PB, the PB–GR-modified GCE (PB–GR/
AA, DA, and UA in the presence of the other two species, or GCE) exhibited excellent electrocatalytic activity toward
simultaneous detection of them in their mixture. both the reduction of hydrogen peroxide and the oxidation of
Guo et al. [119] utilized porphyrin-functionalized GR for hydrazine. Owing to the synergistic effect of PB and GR, a
electrochemical detection of nitroaromatic explosives such twofold current response was observed on PB–GR/GCE. In
as 2,4-dinitrotoluene (2,4-DNT), 2,4,6-trinitrotoluene (2,4,6- addition, the unique zeolite structure of PB allowed molecules
TNT), 1,3-dinitrobenzene (1,3-DNB), and 1,3,5-trinitroben- (H2O2 and hydrazine) with low molecular weight and scale to
zene (1,3,5-TNB). The porphyrin/GR sensor showed high penetrate into the crystal and achieve good electrocatalysis.
sensitivity, good reproducibility, and selectivity, and could Wu et al. [125] investigated the direct electrochemistry of
detect nitroaromatic explosives with an ultratrace detection of GOD integrated with GR and its application to glucose detec-
1, 0.5, 1, and 2 ppb toward 2,4-DNT, 2,4,6-TNT, 1,3,5-TNB, tion. They fabricated the GR/GCE through casting of GR
and 1,3-DNB, respectively. The authors highlighted that the suspension (GR/water) onto the surface of the GCE with a
specifically good adsorptive properties of porphyrin, and the microsyringe. Afterward, the GOD was assembled on GR via
large electroactive surface area and fast charge transfer of GR physical adsorption through incubating the GR/GCE in the
contributed to the ultratrace detection capability. enzyme solution at 4°C. The voltammetric results indicated
Zhang et al. [120] synthesized a water-soluble sulfonated that GOD assembled on GR retained its native structure and
GR (sulfonated-G), for the immobilization and biosensing bioactivity. The proposed electrode exhibited a surface-con-
of horseradish peroxidase (HRP). According to the results, fined process and underwent effective direct electron transfer
the immobilized HRP showed an excellent electrocatalytic reaction with an apparent rate constant (ks) of 2.68 s−1.
activity toward the reduction of H2O2. In addition, the elec- Alwarappan et al. [126] reported enzyme-doped GR
trocatalytic activity of HRP/sulfonated-G/GCE toward NO2− nanosheets for enhanced glucose biosensing. Initially, they
was investigated. Adding NaNO2 to PBS, a new cathodic peak prepared graphitic oxide (GO) from graphite powders accord-
was observed at about –0.8 V, indicating the reduction of NO, ing to a method described by Hummers’ and Offeman, and
produced from the nitrite disproportionation reaction [121]. hydrazine hydrate was utilized to reduce GO [20]. Afterward,
However, no cathodic peak for NO at sulfonated-G/GCE the covalent modification of GR with GOD was carried out.
under the same condition was observed, demonstrating that The authors demonstrated that the presence of various reac-
the immobilized HRP retained high electrocatalytic activity tive functionalities such as ketonic, quinonic, and carboxylic
to nitrite. The authors highlighted that the excellent long-time functional groups on the edge plane of GR can easily bind
stability of the proposed electrode was because of the bio- with the free amine terminals of the GOD and contributed
compatible microenvironment of HRP molecules provided by to strong covalent amide linkage. The polypyrrole (Ppy) film
sulfonated-G. was directly cast onto the GCE surface by performing electro-
Liu et al. [122] developed an amperometric nitrite sen- chemical polymerization of pyrrole using the CV technique.
sor based on immobilization of hemoglobin (Hb) as a model Later, the conjugated GR–GOD was then immobilized onto
redox protein on a GR nanosheet-modified electrode by the Ppy-modified GCE surface and the entire assembly was
CS-dimethylformamide (CS-DMF) hydrogel. They utilized employed for glucose detection.
CS as the building component of the Hb–CS-DMF film Wei et al. [127] developed a label-free electrochemical
because of its good biocompatibility, good film-forming abil- immunosensor for the detection of alpha fetoprotein (AFP) by
ity, and positively charged property at physiological pH [123], immobilizing anti-AFP on GS and thionine (TH)-modified
GCE. They prepared GO powders from graphite by modi- and mercury with short time during the electrodeposition due
fication of Hummers’ method [128]. They prepared GS–TH to their large surface-to-volume ratio. As the deposition time
nanocomposite through mixing different concentrations of increased, the deposit was expected to be thick in a nonuni-
GS and TH solution together by stirring at room temperature. form way. Second, the GR nanosheets catalyzed the growth
The authors highlighted that the proposed method simplified process due to their good conductivity, because the deposition
the immunosensor preparation process, which can make it of was done under conditions of mass transfer control at poten-
wide potential application in clinical diagnosis. tials more negative than –0.95 V [132,133].
Feng et al. [129] studied direct electrochemistry of Song et al. [134] developed a highly efficient enzyme-based
hemo globin on the nanocomposite-based biosensor, in screen-printed electrode (SPE) obtained through covalent
which poly(diallyldimethylammonium chloride) (PDDA)- attachment between 1-pyrenebutanoic acid and succinimidyl
functionalized GR nanosheets (PDDA-G) were prepared ester (PASE) adsorbing on the GR oxide sheets and amines of
and gold nanoparticles (AuNPs) assembled on the surface of tyrosinase-protected gold nanoparticles (Tyr-Au). The authors
PDDA-G film through electrostatic interaction to construct utilized CC as the typical model of phenolic compounds to
the biocompatible interface. The AuNPs/PDDA-G-based sen- investigate the electrochemical sensing characteristics of
sor exhibited prominent electrocatalytic activity toward detec- the Tyr–Au/PASE–GO nanobiocomposite biosensor [135].
tion of H2O2. The Hb/AuNPs/PDDA-G-modified electrode According to the CVs of CC at the Tyr/SPE, Tyr/PASE–GO/
exhibited greatly promoted responses to H2O2 than bare GCE, SPE, and Tyr–Au/PASE–GO/SPE, the reduction peaks at both
PDDA-G, and AuNPs/PDDA-G-modified electrode, indicat- the electrodes were ascribed to the reduction of enzymatic
ing that the AuNPs/GR composite substrate could accelerate reaction product (o-quinone). A pair of reversible redox peaks
the electron transfer on the electrode surface due to its high was obtained at the Tyr/SPE. However, at Tyr/PASE–GO/SPE,
conductivity. According to the obtained results, the cyclic the oxidation peak decreased, the reduction peak increased,
voltammograms of the Hb/AuNPs/PDDA-G-modified elec- and the potential showed a negative shift due to the direct
trode promoted with the addition of hydrogen peroxide. The reduction of o-quinone liberated from the enzyme-catalyzed
Hb/AuNPs/PDDA-G-modified electrode exhibited greatly reaction on the electrode surface. Compared with the cathodic
enhanced responses to H2O2 than bare GCE, PDDA-G, and peak currents of CC at the Tyr/PASE–GO/SPE, the AuNPs
AuNPs/PDDA-G-modified electrode, indicating that the large in Tyr–Au/PASE–GO/SPE dramatically enhanced the electro-
catalytic reduction peak was due to the interaction between chemical response of CC, contributed to an increasing reduc-
immobilized Hb and H2O2. tion current.
Mao et al. [130] prepared a composite of GR sheets/Congo Arvand et al. [136] investigated the electrocatalytic reduc-
red-molecular-imprinted polymers (GSCR-MIPs) synthe- tion of oxadiargyl and its determination on 2-(4-((4-acetyl-
sized through free radical polymerization (FRP) and applied phenyl)diazenyl)phenylamino)-ethanol (ADPE)-modified
as a molecular recognition element to construct DA electro- GR-paste electrode (ADPE/MGRPE). The CVs of PBS (pH
chemical sensors. To prepare GSCR-MIPs, methacrylic acid 3.0) containing 0.1 mg L −1 oxadiargyl at the surface of bare
(MAA), ethylene glycol dimethacrylate (EGDMA), and azobi carbon paste and GR modified carbon paste electrodes are
sisobutyronitrile (AIBN), DA molecules as a template and shown in Figure 27.3a′ and b′ (Figure 27.3 inset). Figure 27.3a
GSCR as a substrate were utilized. For comparison, blank and b show the CVs of 0.1 M PBS in the absence and presence
nonimprinted polymers (GSCR-NIPs) were also prepared in of oxadiargyl at ADPE/MCPE. Figure 27.3c and d are like
the same condition, only without using the DA template mole- Figure 27.3a and b at the surface of ADPE/MGRPE, respec-
cules in the polymerization process. Linear sweep voltammet- tively. No electrochemical response was observed with the
ric studies indicated that the binding capacity of GSCR-MIPs bare carbon paste and GR-modified carbon paste electrodes
was better than MIPs, due to the higher oxidation current of (Figure 27.3 inset). There was also no considerable electroac-
GSCR-MIPs-modified electrode. Moreover, the GSCR-MIPs tivity on the ADPE/MCPE (Figure 27.3a and b). A compari-
had a specific binding capacity for DA molecule which con- son of Figure 27.3a and b shows that the cathodic peak current
tributed to the higher binding capacity of the GSCR-MIPs at the surface of ADPE/MCPE remains almost constant by
than the GSCR-NIPs. adding oxadiargyl to the solution. Comparing the reduction
Li et al. [131] presented a sensing platform for ultrasensitive of oxadiargyl at ADPE/MCPE (Figure 27.3b) and ADPE/
determination of Cd2+ based on the Nafion–GR nanocomposite MGRPE (Figure 27.3d), there was an enhancement of the
(Nafion-G) film modified electrode. The authors investigated cathodic peak current at ADPE/MGRPE relative to the value
the different voltammetric behaviors of the Nafion-G and obtained at ADPE/MCPE. The better performance of the
Nafion-modified electrodes with in situ plated mercury film ADPE/MGRPE may be due to the wrinkled nature of the GR,
electrode using differential pulse anodic stripping voltamme- the nanometer dimensions of the GR, the electronic structure,
try (DPASV). The sharper and higher peak current for Cd2+ and the topological defects present on the GR surfaces. The
was obtained at the Nafion-G modified electrode. Moreover, considerable surface wrinkles could improve the electroactive
the Cd2+ stripping peak appeared at a more negative potential. area of the ADPE/MGRPE.
The authors ascribed this signal enhancement to the following Zhou et al. [137] demonstrated a novel hydrogen peroxide
aspects. First, the interfusion of the GR nanosheets provided biosensor based on Au–GR–HRP–CS biocomposite [133].
a more effective active area for the nucleation of cadmium Cyclic voltammetric study indicated that the direct electron
8
3
peaks were obtained at GCE and CS/GCE and GR–CS/GCE
7 d in blank PBS, indicating that the GR–CS composite is non-
1.5
6 electroactive in the applied potential window. A pair of weak
I/μA 0
5 c redox peaks was observed in the presence of 4-AP at GCE
a′ b
4
–1.5 with peak-to-peak separation (ΔEp) of 193 mV, which was
b′ a
–3 ascribed to the electrochemical behavior of 4-AP. However,
3 350 50 –250–550–850 for CS/GCE, the redox peak currents decreased and the ΔEp
2 E/mV vs. Ag/AgCl increased (237 mV), indicating that the CS film on the GCE
I/μA
1 surface can hinder electron transfer. In comparison with the

0 bare GCE and CS/GCE, the redox peak currents obtained at
–1 GR–CS/GCE were much larger and the ΔEp was much lower
–2 (58 mV), which were attributed to the electrocatalytic activity
–3 of GR toward 4-AP.
–4 Bo et al. [140] developed a DNA biosensor based on oxi-
–5
dized GR and polyaniline nanowires (PANIws) modified
400 150 –100 –350 –600 –850 GCE. To study the electrochemical properties of oxidized GR
E/mV vs. Ag/AgCl and PANIw, the electrode was modified in the same way with
PANI instead of PANIw. PANIw-modified electrode exhib-
FIGURE 27.3 CVs of (a) ADPE/MCPE and (c) ADPE/MGRPE in ited a larger DPV current response compared with PANI. The
a 0.1 M PBS (pH 3.0) at a scan rate of 50 mV s−1 and (b) as (a) and (d) electrode modified with polymer nanowires had a large effec-
as (c) in the presence of 0.1 mg L −1 oxadiargyl. Inset: CVs of (a′) bare tive surface area, as well.
carbon paste electrode and (b′) GR-modified carbon paste electrode Zhou et al. [141] prepared GR-Fe3O4 (G-Fe3O4) composites
in the presence of 0.1 mg L −1 oxadiargyl at a scan rate of 50 mV s−1. with a solvothermal method for the immobilization of HRP
to construct a mediator-free H2O2 biosensor. Electrochemical
transfer of HRP was realized, and the biosensor had an excel- impedance spectroscopy (EIS) study of the biosensor for
lent performance in terms of electrocatalytic reduction toward investigating the electron transfer properties at the surface of
H2O2. According to the results, there was no redox peak cur- the electrode was carried out. The Nyquist plots of the imped-
rent of HRP on the HRP/CS/GCE. This was ascribed to the ance spectroscopy of the bare and modified indium tin oxide
deep embedding of the redox center of HRP in a protein shell, (ITO) electrodes were obtained in the frequency range of 10−1–
which made the distance between the active site of enzyme 104 Hz. In comparison with bare ITO electrode, the diameter
and electrode farther than the distance that electrons could of the semicircle for chitosan (CH) deposited ITO electrode
transfer at a sufficient rate. The insulation of the biopoly- was much larger, indicating that the CH film on the surface of
mer CS further interrupted the electron transfer. However, ITO hindered the electron transfer between the redox probe of
both Au/HRP/CS/GCE and GS/HRP/CS/GCE exhibited a [Fe(CN)6]3–/4– and the electrode surface. However, the diam-
significant cathodic peak current around −0.3 V, indicat- eter decreased significantly in the presence of G-Fe3O4, which
ing that direct electron transfer was realized. In comparison was attributed to the introduction of G-Fe3O4 composites/
with other electrodes, Au/GS/HRP/CS/GCE showed the larg- CH. The G-Fe3O4 composites acted as nanoscale electrodes
est cathodic peak current. In an enzyme molecule, the rate and promoted the electron transfer process. According to the
coefficient of electron transfer between the donator and the results, the Rct increased again after entrapping of HRP, which
acceptor decreased exponentially with the distance between might be ascribed to the blocking of the electron transfer by
the electron transfer centers [138]. The distance between the the macromolecular structure of HRP. The promoted Rct also
active site and the microelectrodes was much smaller than demonstrates the successful immobilization of HRP on the
the original distance between the enzyme and the electrode electrode.
surface after the introduction of GR and Au nanoparticles. Sun et al. [142] demonstrated TiO2-GR (TiO2-Gr) nano-
The nanoscale individual sheets and particles both acted as composite prepared by hydrolysis of titanium isopropoxide
“molecular wires” to connect the active sites with the elec- in colloidal suspension of GR oxide and in situ hydrother-
trode, and increased the electron transfer rate significantly. mal treatment. They investigated the direct electrochemis-
Besides, the high conductivity of GR was also responsible for try and electrocatalysis of hemoglobin at room temperature
the increased current. According to the high quality of crystal IL 1-butyl-3-methylimidazolium hexafluorophosphate, chi-
lattice, electrons could travel without being scattered by lat- tosan (Chit) and TiO2-Gr nanocomposite modified GCE. To
tice imperfections and foreign atoms in GR. These reasons monitor the assembly process, EIS was used. The imped-
made electron transfer in the composite possible. ance spectra represented as Nyquist plots (Zim vs. Z re) has
Yin et al. [139] studied the electrochemical behavior and been recorded for bare GCE (a), Chit/GCE (b), Chit-[bmim]
voltammetric determination of 4-aminophenol (4-AP) on a PF6-TiO2-Gr/GCE (c), and Hb/Chit-[bmim]PF6-TiO2-Gr/
GR–CS composite film-modified GCE. The electrochemical GCE (d), respectively. According to the obtained results, a
behavior of 4-AP was investigated on GCE, CS/GCE, and straight line was observed at bare GCE, which implied low
GR–CS/GCE. According to the obtained results, no redox resistance. However, the diameter of the semicircle increased
significantly at Chit/GCE, which was attributed to the weak Liu et al. [148] investigated simultaneous determination of
conductivity of the Chit film. The diameter of the semicircle hydroquinone and CC based on GR/BMIMPF6 nanocompos-
decreased at Chit-[bmim]PF6-TiO2-Gr/GCE, suggesting that ite-modified electrode. As can be concluded from the results,
the Chit-[bmim]PF6-TiO2-Gr nanocomposite was beneficial to at the bare GCE, two abroad oxidation peaks were observed
the electrons. Moreover, the semicircular diameter increased at 0.366 and 0.238 V, respectively, which were ascribed to
again at Hb/Chit-[bmim]PF6-TiO2-Gr/GCE, which was the oxidation peak of CC and hydroquinone. In compari-
ascribed to the further increase of the thickness of the modi- son with bare GCE, the oxidation peak currents remarkably
fied film and the nonconductive properties of Hb, suggesting increased at GS/GCE. The peak-to-peak separation for CC
that the Chit-[bmim]PF6-TiO2-Gr nanocomposite film was a and hydroquinone was 0.119 V, which indicated that CC
suitable biocompatible microenvironment for Hb. and hydroquinone could be distinguished effectively at GS/
Fu et al. [143] investigated the direct electrochemistry of BMIMPF6/GCE. Moreover, two well-defined cathodic peaks
GOD immobilized in a modified electrode based on a compos- were observed at 0.280 and 0.172 V, which were attributed to
ite film of exfoliated graphite nanosheets (GNSs) and Nafion. the reduction of the oxidation products of CC and hydroqui-
According to the results, both the GNSs and Nafion are elec- none, respectively. In comparison with the GS/GCE, the oxi-
troinactive in this potential window because no obvious redox dation and reduction peak currents further increased at GS/
peaks were observed for the GNSs–Nafion/GCE. However, BMIMPF6/GCE.
at GNSs–Nafion–GOD/GCE, a pair of well-defined redox Fan et al. [149] developed an efficient and facile approach
peaks at –0.437 and –0.409 V was observed. These peaks to yield nanocomposites with TiO2 nanoparticles uniformly
were attributed to direct electron transfer at the immobilized embedded on the GR substrate for electrochemical sensing of
GOD for the conversion of GOD(–FAD) to GOD(FADH2). adenine (A) and guanine (G). The DPV peak potential of G at
The Nafion–GOD/GCE also exhibited a pair of redox peaks the bare GCE and GR/GCE was observed at 0.99 and 0.92 V,
of GOD but the peak currents were significantly smaller than respectively. However, at TiO2-Gr/GCE, the peak potential
those of the GNSs–Nafion–GOD/GCE. The authors high- was negatively shifted to 0.88 V, and the peak current greatly
lighted that the improved direct electrochemical behavior of enhanced. The oxidation peak potential on TiO2-Gr/GCE was
the latter electrode can be ascribed to the presence of abun- also negatively shifted to 1.19 V for A, and the peak current
dant edges sites on GNSs [144]. promoted, which indicated that the proposed nanocomposite
Li et al. [145] studied the electrochemical detection of showed an electrocatalytic activity toward the oxidation of
hydroquinone by GR and Pt–GR hybrid material synthesized G and A.
through a microwave-assisted chemical reduction process. Zeng et al. [150] developed covalently functionalized GR
The electrochemical behaviors of the GR/GCE and Pt–GR/ with CS to improve its biocompatibility and hydrophilicity for
GCE toward hydroquinone were studied by differential pulse the preparation of a glucose biosensor [151]. They recorded
voltammetry (DPV). The weak and broad oxidation peak cyclic voltammograms of 0.5 mM H2O2 at bare GCE, CS–
of hydroquinone on GCE was observed at 0.09 V. In com- GR-modified electrode, and Pd NPs/CS–GR-modified elec-
parison with GCE, the GR/GCE displayed a well-defined trode in 0.05 mM PBS at a scan rate of 100 mV s−1. The Pd
peak at 0.002 V with a much higher current intensity. This NPs/CS–GR-modified electrode exhibited stronger oxidation
phenomenon could be due to the unique properties of GR
that enhanced the active surface area of the electrode and 80
accelerated the electron transfer via improved conductivity
and the good affinity of GR to hydroquinone. The current 60
density and electron transfer promoted in the Pt–GR hybrid
system, which was ascribed to the charge hopping through 40
the metallic Pt nanoparticles and the effective charge migra-
tion through the GR. 20
I/μA
Baby et al. [146] utilized metal decorated GR nanosheets as

an immobilization matrix for amperometric glucose biosen- 0
sors [147]. The cyclic voltammetric response of Nafion–GOD/
f-G, Nafion–GOD/f-G-(Pt–Au), and Nafion–GOD/f-G-Au –20
GC electrodes, respectively, toward the enzymatically liber-
–40
ated H2O2 has been recorded. The carboxylic acid functional
groups introduced after treatment with the acid solution at the –60
edges and surface of GR, allowed them to adsorb the enzyme 0 0.15 0.3 0.45 0.6 0.75 0.9 1.05
of GOD. The link between GR and GOD was more close and E/V vs. Ag/AgCl
stable because this adsorption was fulfilled through a covalent
bond. Owing to the large surface area, GR could enhance a FIGURE 27.4 CVs of Tyr and PC mixture at bare (dotted line), GR
large amount of GOD to be immobilized within the sheets. (dashed line), and MWCNTs-GNS/GCE (solid line). Concentration
Both f-G-Au and f-G-(Pt–Au) Nafion composite films showed of each analyte is 2.0 × 10−5 M; Eacc = 0.0 V; tacc = 180 s; and scan
more pronounced currents than the f-G. rate: 0.1 V s−1.
TABLE 27.2
Sensors Fabricated by Modifying GR
Sensitivity
Electrode Modifier Analyte Method of Synthesis Method pH Real Sample LOD (μM) (μA μM−1) References
Gold Gold nanoflowers–GR-single-stranded Carcinoembryoninc Modified Hummers’ DPV 5.5 Clinical serum 0.01 (ng mL−1) 1.44 (μA mL ng−1) [107]
DNA molecule antigen method
GCE GR–oxygen as functional group Adenine and guanine Modified Hummers’ Amperometry 6.86 – 1.8, 4.79 – [108]
method and DPV
GCE GR–chitosan Horse heart Hummers’ and Offeman Amperometry 7 – – – [109]
cytochrome C method
GCE GOD–GR–chitosan film Glucose Staudenmaier method CV 7.4 – 20 37.93 [110]
Applications of Graphene and Its Derivatives
GCE Gr oxide/prussian blue hybrid film H2O2 Modified Hummers’ Amperometry 6 Human serum 0.122 408.7 (μA mM−1 [112]
method cm−2)
GCE Gold nanoparticle-reduced GR oxide–IL Glucose Modified Hummers’ Amperometry 7.4 – 0.13 (mM) 0.16 [113]
method
GCE Porphyrin-functionalized GR 2,4-DNT, 2,4,6-TNT, – – – – 1, 0.5, 1, 2 – [119]
1,3,5-TNB, and (ppb)
1,3-DNB
GCE Water-soluble sulfonated GR H2O2 Hummers’ method CV 5.7 – 1.17 0.039 (μA mM−1) [120]
GCE Immobilized hemoglobin on GR nanosheet Nitrite Hummers’ and Offeman Amperometry 7 – 0.18 5.33 (µA mM−1) [122]
GCE Prussian blue–GR Hydrogen peroxide Modified Hummers’ and Amperometry 6 – 3, 7 0.53 [124]
and hydrazine Offeman method
GCE GOD integrated with GR Glucose Modified Hummers’ CV 6.9 Human blood 10 110 [125]
method serum
GCE Polypyrrole–GR–GOD Glucose Hummers’ and Offeman CV 7.4 – 3 – [126]
GCE Anti-AFP immobilized on GS and thionine AFP Modified Hummers’ Amperometry 7.4 – 5.77 (pg mL−1) – [127]
method
GCE Nanocomposite film of gold nanopaticles H2O2 – Amperometry 7.4 – 0.39 369.2 [129]
and poly(diallyldimethylammonium
chloride) functionalized GR
GCE GSs/Congo red-molecular-imprinted Dopamine Hummers’ and Offeman Amperometry 7.4 – 0.1 – [130]
polymers method
(Continued)
509
510

Sensors Fabricated by Modifying GR
Sensitivity
Electrode Modifier Analyte Method of Synthesis Method pH Real Sample LOD (μM) (μA μM−1) References
Mercury Nafion–GR nanocomposite Cd2+ Staudenmaier method DPASV – – 500 (μg L−1) – [131]
film
Screen 1-PASE adsorbing on the GR oxide sheets CC Modified Hummers’ Amperometry 7 – 0.024 0.16 [134]
printed and amines of Tyr-Aus method
GR paste 2-(4-((4-acetylphenyl) diazenyl) Oxadiargyl Modified Brodie method DPV 3 – 1.4 (mg L−1) 1.55 (μA mg−1 L) [136]
phenylamino)ethanol
GCE Au nanoparticle–GR–horseradish H2O2 Modified Hummers’ Amperometry 6.5 – 1.7 – [137]
peroxidase–chitosan method
GCE GR–chitosan composite film 4-Aminophenol – DPV 6.3 Paracetamol 0.057 0.033 [139]
tablets
GCE GR oxide–polyaniline nanowire DNA Brodie method DPV 7.3 – 0.325 (pM) – [140]
ITO GR–Fe3O4 nanoparticle–horseradish H2O2 Modified Hummers’ Amperometry 6.5 – 0.6 132 [141]
peroxide method
GCE GR–TiO2 H2O2 – Amperometry 7 – 0.3 0.0707 [142]
nanoparticle–IL–chitosan–hemoglobin
GCE Nafion–GR–GOD Glucose – CV 7 – – 3.4 [143]
GCE Platinum nanoparticle–GR Hydroquinone Hummers’ method DPV 7.4 – 6 3.56 [145]
GCE Pt–Au-functionalized GR–GOD Glucose Staudenmaier method Amperometry 7 – 1.0 – [146]
GCE GR–IL Hydroquinone and CC Modified Hummers’ DPV 5 Local tap water 0.01, 0.02 – [148]
method
GCE GR–TiO2 nanoparticle Adenine and guanine Modified Hummers’ DPV 4.5 Thermally 0.1, 0.15 – [149]
method denatured DNA
GCE GOD–palladium nanoparticle/chitosan- Glucose Modified Hummers’ Amperometry 7.5 – 2.0 31.2 [150]
grafted GR method
GCE CNT–GR Tyrosine and Modified Brodie method DPV 7.5 Human blood 0.19, 0.1 0.0163, 0.0908 [151]
paracetamol serum and
pharmaceuticals
Graphene Science Handbook
TABLE 27.3
Summary of Comparison Study of Graphene as a Modifier with Other Modifiers
Electroanalytical LOD (S/N = 3) Sensitivity
Analyte Electrode Material Technique pH (nM) LR (μM) (μA μM−1) References
L-Dopa GR DPV 6 22 0.04–79 0.043 [23]
Nickel hexacyanoferrate film-modified gold nanoparticle graphite composite electrode DPV 7 530 0.82–2500 0.363 [154]
Applications of Graphene and Its Derivatives
Cobalt hexacyanoferrate/large-mesopore carbon DPV 7.4 17 100–1.9 × 106 0.063 [155]

Polyethylene glycol (PEG)-assisted tungsten oxide nanoparticles Amperometry 7.2 120 0.1–1 – [156]
Nickel hydroxide nanoparticles/MWCNTs composite DPV 7 210 1–672 0.271 [157]
Paracetamol GR SWV 9.3 32 0.1–20 4.055 [92]
Microcrystalline natural graphite–polystyrene composite film-modified electrode DPV 4 34 0.02–100 – [158]
Nafion/TiO2–GR-modified GCE DPV 7 210 1–100 0.1694 [159]
CNTs-modified SPE FIA 2 100 2.5–1000 0.03021 [160]
GCE modified with composite film of poly (4-vinylpyridine) (P4VP) and MWCNTs DPV 7 1.69 0.02–450 1.778 [161]
NADH GPE Amperometry 7.4 300 – 20 [162]
CNSs grown by microwave plasma-enhanced CVD LSV 7 440 0–500 85.8 mA M−1 [163]
Fe3O4 nanoparticles/ MWCNTs composite DPV 7.5 500 50–500 7.67 µA mM−1 [164]
A sheet shape Co3O4 nanooxide Amperometry 6 4250 10–100 1.43 nA μM−1 [165]
Pencil graphite electrode-modified with quercetin Amperometry 7 150 0.5–100 0.034 [166]
UA GR-modified carbon fiber electrode CV 7 132 0.194–49.68 0.246 [167]
Nanoelectrode ensembles formed via PAN-co-PAA self-assembly DPV 6.86 100 0.5–50 0.00527 [168]
Thionine-modified GR oxide Amperometry 7.4 7000 20–4500 – [169]
SPE-modified with Prussian blue Chronoamperometry 8.5 10,000 30–300 – [170]
Note: Differential pulse voltammetry (DPV), cyclic voltammetry (CV), linear sweeping voltammetry (LSV), flow injection analysis (FIA), square wave voltammetry (SWV), limit of detection (LOD), linear
range (LR).
511
current of H2O2 with increasing potential applied on the to these principally advantageous electronic and electrochemi-
electrode. cal properties, GR-based materials have been used to design
Arvand et al. [61] developed a CNT-GR nanosheet nano- and prepare GR-oriented electrodes for a wide range of applica-
composite-modified GCE (MWCNTs-GNS/GCE) for simul- tions in electrochemistry, ranging from electrochemical sens-
taneous voltammetric determination of tyrosine (Tyr) and PC ing, electrocatalysis, and FET devices. Considerable advances
in human blood serum and pharmaceuticals. CNTs and GR in this area have already been made. Nonetheless, GR-related
have their own drawbacks. CNTs have superior mechanical research in electrochemistry still remains an intriguing issue
properties but must be dispersed uniformly and form a net- and needs to be explored further.
work to achieve sufficient percolation for electrical conduc-
tivity. On the contrary, GR has excess of 15,000 cm2 V−1 s−1
electron mobility at room temperature but causes problem of REFERENCES
its restacking property due to the π–π interaction between 1. K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang,
individual GR sheets, and therefore shows lower conductivity S.V. Dubonos, I.V. Grigoriev, A.A. Firsov, Science, 306, 2004,
and specific surface area [152]. According to Figure 27.4, the 666.
peak currents of Tyr and PC at MWCNTs-GNS/GCE (solid 2. A. Geim, Science, 324, 2009, 1530.
line) are remarkably enhanced compared with the bare and 3. L. Xiao, G.G. Wildgoose, R.G. Compton, Sens. Actuators B,
138, 2009, 524.
GR-modified electrodes with oxidation potential of almost
4. M. Pumera, Chem. Eur. J., 15, 2009, 4970.
0.64 and 0.37 V, respectively. Although, these negative 5. F. Chen, N. Tao, J. Acc. Chem. Res., 42, 2009, 429.
shifts were almost the same at the surface of GR-modified 6. M. Pumera, B. Šmíd, K.J. Veltruská, J. Nanosci. Nanotechnol.,
electrode, the higher enhanced peak currents indicated an 9, 2009, 2671.
enhanced electron transfer rate for Tyr and PC at MWCNTs- 7. M. Pumera, Langmuir, 23, 2007, 6453.
GNS/GCE. The porous network structure induced by the 8. X. Li, X. Wang, L. Zhang, S. Lee, H. Dai, Science, 319, 2008,
MWCNTs in MWCNTs-GNS nanocomposite acted as a path- 1229.
way for mass transfer of the chemical reactants and products 9. A. Yu, P. Ramesh, M.E. Itkis, E. Bekyarova, R.C. Haddon, J.
Phys. Chem. C, 111, 2007, 7565.
and as an electrical bridge [153]. These results indicated that 10. X. Li, G. Zhang, X. Bai, X. Sun, X. Wang, E. Wang, H. Dai,
cocontribution of MWCNTs and GR had great improvement Nat. Nanotechnol., 3, 2008, 538.
to the increase of the electrochemical response, which was 11. B.C. Brodie, Phil. Trans., 149, 1859, 249.
due to the excellent characteristics of nanocomposites such 12. M. Endo, C. Kim, K. Nishimura, T. Fujino, K. Miyashita,
as good electrical conductivity, high surface area, high poros- Carbon, 38, 2000, 183.
ity, and more electroactive interaction sites that could provide 13. T. Enoki, M. Suzuki, M. Endo, Graphite Intercalation
increased mass transport and easier accessibility to the active Compounds and Applications, Oxford University Press: New
York, 2003, pp. 26–31.
sites. Literatures on the application of modified GR are sum- 14. A.G.C. Marquez, F.J.R. Macıas, J.C. Delgado, C.G.E.
marized in Table 27.2. Gonzalez, F.T. Lopez, D.R. Gonzalez, D.A. Cullen, D.J.
Smith, M. Terrones, Y.I.V. Cantu, Nano Lett., 9, 2009, 1527.
27.5 COMPARISON STUDY OF GR AS A 15. L. Jiao, L. Zhang, X. Wang, G. Diankov, H. Dai, Nature, 458,
2009, 877.
MODIFIER WITH OTHER MODIFIERS 16. D.V. Kosynkin, A.L. Higginbotham, A. Sinitskii, J.R. Lomeda,
A. Dimiev, K. Price, J.M. Tour, Nature, 458, 2009, 872.
In this chapter, the effects of GR as a modifier on the elec-
17. C. Berger, Z. Song, T. Li, X. Li, A.Y. Ogbazghi, R. Feng, Z.
trochemical experiment results of some analytes are sum- Dai et al., J. Phys. Chem. B, 108, 2004, 19912.
marized in Table 27.3. As can be seen, GR can improve the 18. J.C. Delgado, J.M.R. Herrera, X. Jia, D.A. Cullen, H.
electrochemical responses due to its unique properties. Muramatsu, Y.A. Kim, T. Hayashi et al., Nano Lett., 8, 2008,
277.
19. M. Pumera, Chem. Rec., 9, 2009, 211.
27.6 CONCLUSION 20. W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc., 80, 1958,
GR, a single-layer two-dimensional carbon structure, has 1339.
21. D.C. Marcano, D.V. Kosynkin, J.M. Berlin, A. Sinitskii, Z.
attracted an explosive interest in the recent years. Among all
Sun, A. Slesarev, L.B. Alemany, W. Lu, J.M. Tour, ACS Nano,
the carbon-based materials, GR offers the advantages of unique 4, 2010, 4806.
physical, chemical, and mechanical properties, which may 22. X. Liang, Y. Wang, H. Zheng, Z. Wu, J. Electron. Spectrosc.
open up a new research area for materials science and scope for Relat. Phenom., 196, 2014, 89.
wide-ranging and diversified technological applications. This 23. M. Arvand, N. Ghodsi, J. Solid State Electrochem., 17, 2013,
chapter has addressed advances in the field of GR from the 775.
standpoint of electrochemistry. As known, extraordinary elec- 24. G. Wang, J. Yang, J. Park, X. Gou, B. Wang, H. Liu, J. Yao, J.
Phys. Chem. C, 112, 2008, 8192.
tronic properties in GR are really due to the high quality of its
25. X. Fan, W. Peng, Y. Li, X. Li, S. Wang, G. Zhang, F. Zhang,
electronic structure.GR, as a Dirac fermion system, promises Adv. Mater., 20, 2008, 4490.
intriguing electronic properties such as a high-integer quantum 26. S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas,
Hall effect, the Klein paradox, and an ambipolar electric field A. Kleinhammes, Y. Jia, Y. Wu, S.T. Nguyen, R.S. Ruoff,
effect, along with ballistic conduction of charge carriers. Owing Carbon, 45, 2007, 1558.
27. H. Bai, C. Li, G. Shi, Adv. Mater., 23, 2011, 1089. 63. S.H. Ryu, J.H. Sin, A.M. Shanmugharaj, Eur. Polym. J., 52,
28. C. Xu, X. Wang, J. Zhu, J. Phys. Chem. C, 112, 2008, 19841. 2014, 88.
29. D. Cai, M. Song, J. Mater. Chem., 20, 2010, 7906. 64. J. Geng, H.T. Jung, J. Phys. Chem. C, 114, 2010, 8227.
30. H. Kim, A.A. Abdala, C.W. Macosko, Macromolecules, 43, 65. M. Lotya, Y. Hernandez, P.J. King, R.J. Smith, V. Nicolosi,
2010, 6515. L.S. Karlsson, F.M. Blighe et al., J. Am. Chem. Soc., 131,
31. T. Kuilla, S. Bhadra, D. Yao, N.H. Kim, S. Bose, J.H. Lee, 2009, 3611.
Prog. Polym. Sci., 35, 2010, 1350. 66. A. Ghosh, K. Rao, S. George, C.N.R. Rao, Chem. Eur. J., 16,
32. Y.C. Cao, C. Xu, X. Wu, X. Wang, L. Xing, K. Scott, J. Power 2010, 2700.
Sources, 196, 2011, 8377. 67. W. Tu, J. Lei, S. Zhang, H. Ju, Chem. Eur. J., 16, 2010, 10771.
33. J.M. Yang, S.A. Wang, C.L. Sun, M.D. Ger, J. Power Sources, 68. Y. Xu, L. Zhao, H. Bai, W. Hong, C. Li, G. Shi, J. Am. Chem.
254, 2014, 298. Soc.,131, 2009, 13490.
34. Y. Yu, Z. Chen, S. He, B. Zhang, X. Li, M. Yao, Biosens. 69. L. Tan, K.G. Zhou, Y.H. Zhang, H.X. Wang, X.D. Wang, Y.F.
Bioelectron., 52, 2014, 147. Guo, H.L. Zhang, Electrochem. Comm., 12, 2010, 557.
35. H.K. Jeong, M. Jin, E.J. Ra, K.Y. Sheem, G.H. Han, S. 70. K. Jasuja, J. Linn, S. Melton, V. Berry, J. Phys. Chem. Lett., 1,
Arepalli, Y.H. Lee, ACS Nano, 4, 2010, 1162. 2010, 1853.
36. J.Q. Liu, L. Tao, W.R. Yang, D. Li, C. Boyer, R. Wuhrer, F. 71. J. Shen, M. Shi, N. Li, B. Yan, H. Ma, Y. Hu, M. Ye, Nano
Braet, T.P. Davis, Langmuir, 26, 2010, 10068. Res., 3, 2010, 339.
37. X. Wang, H. Bai, Z. Yao, A. Liu, G. Shi, J. Mater. Chem., 20, 72. Z. Tang, S. Shen, J. Zhuang, X. Wang, Angew. Chem. Int. Ed.,
2010, 9032. 49, 2010, 4603.
38. Y. Lian, Y. Liu, T. Jiang, J. Shu, H. Lian, M. Cao, J. Phys. 73. G. Lu, S. Mao, S. Park, R. Ruoff, J. Chen, Nano Res., 2, 2009,
Chem. C, 114, 2010, 9659 . 192.
39. J. Liu, W. Yang, L. Tao, D. Li, C. Boyer, T.P. Davis, J. Polym. 74. E. Motamedi, M. Talebi Atouei, M.Z. Kassaee, Mater. Res.
Sci. Pol. Chem., 48, 2010, 425. Bull., 54, 2014, 34.
40. J. Liu, Y. Li, Y. Li, J. Li, Z. Deng, J. Mater. Chem., 20, 2010, 75. X. Bin, J. Chen, H. Cao, L. Chen, J. Yuan, J. Phys. Chem.
900. Solids, 70, 2009, 1.
41. Q. Bao, H. Zhang, J.X. Yang, S. Wang, D.Y. Tang, R. Jose, 76. S. Yang, G. Cui, S. Pang, Q. Cao, U. Kolb, X. Feng, J. Maier,
S. Ramakrishna, C.T. Lim, K.P. Loh, Adv. Funct. Mater., 20, K. Müllen, Chem. Sus. Chem., 3, 2010, 236.
2010, 782. 77. V. Chandra, J. Park, Y. Chun, J.W. Lee, I.C. Hwang, K.S. Kim,
42. H.R. Mamaqani, V. Haddadi-Asl, M. Mortezaei, K. Khezri, J. ACS Nano, 4, 2010, 3979.
Appl. Polym. Sci., 131, 2014, 40273. 78. Y. Fan, L. Wang, J. Li, J. Li, S. Sun, F. Chen, L. Chen, W.
43. L. Ren, T. Liu, J. Guo, S. Guo, X. Wang, W. Wang, Jiang, Carbon, 48, 2010, 1743.
Nanotechnology, 21, 2010, 335701. 79. L.S. Zhang, L.Y. Jiang, H.J. Yan, W.D. Wang, W. Wang, W.G.
44. M. Fang, K. Wang, H. Lu, Y. Yang, S. Nutt, J. Mater. Chem., Song, Y.G. Guo, L.J. Wan, J. Mater. Chem., 20, 2010, 5462.
20, 2010, 1982. 80. D.H. Wang, R. Kou, D. Choi, Z.G. Yang, Z.M. Nie, J. Li, L.V.
45. M. Xiao, Y. Lu, S.J. Wang, Y.F. Zhao, Y.Z. Meng, J. Power Saraf et al., ACS Nano, 4, 2010, 1587.
Sources, 160, 2006, 165. 81. J. Yan, Z. Fan, T. Wei, W. Qian, M. Zhang, F. Wei, Carbon,
46. S.V. Rodil, J.I. Paredes, A.M. Alonso, J.M.D. Tascón, J. Mater. 48, 2010, 3825.
Chem., 19, 2009, 3591. 82. X.Y. Zhang, H.P. Li, X.L. Cui, Y. Lin, J. Mater. Chem., 20,
47. Z. Jiang, Q. Li, M. Chen, J. Li, J. Li, Y. Huang, F. Besenbacher, 2010, 2801.
M. Dong, DOI: 10.1039/c3nr00288h. 83. Y. Zhang, H. Li, L. Pan, T. Lu, Z. Sun, J. Electroanal. Chem.,
48. S. Ansari, E.P. Giannelis, J. Polym. Sci. Pol. Phys., 47, 2009, 634, 2009, 68.
888. 84. S. Yang, X. Feng, L. Wang, K. Tang, J. Maier, K. Müllen,
49. H. Kim, C.W. Macosko, Polymer, 50, 2009, 3797. Angew. Chem. Int. Ed., 49, 2010, 4795.
50. A. Satti, P. Larpent, Y. Gun’ko, Carbon, 48, 2010, 3376. 85. S. Chen, J. Zhu, X. Wang, J. Phys. Chem. C, 114, 2010, 11829.
51. H. Kim, Y. Miura, C.W. Macosko, Chem. Mater., 22, 2010, 86. H. Wang, H.S. Casalongue, Y. Liang, H. Dai, J. Am. Chem.
3441. Soc., 132, 2010, 7472.
52. Y. Cao, J. Feng, P. Wu, Carbon, 48, 2010, 3834. 87. C. Guo, H. Yang, Z. Sheng, Z. Lu, Q. Song, C. Li, Angew.
53. Y. Huang, Y. Qin, Y. Zhou, H. Niu, Z.Z. Yu, J.Y. Dong, Chem. Chem. Int. Ed., 49, 2010, 3014.
Mater., 22, 2010, 4096. 88. Y. Lin, K. Zhang, W. Chen, Y. Liu, Z. Geng, J. Zeng, N. Pan,
54. Z. Xu, C. Gao, Macromolecules, 43, 2010, 6716. L. Yan, X. Wang, J.G. Hou, ACS Nano, 4, 2010, 3033.
55. Y. Sun, C. Li, Y. Xu, H. Bai, Z. Yao, G. Shi, Chem. Comm., 89. Y. Ding, Y. Jiang, F. Xu, J. Yin, H. Ren, Q. Zhuo, Z. Long, P.
46, 2010, 4740. Zhang, Electrochem. Comm., 12, 2010, 10.
56. K. Zhang, L.L. Zhang, X.S. Zhao, J. Wu, Chem. Mater., 22, 90. J.Y. Sun, K.J. Huang, S.Y. Wei, Z.W. Wu, F.P. Ren, Colloids
2010, 1392. Surf. B, 84, 2011, 421.
57. Y.B. Wang, D. Kurunthu, G.W. Scott, C.J. Bardeen, J. Phys. 91. F. Li, J. Li, Y. Feng, L. Yang, Z. Du, Sens. Actuators B, 157,
Chem. C, 114, 2010, 4153. 2011, 110.
58. K.S. Choi, F. Liu, J.S. Choi, T.S. Seo, Langmuir, 26, 2010, 12902. 92. X. Kang, J. Wang, H. Wu, J. Liu, I.A. Aksay, Y. Lin, Talanta,
59. H.F. Yang, Q.X. Zhang, C.S. Shan, F.H. Li, D.X. Han, L. Niu, 81, 2010, 754.
Langmuir, 26, 2010, 6708. 93. K. Wang, Q. Liu, X.Y. Wu, Q.M. Guan, H.N. Li, Talanta, 82,
60. J.L. Delgado, M.A. Herranz, N. Martin, J. Mater. Chem., 18, 2010, 372.
2008, 1417. 94. B.C. Brodie, Ann. Chim. Phys., 59, 1860, 466.
61. M. Arvand, T.M. Gholizadeh, Colloids Surf. B, 103, 2013, 84. 95. M. Arvand, M. Anvari, J. Iran Chem. Soc., 10, 2013, 841.
62. E. Yoo, J. Kim, E. Hosono, H. Zhou, T. Kudo, I. Honma, Nano 96. J.R. Lomeda, C.D. Doyle, D.V. Kosynkin, W.H. Hwang, J.M.
Lett., 8, 2008, 2277. Tour, J. Am. Chem. Soc., 130, 2008, 16201.
97. D. Zielińska, B. Pierozynski, J. Electroanal. Chem., 625, 131. J. Li, S. Guo, Y. Zhai, E. Wang, Electrochem. Commun., 11,
2009, 149. 2009, 1085.
98. M. Zhu, C. Zeng, J. Ye, Electroanalysis, 23, 2011, 907. 132. L. Baldrianova, I. Svancara, M. Vlcek, A. Economouc, S.
99. Q. Wang, M. Zheng, J. Shi, F. Gao, F. Gao, Electroanalysis, Sotiropoulos, Electrochim. Acta, 52, 2006, 481.
23, 2011, 915. 133. Y.C. Si, E.T. Samulski, Nano Lett., 8, 2008, 1679.
100. S.F. Wang, F.R. Xie, F. Hu, Sens. Actuators B, 123, 2007, 495. 134. W. Song, D.W. Li, Y.T. Li, Y. Li, Y.T. Long, Biosens.
101. H. Du, J. Ye, J. Zhang, X. Huang, C. Yu, J. Electroanal. Bioelectron., 26, 2011, 3181.
Chem., 650, 2011, 209. 135. E. Dempsey, D. Diamond, A. Collier, Biosens. Bioelectron.,
102. Y.R. Kim, S. Bong, Y.J. Kang, Y. Yang, R.K. Mahajan, J.S. 20, 2004, 367.
Kim, H. Kim, Biosens. Bioelectron., 25, 2010, 2366. 136. M.S. Tehrani, A. Pourhabib, S.W. Husain, M. Arvand,
103. N.I. Kovtyukhova, P.J. Ollivier, B.R. Martin, T.E. Mallouk, Electroanalysis, 24, 2012, 2395.
S.A. Chizhik, E.V. Buzaneva, A.D. Gorchinskiy, Chem. 137. K. Zhou, Y. Zhu, X. Yang, J. Luo, C. Li, S. Luan, Electrochim.
Mater., 11, 1999, 771. Acta, 55, 2010, 3055.
104. F. Schedin, A.K. Geim, S.V. Morozov, E.W. Hill, P. Blake, M.I. 138. X.L. Ren, X.W. Meng, D. Chen, F.Q. Tang, J. Jiao, Biosens.
Katsnelson, K.S. Novoselov, Nat. Mater., 6, 2007, 652. Bioelectron., 21, 2005, 433.
105. M.H. Parvin, Electrochem. Commun., 13, 2011, 366. 139. H. Yin, Q. Ma, Y. Zhou, S. Ai, L. Zhu, Electrochim. Acta, 55,
106. A.L. Higginbotham, D.V. Kosynkin, A. Sinitskii, Z. Sun, J.M. 2010, 7102.
Tour, ACS Nano, 4, 2010, 2059. 140. Y. Bo, H. Yang, Y. Hu, T. Yao, S. Huang, Electrochim. Acta,
107. B. Su, J. Tang, H. Yang, G. Chen, J. Huang, D. Tang, 56, 2011, 2676.
Electroanalysis, 23, 2011, 832. 141. K. Zhou, Y. Zhu, X. Yang, C. Li, Electroanalysis, 23, 2011,
108. H. Xiong, B. Jin, J. Electroanal.Chem., 661, 2011, 77. 862.
109. J.F. Wu, M.Q. Xu, G.C. Zhao, Electrochem. Commun., 12, 142. J.Y. Sun, K.J. Huang, S.F. Zhao, Y. Fan, Z.W. Wu,
2010, 175. Bioelectrochemistry, 82, 2011, 125.
110. X. Kang, J. Wang, H. Wu, I.A. Aksay, J. Liu, Y. Lin, Biosens. 143. C. Fu, W. Yang, X. Chen, D.G. Evans, Electrochem. Commun.,
Bioelectron., 25, 2009, 901. 11, 2009, 997.
111. G. Eda, M. Chhowalla, Nano Lett., 9, 2009, 814. 144. A. Chou, T. Böcking, N.K. Singh, J.J. Gooding, Chem.
112. Y. Zhang, X. Sun, L. Zhu, H. Shen, N. Jia, Electrochim. Acta, Commun., 7, 2005, 842.
56, 2011, 1239. 145. J. Li, C.Y. Liu, C. Cheng, Electrochim. Acta, 56, 2011, 2712.
113. M.H. Yang, B.G. Choi, H.S. Park, T.J. Park, W.H. Hong, S.Y. 146. T.T. Baby, S.S.J. Aravin, T. Arockiadoss, R.B. Rakhi, S.
Lee, Electroanalysis, 23, 2011, 850. Ramaprabhu, Sens. Actuators B, 145, 2010, 71.
114. J. Lu, I. Do, L.T. Drzal, R.M. Worden, I. Lee, ACS Nano, 2, 147. L. Staudenmaier, Ber. Dtsch. Chem. Ges., 31, 1898, 1481.
2008, 1825. 148. Z. Liu, Z. Wang, Y. Cao, Y. Jing, Y.Liu, Sens. Actuators B,
115. W. Hong, H. Bai, Y. Xu, Z. Yao, Z. Gu, G. Shi, J. Phys. Chem. 157, 2011, 540.
C, 114, 2010, 1822. 149. Y. Fan, K.J. Huang, D.J.Niu, C.P. Yang, Q.S. Jing, Electrochim.
116. S. Guo, D. Wen, Y. Zhai, S. Dong, E. Wang, ACS Nano, 4, Acta, 56, 2011, 4685.
2010, 3959. 150. Q. Zeng, J.S. Cheng, X.F. Liu, H.T. Bai, J.H. Jiang, Biosens.
117. F. Li, J. Chai, H. Yang, D. Han, L. Niu, Talanta, 81, 2010, Bioelectron., 26, 2011, 3456.
1063. 151. L.H. Tang, Y. Wang, Y.M. Li, H.B. Feng, J. Lu, J.H. Li, Adv.
118. Y. Bao, J. Song, Y. Mao, D. Han, F. Yang, L. Niu, A. Ivaska, Funct. Mater., 19, 2009, 2782.
Electroanalysis, 23, 2011, 878. 152. S.D. Seo, I.S. Hwang, S.H. Lee, H.W. Shim, D.W. Kim, Ceram.
119. C.X. Guo, Y. Lei, C.M. Li, Electroanalysis, 23, 2011, 885. Int., 38, 2012, 3017.
120. Q. Zhang, Y. Qiao, L. Zhang, S. Wu, H. Zhou, J. Xu, X.M. 153. Y.S. Yun, D. Kim, Y. Tak, H.J. Jin, Synth. Met., 161, 2011,
Song, Electroanalysis, 23, 2011, 900. 2460.
121. S. Zhao, K. Zhang, Y.Y. Sun, C.Q. Sun, Bioelectrochem., 69, 154. P. Prabhu, R. Suresh Babu, S. Sriman Narayanan, Sens.
2006, 10. Actuators B, 156, 2011, 606.
122. P. Liu, X. Zhang, L. Feng, H. Xiong, S. Wang, Am. J. Biomed. 155. X. Yan, D. Pan, H. Wang, X. Bo, L. Guo, J. Electroanal.
Sci., 3, 2011, 69. Chem., 663, 2011, 36.
123. P. Sorlier, A. Denuzière, C. Viton, Biomacromolecules, 2, 156. V. Hariharan, S. Radhakrishnan, M. Parthibavarman, R.
2001, 765. Dhilipkumar, C. Sekar, Talanta, 85, 2011, 2166.
124. Y. Jiang, X. Zhang, C. Shan, S. Hua, Q. Zhang, X. Bai, L. Dan, 157. A. Babaei, A.R. Taheri, Sens. Actuator B, 176, 2013, 543.
L. Niu, Talanta, 85, 2011, 76. 158. A.R. Khaskheli, J. Fischer, J. Brek, V. Vysko, M.I. Bhanger,
125. P. Wu, Q. Shao, Y. Hu, J. Jin, Y. Yin, H. Zhang, C. Cai, Electrochim. Acta, 101, 2013, 238.
Electrochim. Acta, 55, 2010, 8606. 159. Y. Fan, J.H. Liu, H.T. Lu, Q. Zhang, Colloids Surf. B, 85, 2011,
126. S. Alwarappan, C. Liu, A. Kumar, C.Z. Li, J. Phys. Chem. C, 289.
114, 2010, 12920. 160. P.F. Bolado, P.J.L. Ardisana, D.H. Santos, A.C. García, Anal.
127. Q. Wei, K. Mao, D. Wu, Y. Dai, J. Yang, B. Du, M. Yang, H. Chim. Acta, 638, 2009, 133.
Li, Sens. Actuators B, 149, 2010, 314. 161. H. Ghdimi, R.M.A. Tehrani, A.S. Mohamed Ali, N. Mohamed,
128. Z. Liu, J.T. Robinson, X.M. Sun, H.J. Dai, J. Am. Chem. Soc., S.A. Ghani, Anal. Chim. Acta, 765, 2013, 70.
130, 2008, 10876. 162. A. Gasnier, M.L. Pedano, M.D. Rubianes, G.A. Rivas, Sens.
129. Q. Feng, K. Liu, J. Fu, Y. Zhang, Z. Zheng, C. Wang, Y. Du, Actuators B, 176, 2013, 921.
W. Ye, Electrochim. Acta, 60, 2012, 304. 163. Z. Wang, M. Shoji, H. Ogata, Talanta, 99, 2012, 487.
130. Y. Mao, Y. Bao, S. Gan, F. Li, L. Niu, Biosens. Bioelectron., 164. H. Teymourian, A. Salimi, R. Hallaj, Biosens. Bioelectron.,
28, 2011, 291. 33, 2012, 60.
165. C.H. Chen, Y.C. Chen, M.S. Lin, Biosens. Bioelectron., 42, 168. C. Wang, L. Liu, H. Xue, X. Hu, G. Wang, Sens. Actuators B,
2013, 379. 181, 2013, 194.
166. Y. Dilgin, B. Kızılkay, D.G. Dilgin, H.I. Gökc, L. Gorton, 169. Z. Sun, H. Fu, L. Deng, J. Wang, Anal. Chim. Acta, 761, 2013, 84.
Colloids Surf. B, 102, 2013, 816. 170. S. Piermarini, D. Migliorelli, G. Volpe, R. Massoud, A.
167. J. Dua, R. Yue, Z. Yao, F. Jiang, Y. Du, P. Yang, C. Wang, Pierantozzi,C. Cortese, G. Palleschi, Sens. Actuators B, 179,
Colloids Surf. A, 419, 2013, 94. 2013, 170.
Nanoscience & Technology Aliofkhazraei
Ali
Milne
Ozkan
GRAPHENE
Mitura
Gervasoni
SCIENCE

“I am confident in the materials … The wide scope of information covered and the qualifications of the contributors

projects a positive image of the potential quality of the publication.”
—Albert V. Tamashausky, Asbury Carbons Inc.
“This book is a result of an impressive project to collect views from experts in every aspect of graphene science. All popular
topics in the research of this impressive material are covered. This is the best and most complete presentation that has been
published so far for the hottest material of our times. A must-have reference.”
—Ioannis Remediakis, Department of Materials Science and Technology, University of Crete, Greece
HANDBOOK
“…this set of volumes represents a complete handbook showing the state of the art of science and technology related with
graphene. This set of books is written by great specialists and competent experts. For someone who works in this field, this
set of volumes is an essential reference for the characterization and application of graphene.”
— Dr. Alex Axelevitch, Holon Institute of Technology (HIT) Mechanical and
An In-Depth Look at the Outstanding Properties of Graphene
The Graphene Science Handbook is a six-volume set that describes graphene’s special structural, electrical, and
Chemical Properties
chemical properties. The book considers how these properties can be used in different applications (including the
development of batteries, fuel cells, photovoltaic cells, and supercapacitors based on graphene) and produced on a
massive and global scale.
Volume four is dedicated to the mechanical and chemical properties of graphene and covers
• Mechanical properties using a continuum elastic model introduced to describe graphene’s elastic behavior
• Results of theoretical investigations of the mechanical properties of graphene structures
• Mechanical stabilities and properties of graphene under various strains
• Different types of graphene devices for biomolecule and gas sensing
• Printed graphene-based electrochemical sensor technology EDITED BY
• Various types of graphene-based electrochemical sensors
• The chemical vapor deposition of graphene on copper Mahmood Aliofkhazraei • Nasar Ali
• Strategies covering graphene modification
•
•
Graphene in solar cells, including transparent electrodes, active layers, and interface layer
Changes at the micrometric and nanometric scales, and more
K20503
ISBN: 978-1-4665-9123-3
90000
9 781466 591233
K20503_COVER_Final.indd 1 3/17/16 5:43 PM

Mechanical and Chemical Properties: Graphene Science Handbook

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mechanical and Chemical Properties: Graphene Science Handbook

Uploaded by

Copyright:

Available Formats

Nanoscience & Technology Aliofkhazraei

GRAPHENE SCIENCE HANDBOOK

Mechanical and Chemical Properties

K20503_COVER_Final.indd 1 3/17/16 5:43 PM

No claim to original U.S. Government works

International Standard Book Number-13: 978-1-4665-9124-0 (eBook - PDF)

Section I Mechanical Properties

Chapter 1 Mechanical Properties of Graphene......................................................................................................................... 3

Chapter 4 Mechanical Properties of Graphene Sheets........................................................................................................... 61

Section II Chemical Fabrication/Properties

Chapter 8 Graphene-Based Biological and Chemical Sensors............................................................................................. 127

Chapter 9 Printed Graphene-Based Electrochemical Sensors...............................................................................................147

Chapter 12 Chemical Modification of Graphene.................................................................................................................... 207

Chapter 13 Graphene Synthesis by Chemical Vapor Deposition on Copper.......................................................................... 225

Chapter 15 Graphene: Electrochemical Exfoliation and Applications................................................................................... 269

Chapter 16 Modification of Graphene with Polymers via Addition Chemistry..................................................................... 295

Chapter 17 Molecular Theory of Graphene Chemical Modification.......................................................................................317

Section III Recent Progress

Chapter 18 Low-Cost and Simple Method for Graphene Synthesis....................................................................................... 343

Chapter 19 Graphene-Based Solar Cells................................................................................................................................ 361

Chapter 20 Graphene Production from Chlorination of Metallocenes................................................................................... 387

Chapter 21 Chemical Modification of Graphene with Polymers............................................................................................ 405

Chapter 23 Graphene-Based Antibacterial Materials............................................................................................................. 447

Chapter 24 Nanofluidics in Graphene-Based Material Systems............................................................................................. 465

Chapter 25 Nanoporous Graphene Sheets for Gas Separation............................................................................................... 477

Chapter 26 Photorefractive Properties of Graphene-Based Organic Systems........................................................................ 491

Xiao-Rong Li Mihrimah Ozkan Raheel Shah

P. Pasakon Jae-Jin Shim

A. Tuantranont Jing-Juan Xu Liangchi Zhang

Tao Chen and Rebecca Cheung

ABSTRACT and graphene originates from their distinctive electrical prop-

the experimental measurements of graphene’s Young’s mod- Np 2 Np

wave functions and observations of a physical system are

Equivalent stress (GPa)

Armchair direction Zigzag direction Zigzag

Shear modulus (TPa)

Shear stress (GPa)

FIGURE 1.3 Stress–strain relationship of graphene from MD

FIGURE 1.6 Equivalent spring constants of graphene bridges plot-

of 1.03. The tension T 0 comes from various aspects, includ-

Note: CNT, carbon nanotube.

55 energy loss mechanisms. The common mechanisms include

(a) 30 (b) (c)

M eff x ′′(t ) + Γx ′(t ) + M eff ω 20 x(t ) + ζx 2 (t )   d

It can be clearly seen that the dynamic range depends

Note: DFT, density functional theory.

ABSTRACT progress has shown that graphene-based materials can have

Note: AFM, atomic force microscopy; DFT, density functional theory.

(a) 1.4 (b)

300 nm 100 nm Mag = 50.00 K X

2.2 ARCHITECTURE OF THE FFRC Carbon fiber Polymer Wavy CNTs CFF

CONTAINING WAVY CNTs 1, x

Polymer CNTs Carbon fiber

Wavy CNTs Polymer Polymer

Transverse cross-section Longitudinal cross-section

Unwound PMNC lamina

{σ r} = [Cr ]{ r}; r = n, i, and p (2.3)  n   i   p   nc 

 0 0 0 0 0 0 Using Equations 2.7 through 2.9, the local strain vectors

[CPMNC ] = [ T]− T [CNC ][T]−1 (2.17) 2

1f  1PMNC  1CFF  v PMNC C11

[CCFF ] = [C8 ][V9 ]−1 + [C9 ][ V10 ]−1 (2.23) 1 0 0 0 0 0 

σ1CFF  σ1P  σ1   vpC11

to the volume of the FFRC, and have been determined in C13

2.2 ARCHITECTURE OF THE FFRC Carbon fiber Polymer Wavy CNTs CFF

{σ r} = [Cr ]{ r}; r = n, i, and p (2.3)  n   i   p   nc 

1f  1PMNC  1CFF  v PMNC C11

3.3.4 Comparison of Effects between

0.4 3.3.5 Comparison of Effects between