Springer Handbook of Springer Handbook provides a concise compilation of approved key information on methods of research, general principles, and functional relationships in physical and applied sciences. The world’s leading experts in the fields of physics and engineering will be as- signed by one or several renowned editors to write the chapters com- prising each volume. The content is selected by these experts from Springer sources (books, journals, online content) and other systematic and approved recent publications of scientific and technical information. The volumes are designed to be useful as readable desk reference book to give a fast and comprehen- sive overview and easy retrieval of essential reliable key information, including tables, graphs, and bibli- ographies. References to extensive sources are provided. HandbookSpringer of Nanomaterials Robert Vajtai (Ed.)

With 685 Figures and 64 Tables

123 Editor Robert Vajtai Rice University Department of Mechanical Engineering and Materials Science 6100 Main MS-321 Houston, TX 77005-1827 USA

ISBN: 978-3-642-20594-1 e-ISBN: 978-3-642-20595-8 DOI 10.1007/978-3-642-20595-8 Springer Dordrecht Heidelberg London New York

Library of Congress Control Number: 2013942548

c Springer-Verlag Berlin Heidelberg 2013 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Production and typesetting: le-tex publishing services GmbH, Leipzig Senior Manager Springer Handbook: Dr. W. Skolaut, Heidelberg Typography and layout: schreiberVIS, Seeheim Illustrations: le-tex publishing services GmbH, Leipzig; Hippmann GbR, Schwarzenbruck Cover design: eStudio Calamar Steinen, Barcelona Cover production: WMXDesign GmbH, Heidelberg Printing and binding: Stürtz GmbH, Würzburg

Printed on acid free paper

Springer is part of Springer Science+Business Media (www.springer.com)

61/3180/YL 5 4 3 2 1 0 V

Foreword

Nanomaterials are based on structures with character- cup, which was probably created in istic features on the scale of nanometers. This size is Rome during the 4th century AD, small if we compare with normal things around us, but contains embedded of it is not particularly small on the atomic scale. In fact, gold and silver. Because of these distances between individual atoms are typically a tenth particles, the cup normally seems to of a nanometer (an Ångström), so a piece of a mater- be a light color, but it becomes ial with a side of a nanometer may contain hundreds or ruby when light is shone through even a thousand atoms. Therefore a nanomaterial usu- it. The Lycurgus cup is a wonder of ally has some resemblance to a bulk material based craftsmanship from Antiquity, and it on the same atoms, but the normal material has been is based on . modified to reach superior properties such as higher Finally, let’s consider nanoce- mechanical strength, different optical and magnetic per- ramics. The world’s most widely formance, permeability to a fluid, or something else. used artificial material is the nanoce- Prof. Claes-Göran Thus nanomaterials may allow us to obtain properties ramic cement, which was used ex- Granqvist that were previously impossible to achieve, impractical tensively by the Romans in con- Department of to manufacture, or too expensive for use on a scale large structing buildings, baths and aque- Engineering Sciences enough to be significant in daily life. ducts. Furthermore, recent archaeo- Solid State Physics Among the general public, and even the scientific logical discoveries indicate that the Uppsala University Sweden community, nanomaterials are widely perceived as new Romans were not the first, that in many different way – newly invented, newly used by the Macedonians were using cement human cultures, and newly studied. centuries earlier. In fact, nanomaterials are not new at all. Nature Even research on nanomaterials is not as new as it itself is filled with nanofeatures that have evolved in seems. The term apparently began appearing in the titles biological systems, one well known example being of scientific publications only 15 years ago. But today’s moths’ eyes with nanostructured surfaces that provide nano was the subject of an older literature under the antireflection and allow efficient use of feeble light. term ultrafine. Looking at history, we can also see that human be- As the examples above indicate, nanomaterials are ings have been using nanomaterials of various sorts for well rooted in the past. But they are also very much of a very long time. Let us take three examples: nanocar- the future. Let us consider a few specific examples. bons, nanometals, and nanoceramics. Nanocarbons, used for cave painting and the print- Nanocarbons can be created in abundance on the ing of the Gutenberg Bible, are very much in focus nanometer scale when organic matter burns. Such car- today, in the forms of , nanotubes and nanodi- bon nanoparticles were used by humans as far back as amonds, all of which offer a multitude of possibilities forty thousand years ago to depict and decorate. The for future technology. Two-dimensional carbon in the particles were mixed with fat and used for painting in form of graphene has unique properties directly based the caves of Altamira and Lascaux in Spain and France, on quantum physics, and it may have important ap- to mention two especially striking and well known plications in transparent electronics and elsewhere. cases. This kind of carbon, in principle, is also an es- Graphane, its hydrogenated cousin, is exciting in its sential ingredient in ink and printing paste, and it was own right. used by monastic scribes and by Gutenberg and his fol- Nanometals, employed by the Romans to create the lowers to make texts of explosive cultural significance amazing Lycurgus glass cup, are the basis today for and stunning beauty. manifold applications, including thermal collectors that Nanometals have also been utilized for thousands of harness the sun’s energy and innovative plasmonic solar years. An example is the world famous Lycurgus glass cells. Indeed, plasmonics is becoming a household word cup, now in the holdings of the British Museum. This because of its relevance for light-emitting diodes, sen- VI

sors and catalysts for chemical reactions, just to mention issues are of great importance. We should remember the a few technologies. terrible impact that asbestos – in fact, a natural nano- Thus many aspects of nanomaterials are indeed truly material – had on human health before it was widely new and are the subject of intense worldwide interest in banned. We certainly do not want to discover one day today’s academic and industrial research laboratories. that, in our quest for new materials to solve techno- This Handbook, which is a testimony to this growing logical problems, we have unleashed another dangerous body of knowledge, presents welcome and authoritative nanomaterial into the world. surveys over nanocarbons, nanometals and nanoce- The editor and authors are to be congratulated on ramics in its first three parts. Other sections cover the successful completion of this Handbook of Nano- nanocomposites and nanoporous materials, as well as materials. It will surely be a work of great and lasting organic and biological nanomaterials. Applications and importance for the scientific community. impacts are discussed at the end, together with impor- tant questions of toxicology, hazards and safety. These Uppsala, November 2012 Prof. Claes G. Granqvist VII

Foreword

It has been more than a decade since President Bill Clin- science. The main goal of ma- ton talked about the promise of nanotechnology and terials science – macroscopic and the importance of increasing investments in nanoscale nanoscale – is providing new and science and engineering research in a speech at the Cal- improved building blocks for engi- ifornia Institute of Technology on January 21, 2000. neers in all fields. That said, nano- In his remarks, the President recalled Richard Feyn- materials science has distinct fea- man’s American Physical Society talk there in 1959. tures compared to the more mature The following week, in his State of the Union Address, science and engineering of macro- President Clinton announced his 21st Century Research scopic materials, the most salient Fund, a $3 billion budget increase, which included the being its revolutionary nature. New multiagency national nanotechnology initiative (NNI). materials and groups of materials The first year’s budget allocation to NNI was close to with surprising properties continue half a billion US$, nearly doubling what the agencies to be discovered – graphene and Prof. Neal Lane had been spending on nanoscale research; and with the topological insulators are two exam- Malcolm Gillis University continuous support of succeeding administrations the ples from the recent past. As with Professor, budget quadrupled in a decade. This strong federal sup- all exploration at the frontiers of Department of Physics port, initially based on the promise of a revolutionary knowledge, it is impossible to pre- and Astronomy, new technology, was justified by steady scientific and dict what discoveries will be made Senior Fellow, James A. Baker III Institute technological advances at the nanometer scale and by or how those discoveries might lead for Public Policy the growth of commercial applications, especially in to applications, commercial or oth- Rice University biotechnology and , offering new ways erwise. But the history of science Houston, Texas to tackle disease and new industrial tools and toys. and technology suggests that some As President Clinton’s former science advisor, I am of those advances will surpass all our expectations. Al- confident that he is as pleased with the progress in ready we are seeing the benefits of nanotechnology in nanotechnology as are all of us – inside and outside computers and telecommunication devices, computer government – who worked with him to develop and chips and sensors in automobiles, electric car batter- implement the NNI. ies, medicines and sun creams, tablecloths and socks, One way to define nanotechnology, perhaps, is that tennis rackets, boats, golf clubs – and more. Given it is the knowledge and engineering (design and con- the likelihood that ongoing research will yield many trol) of physical, chemical, and biological systems at more nanomaterials, with surprising properties and, at the nanometer (10−9 m) scale – from the size of indi- the same time, the continued exponential growth in the vidual to dimensions of the order 100 nm. number of applications, it seems clear that nanomateri- Nanotechnology is, by its nature, a field of synthesis als will, at some level, transform most aspects of our and synergy often requiring physics, chemistry, biology, lives. It is not too much of a stretch to suggest that and almost all areas of engineering in the performance President Clinton’s policy decision to set up the NNI, of research and engineering design, for example invent- which has supported thousands of scientists and en- ing and optimizing the tools needed to synthesize and gineers working in the field, has indeed helped move manipulate matter at the nanometer scale. As with other us closer to realizing Feynman’s prediction – or, per- new fields, rapid advances in nanotechnology have led haps we should say his vision – of a revolutionary new to specialization into subdisciplines, one of the most technology. In the world of nanomaterials there is still natural and important being nanomaterials. plenty of room at the bottom to use Richard Feynman’s Nanomaterials science and engineering includes the famous words. production, properties, and applications of materials at A handbook, by one definition, is a compilation the nanometer scale; it is a part of nanotechnology of knowledge about a particular field, collected into and at the same time, evidently, a subfield of materials a single volume publication that is convenient to use as VIII

a ready reference. Since nanomaterials science can now other aspects can be followed easily by using the well- be considered a self-sufficient discipline, a handbook is developed index. appropriate and timely. This new Springer Handbook Putting together a handbook in a new field is of Nanomaterials targets several audiences: researchers a formidable challenge. I would like to congratulate the working in industry or academia, as well as gradu- editor and all of the authors who collaborated to plan, ate students studying related fields. The organization collect materials, and write this important groundbreak- of the book follows the usual classification scheme ing Springer Handbook of Nanomaterials. of macroscopic materials science, with information of a materials group – e.g., – collected together; Houston, January 2013 Neal Lane IX

Preface

Those who control materials, control technology, stated a balance between references and scientific results re- Eiji Kobayashi, Senior Advisor of Panasonic Corpora- ported in tables and figures. We describe nanomaterials tion, explaining the importance of materials science and in textbook style for newcomers, encyclopedia-like ele- engineering. I would translate this quote to those who ments and – to follow the fast-space of new results – control nanomaterials control nanotechnology; and, review or research papers for the experienced reader. considering the effect of the development of nanomate- Beyond scientific and moral correctness we also look rials and nanotechnology on our global infrastructure, for clarity by concise and easy-to-follow text, well- it is not too bold to state that those control technol- designed and clear figures which were all professionally ogy at large, too. Nanomaterials have a determinant role drawn by graphics designers. in many of advanced products around us. Stamp-sized The book is divided in Parts A to G and cov- sound recording devices, modern passenger and fighter ers carbon-based nanomaterials: fullerenes, nanotubes, jets, spaceships and space stations, extreme tall build- nanofibers and nanodiamond, noble and common met- ings and long bridges, none of these could be created als and alloys, ceramic materials, crystalline and glassy without these marvelous materials. As one could not and other compounds; composites, hybrid struc- foresee 50 years ago, the fast development that provided tures and solutions as well as porous metals, ceramics the opportunity for these objects to be realized, now we and silicon; organic and bio-nanomaterials, bones and cannot imagine our life without them. fibers and select applications, respectively. This higher The editor considers materials science as the knowl- level structure conforms to the macroscopic classifica- edge of structure; properties of materials predicted or tion of materials and it is composed of chapters. Each explained with the help of this knowledge; experimen- chapter is self-consistent and builds up of similar parts, tal and theoretical tools designed and established for history, definitions, production of the given materials, preparing, characterizing and modifying processes, and properties, and applications. All of these parts are richly last but not least showing application possibilities of illustrated and consist of a balanced ratio of important the resulted materials. After defining nanomaterials we basics and recent results. can simply transpose this description for nanomaterials My pleasant obligation is to thank all of the help science. Materials are considered nanomaterials when I received in planning and implementing the handbook. their structure, processing, characterization or appli- First of all, I need to acknowledge the diligent work of cation differ from the macroscopic materials and this the authors in developing the chapters which involves difference relates to the – normally sub-100 nm – fea- more effort than a review paper, and the reward is not ture size. The description of the nanomaterials in this so immediate and evident. Their expertise, energy and Springer Handbook follows the thorough but concise time are greatly appreciated. I also would like to thank explanation of the synergy of structure, properties, pro- the advices and help of my colleagues at Rice Uni- cessing, and applications. Specifically, our aim was to versity and at Rensselaer Polytechnic Institute; as well point out the distinction between the properties of bulk as Professors Thomas F. George, Bob Curl, Phaedon and nanomaterials and the reasons for these differences. Avouris, Li Song and Jinquan Wei for keeping con- To fulfill these goals, we provide a balanced report tact with many authors. The great workmanship of the of the literature of each materials group. The format Springer publishing team and the continuous support of follows the well-established structure of the Springer the managing editors Mayra Castro and Werner Skolaut Handbooks with chapters as the basic units that are or- are also appreciated. I also need to thank my colleagues ganized into several groups. In each chapter, authors and friends, Laszlo B. Kish, Claes-Goran Granqvist, cover materials of their expertise, however, they focus Pulickel M. Ajayan and Richard W. Siegel that the not only on their own work, but report the interesting collaboration with them oriented me to nanomaterials and important efforts in the community, establishing science. Last, but not least, I thank for the help and pa- X

tience of my wife, Agnes, without her I would have not I hope that it serves as a frequently opened reference been able to finish this job. work. I wish the reader a pleasant and beneficial time when using the Springer Handbook of Nanomaterials, and Houston, November 2012 Robert Vajtai XI

About the Editor

Robert Vajtai is a Faculty Fellow at Rice University, Houston, Texas, in the Depart- ment of Mechanical Engineering and Materials Science. His expertise covers synthesis, processing, characterization of physical and chemical properties of new, advanced ma- terial forms and structures. More specifically Dr. Vajtai’s interests are in nanostructured materials, nanocomposites and nanomaterials; as well as their applications in thermal management, energy storage, microelectromechanical systems, sensors and electronic devices. Dr. Vajtai received his scientific education in physics and his Ph.D. degree in solid- state physics from the University of Szeged (then named Jozsef Attila University), Hungary. From 1987 to 2002 he was a faculty member of the Department of Experi- mental Physics at the University of Szeged, Hungary. He was rewarded by the Bolyai Fellowship of the Hungarian Academy of Sciences for 1999-2000. He spent sabbati- cals as a Fellow of the Swedish Institute in The glosseintragAAngstrom Laboratory in Uppsala, Sweden, in the years 1998 and 1999; as an Eötvös Fellow at the EPFL in Lausanne, Switzerland in 1995/1996 and visited the Max Planck Institute in Göttingen, Germany, in 1993 via a Max Planck Fellowship. Before moving to Rice University in 2008, Dr. Vajtai spent eight years at the Rensse- laer Polytechnic Institute, Troy, New York, where he was a Laboratory Manager at the Rensselaer Nanotechnology Center managing the carbon nanotechnology laboratories. Dr. Vajtai started his research as a physicist studying laser- interaction, melting and oxidation of refractive metals and the nonlinear behavior of the far- from equilibrium processes and systems. Later he developed methods for pulse-probe spectroscopy of biomaterials as well as OH radicals used for the study of organic contamination of the atmosphere by airborne LIDAR systems. His research in ma- terials science started with the synthesis of nanometals and nanosized oxides for the development of sensors. This lead to a new method for the preparation of germanium nanoparticles for building inverse opals used in infrared optical sensing. His most sig- nificant contribution is related to the synthesis of different forms of nanocarbons such as carbon nanotubes, graphene and macroscopic systems designed and built from these carbon allotropes, e.g., electromechanical parts and nanotube wires. Recently, his inter- est extended to various atomically thin layers, hexagonal nitride, transition-metal dichalcogenides and oxides. He has more than 145 journal publications in peer reviewed scientific journals and he delivered numerous invited, keynote and plenary lectures on the topic. Dr. Vajtai is a passionate teacher, he lectured physics, thermodynamics and elec- trodynamics courses with hundreds of experimental demonstrations; introductory and advanced courses of materials science. He received several mentoring awards, among those the Siemens-Westinghouse Mentoring Award. Robert Vajtai is a Faculty Fellow in the Department of Mechanical Engineering & Materials Science at Rice University. He received his undergraduate and Ph.D. degrees from the University of Szeged, then named Jozsef Attila University, Hungary. XIII

List of Authors

Maya Bar-Sadan Rodolfo Cruz-Silva Ben-Gurion University Shinshu University Department of Chemistry Research Center for Exotic Nanocarbons Be’er Sheba , Israel Wakasato e-mail: [email protected] 380-8553 Nagano, Japan Giovanni Barcaro e-mail: [email protected] Italian National Research Council Institute for the Physical and Chemical Processes Pratap Kumar Deheri Via Giuseppe Moruzzi 1 ShayoNano Singapore Pte Ltd. 56124 Pisa, Italy 609969 Singapore e-mail: [email protected]; [email protected] e-mail: [email protected] Paolo Bettotti University of Trento Libo Deng Department of Physics, Nanoscience Laboratory University of Manchester via Sommarive 14 School of Materials 38123 Povo, Italy Oxford Road e-mail: [email protected] Manchester, M13 9PL, UK e-mail: [email protected] Alfredo Caro Los Alamos National Laboratory Materials Science and Technology Division Yi Ding Los Alamos, NM 87544, USA Shandong University e-mail: [email protected] School of Chemistry and Chemical Engineering 27SouthShanDaRoad Eunhyea Chung Jinan, 250100, China Korea Institute of Science and Technology (KIST) e-mail: [email protected] Center for Water Resource Cycle Hwarangno 14-gil 5, Seongbuk-gu Seoul 136-791, Korea Huanli Dong e-mail: [email protected] Chinese Academy of Sciences Institute of Chemistry, Suzanne A. Ciftan Hens Key Laboratory of Organic Solids ITC/International Technology Center Zhongguancun North First Street 2 8100 Brownleigh Road Beijing, 100190, China Raleigh, NC 27617, USA e-mail: [email protected] e-mail: [email protected]

Vicki L. Colvin Mildred S. Dresselhaus Rice University Massachusetts Institute of Technology Office of Research, Chemistry Physics; Electrical Engineering 6100 Main Str. 77 Massachusetts Ave Houston, TX 77005, USA Cambridge, MA 02139, USA e-mail: [email protected] e-mail: [email protected] XIV List of Authors

Dimple P. Dutta Lei Gong Bhabha Atomic Research Centre University of Manchester Chemistry Division School of Materials Mumbai, 400085, India Oxford Road e-mail: [email protected] Manchester, M13 9PL, UK e-mail: [email protected] Hellmut Eckert University of Sao Paulo Takuya Hayashi Department of Physics Shinshu University Av. Trabalhador Saocarlense 400 Department of Electrical and Electronic Sao Carlos, SP 13566-590, Brazil Engineering e-mail: [email protected] 380-8553 Nagano, Japan Morinobu Endo e-mail: [email protected] Shinshu University Wenping Hu Research Centre for Exotic Nanocarbons Key Laboratory of Organic Solids 380-8553 Nagano, Japan Institute of Chemistry, Chinese Academy of e-mail: [email protected] Sciences Adam W. Feinberg Zhongguancun North First Street 2 Carnegie Mellon University Beijing, 100190, China Department of Biomedical Engineering e-mail: [email protected] 700 Technology Dr. Pittsburgh, PA 15219, USA Erik H. Hároz e-mail: [email protected] Rice University Electrical and Computer Engineering Alessandro Fortunelli 6100 Main Str. IPCF Consiglio Nazionale delle Ricerche (CNR) Houston, TX 77005, USA via Giuseppe Moruzzi 1 e-mail: [email protected] 56124 Pisa, Italy e-mail: [email protected] Quentin Jallerat Carnegie Mellon University Yogeeswaran Ganesan Biomedical Engineering Intel Corporation 700 Technology Drive 5200 NE Elam Young Parkway Pittsburgh, PA 15206, USA Hillsboro, OR 97124, USA e-mail: [email protected] e-mail: [email protected] Wei Gao Heli Jantunen Los Alamos National Laboratory University of Oulu Center for Integrated Nanotechnology Department of Electrical Engineering Bikini Atoll Rd Erkki Koiso-Kanttilankatu 3 Los Alamos, NM 87545, USA Oulu 90014, Finland e-mail: [email protected] e-mail: heli.jantunen@oulu.fi Thomas F. George Song Jin University of Missouri-St. Louis University of Wisconsin-Madison Office of the Chancellor, Center for Nanoscience Department of Chemistry One University Boulevard 1101 University Ave. St. Louis, MO 63121, USA Madison, WI 53706, USA e-mail: [email protected] e-mail: [email protected] List of Authors XV

Kaushik Kalaga Gábor Kozma Rice University University of Szeged Department of Mechanical Enginnering & Department of Applied and Environmental Materials Science Chemistry 6100 Main Str. Dugonics tér 13 Houston, TX 77005, USA 6720 Szeged, Hungary e-mail: [email protected] e-mail: [email protected] Jarmo Kukkola Ji-Hee Kim University of Oulu Rice University Department of Electrical Engineering Department of Electrical and Computer Erkki Koiso-Kanttilankatu 3 Engineering Oulu 90014, Finland 6100 Main Str. e-mail: [email protected] Houston, TX 77005, USA e-mail: [email protected] Ákos Kukovecz University of Szeged Yoong A. Kim Department of Applied and Environmental Shinshu University Chemistry Department of Electrical and Electronic Rerrich Béla tér 1 Engineering Szeged, Hungary 380-8553 Nagano, Japan e-mail: [email protected] e-mail: [email protected] Vinod Kumar Banaras Hindu University Ian A. Kinloch Department of Zoology University of Manchester Lanka, Varanasi, 221005, India School of Materials e-mail: [email protected] Oxford Road Manchester, M13 9PL, UK Zoltán Kónya e-mail: [email protected] University of Szeged Department of Applied and Environmental Imre Kiricsi (deceased) Chemistry Rerrich Bela tér 1 6720 Szeged, Hungary Junichiro Kono e-mail: [email protected] Rice University Jaesang Lee Electrical and Computer Engineering & Physics and Korea Institute of Science and Technology (KIST) Astronomy Center for Water Resource Cycle 6100 Main Str. Hwarangno 14-gil 5 Houston, TX 77005, USA Seoul 136-791, Korea e-mail: [email protected] e-mail: [email protected] Krisztián Kordás Seunghak Lee University of Oulu Korea Institute of Science and Technology (KIST) Department of Electrical Engineering Center for Water Resource Cycle Erkki Koiso-Kanttilankatu 3 Hwarangno 14-gil 5 Oulu 90570, Finland Seoul 136-791, Korea e-mail: [email protected].fi e-mail: [email protected] XVI List of Authors

Renat R. Letfullin Guowen Meng Rose-Hulman Institute of Technology Chinese Academy of Sciences Physics and Optical Engineering Institute of Solid State Physics 5500 Wabash Avenue Hefei, Anhui 230031, China Terre Haute, IN 47803-3999, USA e-mail: [email protected] e-mail: [email protected] Younès Messaddeq Roi Levi Université Laval Weizmann Institute of Science Department of Physics Department of Materials and Interfaces 2375, rue de la Terrasse 234 Herzl Street Québec, Québec G1V 0A6, Canada Rehovot 76100, Israel e-mail: [email protected] e-mail: [email protected] Jyri-Pekka Mikkola Åbo Akademi University Zuzanna A. Lewicka Department of Chemical Engineering Rice University Biskopsgatan 8 Department of Chemistry Åbo-Turku 20500, Finland 6100 Main Street e-mail: [email protected]; Houston,TX77005,USA jyri-pekka.mikkola@abo.fi e-mail: [email protected] Melinda Mohl Longtu Li University of Oulu Tsinghua University Department of Electrical Engineering Materials Science & Engineering Erkki Koiso-Kanttilankatu 3 Beijing, 100084, China Oulu 90570, Finland e-mail: [email protected] e-mail: [email protected].fi

Shaily Mahendra Aarón Morelos-Gómez University of California, Los Angeles Shinshu University Civil and Environmental Engineering Institute of Carbon Science and Technology 5732 Boelter Hall 4-17-1 Wakasato Los Angeles, CA 90095, USA 380-8553 Nagano, Japan e-mail: [email protected] e-mail: [email protected] Stephen A. Morin Balaji P. Mandal Harvard University Bhabha Atomic Research Centre Chemistry and Chemical Biology Chemistry Division 12 Oxford Street Mumbai, 400085, India Cambridge, MA 02138, USA e-mail: [email protected] e-mail: [email protected]

Fei Meng Marcelo Nalin University of Wisconsin-Madison Federal University of Sao Carlos Department of Chemistry Department of Chemistry 1101 University Avenue Rodovia Washington Luiz, SP-310 Madison, WI 53706, USA Sao Carlos, Sao Paulo, Brazil e-mail: [email protected] e-mail: [email protected] List of Authors XVII

Sebastien Nanot Sundara Ramaprabhu Rice University Indian Institute of Technology, Madras Electrical and Computer Engineering & Physics and Department of Physics Astronomy Chennai, 600 036, India 6100 Main Street e-mail: [email protected] Houston, TX 77098, USA Jayshree Ramkumar e-mail: [email protected] Bhabha Atomic Research Centre Analytical Chemistry Division Rachelle N. Palchesko Mumbai, 400085, India Carnegie Mellon University e-mail: [email protected] Biomedical Engineering Pittsburgh, PA 15219, USA Arava L.M. Reddy e-mail: [email protected] Rice University Department of Mechanical Engineering and Cary L. Pint Materials Science Vanderbilt University 6100 Main Str. Department of Mechanical Engineering Houston, TX 77005, USA 2301 Vanderbilt Place e-mail: [email protected] Nashville, TN 37215, USA Vincent C. Reyes e-mail: [email protected] University of California, Los Angeles Department of Civil and Environmental Gael Poirier Engineering Federal University of Alfenas 5732 Boelter Hall Institute of Science and Technology Los Angeles, CA 90095, USA Rodovia José Aurélio Vilela 11999 e-mail: vincecreyes@ ucla.edu Poços de Caldas, MG CEP 37715-400, Brazil e-mail: [email protected] Sidney J.L. Ribeiro Sao Paulo State University – UNESP Thalappil Pradeep Institute of Chemistry Indian Institute of Technology Madras Araraquara, SP 14801-970, Brazil Department of Chemistry e-mail: [email protected] Chennai, 600 036, India William D. Rice e-mail: [email protected] Los Alamos National Laboratory National High Magnetic Field Laboratory István Pálinkó Los Alamos, NM 87545, USA University of Szeged e-mail: [email protected] Department of Organic Chemistry Silvia H. Santagneli Dóm tér 8 Sao Paulo State University – UNESP 6720 Szeged, Hungary Institute of Chemistry e-mail: [email protected] Araraquara, SP 14801-970, Brazil e-mail: [email protected] Raju V. Ramanujan Nanyang Technological University Preeti S. Saxena School of Materials Science and Engineering Banaras Hindu University 50 Nanyang Ave. Department of Zoology 639798 Singapore Lanka, Varanasi, Uttar Pradesh 221005, India e-mail: [email protected] e-mail: [email protected] XVIII List of Authors

Olga A. Shenderova John M. Szymanski International Technology Center Carnegie Mellon University 8100 Brownleigh Road Department of Biomedical Engineering Raleigh, NC 27617, USA 700 Technology Drive e-mail: [email protected] Pittsburgh, PA 15219, USA e-mail: [email protected] Rakesh Shukla Bhabha Atomic Research Centre András Sápi Chemistry Division University of Szeged Mumbai, 400085, India Department of Applied and Environmental e-mail: [email protected] Chemistry Shashwat Shukla 1 Rerrich square Nanyang Technological University 6720 Szeged, Hungary 639798 Singapore e-mail: [email protected] e-mail: [email protected] Reshef Tenne Theruvakkattil S. Sreeprasad Weizmann Institute of Science Kansas State University Department of Materials and Interfaces Department of Chemical Engineering Rehovot 76100, Israel 1011 Durland Hall e-mail: [email protected] Manhattan, KS 66502, USA e-mail: [email protected] Humberto Terrones Anchal Srivastava Pennsylvania State University Banaras Hindu University Physics Department Department of Physics S104 Davey Lab Lanka, Varanasi, Uttar Pradesh 221005, India University Park, PA 16802, USA e-mail: [email protected] e-mail: [email protected]

Saurabh Srivastava Mauricio Terrones National Physical Laboratory Pennsylvania State University Biomedical Instrumentation Section, Department of Physics and Materials Science and New Rajender Nagar, New Delhi 110012, India Engineering e-mail: [email protected] University Park, PA 16802, USA Yan Sun e-mail: [email protected] Beihang University (BUAA) School of Biological Science and Medical Nicholas A. Thompson Engineering Rice University 37 Xueyuan Road, Haidian District Department of Physics & Astronomy Beijing, 100191, China 6100 Main Str. e-mail: [email protected] Houston, TX 77005, USA e-mail: [email protected] Mária Szabó University of Szeged Bipul Tripathi Applied and Environmental Chemistry Banaras Hindu University 1 Rerrich Béle tér Department of Physics 6720 Szeged, Hungary Lanka, Varanasi, Uttar Pradesh 221005, India e-mail: [email protected] e-mail: [email protected] List of Authors XIX

Ferdinando Tristán López Xuan Wang Shinshu University Rice University Faculty of Engineering, Research Center for Exotic Department of Electrica and Computer Engineering Nanocarbons 6100 Main Str. 380-8553 Nagano, Japan Houston, TX 77005, USA e-mail: [email protected] e-mail: [email protected]

Qiaoling Xu Avesh K. Tyagi Chinese Academy of Sciences Bhabha Atomic Research Centre Key Laboratory of Materials Physics (CAS), Anhui Chemistry Division Key Laboratory of Nanomaterials and Mumbai, 400085, India Nanostructures e-mail: [email protected] Anhui, 230031, China e-mail: [email protected] Géza Tóth University of Oulu Robert J. Young University of Manchester Department of Electrical Engineering Erkki Koiso-Kanttilankatu 3 School of Materials Oxford Road Oulu 90570, Finland e-mail: [email protected].fi Manchester, M13 9PL, UK e-mail: [email protected]

Robert Vajtai Ling Zhang Rice University Carnegie Mellon University Department of Mechanical Engineering and Department of Biomedical Engineering Materials Science 700 Technology Drive 6100 Main Str. Pittsburgh, PA 15219, USA Houston, TX 77005-1827, USA e-mail: [email protected] e-mail: [email protected] Shaopeng Zhang Tsinghua University Sofia M. Vega Díaz Department of Materials Science and Engineering Shinshu University Beijing, 100084, China Research Center for Exotic Nanocarbons e-mail: [email protected] 380-8553 Nagano, Japan e-mail: [email protected] Zhonghua Zhang Shandong University School of Materials Science and Engineering Aravind Vijayaraghavan Jingshi Road 17923 University of Manchester Shandong, 250061, China School of Computer Science e-mail: zh [email protected] Manchester, M13 9PL, UK e-mail: [email protected]

Xiaohui Wang Tsinghua University Department of Materials Science & Engineering Beijing, 100084, China e-mail: [email protected] XXI

Contents

Foreword by Claes-Göran Granqvist...... V Foreword by Neal Lane ...... VII List of Abbreviations ...... XXIX

1 Science and Engineering of Nanomaterials Robert Vajtai ...... 1 1.1 History and Definition of Nanomaterials...... 2 1.2 Formation of Nanomaterials ...... 6 1.3 Properties of Nanomaterials...... 10 1.4 Typical Applications of Nanomaterials ...... 22 1.5 Concluding Remarks ...... 31 1.6 About the Contents of the Handbook ...... 31 References ...... 31

Part A NanoCarbons

2 Graphene – Properties and Characterization Aravind Vijayaraghavan ...... 39 2.1 Methods of Production ...... 42 2.2 Properties ...... 50 2.3 Characterization ...... 58 2.4 Applications ...... 69 2.5 Conclusions and Outlook...... 74 References ...... 74

3 Fullerenes and Beyond: Complexity, Morphology, and Functionality in Closed Carbon Nanostructures Humberto Terrones ...... 83 3.1 Geometry and Structural Features of Fullerenes ...... 85 3.2 Methods of Synthesis of Fullerenes and Proposed Growth Models .... 88 3.3 Physicochemical Properties of Fullerenes...... 90 3.4 Applications of Fullerenes and Beyond ...... 92 3.5 Conclusions...... 99 References ...... 99

4 Single-Walled Carbon Nanotubes Sebastien Nanot, Nicholas A. Thompson, Ji-Hee Kim, Xuan Wang, William D. Rice, Erik H. Hároz, Yogeeswaran Ganesan, Cary L. Pint, Junichiro Kono ...... 105 4.1 History...... 106 4.2 Crystallographic and Electronic Structure ...... 106 XXII Contents

4.3 Synthesis ...... 111 4.4 Optical Properties ...... 115 4.5 Transport Properties ...... 123 4.6 Thermal and Mechanical Properties...... 128 4.7 Concluding Remarks ...... 135 References ...... 135

5 Multi-Walled Carbon Nanotubes Ákos Kukovecz, Gábor Kozma, Zoltán Kónya ...... 147 5.1 Synthesis ...... 148 5.2 Chemistry of MWCNTs ...... 153 5.3 Properties ...... 157 5.4 Selected Applications...... 163 References ...... 169

6 Modified Carbon Nanotubes Aarón Morelos-Gómez, Ferdinando Tristán López, Rodolfo Cruz-Silva, Sofia M. Vega Díaz, Mauricio Terrones...... 189 6.1 Doped Carbon Nanotubes ...... 191 6.2 Defects in Carbon Nanotubes ...... 193 6.3 Nanotube Chemical Functionalization ...... 197 6.4 Properties of Modified Carbon Nanotubes ...... 203 6.5 Characterization of Modified Carbon Nanotubes...... 208 6.6 Applications of Modified Carbon Nanotubes...... 215 6.7 Toxicity and Biocompatibility ...... 218 6.8 Conclusions...... 220 6.9 Outlook and Perspectives...... 221 References ...... 221

7 Carbon Nanofibers Yoong A. Kim, Takuya Hayashi, Morinobu Endo, Mildred S. Dresselhaus ..... 233 7.1 Similarity and Difference Between Carbon Fibers and Carbon Nanofibers ...... 234 7.2 Growth and Structural Modifications of Carbon Nanofibers ...... 238 7.3 Applications of Carbon Nanofibers...... 251 7.4 Conclusions...... 257 References ...... 258

8 Nanodiamonds Olga A. Shenderova, Suzanne A. Ciftan Hens ...... 263 8.1 Stability of Diamond at the Nanoscale...... 264 8.2 Types of Nanodiamonds and Methods of Nanodiamond Synthesis ... 267 8.3 Detonation Nanodiamond Processing and Modification ...... 278 8.4 Fluorescent Nanodiamonds ...... 284 Contents XXIII

8.5 Applications of Nanodiamond Particles ...... 285 8.6 Future Directions of Production and Applications...... 292 References ...... 293

Part B NanoMetals

9 Noble Metal Nanoparticles Theruvakkattil S. Sreeprasad, Thalappil Pradeep ...... 303 9.1 Historical Perspective of Gold and Silver NPs ...... 304 9.2 Diverse Nanostructures ...... 307 9.3 Common Synthetic Routes for the Preparation of Noble Metal NPs ...... 311 9.4 Properties of Noble Metal Nanoparticles ...... 322 9.5 Postsynthetic Tuning of Properties ...... 324 9.6 Functionalized Metal NPs ...... 343 9.7 Applications of Gold and Silver Nanoparticles ...... 347 9.8 New Gold and Silver Materials – Quantum Clusters ...... 363 9.9 Conclusions...... 366 References ...... 367

10 Nanostructures of Common Metals Melinda Mohl, Krisztián Kordás ...... 389 10.1 Post-Transition Metals...... 390 10.2 Transition Metals...... 392 10.3 Concluding Remarks ...... 398 References ...... 399

11 Alloys on the Nanoscale Giovanni Barcaro, Alfredo Caro, Alessandro Fortunelli ...... 409 11.1 Concepts and Principles ...... 411 11.2 Preparation and Synthesis ...... 413 11.3 Characterization of Nanoparticles and Nanoalloys...... 417 11.4 Properties ...... 424 11.5 Nanostructured Bulk Alloys ...... 450 11.6 Applications ...... 457 11.7 Concluding Remarks ...... 458 References ...... 459

12 Magnetic Nanostructures: Synthesis, Properties, and Applications Shashwat Shukla, Pratap Kumar Deheri, Raju V. Ramanujan...... 473 12.1 Background ...... 474 12.2 Atomic Origin of Magnetism ...... 475 12.3 Magnetic Length Scales and Origin of Nanomagnetic Behavior...... 478 12.4 Magnetic Nanostructures ...... 483 12.5 Conclusions...... 505 References ...... 506 XXIV Contents

Part C NanoCeramics

13 Nanocrystalline Functional Materials Rakesh Shukla, Dimple P. Dutta, Jayshree Ramkumar, Balaji P. Mandal, Avesh K. Tyagi ...... 517 13.1 Synthesis Methods...... 518 13.2 Optical Properties of Oxide Nanomaterials ...... 524 13.3 Sorbent Properties of Oxide Nanomaterials ...... 532 13.4 Catalytic Properties of Oxide Nanomaterials ...... 536 13.5 Oxide Nanomaterials in Ionics...... 538 13.6 Conclusions...... 541 References ...... 542

14 Piezoelectric Nanoceramics Xiaohui Wang, Shaopeng Zhang, Longtu Li ...... 553 14.1 Introduction to BSPT ...... 554 14.2 Synthesis of BSPT Nanopowders via Sol–Gel Method ...... 555 14.3 Sintering of BSPT Nanoceramics...... 556 14.4 Grain Size Effect on the Properties of BSPT Ceramics ...... 563 14.5 Summary ...... 567 References ...... 568

15 Graphite Oxide Wei Gao ...... 571 15.1 Synthesis of Graphite Oxide ...... 572 15.2 Characterization, Chemical Structure and Properties...... 576 15.3 Applications ...... 589 15.4 Concluding Remarks ...... 592 References ...... 592

16 Compound Crystals Roi Levi, Maya Bar-Sadan, Reshef Tenne ...... 605 16.1 Nanostructures...... 605 16.2 Synthetic Methods...... 608 16.3 Physical Properties ...... 618 16.4 Applications ...... 628 16.5 Conclusions...... 630 References ...... 631

17 Growth of Nanomaterials by Screw Dislocation Fei Meng, Stephen A. Morin, Song Jin ...... 639 17.1 Classical Crystal Growth Theories ...... 640 17.2 Theories for Screw-Dislocation-Driven Growth of Nanomaterials ..... 642 17.3 Structural Characterization of these Nanomaterials ...... 645 17.4 Generality of Dislocation-Driven Nanomaterial Growth ...... 649 17.5 Rational Growth of Dislocation-Driven Nanomaterials – General Strategies ...... 658 Contents XXV

17.6 Applications ...... 659 17.7 Summary and Perspectives ...... 660 References ...... 661

18 Glasses on the Nanoscale Hellmut Eckert, Sidney J.L. Ribeiro, Silvia H. Santagneli, Marcelo Nalin, Gael Poirier, Younès Messaddeq ...... 665 18.1 Studying Medium-Range Order in Glasses and Nanoceramics ...... 666 18.2 Nanoceramics ...... 676 18.3 Perspectives and Concluding Remarks ...... 684 References ...... 685

Part D NanoComposites

19 Carbon in Robert J. Young, Libo Deng, Lei Gong, Ian A. Kinloch...... 695 19.1 Materials Basics...... 695 19.2 Composites...... 702 19.3 Graphene Composites ...... 716 19.4 Conclusions...... 722 References ...... 722

20 Dispersions Krisztián Kordás, Jarmo Kukkola, Géza Tóth, Heli Jantunen, Mária Szabó, András Sápi, Ákos Kukovecz, Zoltán Kónya, Jyri-Pekka Mikkola...... 729 20.1 Stabilization of Nanoparticle Dispersions ...... 730 20.2 Nanoparticle Dispersion in Practice ...... 734 20.3 Dispersions of Carbon Nanomaterials ...... 745 20.4 Drying Dispersions on Surfaces ...... 752 20.5 Concluding Remarks ...... 758 References ...... 758

Part E Nanoporous Materials

21 Nanoporous Metals Yi Ding, Zhonghua Zhang ...... 779 21.1 Preparation of Nanoporous Metals ...... 779 21.2 Properties of Nanoporous Metals...... 789 21.3 Applications ...... 808 21.4 Concluding Remarks and Prospects ...... 810 References ...... 811

22 Zeolites István Pálinkó, Zoltán Kónya, Ákos Kukovecz, Imre Kiricsi...... 819 22.1 Common Zeolite Frameworks ...... 822 XXVI Contents

22.2 Zeolite and Zeolite-Related Molecular Sieves ...... 823 22.3 Natural Zeolites: Occurrence and Formation...... 825 22.4 Methods of Identification and Characterization ...... 828 22.5 Synthesis of Zeolitic Materials ...... 830 22.6 Ion Exchange, Sorption, and Diffusion in Microporous Materials ...... 836 22.7 Acid–Base Properties of Zeolites...... 841 22.8 Stability and Modification of Zeolite Structures ...... 843 22.9 Zeolites as Catalysts ...... 846 22.10 Some Special Applications of Zeolites ...... 848 22.11 Conclusions...... 850 References ...... 850

23 Porous Anodic Aluminum Oxide QiaolingXu,GuowenMeng...... 859 23.1 Background ...... 859 23.2 Preparation of AAO Templates ...... 860 23.3 Nanostructures Constructed in AAO Templates ...... 862 23.4 Conclusions and Outlook...... 879 References ...... 879

24 Porous Silicon Paolo Bettotti ...... 883 24.1 Basics of Porous Silicon Electrochemistry and Formation Models ...... 884 24.2 Other Etching Methods ...... 886 24.3 Porous Silicon Structural Properties ...... 887 24.4 Light Emission from Porous Silicon ...... 890 24.5 Thermal and Electrical Properties ...... 891 24.6 The Role of the Surface ...... 891 24.7 Applications of Porous Silicon ...... 892 24.8 Conclusions...... 897 References ...... 898

Part F Organic and Bionanomaterials

25 Organic Nanomaterials Huanli Dong, Wenping Hu ...... 905 25.1 Preparation/Synthesis of Organic Nanomaterials ...... 905 25.2 Properties of Organic Nanomaterials ...... 910 25.3 Applications ...... 925 25.4 Concluding Remarks ...... 930 References ...... 932

26 Nanocomposites as Bone Implant Material Vinod Kumar, Bipul Tripathi, Anchal Srivastava, Preeti S. Saxena...... 941 26.1 The Quest for a Suitable Bone Implant...... 942 26.2 Bone ...... 942 Contents XXVII

26.3 Existing/Conventional Bone Implant Materials and Their Shortcomings ...... 944 26.4 Major Challenges with Existing/Conventional Implant Materials...... 949 26.5 Nanotechnology and Tissue Engineering...... 949 26.6 Future Perspectives ...... 965 References ...... 965

27 Nanofiber Biomaterials Rachelle N. Palchesko, Yan Sun, Ling Zhang, John M. Szymanski, Quentin Jallerat, Adam W. Feinberg...... 977 27.1 Methods of Production ...... 980 27.2 Properties of Nanofiber Biomaterials ...... 986 27.3 Characterization of Nanofiber Biomaterials...... 993 27.4 Applications ...... 999 27.5 Conclusions and Outlook...... 1005 References ...... 1006

Part G Applications and Impact

28 Nanostructured Materials for Energy-Related Applications Arava L.M. Reddy, Sundara Ramaprabhu...... 1013 28.1 Energy-Related Carbon Nanotubes ...... 1013 28.2 CNTs as Support Material for Electrocatalysts in PEMFC ...... 1016 28.3 CNTs as Supercapacitor Electrode Materials ...... 1023 References ...... 1032

29 Nanomaterials in Civil Engineering Jaesang Lee, Seunghak Lee, Eunhyea Chung, Vincent C. Reyes, Shaily Mahendra ...... 1039 29.1 Applications of MNMs in Construction ...... 1041 29.2 Environmental Release of MNMs Used in Construction ...... 1047 29.3 Potential Adverse Biological Impacts and Toxicity Mechanisms ...... 1049 29.4 Mitigation of Environmental and Health Impacts ...... 1052 29.5 Conclusions...... 1054 References ...... 1055

30 Plasmonic Nanomaterials for Renat R. Letfullin, Thomas F. George ...... 1063 30.1 Introduction ...... 1063 30.2 Nanooptics – Lorenz–Mie Formalism ...... 1064 30.3 Optical Properties of Gold Nanoparticles in Biological Media...... 1065 30.4 Kinetics of Heating and Cooling of Nanoparticles ...... 1067 30.5 Spatial Distribution of Temperature Fields Around the Nanoparticle. 1076 30.6 New Dynamic Modes in Selective Plasmonic Nanotherapy ...... 1083 References ...... 1095 XXVIII Contents

31 Carbon Nanotube Membrane Filters Anchal Srivastava, Saurabh Srivastava, Kaushik Kalaga ...... 1099 31.1 Types of Filtration ...... 1100 31.2 Mechanisms of Filtration ...... 1101 31.3 Carbon Nanotube Membrane Filters ...... 1102 31.4 Future Research Perspectives ...... 1112 References ...... 1112

32 Nanomaterial Toxicity, Hazards, and Safety Zuzanna A. Lewicka, Vicki L. Colvin...... 1117 32.1 Engineered Nanomaterials – General Overview...... 1118 32.2 Occurrence of Engineered Nanoparticles in the Environment ...... 1119 32.3 Effects of Nanoparticles on Organisms ...... 1120 32.4 Nanoparticle Physicochemical Characteristics of Relevance for Toxicology...... 1124 32.5 Special Case – Sunscreens...... 1130 32.6 Conclusions...... 1132 References ...... 1133

Acknowledgements...... 1143 About the Authors...... 1145 Detailed Contents...... 1163 Subject Index...... 1181 XXIX

List of Abbreviations

α-SMA α-smooth muscle actin ARPES angle-resolved photoemission p-NP p-nitrophenol spectroscopy 0-D zero-dimensional ASTM American Society for Testing and 1-D one-dimensional Materials 2-D two-dimensional ATQD N-(4-aminophenyl)-N-(4-(3-triethoxy- 2-PAM 2-pyridine-aldoxime methiodide silyl-propyl-ureido)phenyl-1,4-quinon- 2Q double-quantum enediimine) 3-D three-dimensional ATP adenosine-5-triphosphate 3Q triple-quantum ATRP atom-transfer radical polymerization 4Hop 4-hexadecyloxyphenyl AWWA American Water Works Association A B

AA ascorbic acid BASF Badische Anilin und Soda Fabrik AAM anodized aluminum membrane bcc body-centered cubic AAO anodic aluminum oxide BCF Burton–Cabrera–Frank AAO anodized aluminum oxide BCP biphasic calcium phosphate AAS atomic absorption spectroscopy BDAC benzyldimethylammoniumchloride Ab antibody BDNF brain-derived neurotrophic factor AC alternating current BEP Brønsted–Evans–Polanyi relations Acac acetylacetone BES Office of Basic Energy Sciences ACNT aligned carbon nanotube BET Brunauer–Emmett–Teller ACQ aggregation-caused quenching BF bright field AChE acetylcholine esterase BFGF basic fibroblast growth factor ACP amorphous calcium phosphate BG back-gate AD arc discharge BHJ bulk heterojunction AEE aggregation-enhanced emission bioMEMS biological microelectromechanical AES Auger electron spectroscopy system AES 3-(2-aminoethylaminopropyl)trimethoxy- BMG bulk metallic glass silane BN boron nitride AFC alkaline fuel cell BOM bubble overlapping mode AFC antiferromagnetically coupled BP buckypaper AFM BPEA 9,10-bis(phenylethynyl)anthracene AIE aggregation-induced emission BS black silicon AIEE aggregation-induced enhanced BSA bovine serum albumina emission BSI British Standards Institution ALD atomic layer deposition BSPP bis(p-sulfonatophenyl) phenylphosphine AlPO aluminophosphate dihydrate dipotassium AM alveolar macrophage BT barium titanate anti-EGFR anti-epidermal growth factor receptor BT benzenethiol AOC aromatic organic compounds BTCP β-tricalcium phosphate APC antigen-presenting cell APES aminopropyltrimethoxysilane C APPES ambient pressure photoelectron spectroscopy C16TAB hexadecyl trimethyl ammonium bromide APS 3-aminopropyltrimethoxysilane C-PANI conductive camphorsulfonic acid-doped APT atom probe tomography emeraldine PANI APTES (aminopropyl) triethoxysilane C3DT 1,3-Propanedithiol APTS 3-aminopropyltriethoxysilane CA contact angle AR analytical reagent CALPHAD calculation of phase diagrams AR aspect ratio CAM cluster aggregation mode XXX List of Abbreviations

CBED convergent-beam electron diffraction DAPI 4,6-diamidino-2-phenylindole CBEV coordination-dependent bond-energy DAPRAL copolymer of maleic anhydride and variation α-olefin CCDB Cambridge crystallographic data base DBR distributed Bragg mirror CCG chemically converted graphene DC dendritic cell CCT correlated color temperature DC direct current CCVD catalytic chemical vapor deposition DCE 1,2-dichloroethane CD cyclodextrin DD-PTCDI N,N-di(dodecyl)-perylene-3,4,9,10- CFR continuous flow reactor tetracarboxylic diimide CHP cyclohexylpyrrolidone DDA discrete dipole approximation CHT chymotrypsin DDAB didecyldimethylammonium bromide CIE International Commission on DDC N,N-dicyclohexylcarbodiimide Illumination DEFC direct ethanol fuel cell CIP current in the plane DEG diethylene glycol CMG chemically modified graphene DF defluoridation capacity CMOS complementary DF density function metal–oxide–semiconductor DFAC direct formic acid fuel cell CMP chemical–mechanical planarization DFT density functional theory CNF carbon nanofiber DFTB density functional tight binding CNM carbon nanotube membrane DGU density-gradient ultracentrifugation CNT carbon nanotube DI deionized CN-TFMBE 1-cyano-trans-1,2-bis(3,5-bis-trifluoro- DIC differential interference contrast methyl-biphenyl)ethylene DLC diamond-like carbon CO cuboctahedron DLS dynamic light scattering COD 1,5-cyclooctadiene DLVO Derjaguin–Landau–Verwey–Overbeek COLI collagen I DMA dimethylamide COLIV collagen IV DMEU 1,3-dimethyl-2-imidazolidinone COST Cooperation in Science and Technology DMF dimethylformamide COSY correlation spectroscopy DMFC direct methanol fuel cell COT 1,3,5-cyclooctatriene DMPO 5,5-dimethyl-pyrroline N-oxide cp close packed DMSA dimercaptosuccinic acid CP coherent phonon DMSO dimethyl sulfoxide CP cross polarization DNA deoxyribonucleic acid CPP conduction perpendicular to plane DND detonation nanodiamond CPP current perpendicular to the plane DOS density of states CPS collected photo signal DOX doxorubicin CS cross section dpa displacements per atom CS-PCL chitosan-graft-PCL DPPTE 1,2-dipalmitoyl-sn-glycero-3-phospho- CSA chemical shift anisotropy thioethanol CSP colloidal silver preparation DQ double quantum CT charge transfer DR draw ratio CTA+ cetyl-triamine cation DRG dorsal root ganglion CTA cetyltrimethylammonium DRIFT diffuse reflectance infrared CTAB cetyltrimethylammonium bromide Fourier-transform CV crystal violet DSC differential scanning calorimetry CV cyclic voltammetry DT decanethiol CVD chemical vapor deposition DTAB dodecyltrimethylammonium bromide CW continuous-wave DTE desaminotyrosyl-tyrosine ethyl ester CuPC phthalocyanine DWCNT double-walled carbon nanotube CuTCNQ copper tetracyanoquinodimethane DWNT double-walled nanotubes Dox doxorubicin D E D4R double four ring DAAQ 1,5-diaminoanthraquinone ECD electrochemical deposition DAFC direct alcohol fuel cell ECDL electrochemical double layer List of Abbreviations XXXI

ECELL environmental cell FIPOS full isolation by porous oxidized silicon ECM extracellular matrix FIT fluctuation-induced tunneling ECP electronically conducting polymer FITC fluorescein isothiocyanate ECSA electrochemically active surface area FLG few-layer graphene ED electrodialysis FMR ferromagnetic resonance ED electron diffraction FN fibronectin EDAX energy dispersive analysis FND fluorescent carboxylated HPHT ND EDC 1-ethyl-3-(3-dimethylaminopropyl)- FND fluorescently enhanced ND carbodiimide fpRFDR finite-pulse radio frequency-driven EDL electrical double layer recoupling EDLC electric double-layer capacitor EDS energy-dispersive x-ray spectroscopy G EDTA ethylenediaminetetraacetic acid EDX energy-dispersive x-ray spectroscopy GMR giant magnetoresistance EELS electron energy-loss spectroscopy GN gold nanoparticle EFM electrostatic force microscopy GNC gold nanoparticle cluster EG evaporated gold GNP gold nanoparticle EIS electrochemical impedance spectroscopy GNP graphite nanoplatelet EL electroluminescence GNR gold nanorod ELISA enzyme-linked immuno sorbent assay GNR graphene nanoribbon EM electromagnetic GO graphene oxide EMI electromagnetic interference GOX glucose oxidase EOF electroosmotic flow GSH glutathione EPA Environmental Protection Agency GTBMD generalized tight-binding molecular EPR electron paramagnetic resonance dynamics EPS extracellular polymeric substance EQE external quantum efficiency H ESC embryonic stem cell ESR electron spin resonance HA humic acid ESR equivalent series resistance HAADF high-angle annular dark field ETEM environmental TEM HATU 2-(7-aza-1H-benzotriazole-l-yl)-1,1,3,3,- EXAFS extended x-ray absorption fine structure tetramethyluronium hexafluorophosphate ElAP(S)O element aluminophosphosilicate HAZ heat-affected zone EPITH epithelial cells HC hexagonal channel HCCN highly curved carbon nanostructure F HCI highly charged ion hcp hexagonal close packed f-SWCNT functionalized SWCNT HDA hexadecylamine FABMS fast atom bombardment mass HDD 1,2-hexadecanediol spectroscopy HDDR hydrogenation–decomposition– FBI Federal Bureau of Investigation desorption–recombination FBR fluidized bed reactor HDS hydrodesulfurization fcc face-centered cubic HDT hexadecanethiol fct face-centered tetragonal HEPES 4-(2-hydroxyethyl)-1-piperazineethane- FDA Food and Drug Administration sulfonic acid FEB ferrocene/ethanol/benzylamine HEV hybrid electric vehicle FES fluctuation-enhanced sensing HF hydrofluoric acid FESEM field emission scanning electron HG hydrazinium graphene microscope HIV human immunodeficiency virus FET field-effect transistor HL60 human promyelocytic leukemia FF fill factor HMDA hexamethylenediamine FFT fast Fourier transform HMO hydrous dioxide FGO functionalized GO HMOG heavy metal oxide glass FIB fibrinogen HMTA hexamethylenetetramine FIB focused ion beam HNS hot neutron source XXXII List of Abbreviations

HOMO highest occupied molecular orbital ISO International Standards Organization HOPG highly oriented pyrolytic graphite ITO indium tin oxide HP Hall–Petch IZA International Zeolite Association HPA hexylphosphonic acid HPC-Py pyrene-labeled hydroxypropyl cellulose K HPHT high-pressure high-temperature HPMC Hydroxypropylmethyl cellulose KE Kirkendall effect HPSMAP poly(styrene-co-maleic anhydride) KK Kramers–Kronig carrying pyrene HSMA hydrolyzed poly(styrene-co-maleic) L anhydrite h-PSMA hydrolyzed-poly(styrene-alt-maleic LA longitudinal acoustic anhydride) LAM laminin HREM high-resolution electron microscopy LB Langmuir–Blodgett technique HRN helical rosette nanotube LB94 van Leeuwen–Baerends HRP horseradish peroxidase LBL layer-by-layer HRSEM high-resolution scanning electron LCD liquid-crystal display microscope LDA local density approximation HRTEM high-resolution transmission electron LDOS local density of states microscopy LED light-emitting diode HSA human serum albumin LEED low energy electron diffraction hSKMC human skeletal muscle cell LIB lithium-ion battery HTT heat treatment temperature LMP Larson–Miller plot HWHM half-width at half-maximum LN less noble HiPCO high-pressure carbon monoxide LPM large-pore mordenite LPS lipopolysaccharide I LSC limbal stem cells LSP longitudinal surface plasmon IANH International Alliance for NanoEHS LSPR localized surface plasmon resonance (environment, health, safety) LTA Linde type A IC integrated circuit LUMO lowest unoccupied molecular orbital ICP inductively coupled plasma LYM lymphocytes ICP-MS inductively coupled plasma mass spectrometry M IE immersion–electrodeposition IF immunofluorescence M metalloid IF inorganic -like nanoparticle MA mechanical alloying iFF isotactic polypropylene MAE magnetic anisotropy energy IFSS interfacial shear strength MALDI-TOF matrix-assisted laser desorption/ Ig immunoglobulin ionization-time of flight IgG immunoglobulin G MAPO metalaluminophosphate Ih icosahedron MAPSO metalaluminophosphosilicates IKVAV laminin derived self-assembling peptide MAS magic angle spinning I K VAV- PA I K VAV p o l y a c r y l a m i d e MBE molecular beam epitaxy IL interleukin MC metal cluster IL ionic liquid MCFC molten carbonate fuel cell IMR intramolecular rotation MCL maximum contamination limit INCO International Nickel Company MCS ethylene glycol monomethyl ether INT inorganic nanotube MD molecular dynamics IP iminopyrrole MDA malondialdehyde IPCE incident photon to charge carrier MDA mercaptodecanoic acid efficiency MEA membrane electrode assembly iPSC induced pluripotent stem cell MEMS microelectromechanical system IR infrared MF mesoflower ipr isolated pentagon rule MF microfiltration List of Abbreviations XXXIII

MFC microbial fuel cell NGF nerve growth factor MFI melt–flow index NHAP nanohydroxyapatite MFM magnetic force microscopy nHAp nanohydroxyapatite particle MGM metal-graphite multilayer n-HApC nanohydroxyapatite/chitosan MHAP micron particulate hydroxyapatite NHS N-hydroxysuccinimidyl ester ML monolayer NIOSH National Institute for Occupational Safety MN more noble and Health MNM manufactured nanomaterials NIR near infrared MNPM metallic nanoporous material NM noble metal MO methyl orange NMP N-methyl-pyrrolidone MOCVD metalorganic chemical vapor deposition Nmpd N-methylpyridinium MOF metal-organic framework Nmpr N-methylpyrrole MOKE magneto-optical Kerr effect NMR nuclear magnetic resonance MPB morphotropic phase boundary NO-IF nanooctahedra-IF MPC monolayer-protected cluster NP nanoparticle MPCF mesophase pitch-based carbon fiber NPG nanoporous graphite MPS mercaptopropyltrimethoxysilane NPG/GC NPG supported by glassy carbon MPTMS mercaptopropyltrimethoxysilane electrode MR magnetic resonance NPGC nanoporous gold composite MRAM magnetic random-access memory NPM nanoporous metal MRI magnetic resonance imaging NPNT nanoporous nanotube mRNA messenger RNA NPS nanoporous silver MRR material removal rate NR nanorod MRSw magnetic relaxation switching NSC neural stem cell MSA mercaptosuccinic acid NSM nanostructured materials MSC mesenchymal stem cell NT nanotube MSE mercurous sulfate electrode NTS nanostructured transformable steel MTBD [7-methyl-1,5,7-triazabicyclo[4.4.0]dec- NV nitrogen-vacancy 5-ene][bis(perfluoroethylsulfonyl)imide] NW MWCNT multiwalled carbon nanotube MWNT multiwalled nanotubes O

N O/F oxidant-to-fuel OCP open-circuit potential NaBBS sodiumbutylbenzene sulfonate OCT optical coherence tomography NaDDBS sodium dodecylbenzene sulfonate ODA octadecylamine NADH nicotinamide adenine dinucleotide ODE octadecene NaOBS sodium octylbenzene sulfonate ODF orientation distribution function NaPSS polystyrene sulfonate sodium salt ODPA octadecylphosphonic acid NBE near-band-edge ODS octadecyltrimethoxysilane NC nanocrystalline ODS oxide dispersion strengthened nc-AFM noncontact AFM OER oxygen evolution reaction ND nanodiamond OFET organic field-effect transistor NDO ozone-modified nanodiamond Oh octahedron ND-PTCDI N,N-di(nonyldecyl)-perylene-3,4,9,10- OL optical-limiting tetracarboxylic diimide OLC onion-like carbon NEMS nanoelectromechanical system OLED organic light-emitting diode NEUT neutrophils OPD O-phenylenediamine NEXAFS near-edge x-ray absorption fine structure OPH organophosphorus hydrolase NF nanofeatures OPS oxidized PS NF nanofiltration OPV organic photovoltaic NFA nanostructured ferritic alloy ORR oxygen reduction reaction NG natural highly-oriented pyrolytic OSN organic solvent nanofiltration graphite OTM one-temperature model XXXIV List of Abbreviations

P PGLA copolymer of PGA and PLLA PGM platinum group metal P3HT poly(3-hexylthiophene) pIh polyicosahedron P3OT poly(3-octylthiophene) PIPAAm responsive poly(N-isopropylacrylamide) PA peptide amphiphile PL photoluminescence PA-6 prepared a nylon-6 PL-PEG phospholipid polyethylene glycol PAA poly(acrylic acid) PLA poly-ethylene oxide PABS polyaminobenzene sulfonic acid PLA pulsed laser ablation PAFC phosphoric acid fuel cell PLE photoluminescence excitation PAGE polyacrylamide gel electrophoresis PLGA poly(lactic-co-glycolic) acid PAH polycyclic aromatic hydrocarbon PLLA poly(l-lactic) acid PAN polyacrylonitrile PM dipropylene glycol monomethylether PANI polyaniline PMMA poly-methyl methacrylate PATS polythiophene derivatives PMN-PT PbMg1/3Nb2/3 O3-PbTiO3 PBO poly(p-phenylene benzobisoxazole) PmPV poly(m-phenylenevinylene-co-2,5- PBS phosphate buffered saline dioctoxy-p-phenylenevinylene) PC pentagonal column PN phosphorus-nitrogen PC photonic crystal PNIPAm poly(N-isopropyl acrylamide) PC polycarbonate PNP plasmonic PC principal component PP polypropylene PCA principal component analysis pp peak-to-peak PCB polychlorinated biphenyl PPCP 1,2,3,4,5-pentaphenyl-1,3- PCE power conversion efficiency cyclopentadiene PCF photonic crystal fiber PT PbTiO3 PCL poly(ε-caprolactone) PPE poly-p-phenyleneethynylene PCL-G PCL-gelatin PPF propylene fumarate PDDA poly(diallyldimethyl)ammonium PPTA poly phenylene terephthalamide chloride PPV poly-p-phenylenevinylene PDDP 1-phenyl-3-((dimethylamino)styryl)-5- PPy polypyrrole ((dimethylamino)phenyl)-2-pyrazoline PRR pattern recognition receptor PDEAEMA poly(2-diethylaminoethyl methacrylate) PS polystyrene PDGF platelet-derived growth factor PS porous silicon PDLC polymer-dispersed liquid-crystal PS-PFS poly(styrene-b-ferrocenyldimethylsilane) PDMS polydimethylsiloxane PSD photo signal detector PDOS phonon density of states PSS poly(sodium 4-styrenesulfonate) PE photoelectron PSS polystyrene sulfonate PE polyethylene PSU polysulfonate PEC photoelectrochemical PSU polysulfone PECVD plasma-enhanced CVD PSVPh poly(styrene-co-vinyl phenol) PEDOT poly(3,4-ethylenedioxythiophene) Pt-NPG platinum-decorated nanoporous gold PEEk produced poly(ether ether ketone) Pt-NPGL platinum-plated nanoporous gold leaf PEG polyethylene glycol PTCDI N,N-di(propoxyethyl)perylene-3,4,9,10- PEI polyethyleneimine tetracarboxylic diimide PEL permissible occupational exposure PTCE track-etched polycarbonate limit PTFE polytetrafluoroethylene PEMFC proton exchange membrane fuel cell PU polyurethane PEN Project on Emerging PV pervaporation PEO poly(ethylene oxide) PV photovoltaic PES potential energy surface PVA polyvinyl alcohol PET polyethylene terephthalate PVC polyvinylchloride PFG pulsed-field-gradient PVD physical vapor deposition PFM piezoelectric force microscopy PVDF polyvinyldifluoride PG PCL–gelatin PVP polyvinyl pyrrolidone PG proteoglycan PW plane wave PGA poly(glycolic acid) pzc point of zero charge List of Abbreviations XXXV

PZN-PT PbZn1/3Nb2/3O3-PbTiO3 SBU secondary building unit PZT Pb(Zr,Ti)O3 SC simple cubic SC sodium cholate Q SCC stress corrosion cracking SCE saturated calomel electrode QC quantum cluster SCR space-charge region QD quantum dot SD standard deviation QEXAFS quick EXAFS SDBS sodium dodecylbenzene sulfate QHE quantum Hall effect SDCH samaria-doped ceria SDS sodium dodecyl sulfate R SEC size exclusion chromatography SEI solid–electrolyte interphase R6G rhodamine 6G SEIRA surface-enhanced infrared absorption RA right angle SEM scanning electron microscopy RBM radial breathing mode SES scanning electron spectroscopy RCF rabbit corneal fibroblast SERS surface-enhanced Raman scattering RE rare-earth SET single-electron transistor rebar reinforcement bar SF silk fibroin REDOR rotational echo double resonance SFF solid freedom fabrication RF radio frequency SFG sum-frequency generation RFDR radiofrequency-driven recoupling SFM scanning force microscopy RFID radiofrequency identification SGS spaced superconducting electrode RGB red green SHE standard hydrogen electrode RGD Arg-Gly-Asp SIM structured illumination microscopy RGO reduced graphene oxide SIMS secondary-ion mass spectrometry rhBMP-2 recombinant human bone morphogenic siRNA silenced RNA protein-2 SL superlattice RHE reversible hydrogen electrode SLS solution–liquid–solid RIA radioimmuno assay SMA shape-memory alloy RIE reactive-ion etching SMAD solvated metal atom dispersion RIR restriction of intramolecular rotation SNR signal-to-noise ratio RJS rotary jet spinning SOCT sodium octanoate RKKY Rudermann–Kittel–Kasuya–Yosida SOFC solid oxide fuel cell RM reactive milling SOI silicon-on-insulator RMS microscale surface roughness SP surface plasmon RNA ribonucleic acid SP-STM spin-polarized scanning tunneling RO reverse osmosis microscopy ROS reactive oxygen species SPM scanning probe microscopy RPC retinal progenitor cells SPM small-pore mordenite RRR replacement reaction SPP surface plasmon polariton RRS resonant Raman scattering SPR surface plasmon resonance RT room temperature SPS spark plasma sintering RT-PCR real-time polymerase chain reaction SQ single quantum R&D research and development SQUID superconducting quantum interference device S SRNF solvent resistant nanofiltration SS stainless steel S–W Stone–Wales SSA specific surface area S/L solid/liquid SSNMR solid-state nuclear magnetic resonance SA sliding angle STEM scanning transmission electron SA solar ablation microscopy SAED selected-area electron diffraction STM scanning tunneling microscopy SAM self-assembled monolayer STORM stochastic optical reconstruction SANS small -angle neutron scattering microscopy SAPO silicoaluminophosphate STS scanning tunneling spectroscopy SAXS small-angle x-ray scattering SWCNT single-walled carbon nanotube XXXVI List of Abbreviations

SWNH single-wall nanohorn U SWNT single-walled nanotube SXRD surface x-ray diffraction UF ultrafiltration ShdH Shubnikov–de Haas UHP ultrahigh pressure Si-MEMS silicon microelectromechanical system UHV ultrahigh vacuum Si-nc silicon nanocrystal UNCD ultrananocrystalline diamond UPD underpotential deposition T UV ultraviolet UV-VIS ultraviolet-visible TA thioctic acid UVR ultraviolet radiation TA transverse acoustic TAMRA tetramethylrhodamine V TASA template-assisted self-assembly TCNQ tetracyanoquinodimethane vdW van der Waals TCO transparent conductive oxide VGCF vapor-grown carbon fiber TDABr tetradodecylammonium bromide VHS van Hove singularity TDDFT time-dependent density-functional- VLS vapor–solid–liquid theory VPC vacuum pyrolysis/carbothermal TDPA tetradecylphosphonic acid VRH variable range hopping TE element VS vapor–solid TEG tetra(ethylene glycol) VSFG vibrational sum-frequency generation TEM transmission electron microscopy VSM vibrating sample magnetometry TEOS tetraethyl orthosilicate VSS vapor–solid–solid TEP thermoelectric power Van vancomycin TFT thin-film transistor TG top gate W TGA thermogravimetric analysis TGA thioglycolic acid WAXD wide angle x-ray diffraction TGF-β transforming growth factor WC carbide THF tetrahydrofuran WG waveguide THPC tetrakismethyl)phosphonium chloride WHO World Health Organization TIC toxic industrial chemical TIPS thermally induced phase separation X TL transition-metal element TMAH tetramethylammonium hydroxide XANES x-ray absorption near-edge spectroscopy TMR tunnel magnetoresistance XAS x-ray absorption spectroscopy TNF-α tumor necrosis factor xc exchange–correlation TNT 2-methyl-1,3,5-trinitrobenzene XPS x-ray photoelectron spectroscopy TO truncated octahedron XRD x-ray diffraction TOAB tetraoctylammonium bromide TOF turnover frequency Y TOP trioctylphosphine TOPO trioctylphosphine oxide YAB YAl 3(BO3)4 TPA tetrapropylammonium YAM Y 4Al2O9 TPD temperature programmed desorption Y-CNT Y-shaped carbon nanotube TPI 2,4,5-triphenylimidazole TPL two-photon luminescence Z TPP 1,3,5-triphenyl-2-pyrazoline TSP transverse surface plasmon ZAP zone axis pattern TSW Thrower–Stone–Wales ZHDS hydroxydodecylsulfate TTCP tetracalcium phosphate ZHS zinc hydroxysulfate TWC three-way catalyst ZLC zero-length-column ThT thioflavin T ZSM zeolite sieve of molecular porosity 1

1. Science and Engineering Introduction Scienceof Nanomaterials and E Robert Vajtai

1.1 History and Definition of Nanomaterials . 2 Nanomaterials possess different properties com- 1.1.1 History of Nanomaterials...... 2 pared with macroscopic (bulk) materials built up 1.1.2 Definition of Nanoscale from the same atoms or compounds. The produc- and Nanomaterials...... 4 tion routes, characterization, and applications of 1.2 Formation of Nanomaterials...... 6 materials sized on the nanometer scale also differ from the bulk. 1.3 Properties of Nanomaterials ...... 10 In this chapter we define nanomaterials and 1.3.1 Morphology of Nanomaterials...... 10 1.3.2 Bonds and Structures...... 10 the specific science that describes them, and collect 1.3.3 Mechanical Properties examples of synthesis and applications of a range of Nanomaterials ...... 12 of these materials; we also dedicate an extended 1.3.4 Electrical, Magnetic, and Optical part of the chapter to material properties, e.g., Properties ...... 14 morphology, mechanical, electrical, magnetic, 1.3.5 Thermal Properties ...... 18 and optical properties. In both the general and 1.3.6 Chemical Properties, Reactivity, specific parts of the chapter, emphasis is placed on and Functionalization...... 20 the differences from the bulk phase of the same 1.3.7 Behavior of Nanomaterials material and, if possible, the size dependence of in Corrosive Environments...... 22 the various material properties. 1.4 Typical Applications of Nanomaterials .... 22 Following the handbook format, the chapter 1.4.1 Catalysts and Catalyst Templates .... 26 is concise and covers various common properties 1.4.2 Energy Conversion and Storage ...... 26 of nanomaterials and correlations with which 1.4.3 Sensors Based on Nanomaterials.... 28 nanoscientists work; however, we insert specific 1.5 Concluding Remarks ...... 31 parts which have some curiosity value, as well as 1.6 About the Contents of the Handbook ..... 31 several aspects of our own research. References ...... 31

The techniques, recipes, and later science of making and atoms built into a cage similar to carbon fullerene), and testing tools, artistic objects, and weapons are the main synthesis should follow precisely this design. criteria for the characterization and classification of hu- In this chapter, we collect two sorts of information man historical ages – from the Stone Age through the about nanomaterials. To a lesser extent, the common Bronze and Iron Ages to the Silicon Age we are living properties of nanomaterials that are discussed in the in. During the history of mankind, the role of nanomate- following chapters are described, whereas in the main rials has continuously increased, and the growth of this part of the chapter, the features and properties of vari- field is accelerating. ous groups of materials are discussed. This introductory In any modern scientific approach, one should de- chapter is more exemplary than universal, and in most sign and synthesize materials with a high level of con- cases not detailed. More detailed description of the trol. The ultimate material design is when one can plan properties and materials discussed here can be found in the structure of the produced materials atom by atom, the later chapters. For navigation in the handbook please with every defect, bond length, etc. (for example, 60 car- use the extended version of the table of contents and the bon atoms, or 72, 80, or even a few million, and 80 boron well-detailed subject index.

R. Vajtai (Ed.), Springer Handbook of Nanomaterials, DOI 10.1007/978-3-642-20595-8_1, © Springer-Verlag Berlin Heidelberg 2013 2 Introduction

Introduction 1.1 History and Definition of Nanomaterials In this part of the chapter, we briefly mention several miniaturization and integration of these devices in the important and interesting events from both ancient and first chips [1.12]. In the same decade when the first modern ages for nanomaterials production and appli- integrated circuit (IC) was fabricated, Richard Feyn- cation. We also present how our fellow scientists and man presented his famous lecture [1.13] about the government organizations describe and define nanoma- huge information storage capacity of materials if and terials. (as he rather saw it) when one goes to the atomic scale, to store ultimately one bit of information in 1.1.1 History of Nanomaterials every atom. He calculated that all of the informa- tion stored in the Encyclopedia Britannica would fit Although we indubitably live not only in the Silicon but onto the head of a pin – assuming 120 dpi resolution also in the beginning of the Nano Age, almost every used in the original print and that the resolution is chapter in this book, as well as most of the comprehen- increased 25 000-fold – without breaking the known sive review papers in the literature, include historical rules of physics, and he even envisioned several meth- aspects. Usage of nanomaterials remounts to traditional ods for writing, multiplying, and reading information Chinese medicine [1.6], and Mayan [1.7] and medieval at that density. In Fig. 1.1d,e, examples of nanomate- Italian paints [1.8]. Nanomaterials such as the color- rials from the last 50 years are displayed. Figure 1.1d ful and magically healing inks made of colloid-sized shows chromium nanocrystals [1.3] synthesized by the (nano) gold particles were also used in artistic applica- inert gas deposition method [1.3, 14]. Figure 1.1e dis- tions such as the Lycurgus cup [1.9] and for producing plays schematics for the newly discovered allotropes both small and large stained-glass windows for castles of carbon, i. e., nanodiamond, fullerene, nanotubes, and and cathedrals. graphene; these materials form such an important part In Fig. 1.1, we have collected several exam- of today’s nanomaterial science that at least one chap- ples of famous nanomaterials in chronological order ter in this handbook summarizes knowledge about each (Fig. 1.1a–f). The timescale in Fig. 1.1gshowsthese of them. Figure 1.1f shows a transmission electron mentioned examples and some other important mater- microscopy (TEM) image of gold nanorods and their ials as well as important events. Figure 1.1a–c shows shape distribution [1.5]; these nanorods are useful in pre-modern era examples of nanomaterial applications. nanomedicine, e.g., for imaging and curing cancer. The The Lycurgus cup (Fig. 1.1a) includes gold nanoparti- bottom part of Fig. 1.1g shows a collection of several cles which make its color green when we look at it historical and novel nanomaterials and related events in the usual way, in reflected light, but red in trans- displayed on a logarithmic scale, dating back from the mitted light [1.9]. Similarly, metal nanoparticles were present to 6000 BC. Even without displaying many im- used in the magnificent south rose window of Notre portant discoveries in the last decades, Nobel and Kavli Dame Cathedral (Fig. 1.1b). The specific nanostruc- Prizes awarded for achievements in nanomaterial sci- ture with cementite and carbon nanotubes ence, and milestones of commercialization, an obvious of damascene-style steels (Fig. 1.1c) [1.10]inventedin exponential acceleration is visible. the Mediterranean, and similar techniques used in Swe- den, are representative examples of the first manmade Fig. 1.1a–g History of nanomaterial use. (a–c) Premodern engineered nanomaterials [1.10]. Of course, ancient examples for nanomaterial applications. (a) The Lycurgus and premodern technologies could not control material cup photographed in reflective light and in transmission properties on the basis of knowledge of nanometer-scale (after [1.1]). The south rose window of Notre Dame properties; however, the recipes and methods for nano- Cathedral in Paris, France. (c) Damascus saber (photo material production were successful enough to provoke Tina Fineberg) (after [1.2]). (d,e) Modern-age milestone admiration and sometimes fear in scientists of their age. examples of nanomaterials. (d) Metal nanoparticles, Cr At the same time, even with our modern knowledge, shown here, produced by the inert gas deposition tech- the admiration for these first craftsmen of nanomaterials nique (after [1.3]); (e) : nanodiamond, still holds. fullerenes, nanotubes, and graphene [1.4]; and (f) gold The history of nanomaterials took an interesting nanorods [1.5]. (g) Timescale – measured back from to- turn in the 20th century. First, the discovery of semi- day to 6000 BC on a logarithmic scale – of nanomaterial conductor-based transistors [1.11] opened the road for production and events related to nanomaterials  Science and Engineering of Nanomaterials 1.1 History and Definition of Nanomaterials 3

In this fast-paced and accelerating scientific and en- and in the whole Springer Handbook of Nanomateri- Introduction gineering nanoworld, one needs to follow several dozen als, we try to provide a compass to identify the most journals which deal with nanomaterials. In this chapter important features.

a) b) c)

d) e) f)

Diamond C60 Buckminsterfullerene

Abundance (%)

30 20 10 2 Cr # 520 20 nm Graphite (10, 10) Tube 2.0 2.53.0 3.5 4.0 4.5 Aspect ratio g) Ancient carbon black inks Healing nanogold Lycurgus cup Stained-glass windows Damascene saber Faraday's gold colloid First Nobel Prize for nanomaterials Feynman's speech Metal nanoparticles prepared Quantum dots

C60 discovered First nano journal MWNT recognized SWNT produced Years since event (plotted in 2012) Dawn of graphene 550500 5000 4 Introduction Introduction The scale of things – Nanometers and more

Things natural Things manmade

Dust mite Ant Head of a pin The challenge

10–2 m 1cm 10 mm 1–2 mm ≈ 5mm 200 μm Microelectromechanical system (MEMS) devices 10–3 m 1 000 000 nm Human hair Fly ash =1mm

≈ 60–120 μm wide –4 0.1 mm 10 m 100 μm Pollen grain ≈10–20μm

Red blood cells 10–5 m 0.01 mm 10 μm 10–100 μm wide Fabricate and combine Red blood cells nanoscale building blocks to make useful devices, e.g., a photo- 10–6 m 1000 nm = 1 μm Zone plate synthetic reaction x-ray lens center with integral ≈7–8 μm semiconductor storage.

10–7 m 0.1 μm ATP synthase 100 nm Self-assembled, Nanotube nature-inspired Outer ring spacing electrode structure ≈ 35 nm Ultraviolet –8 0.01 μm 10 m 10 nm Nanoworld Microworld 10 nm diameter 10–9 m 1nm Many 10s of nm 1 μm

DNA Atoms of silicon Quantum corral of 48 iron atoms Soft x-rayon copper Visiblesurface Infrared positioned Microwave one Carbon –10 0.1 mm 10 m at a time with an STM tip Carbon buckyball nanotube spacing 0.078 nm

≈ 2–1/2 nm ≈1nm diameter diameter

Corral diameter ≈14 nm ≈1.3 nm diameter

Fig. 1.2 Scale of things chart designed by the Office of Basic Energy Sciences (BES) for the US Department of Energy (ATP: adenosine-5-triphosphate (after [1.15]))

1.1.2 Definition of Nanoscale materials need to have one, most obvious feature: and Nanomaterials at least one dimension should be on the nanoscale. In this section, we elaborate what nanoscale means As mentioned earlier, nanomaterials differ from the and describe the common properties and features that corresponding bulk materials in many ways, and this make nanomaterials different from their well-known dissimilarity is the reason for the formation of a new macroscopic (bulk) counterparts. We also give def- discipline that describes the properties and behavior of initions of nanosize, nanoscale, and nanostructured nanomaterials. To exhibit these unique nanoproperties, materials. Science and Engineering of Nanomaterials 1.1 History and Definition of Nanomaterials 5

Sizes Are Important: Usual Range of Values Different from Bulk Introduction First of all we need to define the size range of struc- The most important question is why materials on this tures. The term nano comes from Greek and means 1–100 nm scale are distinguished and have their own dwarf ; when used together with units of physical quan- science and engineering. In the size range below 1 nm tities, it expresses 10−9 times smaller than the unit. we can find molecules, atoms, elementary particles, etc., For nanomaterials, the nanoscale mainly means that which are different from the bulk but already have their length (e.g., size, diameter, edge) is measurable on the own disciplines. Multiple atoms or molecules behave nanometer scale. Obviously, other physical quantities, differently from individual ones for some obvious rea- such as area, volume, mass, and energy, may be very sons; e.g., larger size results in different behavior in far from this particular prefix when we use the corre- collisions with atoms. Other changes with increasing sponding SI units. More exactly, most of the definitions number of atoms in the particle include perturbation place nanomaterials between the approximate limits of oftheatomicenergylevelsleadingtoenergyband 1 and 100 nm. One nanometer is also equal to 10 Å, us- structures with fine structure; e.g., the energy differ- ing the Angstrom, which is widely used in microscopy ence between neighboring states is E/2N, where E is and atomic/molecular physics. the width of the energy band and N is the number of In Fig. 1.2, a comparative scale is displayed to show atoms included. The distinction between molecules and objects with sizes that fall into our common under- nanomaterials, however, is not sharp. C60 is considered standing (e.g., ants, pinhead, piece of human hair) to a and nanomaterial at the same time, and other nanomaterials with few-nanometer feature size (e.g., large molecules are also used as materials in molecular fullerene, nanotube, and DNA (deoxyribonucleic acid)). electronics. The figure also shows objects on the intermediate At the other end of the size scale, materials start to micrometer scale (red blood cells and microelectrome- be different from bulk below 100 nm size, because the chanical systems (MEMSs)), and for comparison with effects of quantum confinement on electrical, thermal, another well-known scale, the different ranges of and optical properties become significant at about this electromagnetic waves. On this scale, nanomaterials size. can be positioned between the wavelengths of visi- Another, very important common feature of nano- ble/ultraviolet light and x-ray radiation. Being smaller materials (e.g., nanoparticles and nanocrystals) is that than the shortest visible wavelengths limits the methods they have a very high fraction of their atoms on their available for determining the shapes of nanomaterials surfaces. These atoms behave differently from the ones by conventional light microscopy. located inside the object or in ideal bulk crystals (in

Fig. 1.3 Schematic of the top-down and bottom-up nanomaterial produc- tion procedures 6 Introduction

Introduction Fig. 1.4a–c Examples of top-down a) b) nanostructure fabrication methods. (a),(b) Nanoletters generated by direct writing with an AFM tip. (a) The letters are made of SiO2 on a silicon wafer. (b) Structure after selective etching to remove the oxide [1.16]. (c) High-aspect-ratio silicon pillars created by pattern generation by Ga- ion implantation in a focused ion 100 nm beam (FIB) setup and reactive plasma c) etching of the wafer [1.17]

2 μm 1 μm 200 nm

fact, the definition of an ideal crystal includes no phys- changes mechanical, electrical, and thermal properties; ical boundary or surface, as the crystal exhibits infinite e.g., the melting point of small particles can be con- periodicity) owing to the asymmetrical forces acting siderably lower than the bulk value. In the remainder on them; there is a force acting on these atoms which of this chapter we describe in slightly more detail is directed into the particle. The integrated effect of these changes in the physical properties of nanoma- the forces on every surface atom provides a surface terials as a function of feature size; later chapters in tension, the related pressure being so high that it can this handbook provide more data for individual material change bond lengths in crystals. Through this, it also groups.

1.2 Formation of Nanomaterials Formation of nanosized materials – nanoparticles, nano- nanomaterials are built up from their ultimate building porous or nanostructured macroscopic materials – is blocks – atoms and molecules – via self-assembly pro- achieved by two basic routes, namely the so-called top- cesses. Figure 1.3 shows a comparative representation down and bottom-up methods [1.20, 21]. In the former, of top-down and bottom-up methods. In the top-down macroscopic materials are used to fabricate nanomateri- method we start from bulk materials and fabricate struc- als and nanostructures using – typically – very sophis- tures or particles on the nanoscale, analogously to the ticated methods. In the latter, nanoparticles and other carving of a (or many small ones) from a mar-

Fig. 1.5a–f Examples for bottom-up assembly of nanoparticles. (a–d) Different representations of an AuPd nanoparticle: (a) AFM topography of the surface of the sample covered by the nanoparticles; (b) high-resolution transmission electron microscopy (HRTEM) image of an individual nanoparticle; (c) HRTEM simulation of the same nanoparticle; (d) solid- ball atom model of an icosahedral nanoparticle (after [1.18]). (e,f) Explanation of the correlation between particle size distribution and the flow properties of the carrier gas. (e) Size distribution at medium inert gas drift conditions; (f) size and size distribution as a function of drift intensity (after [1.19])  Science and Engineering of Nanomaterials 1.2 Formation of Nanomaterials 7 Introduction z (nm) 1000 a) b)

800

600

400

200

200400 600 800 x (nm) 1nm

c) d)

1nm 1nm e) Number of particles f) Normalized geometric mean (Š/L2) 10

3500 Geometric standard deviation ( N =105 L =10 3000 L =50 L = 200 1 Š/L2 L = 100 2500 δ = 0.02 Š 0.1 L = 200 2 2000 = 9244 L = 500 σ = 1.32 1500 0.01 σ 1000 0.001 σ

500 ) Critical drift 0 0.0001 1 05000 10 000 15 000 20 000 25 000 30 000 35 000 0.01 0.11 10 100 1000 10 000 δ δ Particle radius (arb. units) Normalized drift ( / 0) 8 Introduction Introduction a) Nature of Spheres Cylinders Double gyroid Double diamond Lamellae patterns (SPH) (3-D) (CYL) (2-D) (DG) (3-D) (DG) (3-D) (LAM) (1-D) A Space group lm3m p6mm la3d Pn3m Pm ab c d

Blue domains: B A block ef g h

Volume C fraction 0–21% 21–33% 33–37% 37–50% of A block ij k l b)

150 nm

Fig. 1.6a,b Block copolymer self-assembly used in bottom-up and top-down fabrication of nanostructures. (a) Schematic showing various block copolymer morphologies; blue color represents minority phase, with the matrix surrounding; schematic of a linear triblock copolymer; triblock copolymer structures displayed by coding colors. (b) SEM image of the Co dot array developed by poly(styrene-b-ferrocenyldimethylsilane) (PS-PFS) BCPs and Ne ion-beam etching (after [1.22–24])

ble boulder; in the bottom-up method we use techniques ogy of integrated circuit fabrication (e.g., photolithog- to build up materials from their component parts, i. e., raphy developed to < 20 nm resolution, nanoimprint atoms and molecules, analogously to the building of lithography) along with several completely new meth- a cathedral or forming a sculpture from clay. ods [e.g., e-beam and focused ion beam (FIB) lithog- As procedures used in synthesis and characteriza- raphy, direct writing with atomic force microscopy tion are normally considered parts of nanotechnology (AFM), and scanning tunneling microscopy (STM)]. rather than nanomaterials science, being discussed in Figure 1.4 shows examples of these revolutionary meth- detail in the nanotechnology literature, we only give ods. Figure 1.4a presents a demonstration of nanoscale a very short description for the sake of completeness. oxidation of silicon in the shape of letters using an Top-down approaches include evolutionary tech- AFM tip [1.16], and the same pattern after etching out niques which are similar to those used in microtechnol- the oxide by hydrofluoric acid (HF)(Fig.1.4b). The Science and Engineering of Nanomaterials 1.2 Formation of Nanomaterials 9 Introduction AFM tip a) b) ECD of NWs CVD of Y-CNTs

Y-CNTs

Metal NW NW Writing layer direction Molecular transport 50 nm

Water meniscus 200 nm

Au substrate 200 nm 5 nm 5 μm

c)

Alumina template

Connected cupSeperated cup Nanoring structure

Fig. 1.7a–c Examples of nanomaterial production by methods which merge top-down and bottom-up features. (a) Dip- pen , AFM structure deposition, and ink self-assembly on the surface (after [1.25]). (b) TEM image of gold nanowires grown in an alumina template (after [1.26]); complex shapes and structures generated in predesigned and fabricated alumina template junctions (after [1.27]). (c) Schematics of low-aspect-ratio nanocup preparation on alumina template and SEM images of the two sides of the freestanding nanocup layer (ECD: electrochemical deposition; NW: nanowire; Y-CNT: Y-shaped carbon nanotube) (after [1.28]) printing resolution of the letters corresponds to almost position method [1.3, 14, 18, 30, 31], also play a very 2 000 000 dpi, which is in the range of Feynman’s calcu- important role in determining the mean particle size lations. Figure 1.4c shows high-aspect-ratio structures and distribution width [1.19, 32]. One of the most fabricated by cryogenic plasma etching of gallium ion- widely used bottom-up methods is chemical vapor de- implanted regions of a silicon wafer [1.17]. The authors position (CVD), where nanomaterials grow on catalyst defined the pattern by FIB lithography and used reac- layers; references to illustrate such materials can be tive etching, resulting in structures with 65 and 40 nm taken from thousands of related papers for single-walled diameters and 600 nm height (aspect ratio up to 15). nanotubes (SWNTs) [1.33, 34], multi-walled nanotubes Bottom-up approaches are quite similar to the for- (MWNTs) [1.35], and h-BN [1.36] deposition. mation of macroscopic crystals. However, the thermo- Solution-based methods are governed by the con- dynamics of the processes involved for nanomaterials centration of the components used, the temperature of differ from the bulk; as we show later, the free en- the reaction, and the amount of surfactant; examples ergy, chemical potential, phase diagrams, and kinetics include seed-mediated growth of copper nanoparti- of phase transformations are distinct. Nucleation [1.29] cles [1.37], synthesis of shape-controlled gold nanopar- and diffusion are the key elements in the formation ticles [1.38], and specifically the Kirkendall effect of particles; however, other parameters, such as forced applied for preparation of hollow particles [1.39, 40]. or natural drift of the carrier gas in the synthesis Sonothermal synthesis [1.41] and exfoliation of (Figs. 1.1 and 1.5) of nanoparticles by the inert gas de- two-dimensional (2-D) materials – for making nanoma- 10 Introduction

Introduction terials from bulk – are considered bottom-up methods, this technique are displayed in Fig. 1.6. Figure 1.7a as we do not have control over the exact manner of shows schematics of dip-pen lithography; in this case, formation or properties of each of the particles via the AFM manipulation is the top-down component macroscopic equipment or tools. while self-assembly of the printed materials is the Another possibility is to combine the advantages bottom-up component. In the widely used method of the bottom-up and top-down methods. A few of membrane-based deposition [1.26], we have con- such examples are displayed in Figs. 1.6 and 1.7. trol over the shape of the alumina template trough An interesting way of increasing the resolution of parameters such as the etching time, acid type, photolithography beyond its physical limitations is and temperature; as examples, nanowire, nanojunc- by using directed self-assembly and self-orientation tion nanocup, and nanoring fabrication are collected in of block copolymers [1.42]; structures derived from Fig. 1.7b–c.

1.3 Properties of Nanomaterials When nanomaterials are studied, with few exceptions, in Fig. 1.8c is a GaAs nanowire mat; the inset shows the goal of the investigation is to reveal the values of an individual wire [1.47]. The most famous 1-D mater- physical properties similar to those used to character- ials are the single-walled nanotubes (SWNTs) [1.48,49] ize bulk phases of materials. In most cases though, the and multi-walled carbon nanotubes (MWCNTs) [1.50]. values of these properties are very different from those Two-dimensional materials include graphene [1.51], in the bulk, depending on the size and morphology of boron nitride (h-BN)[1.36], and graphene oxide [1.52]. the nanomaterials; this is similar to the way in which In Fig. 1.8d, TEM images of BCN of two and three the properties of bulk materials also depend on their atomic layer thickness are shown [1.53]. Typical 3-D crystal system (e.g., face-centered cubic (fcc) versus nanostructured materials are nanoporous metals and ce- body-centered cubic (bcc)). ramics, aerogels, and zeolites [1.54]. Figure 1.8eshows Gleiter schematics of different kinds of 3-D mater- 1.3.1 Morphology of Nanomaterials ials [1.55], and Fig. 1.8e,f illustrates the self-assembly of microparticles from azobenzene thiol-functionalized Not only the overall size, but also the shape of nanoparticles [1.56]. nanomaterials is a factor governing other proper- ties; the aspect ratio, porosity, and surface roughness 1.3.2 Bonds and Structures all change the surface-to-volume ratio and thereby other properties. Shape is such an important prop- The compressive strain in bulk solid materials is rel- erty that classification of nanomaterials can be based atively small, even in the case of high applied stress, on their dimensionality or aspect ratio. Figure 1.8 because the bonds in solids are strong and changing shows schematics of nanomaterials limited to zero, their length requires high pressure and a substantial one, two, and three dimensions. Zero-dimensional (0-D) amount of energy. Quite amazingly, the asymmetry objects have nanometer feature size in every direc- caused by the forces near to surface atoms in solids tion; one-dimensional (1-D) objects have nanometer change the bond length of nanomaterials at least on the size in two directions but larger, e.g., micrometer, scale that we are able to achieve by applying outside length in the third; two-dimensional (2-D) objects forces. The importance of the effect of these changes in are atomically thin sheets of materials, while three- bond length increases with the surface atom to volume dimensional (3-D) nanomaterials are nanoporous or atom ratio (dispersion) and accordingly on the particle nanostructured materials. Figure 1.8b, c, and d show size; it is determinant in clusters and nanoparticles at the examples of 0-D, 1-D,and2-D materials, respec- lower end of the nanoscale, normally between 1–5 nm tively. Quantum dots [1.43], nanoparticles [1.3, 14], (about 50–5000 atoms). and fullerenes [1.44] are 0-D objects; the example Table 1.1 presents a collection of data for bond shown here is a series of TEM images of iron ox- length relaxation [1.57], data based on references [1.58– ide nanoparticles with particle diameters of 6–13 nm, 70]. It is clear that bond contraction is between 4% where the diameter was controlled with 1 nm accu- and 30%, and, surprisingly for a bulk material scientist, racy [1.45, 46]. The one-dimensional material shown there is a 30% change in the bond length of diamond, Science and Engineering of Nanomaterials 1.3 Properties of Nanomaterials 11 Introduction a)

3-D 2-D 1-D 0-D

c) d) b)

e)Families of NSM f) Composition of Crystallites Chemical Different boundaries and dispersed in matrix composition Same for different crystallites of different of crystallites crystallites different composition Shape of crystallites

Layer-shaped

Rod-shaped

Equiaxed Categories of NSM Categories crystallites

Fig. 1.8a–f Classification of nanomaterials based on their shape, dimensionality, and structure. (a) Simple schematics showing 1-D, 2-D,and3-D nanomaterials. (b) 0-D structures: TEM images of iron oxide nanoparticles with particle diameters of 6, 7, 8, 9, 10, 11, 12, and 13 nm (after [1.45, 46]). (c) 1-D structures: SEM images of GaAs nanowires with an inset of the TEM image of an individual nanowire (after [1.47]). (d) 2-D structures: atomic layers (two and three) of BCN (after [1.53]). (e) Schematics of classification of different 3-D (bulk-like) nanostructured materials (NSM) (after [1.55]). (f) 3-D structure: a microobject built up from self-organized nanoparticles (after [1.56, 71]) too. The table summarizes data for covalent, metallic, momentum, and hardness – caused by the bond length and ionic bonds. While the contraction of ionic bonds contraction and reported in the referenced literature. is slightly less than that for the other two bond types, it Figure 1.9 illustrates how the bond length and bond is also obvious that the phenomenon is universal for all potential depend on nanoparticle size and shape. Fig- kinds of bonds. Table 1.1 also names the changes in the ure 1.9a shows the relative change in the lattice constant physical properties – energy related to bonds, magnetic for several metal nanoparticles – gold, copper, platinum, 12 Introduction

Introduction Table 1.1 Bond lengths for typical bonds (covalent, metallic, and ionic) and effect on several physical properties (af- ter [1.57], data from [1.58–70])

Bond nature Medium c1 = d1/d Effect Covalent Diamond {111} [1.72] 0.7 Surface energy decrease Metallic Ru [1.73]andCo[1.74] 0.9 Re [1.75] (1010) surfaces 0.9 Atomic magnetic momentum is increased by 25–27% [1.76–78] Fe-W, Fe-Fe [1.79] 0.88 Fe(310) [1.77], N(210) [1.78] 0.88 Al(001) [1.80] Cohesive energy rises by 0.3eV/bond [1.81] Ni, Cu, Ag, Au, Pt and 0.85–0.9 Pd dimer bond [1.80] Single-bond energy increases 2–3-fold [1.80] Ti, Zr [1.80] 0.7 V[1.80] 0.6 Ionic O−Cu(001) [1.82–84] 0.88–0.96 O−Cu(110) [1.82, 83] 0.9 N−Ti/Cr [1.85] 0.86–0.88 N-TiCr surface is 100% harder that the bulk [1.85] Extraordinary (Be, Mg) (0001) Zn, Cd and >1 No indication of effects on physical properties is yet given cases Hg dimer bond [1.80]

silver, and aluminum – in the size range of 1–20 nm. grate the crystal. On the nanoscale, the hexagonal sheets Figure 1.9b illustrates how the shape and position of of carbon are separated, and strength is measured on in- the bond potential change for ≈ 15% lattice contrac- dividual flakes and domains. The role of point defects tion. The shorter bond length corresponds to a deeper in nanomaterials, from the point of view of mechanical and narrower potential well, resulting in stronger ma- strength, is not as critical as for discontinuities in bulk terials and lower thermal expansion. Figure 1.9cshows materials. Graphene and carbon nanotubes have high nanochain formation for gold, in which chain formation strength and Young’s modulus owing to these strong is successful, and for copper, in which chains do not in-plane bonds. form. The bond energy for atoms in the nanometal chain It has been shown that metals with grain structures structures is 2–3 times larger than for bulk materials. have an optimum grain size for minimum creep [1.89]; the optimum grain size for most metals is on the 1.3.3 Mechanical Properties few-nanometer scale, corresponding to the limit of of Nanomaterials Hall–Petch hardening by grain size reduction. Similarly, in the case of ceramics, the maximum Vickers hardness Nanomaterials and nanostructured materials have dis- is measured at the optimum grain size that falls in the tinct mechanical properties for several reasons: Firstly, range of 10–50 nm [1.90, 91]. the shorter bond length results in stronger and stiffer Figure 1.10 demonstrates several basic mechani- materials, as mentioned in the previous section; Fur- cal properties of nanomaterials. Figure 1.10aandb thermore, the limited size of the units of the material show the particle size dependence of hardness for diminishes the probability of certain defects; e.g., grain boundaries are very rare in small nanoparticles. It is well Fig. 1.9a–c Typical effects of properties of bonds in known that carbon–carbon bonds in the hexagonal lat- nanoparticles as a function of particle size. (a) Dependence tice of graphite are the strongest ones in any solid, and of the bond length on particle diameter for several metals second only to the N−O bond overall. On the macro- (after [1.86]). (b) Increase of the bond energy density with scopic scale, however, we do not consider graphite an decreasing particle size and the consequently shorter bond extraordinarily strong material as its bonds in the c- length (after [1.57, 87]). (c) Bond strength as a function of crystal direction are weak, and the lateral strength in dimensionality. Chain formation of gold atoms and failure the a–b plane is diminished greatly by the fact that the to achieve the same with copper. Binding energies of bulk layer is not continuous, so pulling forces easily disinte- and atomic 1-D chain structures (after [1.88])  Science and Engineering of Nanomaterials 1.3 Properties of Nanomaterials 13 Introduction a)ε (%) b) Atomic potential

0.5 0

0 –1 E Au C –2 –0.5 –m ci

–3 –1 EB (d) 0 EB (di) –1.5 0.6 0.8 1 1.2 1.4 1.6 1.8 2 –1 c)

–2 Cu

–3 Binding energy per bond (eV) 0.0 1 0.5 Bulk fcc –0.3 0 –0.5 –0.6 –1 Pt –1.5 –0.9 Au Cu –2 Pt Pd –1.2 –2.5 Ni 0.0 Ag –32 2.2 2.6 2.4 2.8 3.0 3.2 3.4 Nearest-neighbor bond length (Å) –0.5

Ag –1.0

Binding energy per bond (eV) 1 –1.5 0.5 Chain –0.2 0 –0.5

–0.4 –1 Au –1.5 Cu Al Pt –2 Pd –0.6 Ni –2.5 Ag –3 02051510 2 2.2 2.6 2.4 2.8 3.0 3.2 3.4 D (nm) Nearest-neighbor bond length (Å) 14 Introduction

Introduction metals and intermetallic alloys [1.92], respectively. In the high aspect ratio of 1-D structures is the field en- the case of metals, there is an obvious trend for the hancement of the electrostatic potential, which enables hardness to be 3–10 times higher for the smallest a low turn-on voltage for field emission [1.98, 99]. nanoparticles than for those larger than 100 nm, lat- Nanomaterials, especially 1-D structures such as ter is the value corresponding to the bulk material. nanowires and nanotubes, conduct electrical current dif- For the alloys, a trend of increasing hardness is vis- ferently compared with bulk conductors. The limited ible for the 100 to ≈ 40 nm range; however, in many size in the direction perpendicular to the axis and the cases this trend becomes softening for further diminish- atomically organized structure of the 1-D objects – ing nanoparticle sizes. Figure 1.10c,d shows buckling crystals or nanotubes, e.g., molecules – result in weak of carbon nanotubes [1.93–95]. While SWNTshave phonon–charge carrier interaction and ballistic trans- very high Young’s modulus and relatively high stiff- port. In the ballistic regime of electrical conduction, the ness against tensile forces [1.96], they can be deformed resistance does not depend on the length of the conduc- by compressive forces, which makes them ideal high- tor but rather on the number of conductance channels resolution yet gentle and forgiving AFM tips [1.95]. used, following Landauer’s law [1.100]. Often, there is Figure 1.10e,f demonstrates the in-plane mechanical a single channel, and the quantum nature of the conduc- strength of a graphene sheet based on nanoidentation tance as a function of the number of channels can be carried out by an AFM tip [1.97]. demonstrated, e.g., for carbon nanotubes [1.101]. How- ever, in other reports, e.g., on nanotubes used to build 1.3.4 Electrical, Magnetic, field-effect transistors (FETs), the conductance mecha- and Optical Properties nism is considered to be diffusive [1.102]. Another important effect in devices made of nano- Nanoscale materials, especially with dimensionality of sized conductors and semiconductors is the Coulomb zero to two, exhibit intriguing electromagnetic behav- blockade that occurs when new charge carriers cannot ior, as the most important effects and properties are enter the conduction channel while another charge car- derived from the quantum confinement of the wave rier occupies it, as demonstrated for single-molecule function of the charge enclosed in the particles. In the devices [1.103]. This phenomenon enables the function- nanometer-sized dimensions of the material, the wave ing of single-electron transistors (SETs) even at room function has a limited number of solutions. This re- temperature [1.104]. SETs are also useful to investigate sults in changes in several electrical properties: with spin–spin interactions between localized and mobile the decreasing feature size the bandgap becomes wider, electrons, i. e., the Kondo effect, more effectively than the conductivity decreases and the density of states de- in macroscopic systems [1.105]. creases. Optical properties change accordingly, as the High current-carrying capacity – i. e., no failure absorbed and emitted photons depend on the energy caused by electromigration – of individual carbon nano- difference between the states among the bands or in tubes, both single-walled (4 × 109 Acm2)[1.106]and their fine structure. Beyond this, quasiparticles are gen- multi-walled (> 1010 Acm2 at 250 ◦C) [1.107], and wires erated due to the confined space. An electron–hole pair, made of nanotubes has also been demonstrated. Since i. e., an exciton, carries energy in nanomaterials with- carbon atoms are bound into the nanotube by covalent out motion of net charge. Plasmons, surface plasmons bonds, the charge carriers and high temperature cannot (polaritons), and polarons play a key role in nanoma- easily move them away from their original location. The terial interactions with optical phonons, e.g., causing state-of-the-art specific conductivity of carbon nanotube the color dependence of gold colloids as a function of particle size. Magnetic properties are also governed by Fig. 1.10a–d Selected mechanical properties of nanoma- the size of the nanomaterials; e.g., superparamagnetic terials. (a,b) Hardness of metals and intermetallic alloys materials exist below the size of the domain size. as a function of grain size in nanophase materials (af- ter [1.92]). (c),(d) Molecular dynamics computation of the Electrical Properties behavior of a carbon nanotube under compressive force. Electrostatic forces play a major role in nanoparticle The structure in the strain energy curve corresponds to formation and also have an important effect on nano- the modes of buckling (after [1.93–95]). (e,f) Measure- material properties [1.71]. Most of the larger structures ment and values of the Young’s modulus for a suspended built up from nanomaterials are held together by van graphene sheet on an alumina template by AFM (af- der Waals forces. One of the advantages attributed to ter [1.97])  Science and Engineering of Nanomaterials 1.3 Properties of Nanomaterials 15 Introduction a) b) 100 15 5 d (nm) 100 15 5 d (nm) 12 14 Pd TiAl Cu 10 12 TiAl Ag Nb3Sn Fe 10 Nb3Al 8 Fe Nb3Al Ni 8 TiAlNb Ni 6 Ni-P 6 Cu Ni-P 4 Se FeCuSiB Hardness (GPa) Hardness (GPa) 4 FeMoSiB 2 2 FeMoSiB FeSiB

0 0 NiZr2 0 0.1 0.2 0.3 0.4 0.5 0.6 0 0.1 0.2 0.3 0.4 0.5 0.6 NiZr2 d –1/2 (nm–1/2) d –1/2 (nm–1/2) c) d) Strain energy 0.006

0.004

0.002

0 0 0.05 0.10 0.15 Strain ε e) f) Effective Young’s modulus (TPa) 0.7 0.80.9 1.0 1.1 1.2 1.3 1.4 15

0.5 μm

1 μm1.5μm 10 Counts

5

0 235 268302 335 369 402 436 469 E2-D(N/m) 16 Introduction Introduction a) b) c) Energy I (μA) Conductance (G ) 200 5.0 0

Metal Semiconductor 4.0 Bulk Nanocrystal Atom Bulk Nanocrystal Atom 150 3.0 100 2.0

50 1.0

0 0 Density of states 0 0.05 0.10 0.15 0.20 0 1000 2000 3000 Δt (s) Depth (nm) Density of states Initial Thinned (3 steps) 3-D 2-D 1-D 0-D Thinned (10 steps)

Energy

2 d) G (μS) e) Mobility (cm /Vs) f) Current gain (h21) 6000 100 Dev1: V = 2.5 V 500 d TG 5000 Dev1: Vd = 2.0 V D 4.2 K Dev2: V = 2.0 V 400 S d 4000 Dev3: Vd = 2.0 V

300 3000 1/f BG 10 350 K 200 2000 20 VBG 15 VBG 10 V 5V 100 BG BG 1000 –5 V –10 V BG BG Holes Electrons –15 VBG –20 VBG 0 0 1 –2.5 –1.5 –0.5 0.5 1.5 2.5 3.5 4.5 –3 –2 –1 0 1 2 3 1 10 100 12 –2 VTG (V) Carrier density (10 cm ) Frequency (GHz) Fig. 1.11a–f Examples for electrical properties of nanomaterials. (a) Schematics to represent changes in the DOS of nanomaterials as a function of dimensionality (after [1.109]). (b) Quantized decrease of the current when shells of a multi-walled nanotube burned down consecutively and the schematics of the nanotube with partially removed walls (after [1.110]). (c) Conductance of carbon nanotubes when a parallel circuit was built from one to four individual tubes. As the voltage was low (100 mV), each nanotube added only one channel to the conductance and the measured values were nG,wheren is the number of nanotubes/channels. A larger number of nanotubes resulted in conductance up to 1000 G (after [1.111]). (d) Transistor characteristics of a FET device where the active element was a single-layer graphene channel. The figure displays the transconductance, charge carrier mobility, (e) and current gain (f) as a function of top and bottom gate voltage, carrier density and temperature, frequency and drain voltage, respectively (after [1.112])

fibers surpasses the values for all conventionally applied of the density of states (DOS) on the dimensionality of metallic conductors (Cu, Al, Au, etc.) [1.108]. the nanomaterials; the larger the number of quantum- Figure 1.11 shows a collection of electrical proper- confined dimensions, the higher the number of discrete ties. Figure 1.11a shows a schematic of the dependence states in the DOS [1.109]. Figure 1.11b shows how the Science and Engineering of Nanomaterials 1.3 Properties of Nanomaterials 17

resistance of a carbon nanotube increases as it loses car- perparamagnetic materials does not show hysteresis, Introduction bon layers from its wall; the resistance, and accordingly and the exact shape depends on the particle size or its the conductivity, also changes in quanta, illustrating distribution. the independence of the conductance channels in each By introducing defects into the lattice of carbon, layer [1.110]. Results of a similar experiment are shown e.g., implanting nitrogen or carbon atoms into nanodi- in Fig. 1.11c. Here, the conductance of a nanotube amond, ferromagnetic behavior is observable [1.115]. bundle was measured when one, two, three, and four Also, ferromagnetism is measured on monoatomic nanotubes were included sequentially in a circuit in- cobalt structures at low temperature (10 K), showing the cluding a contact made of mercury [1.111]. In both interaction between neighboring atoms [1.116]. During cases, the size of the quantized steps of the conduc- Mackay transformation and Bain transformation, the tance change – following the Landauer law – is equal to crystal structure is changing, and the magnetic proper- G0. By immersing the nanotube bundle further into the ties follow the crystal symmetry [1.117], as shown in mercury electrode, the authors measured conductance Fig. 1.12c. as high as ≈ 1000 G0.InFig.1.11d–f, the character- Figure 1.12d shows the grain size dependence of istics of a FET based on a single-layered graphene the magnetization of Ni films of different grain sizes, sheet are shown [1.112]. First, the structure of the i. e., the magnetization hysteresis loops, the coercive device and the conductance as a function of the top force values, and the relative change in the magne- gate (TG) voltage are displayed at different back-gate tization [1.118]. In all three cases, the changes are (BG) voltages. The electron and hole mobility were cal- significant below feature size of 10 nm. culated from Hall-effect measurements, being higher than 5000 cm2 V−1 s−1 for low temperatures and car- Optical Properties rier densities, and around 3000 cm2 V−1 s−1 at room It is a common misbelief that nanomaterials cannot be temperature (RT). The high-frequency behavior of the seen by the naked eye or be detected by optical mi- device shows 1/ f dependence of the current gain as croscopy. This misbelief is based on the fact that the a function of frequency, and the cutoff frequency is in nanoscale is defined as 1–100 nm while the wavelength the range of 100 GHz. of visible light ranges between 400 to 800 nm; however, this scale difference only means that size and shape can- Magnetic Properties not be resolved by visible-range photons. In fact, there The magnetic properties of nanomaterials depend are plenty of interactions between atoms, molecules, strongly on the characteristic size, i. e., the diameter of and nanomaterials with light, and some of these inter- nanoparticles or the grain size of nanostructured ma- actions are applicable to detect the size and morphology terial, which is normally small in comparison with the of nanomaterials. All of these detection methods need to magnetic domain size of the material. The configura- be first used in parallel with other approaches to mea- tion of thin films and few-domain nanoparticles was sure, e.g., nanoparticle size, and after this calibration, investigated by Kittel [1.113]. Figure 1.12a presents the optical method itself can serve as a fast and easy these configurations and the energy density for the measurement approach. In the case of gold nanoparti- various configurations in thin films, showing that (for cles or thin layers of graphene, practiced researchers can small size) the single-domain system is energetically fa- tell their size or thickness by eye alone, in the latter case vorable. Nanomaterials can be classified according to with subnanometer, atomic accuracy. In this section, the the type of interaction among the magnetic particles, main features of the optical properties of nanomateri- extending from no interaction in a well-distributed als are described, while detailed description of these nanoparticle system to strongly interacting nanostruc- properties for specific nanomaterials is included in the tured materials [1.114](Fig.1.12b). Ferrofluids consist relevant chapters of this handbook. of particles surrounded by surfactants, capping agents, Optical properties – similarly to the electrical ones – and solvents, forming an independent system, while al- are governed by quantum confinement: the lower di- loys with magnetic constituents are strongly correlated. mensionality and smaller size result in a larger energy Superparamagnetic materials behave similarly to difference between neighboring discrete energy levels paramagnetic materials. However, the magnetization in the DOS, and accordingly higher excitation energy value is considerably higher and the particles keep (Fig. 1.13)[1.109]. Inasmuch, the smaller the particle, their magnetization values for a measurable timescale, the shorter the wavelength in the absorption spectrum, e.g., several minutes. The magnetization curve of su- and the color shifts from red to blue. This interac- 18 Introduction

Introduction tion with photons can be interpreted by introducing Fig. 1.12a–d Selected magnetic features in nanomateri- a quasiparticle known as the surface plasmon (SP)to als. (a) Domain configurations studied by Kittel. For thin represent the oscillations in the confined space [1.43]. films the single-domain configuration has lowest energy The absorption spectrum has a maximum when the SP– for thickness below 300 nm. Several domain configura- photon interaction is strongest, at the surface plasmon tions for nanoparticles also were suggested in this paper, resonance (SPR) wavelength. The SPR for different which was published in 1946 [1.113]. (b) Schematics for nanoparticles depends – beyond the size – also on the the main types of nanostructures for magnetic behavior, material and shape of the nanoparticles. Figure 1.13a including independent ultrafine particles (e.g., ferroflu- demonstrates visible-range colors caused by SPRsof ids), core–shell particles with magnetic core, particles as different energy for several basic shapes of gold, sil- fillers in a matrix, and small crystallites in a noncrys- ver, and alloy nanoparticles [1.119]. Figure 1.13bshows talline matrix (after [1.114]). (c) Mackay transformation a particular case from [1.119], where the resonance shows the magnetization changes as a function of crys- wavelength shifts when the silver nanoshell particles tal structure. Schematics of icosahedron, cuboctahedron, are coated with a gold layer of increasing thickness. and fcc with inscribed body-centered tetragonal Bain cell Figure 1.13c presents a silver nanoprism preparation and the magnetization as a function of cluster size are method where the size distribution and accordingly the displayed (after [1.117]). (d) Magnetic hysteresis loops, absorption spectrum are well controlled by the wave- coercive forces, and relative changes compared with the length of light used in the photo-induced reaction. bulk value for Ni film with different grain sizes is displayed Another interesting phenomenon is the shift in the ab- (after [1.118])  sorption of gold nanotriangles from 800 nm to 600 nm with the extent of the truncation of the tips [1.120]. of thermal properties that strongly depend on the par- Along with other microscopy techniques, the opti- ticle or feature size of nanomaterials. Phase diagrams cal property corresponding to Raman scattering [1.121, of nanomaterials depend on the particle size, shape, and 122] provides insightful information into the local even their environment. bonding system of nanomaterials through phonon– The Gibbs–Thomson equation describes the de- photon interactions. It is used for characterizing a wide creasing tendency of melting point with decreasing range of materials from metals to oxides; one of the particle size. For spherical particles, for instance, it characteristic examples is its application in studying predicts that the difference in the melting point has carbon nanomaterials. Figure 1.13d shows spectra of inverse dependence on the diameter of the particle. Fig- carbon structures for comparative analysis. Graphene ure 1.14a shows the melting point of gold particles as has two characteristic peaks, the ratio of the intensity a function of particle size D [1.90, 126]. Below 20 nm of the G and G peaks depends on the number of layers the decrease is observable, and for particles smaller than in the 2-D structure. In the case of carbon nanotubes, the 3 nm the melting temperature decreases by ≈ 600 K. D and G peaks dominate the spectra, while for SWNTs, Figure 1.14b illustrates an unusual behavior. In specific the radial breathing mode (RBM) is also present and cases, indium particles embedded into an aluminum ma- provides information about the diameter and (to some trix show a melting point increase with decreasing grain extent) the chirality of the nanotubes. Shifts, shoulders, size; this exceptional behavior is caused by the interac- and new peaks in the spectra can be interpreted as sig- tion with the matrix: here, for smaller grain size, the nals caused by defects in the bonds or in the structure. surface energy of the grain decreases. Fluorescence [1.123] and bandgap photolumines- Figure 1.14c shows experimental demonstrations cence are also characteristic properties of nanomaterials, of the theoretically predicted, unusually high ther- the latter being applied, e.g., to study the diameter mal conductivity [1.127, 128] of carbon nanotubes and chirality of SWNTs in solutions [1.124, 125]. The and graphene, ≈ 6000 W/(m K). Values of thermal schematics of the band structure with the first and sec- conductivity measured on a thick layer of aligned ond van Hove optical transitions and the measured SWNTs range up to 240 W/(m K) in the direction of values on a Kataura plot are displayed in Fig. 1.13e. alignment, and approximately ten times lower in the perpendicular direction. A similar trend was observed 1.3.5 Thermal Properties for the heat conductivity of aligned carbon nanotube forests [1.129]. For the measurement of thermal con- Thermal conductivity, specific heat, melting point, and ductivity in graphene sheets, the shift in the position glass-transition temperature are just a few examples of the Raman G peak was investigated. This shift Science and Engineering of Nanomaterials 1.3 Properties of Nanomaterials 19 Introduction a) 45° Energy per unit area (erg/cm2) I T 1000

D 100 ––++ –– ++ Case II L II Case I 10 ++–– ++ –– 3×10–5 cm Case III

III –7 –6 –5 –6 –3 10 10 10 10 10 L Film thickness (cm) b) c)

A

A A c B B B c' C a' C b' b A C a μ Magnetic moment ( B/atom) 3.5 Bcc Mackay transformed Icosahedral B 3.0 Cuboctahedral Bcc (nonmagic) C 2.5 Bcc Icosahedral bcc (bulk) 2.0 Exp.

D 1.5 0 200 400 600 Cluster size (atoms) d) 4πM (kG) MC (kOe)

0.3 4 D =10nm D = 7.9 nm D = 5.5 nm 0.2 D = 3.2 nm 2 D = 3.1 nm 0.1

0.0 0 MS tailoring (%) 0.25 57.510 D (nm) 0

–20 –2 –40

–60

–4 –80

–100 –1.0 –0.50.0 0.5 1.0 0 52010 15 25 H (kOe) D (nm) 20 Introduction

Introduction corresponds to ≈ 5000 W/(m K) heat conductivity. An- Fig. 1.13a–e Selected optical properties of nanomaterials. other interesting hybrid system is displayed in panel (a) Typical surface plasmon resonance spectral ranges Fig. 1.14d, where covalently bonded carbon nanotube of silver and gold nanoparticles having various mor- pillars connect parallel graphene sheets to each other. phologies, compositions, and structures (after [1.119]). By changing the geometry of this unique system, i. e., (b) Ultraviolet-visible (UV-Vis) extinction spectra and the length and separation of the pillars, one can tune for photo of solution of silver nanoshells coated with different different vertical and lateral thermal conductivities. thicknesses of gold, and solid gold colloids; TEM image of the shell and solid particles (after [1.119]). (c) TEM image 1.3.6 Chemical Properties, Reactivity, and of silver nanoprisms and extinction spectra of nanoprisms Functionalization prepared with illumination by various laser wavelengths (after [1.130]). (d) Raman shift measured on different types Chemical approaches are used in almost all nanomate- of graphene-related nanocarbons. The main features (RBM rial preparation processes. As prepared, nanomaterials and disorder-induced D, D,andD+ D bands; first-order have extraordinary properties; however, for many ap- Raman-allowed G band; and second-order Raman over- plications they need to be modified, and accordingly tones G (2iTO) and 2G) are labeled in some spectra, but further chemical treatment is needed. Some nanomateri- the assignment applies to all of them. The analysis of the als, e.g., fullerenes and nanotubes, can be considered as frequency, line shape, and intensity of these features at molecules, and their chemistry is quite well defined. In different exciting laser wavelengths provides a great deal the case of metallic and ceramic nanomaterials, the re- of information about each respective sp2 carbon structure actions and products depend on the size and exact shape (after [1.122]). (e) Method for identifying single-walled of the nanomaterials used. In the narrower definition, carbon nanotube diameters by spectrofluorometry (Kataura we consider only modifications to the strong bonds – plots). First and second van Hove optical transitions are covalent, ionic, and metallic – of the materials; in the displayed as a function of structure for semiconducting wider definition, modifications to weak interactions – nanotubes (after [1.125]) (HOPG: highly oriented pyrolytic hydrogen bonding and van der Waals interaction – graphite; SWNH: single-walled nanohorn)  among parts of the new nanomaterial are also consid- ered. Exchange reactions and chemical transformations alkali ligands, which has been suggested for use as an of nanoalloys also represent a vast field of nanomaterial effective hydrogen storage material (Fig. 1.15b) [1.134]. chemistry [1.131]. Carrying out covalent chemistry of SWNTsis In the scope of this short introduction, a few necessary for achieving properties important for the example reactions are presented, where changes in co- formation of composites, preparation of solutions, valent bonds are dominant. Small pieces of graphene, etc. [1.135]. In most cases, the first chemical reaction also called graphene molecules, can be handled by aims to break the strong carbon–carbon sp2 bonds in the typical methods of organic chemistry; they can be dec- sidewalls of the nanotubes by oxidation or fluorination. orated, or connected to each other, and their lateral size Afterwards, the attached OH or F groups can be substi- can be controllably increased or decreased [1.132]. In tuted with alkyl or aromatic groups. Figure 1.15 shows Fig. 1.15a, schemes of several graphene molecules are different kinds of nanotube reaction products. Various displayed, demonstrating differently sized, bare, and R- groups can be introduced onto the sidewall or defect functionalized samples. The figure also presents a larger locations in the sidewall, or in the cap; surfactants or molecule, namely a nanopropeller, which is built from can be attached to the surface via noncovalent three graphene flakes connected to each other. The color bonds, and the cage of the nanotube can also be filled of these materials depends on the size of the molecules; with different chemicals (in this schematic, C60). with increasing size, the color shifts from blue in the Producing nanotubes with given chirality by direct direction of red. growth or separation methods is still an underdevel- Another example of well-defined molecules which oped area and a challenging task. Functionalization are considered nanomaterials is the fullerenes. These of metallic nanotubes [1.136] with side-groups, which can be modified in two ways: additional atoms are ei- makes them heavier than unfunctionalized semicon- ther included inside the cage of the fullerene or added as ductor nanotubes, offers a way for more selective and outside ligands, being called endohedral (inner) and ex- higher yield separation. ohedral (outer) fullerenes, respectively. An uncommon Another important area of nanocarbon chemistry example is that of B80 [1.133], a fullerene decorated by is the oxidation, reduction, and functionalization of Science and Engineering of Nanomaterials 1.3 Properties of Nanomaterials 21 Introduction a) 400 nm Light color 750 nm

Silver rods Silver spheres Gold rods Gold spheres Gold shell with hollow interiors Gold/silver alloyed spheres

Silver cubes Silver plates b) Extinction (arb. units) (a) (b) 0.7 (c) 0.6 (d) 0.5 (e) 0.4 0.3 0.2 10 nm 300 400 500 600 700 800 (a)(b) (c) (d) (e) Wavelength (nm) c) Extinction

470 488 500 514 550 600 633

470488 500 514 550 600 633 400 600 800 1000 d) λ (nm)

Intensity e) Energy Eii (eV) 4.5 G G' 4.0 Graphene C2 3.5 E S HOPG 44 3.0 C1 E S RBM 33 SWNT 2.5 E S E S D 11 22 G D+D' 2.0 G' E S D' 2G V 22 1 1.5 Damaged graphene

V2 1.0 SWNH 0.5 S E11 Amorphous carbon 0.0 01000 2000 3000 4000 Density of electronic states 0.51.0 1.5 2.0 2.5 Raman shift (cm–1) Nanotube diameter (nm) 22 Introduction

Introduction graphene. Graphite oxide is often produced to aid ex- Fig. 1.14a–d Thermal properties of nanomaterials. (a) Melt- foliation of layers of graphene into graphene oxide ing point of gold nanoparticles as a function of size, (GO), after which reduction procedures are needed to showing decrease with smaller sizes (after [1.90, 126]). yield graphene as a product (reduced graphene oxide, (b) Melting point of indium nanograins in aluminum ma- RGO)[1.137, 138]. Figure 1.15d displays the consec- trix; the direction of deviation from the bulk value depends utive steps applied in one of the possible reduction on the method of preparation (after [1.139]). (c) Ther- procedures of GO, as well as a photo of the solutions at mal conductivity of SWNTs and graphene. The thermal each step and the C 1s x-ray photoelectron spectroscopy conductivity of aligned nanotubes is within an order of (XPS) peak in the spectra of these materials. magnitude of in-plane values for graphite or diamond (af- ter [1.140, 141]). Raman shift of the G peak position of 1.3.7 Behavior of Nanomaterials a graphene sample versus change in total dissipated power. in Corrosive Environments Values calculated for thermal conductivity are higher than for CNTs(after[1.142]). (d) Pillared graphene structure Some of the ancient and medieval applications of nano- proposed originally for hydrogen storage (after [1.143]). materials pointed towards corrosion prevention. The Thermal conductivity of the structure along the graphene Chinese heiqigu are black bronze mirrors with a surface plane direction is decreased while the vertical thermal con- coating made of SnO2 nanoparticles, most probably ductivity is increased by increasing the number of pillars doped with Cu, Fe, Pb, and Si [1.145,146]. Similarly, as (after [1.144])  mentioned in Sect. 1.1.1, Mayan blue paint was not only a rare and beautiful color for its time, but also had corro- onto the surface of the specimen. The SEM and TEM sion resistance and retained its properties for centuries images show the morphology of the deposited layer; it while buried in soil [1.7]. has a microgranular structure built up from sub-10 nm Figure 1.16a shows an example of preventing an particles. The samples were oxidized in air at high tem- aluminum alloy from corroding in an NaCl-containing perature, and the oxidation kinetics showed parabolic water environment [1.147]. The scanning electron growth of the oxide layer (Wagner regime) for both the microscopy (SEM)andTEM images show the mor- uncoated and coated alloy, however the rate of oxidation phology of halloysite (aluminosilicate) nanocontainers was higher for the uncoated sample at all tempera- which hold and release anticorrosion agents, normally tures investigated. The surface of the oxidized sample polymers, onto the surface when the pH of the envi- was coated by nanowire structures, mainly alumina and ronment changes. The photos show extensive surface TiO2 nanowires, while the uncoated sample had a mi- corrosion of the specimen in a dilute solution when crogranular oxide layer. In a hot salt environment, the coated without the nanomaterial included but minimal trends were different: the coated sample started to cor- damage in ten times higher concentration salt solution rode faster, and after an initial fast scaling, the process with the filled halloysite additive. slowed down. Superalloys are designed and manufactured not only In several cases, nanomaterials exhibit the same or for mechanical strength and creep resistance but also worse corrosion resistance than bulk materials. This ef- to avoid corrosion in high temperature and harsh en- fect is mainly attributed to the high concentration of vironmental conditions. They themselves may or may defects in these nanoparticles and nanowires. In [1.149], not consist of nanostructured materials. Figure 1.16b silver nanowires were investigated, and the authors shows an effective nanomaterial coating for corrosion found faster sulfide formation in nanowires than for prevention of a Ni superalloy (K52) [1.148] deposited bulk silver and that the reaction mechanism was the by sputtering a nanostructured layer of the same alloy same.

1.4 Typical Applications of Nanomaterials Nanomaterials are not a sanctified object of nano- engineering to bioengineering, we use nanomaterials science; they have already found applications in a very regularly in our computers, stain-resistant clothes, and wide range of engineering fields. From mechanical suntan creams. Other applications such as nanomaterial- Science and Engineering of Nanomaterials 1.4 Typical Applications of Nanomaterials 23 Introduction a)T (K) b) Tm (°C) 220

Melting point bulk Melt-spun 1300 200 Ball-milled

1000 180

160 T0 =156.6°C

140 500

300 120 050 100 150 200 020 40 60 80 100 D (Å) D (nm) c) (W/(mK)) Intensity (arb. units) G peak position shift (cm–1) 5000 250 6 Suspended graphene ≈1583 cm–1 Suspended Experimental data H-aligned SWNTs 4 graphene Linear extrapolation 200 4000 Linear regression Y = A + B × X Laser power (mW) Parameter Value Error 2 Aligned G peak A 1.19548 0.27638 150 B –1.29207 0.11447 3000 0.950 0 2.168 100 –2 2000 –4 50 Unaligned Excitation: 488 nm 1000 –6 Ambient: room temperature 0 0 50 100 150 200 250 300 350 400 1500 1550 16001650 0 1 2 3 4 T (K) Raman shift (cm–1) Power change on the sample (mW) d) Thermal conductivity (W/(m K)) 500 450 400 Thermal conductivity (W/(m K)) 350 100 300 80 250 60 40 200 20 150 0 0 5152010 25 100 Minimum interpillar distance (Å) 50 0 2030 40 50 60 70 80 90 100 110 Graphite sheet length (nm) 24 Introduction Introduction a) b)

H AM B

B80Li(H2)2 B80Na(H2)6 B80K(H2)8

R R R R R R R R R

R R R R R R

RR RR R RRR R

C C C 42 60 78 B80Na12(H2)72 B80K12(H2)72

c) d) GO: 4.08×10–1 S/m CCG1: 8.23×101 S/m

NaBH4/H2O 80 °C, 1 h

Conc. H2SO4 180 °C, 12 h CCG3: 2.02×104 S/m CCG2: 1.66×103 S/m 1100 °C Annealing SWNT in Ar/H2 15 min

GP CCG3 CCG2 CCG1 GO

278 280 282 284 286 288 290 292 294 296 298 Fig. 1.15a–d Chemical properties of nanomaterials. (a) Graphene molecules demonstrating various sizes and shapes. Graphene molecules are mostly 2-D materials; however, they can be organized into larger, 3-D structures; a nanopro- peller is shown as an example. The size dependence of the color of planar polycyclic aromatic hydrocarbons (PAHs) is also demonstrated (after [1.132]). (b) Modifications of the B80 molecule by forming complexes with alkali elements. The complexes are suggested as effective hydrogen storage materials up to 11 wt. % capacity (after [1.134]). (c) Various ways of functionalization of SWNTs: attaching molecules to the defect locations in the side-walls, connecting molecules by covalent bonds, functionalization with surfactants, functionalization with polymers, and endohedral functionalization with fullerenes (after [1.150]). (d) Steps of a GO reduction procedure showing the schematics, the corresponding bond- ing energies in XPS, and photos of the solutions for each step. The vials contain GO in deionized (DI) water, GO in dimethylformamide (DMF), CCG3(CCG: chemically converted graphene) in DI water, and CCG3 in DMF from left to right (after [1.137]) Science and Engineering of Nanomaterials 1.4 Typical Applications of Nanomaterials 25

Fig. 1.16a,b Application Introduction a) of nanomaterials for preventing corrosion. (a) SEM and TEM images of halloysite nanotube nanocontainers (top row). Photos of aluminum spec- imens after 2 weeks of exposure to corrosive medium (bottom row): a1) blank aluminum alloy immersed in 0.3% NaCl, a2) blank aluminum al- 1 μm 200 nm 100 nm loy immersed in 0.3% NaCl saturated with 2- a1) a2) a3) mercaptobenzothiazole, and a3) aluminum alloy coated with halloysite- doped sol–gel film after immersion in 3% NaCl solution (after [1.147]). (b) Top row: Layers of nanostructured materials on its bulk substrate for K52 alloy. SEM images b) for top b1) and cross- section b2) morphologies b1) b2) b3) are shown along with the TEM showing the structure of the sputtered nanocoating b3). Bottom row: Oxidation kinetics of the bulk and coated alloys 20 μm 10 μm 100 nm exhibited faster oxidation rate for the bulk samples Mass change (mg/cm2) Mass change (mg/cm2) Mass change (mg/cm2) at the three tempera- 0.4 1.0 4.0 ◦ ◦ 800 °C 900 °C 1000 °C tures (800 C, 900 C, 3.5 1000 ◦C) investigated 0.8 (after [1.148]) 0.3 Cast alloy 3.0 Cast alloy 2.5 Cast alloy 0.6 0.2 2.0 Nanocoating Coating 0.4 1.5 Nanocoating 0.1 1.0 0.2 0.5

0.0 0.0 0.0 2040 60 80 100 020 40 60 80 100 020 40 60 80 100 Time (h) Time (h) Time (h) 26 Introduction

Introduction filled composites and medicines released from nanocap- conversion of Li ions to lithium oxide during discharg- sules will soon belong to this category. In scientific ing, and lithium oxide to lithium ions during charging communications, a vast number of applications have of the battery. In the nanosized alloy catalyst the plat- been proposed in diverse fields. Several chapters in inum part is responsible for the reduction reaction and the final part of this handbook are devoted to ap- the gold part helps the oxidation. plications, and in all the other parts, each chapter contains a short section showcasing applications of 1.4.2 Energy Conversion and Storage the nanomaterials considered. Here, we provide only a glimpse on applications, grouped into the categories Harvesting clean energy – mainly by converting nat- of catalysts and catalyst templates, nanomaterials in urally present energy to electricity – from abundant energy conversion, and storage and sensors based on resources is one of the most important tasks to be ac- nanomaterials. complished by modern engineering. As clean energy sources (e.g., solar, wind) are not correlated with en- 1.4.1 Catalysts and Catalyst Templates ergy demand in time and non-localized energy sources are needed in transportation, too, efficiency in both Heterogeneous catalytic reactions need high surface energy harvesting and energy storage is necessary. En- area for obvious reasons, as the process can be real- ergy harvesting involves macroscopic equipment, e.g., ized only at surfaces, and all materials in the volume wind turbines, however in many methods nanomaterials represent unnecessary weight and cost. There is another are used to improve efficiency. The two most relevant reason why nanomaterials can behave differently from fields are thermo- and piezoelectric conversion and solar their larger-grain counterparts; i. e., they have different energy conversion by solar cells. Thermoelectric nano- selectivity features. In surface it is well known materials convert thermal energy between two locations that different crystal planes and different defect sites on at different temperatures. The measure of the efficiency the surface (e.g., steps, kinks) promote different reac- of a material in thermoelectric conversion is the ZT tions for the same conditions [1.151]. It has also been parameter, the ratio of the Seebeck coefficient and the pointed out that, in homogeneous catalysis, edge and thermal and electrical conductivity of the material. Su- corner atoms play an exceptional role; however, their perlattices are the most efficient conversion devices, ratio can change during the reaction, and in fact surface however their fabrication is costly and scaling up is reconfiguration is common [1.155]. In nanomaterials, complicated. Work has been carried out for efficient the number of this kind of defects is naturally higher, use of simpler systems, e.g., silicon nanowires [1.79]. depending also on their size and shape, so control of Another way to apply nanomaterials in energy con- these parameters provides well-regulated reactions in version is by using quantum dots – nanoparticles – in a multichannel process. Figure 1.17a displays an exam- solar cells in order to trap more photons or capture ple of how the size and shape of platinum nanoparticles more energy from photons [1.156]. Particular cases of govern the completeness of a pyrrole hydrogenation reaction [1.151]. Figure 1.17b shows how the size of Fig. 1.17a–c Application of nanomaterials as catalysts. a cobalt catalyst changes the yields of Fischer–Tropsch (a) Pyrrole hydrogenation reaction catalyzed by platinum synthesis [1.152]. The yield is proportional to the dis- nanoparticles having different size; the reaction product is persion of cobalt atoms, and the site-time yield is a mix of pyrrolidine and n-butylamine for small (1–2 nm) accordingly constant, which means that the size of the particles and mostly n-butylamine for larger (3–5 nm) par- particles has an important effect; however, it does not ticles; the amount of butane and ammonia slowly increases change the surface features, so the authors could not with increasing particle size (after [1.151]). (b) CO con- find a shape dependence version time yield and turnover rate in a Fischer–Tropsch Figure 1.17c introduces a specific, electrochemical, reaction as a function of cobalt dispersion [1.152]. (c) En- use of gold–platinum nanocatalyst used in Li–air and ergy density for several commonly investigated batteries; Zn–air batteries [1.153]. It is clear that these batteries schematic of a possible configuration for Li–air or Zn–air have much higher specific energy storage capacity than battery, showing the location of the catalyst; charge– conventionally used ones. The battery works similarly discharge characteristics of a Li–air cell displaying the role to Li-ion batteries inside, however they have a special of the PtAu catalyst particle in the reduction and oxida- porous anode where the air – in some cases oxygen or tion steps (ORR: oxygen reduction reaction; OER: oxygen ozone – enters. Here, a catalyst is needed for efficient evolution reaction) (after [1.153, 154])  Science and Engineering of Nanomaterials 1.4 Typical Applications of Nanomaterials 27 Introduction a) H H b) Cobalt-time yield (104 s–1) N +2H2 N +H2 +H2 +H2N +NH3 30 TiO SiO Al O Other Pyrrole Pyrrolidine n-Butylamine Butane and 2 2 2 3 Co Ammonia Selectivity (%) TiO2 SiO2 Co-Ru 25 100 90 n-Butylamine 80 20 70 60 15 50 40 10 30 Pyrrolidine 20 5 10 Butane and Ammonia 0 0 0 1 2 3 4 5 0 0.02 0.04 0.05 0.08 0.1 0.12 0.14 Selectivity (%) Pt size (nm) Site-time yield (103 s–1) Co dispersion 100

90 n-Butylamine 70

50 5 nm Pt nanocubes 10 Pyrrolidine 5 nm Pt nanopolyhedra 30 TiO2 SiO2 Al2O3 Other Co TiO SiO 10 Butane and Ammonia 2 2 C-Ru

1 380 390 400 410 420 0 0.02 0.04 0.05 0.08 0.1 0.12 0.14 0.16 0.18 Pt size (nm) Cobalt fractional dispersion c) E vs. Li (V ) 5.0 Li 12 000 (a) (b) 10 000 Theoretical specific energy (Wh/kg) Practical specific energy (Wh/kg) 4.5 8000 Ar-filled + – 6000 xLi +O2+xe OER @ Pt Carbon 4000 4.0 L O O -filled 2000 x 2 2 0 3.5

Li-air Ni/Cd Ni/MH Li-ion Zn-air Lead/acid 3.0 Discharge PtAu/C e– e– 2.5 Lithium Organic xLi++O +xe– metal electrolyte 2 ORR @ Au O2 + Air Li 2.0 LxO2

Porous carbon 0800 800 1200 –400 0 Catalysts Q(mAh/gcarbon) 28 Introduction

Introduction clean energy conversion that use nanomaterials are dis- Fig. 1.18a–c Application of nanomaterials in energy con- played in Fig. 1.18a. The measurements show a strong version and storage. (a) Dependence of photoreduction dependence of the reduction rate achieved with TiO2 efficiency of TiO2 particles on the size of a gold on the gold particle size; the gold template modifies the nanoparticle used as a substrate. The smaller the par- Fermi level, and the shift in the Fermi energy increases ticle, the higher the reduction efficiency. Schemes of with decreasing particle size. This reaction is used in (H2PCnMPC + C60)m configurations are also displayed. photoelectrochemical energy conversion. It was also These particles are applicable for photo-induced cur- shown that the solar energy conversion is dependent on rent; the photocurrent for the visible spectrum is dis- the configuration of the optically transparent electrode played for different configurations of the structure of (OTE/SnO2/(H2PCnMPC + C60)m), where the length OTE/SnO2/(H2PCnMPC + C60)m with parameters of of the linker (the number of fullerene molecules) plays [H2P] = 0.19 mM, n = 15, Cn = [C60] = 0.38 mM (af- the most important role [1.72]. For energy storage, ter [1.72, 74]). (b) Evolution of MnO2 morphology as we have collected examples from our previous work a function of the dwell time of hydrothermal synthe- in Fig. 1.18b. The SEM images show an MnO2 struc- sis. Supercapacitors prepared with nanowire electrodes ture prepared by hydrothermal synthesis [1.73]. The have 150–200 F/g specific capacitance (after [1.73]). nanowires obtained have potential applications in super- (c) Physical appearance of nanomaterial supercapacitors; capacitor devices. We also present an atomically thin, transparent, flexible supercapacitor with graphene elec- transparent, flexible supercapacitor based on single- or trodes (after [1.75]) and laser-defined GO electrode–RGO few-layer graphene electrodes. The transparency of the electrolyte (after [1.76]) (IPCE: incident photon to charge GO and RGO layers are demonstrated placed them onto carrier efficiency; PVA: polyvinyl alcohol)  a printout; here the thickness of the multilayered films was set to ≈ 10 nm by consecutive dip-coating steps. method is based on measurement of a signal as a func- The last part (Fig. 1.18b) shows a supercapacitor array tion of time, analysis of this signal, and in most cases designed on a monolithic GO piece. Parts of the GO fast Fourier transformation (FFT), and finally interpre- were reduced by a laser beam, and these RGO parts tation of the power spectral density in characteristic serve as electrodes while the unreduced GO acts as frequency ranges to identify the chemical environ- a solid electrolyte. ment [1.78, 81]. This interpretation can be a simple linearization, application of a neural network, or use 1.4.3 Sensors Based on Nanomaterials of principal component analysis (PCA) plots. All these interpretation methods allow selectivity for different Nanomaterials are excellent choices for use as active gases and for different concentrations. Any kind of sen- elements of chemical (gas) sensors. To achieve high sor can be essentially used with this method; however, sensitivity, the material needs to have high dispersion, sensors made of nanomaterials have the advantage of as surface atoms interact and undergo changes whereas much stronger FES signals and significantly more in- atoms inside the material do not change, meaning that, formation content [1.80, 82, 83]. Carbon nanotubes are the higher the surface atom ratio, the larger the relative especially useful and important building blocks of FES change in the measured physical property of the ma- devices, as they can be functionalized with appropriate terial. Similarly to applications as catalysts, in sensors chemical groups that make them selective for particu- not only the size of the particles is important but also lar materials [1.84]. Functionalization is especially well their shape. As the shape, number, and configuration studied in the case of CNTs for use in biosensors [1.85]. of edge and corner atoms change, further differences Figure 1.19b displays a print made with a carbon nano- occur in the signals measured. Figure 1.19a demon- tube ink. A similar printing method and ink were used strates the effect of various shapes when nanoparticles, to create the sensors for the devices providing the sig- nanowires, and nanoplatelets of tungsten oxide were nals in the lower row. The fluctuation power spectra used [1.77]. In this case, the nanowires have the high- were measured using a Taguchi-type sensor, and show est sensitivity, followed by the platelets and finally the measurable differences between three different gases nanoparticles. compared with a synthetic air reference. The PCA plots Fluctuation-based sensing or fluctuation-enhanced demonstrate the sensitivity and selectivity of the printed sensing (FES) provides selectivity via investigation carbon nanotube sensors. Signals of 30 ppb concentra- of fluctuations in resistance, voltage or current. The tion are shown to be selective for the measured gases. Science and Engineering of Nanomaterials 1.4 Typical Applications of Nanomaterials 29 Introduction a) Reduction efficiency (%) 25 Au

TiO2 20

15

10 Au 5

0 No Au 8 nm Au 5 nm Au 3 nm Au 300 nm

IPCE (%) 60 17.3 Å

50 d

50 c

NHCO 30.94 Å NHCO 50 b (CH2)5 (CH2)5 4.24 Å 50 10.05 Å 10 a 0 400 600 800 1000 Wavelength (nm) b) 1h 6h 18h

X50.000 100 nm X50.000 100 nm X50.000 100 nm c) Electrical contact

Gold current collector GO film RGO film Graphene acitor 2-D supercapacitor 2-D su PVA-H3PO4 (Gel electrolyte) acitor 2-D supercapacitor 2-D su Graphene acitor 2-D supercapacitor 2-D su Gold current collector acitor 2-D supercapacitor 2-D su 30 Introduction Introduction a) 2000 1000 ppm 1600 500 ppm 1200 800

Response 100 ppm 400 50 ppm 10 ppm 1 ppm 0 0 400 800 1200 1600 2000 Time (s) 3500 WO nanowires 3000 2.72 WO nanoplatelets 2500 3 100 nm 300 nm 2000 WO 3 nanoparticles 1500 Response 1000 500 0 50 100 150 200 250 Temperature (°C) 3.5 5nm 3.0 2.5 2.0 Log (response) 1.5

0 1 23 50 nm Log (concentration (ppm))

–2 2 2 b) R normalized power spectrum (V /Ω ) 10–19 H2 500 ppm 10–21 NO 10 ppm –23 10 Synthetic air

10–25

SO2 500 ppm 10–27 110100 Frequency (Hz) PC–2×10–14 (0.16 %) PC–2×10–15 (0.34 %)

5 5 30 ppb 0 0.1 ppm 0.1 ppm 0 N2O H2S 30 ppb N2O –5 –5 H2S

–15 –10 –5 0 5 10 15 –20 –15 –10 –5 0 5 10 PC–1×10–14 (99.41 %) PC–1×10–15 (98.59 %) Science and Engineering of Nanomaterials References 31

Fig. 1.19a,b Application of nanomaterials in sensors. (a) SEM and TEM images of tungsten oxide nanoplatelets and Introduction TEM of WO2.72 nanowires. Measuring different concentrations of H2S using a nanoplatelet sensor. Gas sensing mea- ◦ surement: sensor signal as a function of H2S concentration at 250 C. Comparative graphs showing sensitivity of different nanostructures as a function of temperature and concentration (after [1.77]). (b) Demonstration of selectivity achieved by the fluctuation-enhanced sensing method. An image printed using nanotube solution as ink, similar to the ones used to print circuits for sensors (after [1.157]). Linearized spectra measured on thick-film nanoparticle sensors (after [1.81]). PCA plots of FES of N2O(circles)andH2S(squares) measured with an MWCNT sensor showing both sensitivity and selectivity (PC: principal component) (after [1.83]) 

1.5 Concluding Remarks Nanomaterials differ in many regards from their macro- properties of materials as a function of their charac- scopic – bulk – allotropes, and accordingly they call teristic size clearly shows that nanomaterials represent for special scientific and engineering approaches. The a one-of-a-kind, coherent group of materials. The later large number of scientific articles, encyclopedias, and chapters in this handbook elaborate on the synthesis, handbooks in related topics, as well as textbooks and specific properties, and applications of the various ma- full courses, show the growing importance of this terial groups. Our main objective in this chapter was field. to collect and present material properties in a way that In this chapter we could give only a glimpse of would introduce novice readers to a new topic, and at the numerous unique and interesting, mainly univer- the same time serve as a reference for the everyday work sal properties of nanosized materials. This similarity of of well-established scientists and engineers.

1.6 About the Contents of the Handbook The structure of the handbook follows the classification and building blocks, as well as solutions, so-called of important material groups. Part A describes carbon- nanoinks. Part E deals with porous nanomaterials, met- based nanomaterials from fullerenes to nanotubes and als, ceramics, and silicon. Part F presents examples of nanofibers, as well as nanodiamonds. Part B introduces organic and bio-nanomaterials, bones, and fibers. Fi- noble and common metals and alloys. Part C describes nally, Part G contains several chapters on the topics of ceramic materials, crystalline and glassy oxides, and applications in nanomedicine, civil engineering, toxi- other compounds. Part D describes composites and cology, and hazards, as well as energy harvesting and hybrid structures where nanomaterials serve as fillers energy storage.

References

1.1 E. Rzepka: Raman scattering (2009) ment of over a million compared to the gold metal, http://www.raman-scattering.eu/raman/index.php Chem. Phys. Lett. 317(6), 517–523 (2000) 1.2 H. Fountain: Antique Nanotubes, The New York 1.6 M.A. Peiyan, F.U. Zhengyi, S.U. Yanli, M.A. Jingjing: Times, November 28 (2006), available online at The nano pulverization of traditional Chinese http://www.nytimes.com/2006/11/28/science/ medicine Liuwei Dihuang, J. Wuhan Univ. Technol. 28observ.Html?_r=1 Mater. Sci. Ed. 21(2), 105–108 (2006) 1.3 C.G. Granqvist, R.A. Buhrman: Ultrafine metal par- 1.7 M. José-Yacaman, L. Rendón, J. Arenas, M.C.S. Puche: ticles,J.Appl.Phys.47(5), 2200–2219 (1976) Maya blue paint: An ancient nanostructured mater- 1.4 M.S. Dresselhaus, C. Dresselhaus, P.C. Eklund: Sci- ial, Science 273(5272), 223–225 (1996) ence of Fullerenes and Carbon Nanotubes (Academic, 1.8 G. Padeletti, P. Fermo: How the masters in Umbria, San Diego 1995) Italy, generated and used nanoparticles in art fab- 1.5 M.B. Mohamed, V. Volkov, S. Link, M.A. El-Sayed: The rication during the Renaissance period, Appl. Phys. ‘lightning’ gold nanorods: Fluorescence enhance- A 76(4), 515–525 (2003) 32 Introduction

Introduction 1.9 M.-C. Daniel, D. Astruc: Gold nanoparticles: As- 1.25 R.D. Piner, J. Zhu, F. Xu, S.H. Hong, C.A. Mirkin: “Dip- sembly, supramolecular chemistry, quantum-size- pen” nanolithography, Science 283(5402), 661–663 related properties, and applications toward biology, (1999) catalysis, and nanotechnology, Chem. Rev. 104(1), 1.26 C.R. Martin: Membrane-based synthesis of nano- 293–346 (2003) materials, Chem. Mater. 8(8), 1739–1746 (1996) 1.10 M. Reibold, P. Paufler, A.A. Levin, W. Kochmann, 1.27 G.W.Meng,F.M.Han,X.L.Zhao,B.S.Chen, N. Patzke, D.C. Meyer: Materials: Carbon nanotubes D.C. Yang, J.X. Liu, Q.L. Xu, M.G. Kong, X.G. Zhu, in an ancient Damascus sabre, Nature 444(7117), Y.J. Jung, Y.J. Yang, Z.Q. Chu, M. Ye, S. Kar, 286–286 (2006) R. Vajtai, P.M. Ajayan: A general synthetic ap- 1.11 J. Bardeen, W.H. Brattain: The transistor, a semi- proach to interconnected nanowire/nanotube and conductor triode, Phys. Rev. 74(2), 230–231 (1948) nanotube/nanowire/nanotube heterojunctions with 1.12 J.S. Kilby: Invention of integrated-circuit, IEEE Trans. branched topology, Angew. Chem. Int. Ed. 48(39), Electron Devices 23(7), 648–654 (1976) 7166–7170 (2009) 1.13 R.P. Feynman: There’s plenty of room at the bottom, 1.28 H. Chun, M.G. Hahm, Y. Homma, R. Meritz, K. Ku- Eng. Sci. 23(5), 22–36 (1960) ramochi, L. Menon, L. Ci, P.M. Ajayan, Y.J. Jung: 1.14 R.W. Siegel, S. Ramasamy, H. Hahn, Z.Q. Li, T. Lu, Engineering low-aspect ratio carbon nanostruc- R. Gronsky: Synthesis, characterization, and proper- tures: Nanocups, nanorings, and nanocontainers, ties of nanophase TiO2,J.Mater.Res.3(6), 1367–1372 ACS Nano 3(5), 1274–1278 (2009) (1988) 1.29 A.R. Tao, S. Habas, P.D. Yang: Shape control of col- 1.15 Office of Basic Energy Sciences (BES): The Scale loidal metal nanocrystals, Small 4(3), 310–325 (2008) of Things (US Department of Energy, Washington 1.30 J. Banhart: Manufacture, characterisation and ap- D.C. 2006), available online at http://science.energy. plication of cellular metals and metal foams, gov/bes/news-and-resources/scale-of-things- Progress Mater. Sci. 46(6), 559–632 (2001) chart/ 1.31 C. Suryanarayana: Nanocrystalline materials, Int. 1.16 P. Avouris, R. Martel, T. Hertel, R. Sandstrom: AFM- Mater. Rev. 40(2), 41–64 (1995) tip-induced and current-induced local oxidation of 1.32 J. Soderlund, L.B. Kiss, G.A. Niklasson, C.G. Granqvist: silicon and metals, Appl. Phys. A 66, S659–S667 Lognormal size distributions in particle growth pro- (1998) cesses without coagulation, Phys. Rev. Lett. 80(11), 1.17 N. Chekurov, K. Grigoras, A. Peltonen, S. Franssila, 2386–2388 (1998) I. Tittonen: The fabrication of silicon nanostructures 1.33 K. Hata, D.N. Futaba, K. Mizuno, T. Namai, M. Yu- by local gallium implantation and cryogenic deep mura, S. Iijima: Water-assisted highly efficient reactive ion etching, Nanotechnology 20(6), 065307 synthesis of impurity-free single-waited carbon (2009) nanotubes, Science 306(5700), 1362–1364 (2004) 1.18 E. Perez-Tijerina, M.A. Gracia-Pinilla, S. Mejia- 1.34 H.W. Zhu, C.L. Xu, D.H. Wu, B.Q. Wei, R. Vajtai, Rosales, U. Ortiz-Mendez, A. Torres, M. Jose- P.M. Ajayan: Direct synthesis of long single-walled Yacaman: Highly size-controlled synthesis of Au/Pd carbon nanotube strands, Science 296(5569), 884– nanoparticles by inert-gas condensation, Faraday 886 (2002) Discuss. 138, 353–362 (2008) 1.35 B.Q. Wei, R. Vajtai, Y. Jung, J. Ward, R. Zhang, 1.19 L.B. Kiss, J. Söderlund, G.A. Niklasson, C.G. Granqvist: G. Ramanath, P.M. Ajayan: Organized assembly of The real origin of lognormal size distributions of carbon nanotubes – cunning refinements help to nanoparticles in vapor growth processes, Nanos- customize the architecture of nanotube structures, truct. Mater. 12(1–4), 327–332 (1999) Nature 416(6880), 495–496 (2002) 1.20 D.D. Majumder, R. Banerjee, C. Ulrichs, I. Mewis, 1.36 L. Song, L. Ci, H. Lu, P.B. Sorokin, C. Jin, J. Ni, A. Goswami: Nano-materials: Science of bottom-up A.G. Kvashnin, D.G. Kvashnin, J. Lou, B.I. Yakobson, and top-down, IETE Tech. Rev. 24(1), 9–25 (2007) P.M. Ajayan: Large scale growth and characteriza- 1.21 K. Ariga, J.P. Hill, M.V. Lee, A. Vinu, R. Charvet, tion of atomic hexagonal boron nitride layers, Nano S. Acharya: Challenges and breakthroughs in recent Lett. 10(8), 3209–3215 (2010) research on self-assembly, Sci. Technol. Adv. Mater. 1.37 B. Nikoobakht, M.A. El-Sayed: Preparation and 9(1), 014109 (2008) growth mechanism of gold nanorods (NRs) us- 1.22 C. Park, J. Yoon, E.L. Thomas: Enabling nanotechnol- ing seed-mediated growth method, Chem. Mater. ogy with self assembled block copolymer patterns, 15(10), 1957–1962 (2003) Polymer 44(22), 6725–6760 (2003) 1.38 M. Grzelczak, J. Perez-Juste, P. Mulvaney, L.M. Liz- 1.23 F.S. Bates, G.H. Fredrickson: Block copolymers – de- Marzan: Shape control in gold nanoparticle synthe- signer soft materials, Phys. Today 52(2), 32–38 (1999) sis, Chem. Soc. Rev. 37(9), 1783–1791 (2008) 1.24 J.Y. Cheng, C.A. Ross, V.Z.H. Chan, E.L. Thomas, 1.39 Y.D. Yin, R.M. Rioux, C.K. Erdonmez, S. Hughes, R.G.H. Lammertink, G.J. Vancso: Formation of G.A. Somorjai, A.P. Alivisatos: Formation of hol- a cobalt magnetic dot array via block copolymer low nanocrystals through the nanoscale Kirkendall lithography, Adv. Mater. 13(15), 1174–1178 (2001) effect, Science 304(5671), 711–714 (2004) Science and Engineering of Nanomaterials References 33

1.40 H.J. Fan, U. Gosele, M. Zacharias: Formation of 1.57 C.Q. Sun, B.K. Tay, X.T. Zeng, S. Li, T.P. Chen, J. Zhou, Introduction nanotubes and hollow nanoparticles based on Kirk- H.L. Bai, E.Y. Jiang: Bond-order-bond-length- endall and diffusion processes: A review, Small 3(10), bond-strength (bond-OLS) correlation mechanism 1660–1671 (2007) for the shape-and-size dependence of a nanosolid, 1.41 A. Gedanken: Using sonochemistry for the fabrica- J. Phys. D 14(34), 7781–7795 (2002) tion of nanomaterials, Ultrason. Sonochem. 11(2), 1.58 T. Halicioglu: Calculation of surface energies for low 47–55 (2004) index planes of diamond, Surface Sci. 259(1–2), L714– 1.42 C.J. Hawker, T.P. Russell: Block copolymer lithog- L718 (1991) raphy: Merging “bottom-up” with “top-down” 1.59 A. Baraldi, S. Lizzit, G. Comelli, G. Paolucci: Oxygen processes, MRS Bulletin 30(12), 952–966 (2005) adsorption and ordering on Ru (10¯10),Phys.Rev.B 1.43 N.J. Halas, S. Lal, W.S. Chang, S. Link, P. Nordlander: 63(11), 115410 (2001) Plasmons in strongly coupled metallic nanostruc- 1.60 H. Over, G. Kleinle, G. Ertl, W. Moritz, K.H. Ernst, tures, Chem. Rev. 111(6), 3913–3961 (2011) H. Wohlgemuth, K. Christmann, E. Schwarz: A leed 1.44 H.W. Kroto, J.R. Heath, S.C. O’Brien, R.F. Curl, structural-analysis of the Co(1010) surface, Surface R.E. Smalley: C60: Buckminsterfullerene, Nature Sci. 254(1–3), L469–L474 (1991) 318(6042), 162–163 (1985) 1.61 H.L. Davis, D.M. Zehner: Structure of the clean 1.45 J. Park, J. Joo, S.G. Kwon, Y. Jang, T. Hyeon: Synthe- re(1010) surface, J. Vac. Sci. Technol. 17(1), 190–193 sis of monodisperse spherical nanocrystals, Angew. (1980) Chem.Int.Ed.46(25), 4630–4660 (2007) 1.62 X. Qian, W. Hubner: First-principles calculation of 1.46 J.Park,E.Lee,N.-M.Hwang,M.Kang,S.C.Kim, structural and magnetic properties for Fe mono- Y. Hwang, J.-G. Park, H.-J. Noh, J.-Y. Kim, layers and bilayers on W(110), Phys. Rev. B 60(23), J.-H. Park, T. Hyeon: One-nanometer-scale size- 16192–16197 (1999) controlled synthesis of monodisperse magnetic iron 1.63 W.T. Geng, A.J. Freeman, R.Q. Wu: Magnetism at oxide nanoparticles, Angew. Chem. Int. Ed. 44(19), high-index transition-metal surfaces and the effect 2872–2877 (2005) of metalloid impurities: Ni(210), Phys. Rev. B 63(6), 1.47 X.F. Duan, C.M. Lieber: General synthesis of com- 064427 (2001) pound semiconductor nanowires, Adv. Mater. 12(4), 1.64 W.T. Geng, M. Kim, A.J. Freeman: Multilayer relax- 298–302 (2000) ation and magnetism of a high-index transition 1.48 S. Iijima, T. Ichihashi: Single-shell carbon nano- metal surface: Fe(310), Phys. Rev. B 63(24), 245401 tubes of 1-nm diameter, Nature 363(6430), 603–605 (2001) (1993) 1.65 I.P. Batra: Lattice-relaxation in aluminum mono- 1.49 D.S. Bethune, C.H. Kiang, M.S. Devries, G. Gorman, layers,J.Vac.Sci.Technol.Vac.Surf.andFilms3(3), R. Savoy, J. Vazquez, R. Beyers: Cobalt-catalyzed 1603–1606 (1985) growth of carbon nanotubes with single-atomic- 1.66 P.J. Feibelman: Relaxation of hcp(0001) surfaces: layerwalls, Nature 363(6430), 605–607 (1993) A chemical view, Phys. Rev. B 53(20), 13740–13746 1.50 S. Iijima: Helical microtubules of graphitic carbon, (1996) Nature 354(6348), 56–58 (1991) 1.67 C.Q. Sun: O-Cu(001): I. Binding the signatures of 1.51 A.K. Geim, K.S. Novoselov: The rise of graphene, Nat. LEED, STM and PES in a bond-forming way, Surf. Rev. Mater. 6(3), 183–191 (2007) and Lett. 8(3–4), 367–402 (2001) 1.52 S. Park, R.S. Ruoff: Chemical methods for the 1.68 C.Q. Sun: O-Cu(001): Ii. Vleed quantification of the production of graphenes, Nat. Nanotechnol. 4(4), four-stage Cu3O2 bonding kinetics, Surf. Rev. Lett. 217–224 (2009) 8(6), 703–734 (2001) 1.53 L. Ci, L. Song, C.H. Jin, D. Jariwala, D.X. Wu, Y.J. Li, 1.69 C.Q. Sun: Time-resolved vleed from the O-Cu(001): A. Srivastava, Z.F. Wang, K. Storr, L. Balicas, F. Liu, Atomic processes of oxidation, Vacuum 48(6), 525– P.M. Ajayan: Atomic layers of hybridized boron 530 (1997) nitride and graphene domains, Nat. Mater. 9(5), 1.70 C.Q. Sun, B.K. Tay, S.P. Lau, X.W. Sun, X.T. Zeng, S. Li, 430–435 (2010) H.L. Bai, H. Liu, Z.H. Liu, E.Y. Jiang: Bond contraction 1.54 J. Zhu, Z. Konya, V.F. Puntes, I. Kiricsi, C.X. Miao, and lone pair interaction at nitride surfaces, J. Appl. J.W. Ager, A.P. Alivisatos, G.A. Somorjai: Encap- Phys. 90(5), 2615–2617 (2001) sulation of metal (Au, Ag, Pt) nanoparticles into 1.71 D.A.Walker,B.Kowalczyk,M.O.delaCruz,B.A.Grzy- the mesoporous SBA-15 structure, Langmuir 19(10), bowski: Electrostatics at the nanoscale, Nanoscale 4396–4401 (2003) 3(4), 1316 (2011) 1.55 H. Gleiter: Nanostructured materials: Basic concepts 1.72 P.V. Kamat: Meeting the clean energy demand: and microstructure, Acta Mater. 48(1), 1–29 (2000) Nanostructure architectures for solar energy conver- 1.56 K.J.M. Bishop, C.E. Wilmer, S. Soh, B.A. Grzybowski: sion, J. Phys. Chem. C 111(7), 2834–2860 (2007) Nanoscale forces and their uses in self-assembly, 1.73 V. Subramanian, H.W. Zhu, R. Vajtai, P.M. Ajayan, Small 5(14), 1600–1630 (2009) B.Q. Wei: Hydrothermal synthesis and pseudocapac- 34 Introduction Introduction itance properties of MnO2 nanostructures, J. Phys. bio, I. Kiricsi: Carbon nanotube based sensors and Chem. B 109(43), 20207–20214 (2005) fluctuation enhanced sensing, Phys. Status Solidi (c) 1.74 T.Hasobe,H.Imahori,P.V.Kamat,T.K.Ahn, 7(3–4), 1217–1221 (2010) S.K. Kim, D. Kim, A. Fujimoto, T. Hirakawa, 1.85 J. Wang: Carbon-nanotube based electrochemi- S. Fukuzumi: Photovoltaic cells using composite cal biosensors: A review, Electroanalysis 17(1), 7–14 nanoclusters of porphyrins and fullerenes with gold (2005) nanoparticles, J. Am. Chem. Soc. 127(4), 1216–1228 1.86 Q. Jiang, L.H. Liang, D.S. Zhao: Lattice contraction (2005) and surface stress of fcc nanocrystals, J. Phys. Chem. 1.75 J.J. Yoo, K. Balakrishnan, J.S. Huang, V. Me- B 105(27), 6275–6277 (2001) unier, B.G. Sumpter, A. Srivastava, M. Conway, 1.87 Q.S. Chang, X.W. Sun, B.K. Tay, S.P. Lau, H.T. Huang, A.L.M. Reddy, J. Yu, R. Vajtai, P.M. Ajayan: Ultrathin S. Li: Dielectric suppression and its effect on pho- planar graphene supercapacitors, Nano Lett. 11(4), toabsorption of nanometric semiconductors, J. Phys. 1423–1427 (2011) D 34(15), 2359 (2001) 1.76 W. Gao, N. Singh, L. Song, Z. Liu, A.L.M. Reddy, L.J. Ci, 1.88 S.R. Bahn, K.W. Jacobsen: Chain formation of metal R. Vajtai, Q. Zhang, B.Q. Wei, P.M. Ajayan: Direct atoms, Phys. Rev. Lett. 87(26), 266101 (2001) laser writing of micro-supercapacitors on hydrated 1.89 E. Arzt: Overview no. 130 – size effects in materials graphite oxide films, Nat. Nanotechnol. 6(8), 496– due to microstructural and dimensional constraints: 500 (2011) A comparative review, Acta Mater. 46(16), 5611–5626 1.77 C.S. Rout, M. Hegde, C.N.R. Rao: H2s sensors based (1998) on tungsten oxide nanostructures, Sens. Actuator B 1.90 R.A. Andrievski: Size-dependent effects in proper- 128(2), 488–493 (2008) ties of nanostructured materials, Rev. Adv. Mater. 1.78 L.B. Kish, R. Vajtai, C.G. Granqvist: Extracting in- Sci. 21(2), 107–133 (2009) formation from noise spectra of chemical sensors: 1.91 N. Dubrovinskaia, V.L. Solozhenko, N. Miyajima, Single sensor electronic noses and tongues, Sensor V. Dmitriev, O.O. Kurakevych, L. Dubrovinsky: Super- Actuat or B 71(1–2), 55–59 (2000) hard nanocomposite of dense polymorphs of boron 1.79 A.I. Boukai, Y. Bunimovich, J. Tahir-Kheli, J.K. Yu, nitride: Noncarbon material has reached diamond W.A. Goddard, J.R. Heath: Silicon nanowires as ef- hardness, Appl. Phys. Lett. 90(10), 101912–1–101912–3 ficient thermoelectric materials, Nature 451(7175), (2007) 168–171 (2008) 1.92 R.W. Siegel, G.E. Fougere: Mechanical properties of 1.80 H. Haspel, R. Ionescu, P. Heszler, A. Kukovecz, nanophase metals, Nanostruct. Mater. 6(1–4), 205– Z. Konya, Z. Gingl, J. Maklin, T. Mustonen, K. Ko- 216 (1995) rdas, R. Vajtai, P.M. Ajayan: Fluctuation enhanced 1.93 B.I. Yakobson, P. Avouris: Mechanical properties of gas sensing on functionalized carbon nanotube carbon nanotubes, Carbon Nanotub. 80,287–327 thin films, Phys. Status Solidi (b) 245(10), 2339–2342 (2001) (2008) 1.94 B.I. Yakobson, C.J. Brabec, J. Bernholc: Nanome- 1.81 L.B. Kish, Y.F. Li, J.L. Solis, W.H. Marlow, R. Vajtai, chanics of carbon tubes: Instabilities beyond C.G. Granqvist, V. Lantto, J.M. Smulko, G. Schme- linear response, Phys. Rev. Lett. 76(14), 2511–2514 ra: Detecting harmful gases using fluctuation- (1996) enhanced sensing with taguchi sensors, IEEE Sens. 1.95 B.I. Yakobson, R.E. Smalley: Fullerene nanotubes: C- J. 5(4), 671–676 (2005) 1000000 and beyond, Am. Sci. 85(4), 324–337 (1997) 1.82 J. Maklin, T. Mustonen, N. Halonen, G. Toth, K. Ko- 1.96 M.M.J. Treacy, T.W. Ebbesen, J.M. Gibson: Ex- rdas, J. Vahakangas, H. Moilanen, A. Kukovecz, ceptionally high young’s modulus observed for Z. Konya, H. Haspel, Z. Gingl, P. Heszler, R. Vajtai, individual carbon nanotubes, Nature 381(6584), P.M. Ajayan: Inkjet printed resistive and chemical- 678–680 (1996) fet carbon nanotube gas sensors, Phys. Status Solidi 1.97 C. Lee, X.D. Wei, J.W. Kysar, J. Hone: Measurement (b) 245(10), 2335–2338 (2008) of the elastic properties and intrinsic strength of 1.83 D. Molnar, P. Heszler, R. Mingesz, Z. Gingl, monolayer graphene, Science 321(5887), 385–388 A.Kukovecz,Z.Konya,H.Haspel,M.Mohl,A.Sapi, (2008) I. Kiricsi, K. Kordas, J. Maklin, N. Halonen, G. Toth, 1.98 W.A.Deheer,A.Chatelain,D.Ugarte:Acarbon H. Moilanen, S. Roth, R. Vajtai, P.M. Ajayan, Y. Pouil- nanotube field-emission electron source, Science lon, A. Rubio: Increasing chemical selectivity of 270(5239), 1179–1180 (1995) carbon nanotube-based sensors by fluctuation-en- 1.99 Y.N. Xia, P.D. Yang, Y.G. Sun, Y.Y. Wu, B. Mayers, hanced sensing, Fluct. Noise Lett. 9(3), 277–287 B. Gates, Y.D. Yin, F. Kim, Y.Q. Yan: One-dimensional (2010) nanostructures: Synthesis, characterization, and 1.84 Á. Kukovecz, D. Molnár, K. Kordás, Z. Gingl, H. Moila- applications, Adv. Mater. 15(5), 353–389 (2003) nen, R. Mingesz, Z. Kónya, J. Mäklin, N. Halonen, 1.100 R. Landauer: Electrical resistance of disordered one- G.Tóth,H.Haspel,P.Heszler,M.Mohl,A.Sápi, dimensional lattices, Philos. Mag. 21(172), 863–867 S. Roth, R. Vajtai, P.M. Ajayan, Y. Pouillon, A. Ru- (1970) Science and Engineering of Nanomaterials References 35

1.101 C.T. White, T.N. Todorov: Carbon nanotubes as 1.117 G.Rollmann,M.E.Gruner,A.Hucht,R.Meyer,P.En- Introduction long ballistic conductors, Nature 393(6682), 240–242 tel, M.L. Tiago, J.R. Chelikowsky: Shellwise mackay (1998) transformation in iron nanoclusters, Phys. Rev. Lett. 1.102 R. Martel, T. Schmidt, H.R. Shea, T. Hertel, P. Avouris: 99(8), 083402 (2007) Single- and multi-wall carbon nanotube field- 1.118 W.H. Zhong, C.Q. Sun, S. Li: Size effect on the effect transistors, Appl. Phys. Lett. 73(17), 2447–2449 magnetism of nanocrystalline Ni films at ambient (1998) temperature, Solid State Commun. 130(9), 603–606 1.103 J. Park, A.N. Pasupathy, J.I. Goldsmith, C. Chang, (2004) Y. Yaish, J.R. Petta, M. Rinkoski, J.P. Sethna, 1.119 Y.G. Sun, Y.N. Xia: Gold and silver nanoparticles: H.D. Abruna, P.L. McEuen, D.C. Ralph: Coulomb A class of chromophores with colors tunable in the blockade and the Kondo effect in single-atom tran- range from 400 to 750 nm, Analyst 128(6), 686–691 sistors, Nature 417(6890), 722–725 (2002) (2003) 1.104 H.W.C.Postma,T.Teepen,Z.Yao,M.Grifoni, 1.120 K.L. Kelly, E. Coronado, L.L. Zhao, G.C. Schatz: The C. Dekker: Carbon nanotube single-electron transis- optical properties of metal nanoparticles: The in- tors at room temperature, Science 293(5527), 76–79 fluence of size, shape, and dielectric environment, (2001) J. Phys. Chem. B 107(3), 668–677 (2003) 1.105 D. Goldhaber-Gordon, H. Shtrikman, D. Ma- 1.121 M.S. Dresselhaus, G. Dresselhaus, R. Saito, A. Jo- halu, D. Abusch-Magder, U. Meirav, M.A. Kastner: rio: Raman spectroscopy of carbon nanotubes, Phys. Kondo effect in a single-electron transistor, Nature Rep.-Rev. Soc. Phys. Lett. 409(2), 47–99 (2005) 391(6663), 156–159 (1998) 1.122 M.S. Dresselhaus, A. Jorio, M. Hofmann, G. Dressel- 1.106 A. Javey, P.F. Qi, Q. Wang, H.J. Dai: Ten- to 50- haus, R. Saito: Perspectives on carbon nanotubes nm-long quasi-ballistic carbon nanotube devices and graphene raman spectroscopy, Nano Lett. 10(3), obtained without complex lithography, Proc. Nat. 751–758 (2010) Acad. Sci. USA 101(37), 13408–13410 (2004) 1.123 B.K. Gupta, P. Thanikaivelan, T.N. Narayanan, 1.107 B.Q. Wei, R. Vajtai, P.M. Ajayan: Reliability and cur- L. Song, W. Gao, T. Hayashi, A. Leela Mo- rent carrying capacity of carbon nanotubes, Appl. hana Reddy, A. Saha, V. Shanker, M. Endo, Phys. Lett. 79(8), 1172–1174 (2001) A.A. Martí, P.M. Ajayan: Optical bifunctional- 1.108 Y. Zhao, J.Q. Wei, R. Vajtai, P.M. Ajayan, E.V. Barrera: ity of europium-complexed luminescent graphene doped carbon nanotube cables exceeding nanosheets, Nano Lett. 11(12), 5227–5233 (2011) specific electrical conductivity of metals, Sci. Rep. 1.124 M.J. O’Connell, S.M. Bachilo, C.B. Huffman, 1(83), 1–5 (2011) V.C. Moore, M.S. Strano, E.H. Haroz, K.L. Rialon, 1.109 A.P. Alivisatos: Semiconductor clusters, nanocrys- P.J.Boul,W.H.Noon,C.Kittrell,J.P.Ma,R.H.Hauge, tals, and quantum dots, Science 271(5251), 933–937 R.B. Weisman, R.E. Smalley: Band gap fluorescence (1996) from individual single-walled carbon nanotubes, 1.110 P. Avouris, J. Appenzeller, R. Martel, S.J. Wind: Car- Science 297(5581), 593–596 (2002) bon nanotube electronics, Proc. IEEE 91(11), 1772–1784 1.125 R.B. Weisman, S.M. Bachilo: Dependence of optical (2003) transition energies on structure for single-walled 1.111 P. Poncharal, S. Frank, Z.L. Wang, W.A. de Heer: carbon nanotubes in aqueous suspension: An em- Conductance quantization in multiwalled carbon pirical Kataura plot, Nano Lett. 3(9), 1235–1238 (2003) nanotubes, Eur. Phys. J. D 9(1–4), 77–79 (1999) 1.126 P. Buffat, J.P. Borel: Size effect on the melting 1.112 P. Avouris: Graphene: Electronic and photonic prop- temperature of gold particles, Phys. Rev. A 13(6), erties and devices, Nano Lett. 10(11), 4285–4294 2287–2298 (1976) (2010) 1.127 S. Berber, Y.K. Kwon, D. Tomanek: Unusually high 1.113 C. Kittel: Theory of the structure of ferromagnetic thermal conductivity of carbon nanotubes, Phys. domains in films and small particles, Phys. Rev. Rev. Lett. 84(20), 4613–4616 (2000) 70(11–1), 965–971 (1946) 1.128 J. Hone, M. Whitney, C. Piskoti, A. Zettl: Thermal con- 1.114 D.L. Leslie-Pelecky, R.D. Rieke: Magnetic proper- ductivity of single-walled carbon nanotubes, Phys. ties of nanostructured materials, Chem. Mater. 8(8), Rev. B 59(4), R2514–R2516 (1999) 1770–1783 (1996) 1.129 T. Borca-Tasciuc, S. Vafaei, D.A. Borca-Tasciuc, 1.115 S. Talapatra, P.G. Ganesan, T. Kim, R. Vajtai, B.Q. Wei, R. Vajtai, P.M. Ajayan: Anisotropic thermal M. Huang, M. Shima, G. Ramanath, D. Srivastava, diffusivity of aligned multiwall carbon nanotube ar- S.C. Deevi, P.M. Ajayan: Irradiation-induced mag- rays, J. Appl. Phys. 98(5), 054309–054303–6 (2005) netism in carbon nanostructures, Phys. Rev. Lett. 1.130 C. Xue, C.A. Mirkin: pH-switchable silver nanoprism 95(9), 097201 (2005) growth pathways, Angew. Chem. Int. Ed. 46(12), 1.116 P. Gambardella, A. Dallmeyer, K. Maiti, M.C. Malagoli, 2036–2038 (2007) W. Eberhardt, K. Kern, C. Carbone: Ferromagnetism 1.131 G.D. Moon, S. Ko, Y. Min, J. Zeng, Y. Xia, U. Jeong: in one-dimensional monatomic metal chains, Na- Chemical transformations of nanostructured mater- ture 416(6878), 301–304 (2002) ials, Nano Today 6(2), 186–203 (2011) 36 Introduction

Introduction 1.132 A.C. Grimsdale, K. Mullen: The chemistry of organic 1.145 G.D. Smith, R.J.H. Clark: Raman microscopy in ar- nanomaterials, Angew. Chem. Int. Ed. 44(35), 5592– chaeological science, J. Archaeol. Sci. 31(8), 1137–1160 5629 (2005) (2004) 1.133 N.G. Szwacki, A. Sadrzadeh, B.I. Yakobson: B-80 1.146 C. Wang, B. Lu, J. Zuo, S. Zhang, S. Tan, M. Suzuki, fullerene: An ab initio prediction of geometry, sta- W.T. Chase: Structural and elemental analysis on bility, and electronic structure, Phys. Rev. Lett. the nanocrystalline SnO2 in the surface of ancient 98(16), 166804–1–166804–4 (2007) chinese black mirrors, Nanostruct. Mater. 5(4), 489– 1.134 Y. Li, G. Zhou, J. Li, B.-L. Gu, W. Duan: Alkali- 496 (1995) metal-doped B80 as high-capacity hydrogen storage 1.147 D.G. Shchukin, S.V. Lamaka, K.A. Yasakau, M.L. Zhe- media, J. Phys. Chem. C 112(49), 19268–19271 ludkevich, M.G.S. Ferreira, H. Mohwald: Active (2008) anticorrosion coatings with halloysite nanocontain- 1.135 J.L. Bahr, J.M. Tour: Covalent chemistry of single- ers, J. Phys. Chem. C 112(4), 958–964 (2008) wall carbon nanotubes, J. Mater. Chem. 12(7), 1952– 1.148 F.Wang,X.Tian,Q.Li,L.Li,X.Peng:Oxidationand 1958 (2002) hot corrosion behavior of sputtered nanocrystalline 1.136 D. Wunderlich, F. Hauke, A. Hirsch: Preferred func- coating of superalloy K52, Thin Solid Films 516(16), tionalization of metallic and small-diameter single 5740–5747 (2008) walled carbon nanotubes via reductive alkylation, 1.149 J.L. Elechiguerra, L. Larios-Lopez, C. Liu, D. Garcia- J. Mater. Chem. 18(13), 1493–1497 (2008) Gutierrez, A. Camacho-Bragado, M.J. Yacaman: 1.137 W. Gao, L.B. Alemany, L. Ci, P.M. Ajayan: New in- Corrosion at the nanoscale: The case of silver sights into the structure and reduction of graphite nanowires and nanoparticles, Chem. Mater. 17(24), oxide, Nat. Chem. 1(5), 403–408 (2009) 6042–6052 (2005) 1.138 D.R. Dreyer, S. Park, C.W. Bielawski, R.S. Ruoff: The 1.150 A. Hirsch: Functionalization of single-walled carbon chemistry of graphene oxide, Chem. Soc. Rev. 39(1), nanotubes, Angew. Chem. Int. Ed. 41(11), 1853–1859 228–240 (2010) (2002) 1.139 K. Lu, Z.H. Jin: Melting and superheating of low- 1.151 Y. Li, G.A. Somorjai: Nanoscale advances in catalysis dimensional materials, Curr. Opin. Solid State Mater. and energy applications, Nano Lett. 10(7), 2289–2295 Sci. 5(1), 39–44 (2001) (2010) 1.140 J. Hone, M.C. Llaguno, M.J. Biercuk, A.T. John- 1.152 E. Iglesia: Design, synthesis, and use of cobalt- son, B. Batlogg, Z. Benes, J.E. Fischer: Thermal based Fischer-Tropsch synthesis catalysts, Appl. properties of carbon nanotubes and nanotube- Catal. A 161(1–2), 59–78 (1997) based materials, Appl. Phys. A 74(3), 339–343 1.153 J.-S. Lee, S. Tai Kim, R. Cao, N.-S. Choi, M. Liu, (2002) K.T. Lee, J. Cho: Metal-air batteries with high energy 1.141 J. Hone, M.C. Llaguno, N.M. Nemes, A.T. Johnson, density: Li-air versus Zn-air, Adv. Energy Mater. 1(1), J.E. Fischer, D.A. Walters, M.J. Casavant, J. Schmidt, 34–50 (2011) R.E. Smalley: Electrical and thermal transport prop- 1.154 Y.C. Lu, Z.C. Xu, H.A. Gasteiger, S. Chen, K. Hamad- erties of magnetically aligned single walt carbon Schifferli, Y. Shao-Horn: Platinum-gold nanoparti- nanotube films, Appl. Phys. Lett. 77(5), 666–668 cles: A highly active bifunctional electrocatalyst for (2000) rechargeable lithium-air batteries, J. Ame. Chem. 1.142 A.A. Balandin, S. Ghosh, W.Z. Bao, I. Calizo, Soc. 132(35), 12170–12171 (2010) D. Teweldebrhan, F. Miao, C.N. Lau: Superior thermal 1.155 C. Burda, X. Chen, R. Narayanan, M.A. El-Sayed: conductivity of single-layer graphene, Nano Lett. Chemistry and properties of nanocrystals of different 8(3), 902–907 (2008) shapes, Chem. Rev. 105(4), 1025–1102 (2005) 1.143 G.K. Dimitrakakis, E. Tylianakis, G.E. Froudakis: Pil- 1.156 A.J. Nozik: Nanoscience and nanostructures for pho- lared graphene: A new 3-d network nanostructure tovoltaics and solar fuels, Nano Lett. 10(8), 2735–2741 for enhanced hydrogen storage, Nano Lett. 8(10), (2010) 3166–3170 (2008) 1.157 K. Kordas, T. Mustonen, G. Toth, H. Jantunen, M. La- 1.144 V. Varshney, S.S. Patnaik, A.K. Roy, G. Froudakis, junen, C. Soldano, S. Talapatra, S. Kar, R. Vajtai, B.L. Farmer: Modeling of thermal transport in pil- P.M. Ajayan: Inkjet printing of electrically conduc- lared-graphene architectures, ACS Nano 4(2), 1153– tive patterns of carbon nanotubes, Small 2(8–9), 1161 (2010) 1021–1025 (2006) 37 NanoCarbPart A Part A NanoCarbons

2 Graphene – Properties and Characterization Aravind Vijayaraghavan, Manchester, UK 3 Fullerenes and Beyond: Complexity, Morphology, and Functionality in Closed Carbon Nanostructures Humberto Terrones, University Park, USA 4 Single-Walled Carbon Nanotubes Sebastien Nanot, Houston, USA Nicholas A. Thompson, Houston, USA Ji-Hee Kim, Houston, USA Xuan Wang, Houston, USA William D. Rice, Los Alamos, USA Erik H. Hároz, Houston, USA Yogeeswaran Ganesan, Hillsboro, USA Cary L. Pint, Nashville, USA Junichiro Kono, Houston, USA 5 Multi-Walled Carbon Nanotubes Ákos Kukovecz, Szeged, Hungary Gábor Kozma, Szeged, Hungary Zoltán Kónya, Szeged, Hungary 6 Modified Carbon Nanotubes Aarón Morelos-Gómez, Nagano, Japan Ferdinando Tristán López, Nagano, Japan Rodolfo Cruz-Silva, Nagano, Japan Sofia M. Vega Díaz, Nagano, Japan Mauricio Terrones, University Park, USA 7 Carbon Nanofibers Yoong A. Kim, Nagano, Japan Takuya Hayashi, Nagano, Japan Morinobu Endo, Nagano, Japan Mildred S. Dresselhaus, Cambridge, USA 8 Nanodiamonds Olga A. Shenderova, Raleigh, USA Suzanne A. Ciftan Hens, Raleigh, USA 39

2. Graphene – Properties and Characterization A Part Graphene –2 P Aravind Vijayaraghavan

2.1.3 Decomposition of Carbides ...... 44 Graphene is the two-dimensional allotrope of car- 2.1.4 Exfoliation by a Solvent ...... 46 bon, consisting of a hexagonal arrangement of 2.1.5 Synthetic Production Route ...... 49 carbon atoms on a single plane. This chapter ex- 2.1.6 Graphene Nanoribbon (GNR)...... 49 plores the history of graphene, as the theoretical 2.1.7 Derivatives of Graphene...... 50 building block for other carbon allotropes as well as 2.2 Properties ...... 50 its rise as a material in its own right in recent years. 2.2.1 Structure and Physical Properties ... 50 Graphene can be fabricated by different methods 2.2.2 Mechanical Properties...... 51 including mechanical exfoliation, chemical vapor 2.2.3 Electronic Properties...... 52 deposition, and decomposition of SiC, although 2.2.4 Optical Properties...... 55 bulk-quantity production of pristine graphene 2.2.5 Thermal and Thermoelectric remains a challenge. The atomic and electronic Properties ...... 56 structure of graphene is described, highlighting 2.2.6 Chemical Properties...... 56 the strong correlation in graphene between struc- 2.2.7 Properties of Graphene Derivatives. 57 ture and properties, as is the case with other carbon allotropes. Graphene exhibits a number 2.3 Characterization ...... 58 2.3.1 Optical Characterization ...... 58 of unique and superlative electronic and optical 2.3.2 Transmission Electron Microscopy... 58 properties. The intrinsic properties of graphene 2.3.3 Scanning Probe Techniques ...... 60 can be tailored by nanofabrication, chemistry, 2.3.4 Angle-Resolved Photoemission electromagnetic fields, etc. Various applications Spectroscopy (ARPES)...... 60 of graphene have been proposed in electronic, 2.3.5 Raman Spectroscopy...... 61 optoelectronic, and mechanical products. In ad- 2.3.6 Electrical Characterization ...... 63 dition, graphene has emerged as a candidate in 2.3.7 Photocurrent Microscopy ...... 69 chemical, biochemical, and biological applica- tions. Derivatives of graphene such as graphene 2.4 Applications ...... 69 oxide or graphane are also of interest in terms of 2.4.1 Structural and Electrical Composites 69 both fundamental properties and applications. 2.4.2 Transparent Conducting Films ...... 69 2.4.3 Sensors ...... 72 2.4.4 Electronic Applications...... 73 2.4.5 Photonics and Optoelectronics ...... 74 2.1 Methods of Production...... 42 2.1.1 Micromechanical Cleavage 2.5 Conclusions and Outlook ...... 74 of Graphite (Scotch Tape Technique) 42 2.1.2 Chemical Vapor Deposition (CVD) .... 43 References ...... 74

As a three-dimensional (3-D) material, carbon exists as Carbon atoms in diamond are all sp3 hybridized and three predominant allotropes: diamond, graphite, and arranged in diamond cubic structure which comprises amorphous carbon (historically knows as carbon black). two interpenetrating face-centered cubic (fcc) lattices. These are distinguished by their crystalline structure Graphite has a layered structure, where the sp2 hy- and the hybridization of the carbon atoms therein. bridized carbon atoms are arranged in a hexagonal

R. Vajtai (Ed.), Springer Handbook of Nanomaterials, DOI 10.1007/978-3-642-20595-8_2, © Springer-Verlag Berlin Heidelberg 2013 40 Part A NanoCarbons atA Part a) b) c) d) 2

Fig. 2.1a–d Allotropes of sp2 carbon: (a) graphite (3-D), (b) graphene (2-D), (c) carbon nanotube (1-D), and (d) fullerene (0-D) (courtesy K.S. Novoselov)

lattice in each plane, while the planes themselves are ever, had been previously synthesized and described, AB (Bernal) stacked and held together by van der including epitaxial graphene which has been known Waals forces. Amorphous carbon, as the name indicates, since the 1970s. Chemical derivatives of graphite such does not have long-range crystalline order, although as graphite oxide can be traced back to the 1950s and locally the atoms are bound together covalently and can exist as single layers (graphene oxide), although comprise a mix of sp2 and sp3 carbons. While di- these are not truly two-dimensional layers due to out- amond can be reduced in size to the nanoscale to of-plane atoms which stabilize their structure. form nanodiamond, it is graphite that can be truly re- Since graphene occurs naturally as a constituent of duced to lower-dimensional allotropes. A single layer of bulk graphite, it appeared to be the logical place to start graphite is defined as graphene, the topic of this chapter. the hunt for freestanding graphene. This effort culmi- Graphene has been used as the building block to con- nated in the successful exfoliation of a single sheet of ceptually visualize carbon allotropes such as graphite, carbon atoms by Andre Geim and Kostya Novoselov carbon nanotubes, and fullerenes; it was believed that at the University of Manchester in 2004 [2.4], using such a freestanding, two-dimensional (2-D) structure a technique referred to as micromechanical cleavage, would not be stable. Carbon nanotubes (CNT) form its or colloquially, the Scotch tape method. This discov- one-dimensional (1-D) counterpart, while fullerenes are ery and the subsequent investigations into the properties the zero-dimensional (0-D) allotropes. These various of graphene were rewarded with the Nobel Prize in forms of carbon are summarized in Fig. 2.1.Itisim- Physics in 2010 for Geim and Novoselov. The existence portant to note that CNTs or fullerenes are not unique of graphene, however, does not contradict the physics structures, but rather describe a family of structures, that predicted that it could not exist. It was discovered which are described in detail in subsequent chapters. that graphene is not truly flat; there exist atomic-scale Nonetheless, their structure and properties are all de- ripples in the carbon sheet, which accommodate the rived from graphene. excess surface energy, thereby stabilizing the 2-D struc- Despite serving as the fundamental building block ture of graphene. For this reason, it might be argued that for these carbon allotropes, graphene remained a con- graphene is a quasi-2-D material. However, it has been cept until 2004. It was substantially predated by its shown to exhibit a range of properties that are unique to related allotropes, fullerenes being discovered and de- 2-D physics, and therefore graphene will be identified scribed in 1985 [2.1], while carbon nanotubes were as a 2-D material for the rest of this chapter, without synthesized and their atomic structure elucidated in qualification. 1991 [2.2]. The term “graphene” was coined in A surge of research into the structure and properties 1987 [2.3], to describe one of the two alternating lay- of graphene ensued, and graphene did not disappoint. ers in graphite intercalation compounds (GIC), the other Almost immediately, an anomalous quantum Hall effect layer being the intercalating agent. It was postulated that was reported in graphene, which also serves as direct independent, freestanding graphene would not become experimental evidence for the electrons in graphene a physical reality since it would voluntarily transform behaving as massless Dirac fermions, confirming theo- into a more stable allotrope in an attempt to minimize its retical predictions. Graphene yielded record high values surface energy. Supported monolayers of carbon, how- for various properties, such as tensile strength, car- Graphene – Properties and Characterization 41 atA Part

Carbon nanotubes a) Graphene

Fullerenes 2

8000

Sumio Iijima reports Andre Geim and 6000 Robert Curl, Harold measurements of Kostya Novoselov Kroto, and Richard multiwalled carbon Curl, Kroto, and report measurements Geim and Novoselov win Nobel Prize in 4000 Smalley report discovery nanotubes, sparking Smalley win Nobel on graphene flakes widespread interest Physics of C60 (fullerenes) Prize in Chemistry

Publications per year 2000

0 1985 1990 1995 2000 2005 2010

b) Carbon nanotubes

Graphene

Fullerenes 2000 Today's major 1600 graphene Graphene's future Carbon nanotubes Four of today's manufacturers • Touchscreens (CNTs) used in • Capacitors 1200 large-scale CNT founded CNT composites lithium-ion batteries • Fuel cells manufacturers used in car parts and sporting • Batteries founded and tested in aircraft 800 equipment • Sensors • High-frequency circuits 400 • Flexible electronics

Patent applications per year Patent 0 1990 1995 2000 2005 2010 2015

Fig. 2.2a,b Time line of nanocarbon allotropes and the (a) number of publications and (b) number of patent applications per year on each topic (after [2.5]) rier mobility, thermoelectric power, etc. Scientists have Any chapter on graphene would not be complete also succeeded in transferring epitaxial graphene from without discussion of its other, quasi-2-D derivatives its native metal or silicon carbide substrate onto other based on the fundamental graphene structure. Perhaps substrates of interest or as freestanding structures. the most interesting of these is the case where two Since the early 1990s, carbon nanotubes have been graphene layers are AB stacked to form a bilayer. described as rolled-up graphene sheets. This descrip- Unlike graphene, which has zero electronic bandgap tion has come full circle, with the recent unzipping of and is therefore a quasimetal, this bilayer structure carbon nanotubes to yield graphene. While the num- can have a bandgap. A bandgap is critical for elec- ber of publications on carbon nanotubes has leveled tronic applications, and one of the most active areas off in recent years at about 8000 papers per year, the of research in graphene is currently the generation of publications on graphene are only just starting their ex- a stable, true, electronic bandgap in graphene. The ponential growth at a rate faster than that which was oxidized derivative, graphene oxide (GO), has been al- enjoyed by carbon nanotubes in their first few years luded to before, and serves as a useful intermediate (Fig. 2.2; over 3000 papers were published in the field in a chemical route for graphite exfoliation. Hydro- of graphene in 2010 [2.5]). A similar trend is also ob- genated graphene, christened graphane, and fluorinated served in patent applications based on carbon nanotubes graphene or fluorographene have been recently pro- and graphene. duced and characterized. 42 Part A NanoCarbons atA Part 2.1 Methods of Production Historically, graphene supported on substrates such as nity, until the interest of condensed matter physicists

2.1 metals and SiC was synthesized first; however, such turned to graphene and other lower-dimensional car- graphene was not liberated from the substrate support bons. The first successful thinning down of graphite to form a truly two-dimensional structure. Freestand- to its monolayer graphene form involved a wet/dry ing graphene was not a reality until the ground-breaking method [2.4]. The surface of highly oriented pyrolytic publication of Novoselov and Geim in 2004. Con- graphite (HOPG) was first patterned into square mesas, sequently, this micromechanical cleavage method is which were pressed into wet photoresist. After baking, presented first, followed by the epitaxial synthesis of the mesas attached to the now dry photoresist, and could graphene as well as chemical vapor deposition on be detached from the bulk of the HOPG. Scotch tape metallic substrates. Graphene can also be exfoliated was used to repeatedly peel off layers of graphite from from graphite by sonochemical means in a solvent, the mesas, thinning them down, until only very thin followed by a purification step to extract the mono- layers remained in the photoresist. The photoresist was layers. Recently, islands of nanographene have also then dissolved in acetone to release these thin flakes, been synthesized by chemical routes, and this bottom- which float on the solvent surface. The flakes were up approach is discussed. Finally, separate sections are collected onto Si/SiO2 wafer pieces dipped into the sol- devoted to production of graphene nanoribbons (GNR) vent. Thicker flakes adhering to the silicon could be and derivatives of graphene. cleaned off by sonication in 2-propanol, while thinner flakes were reported to adhere strongly to the substrate 2.1.1 Micromechanical Cleavage of Graphite due to capillary forces. (Scotch Tape Technique) In the resultant sample, the flakes of graphene, bi- layer graphene, and few-layer graphene (FLG)mustbe Micromechanical cleavage (or exfoliation), as the name distinguished from among a sea of flakes of various implies, refers to thinning down of graphite by me- thicknesses. Fortunately, at certain thicknesses of the chanically reducing the number of layers in a repeated SiO2 layer that covers the silicon wafer, for example, fashion. Graphite is known to cleave preferentially 300 nm, the interference contrast generated by graphene along the interlayer direction where layers are held to- flakes on the surface makes them relatively easy to gether by weak van der Waals forces, rather than across spot and identify by optical microscopy. A trained the strong covalent bonds that bind the atoms within eye can, in fact, distinguish between graphene, bilayer a layer. The most common procedure to accomplish graphene, and thicker flakes. This is described in de- this is using adhesive tape. Attaching a thick graphite tail in Sect. 2.2.4 on optical properties of graphene. flake to adhesive tape on both its exposed faces, and Figure 2.3 shows the appearance of micromechani- then peeling the two pieces of tape apart, results in cally cleaved graphene flakes of different thicknesses in two thinner flakes of graphite, stuck to each of the ad- an optical microscope [2.7], atomic force microscope hesive tape pieces. Repeating this process a sufficient (AFM)[2.8], and transmission electron microscope number of times would in principle result in a single (TEM)[2.9]. These, along with Raman spectral map- sheet of carbon atoms, i. e., graphene, adhered to the ping, are routinely used to characterize graphene flakes, adhesive tape. This is still not freestanding graphene, and are described in Sect. 2.3 dedicated to graphene since it is supported by the tape, and must be transferred characterization. to a suitable weakly coupling substrate or suspended The procedure has since evolved into a completely across supports. The genericized Scotch trademark for dry technique. It has been shown that rubbing the transparent adhesive tape has subsequently lent its name freshly cleaved surface of a layered material such as to the micromechanical exfoliation of graphite. Differ- graphite on another solid surface results in a vari- ent variations of this method exist. ety of flakes, among which monolayer flakes can be One of the earliest such efforts was undertaken by invariably found [2.8]. Alternatively, HOPG is me- Fernandez-Moran, who succeeded in thinning graphite chanically cleaved repeatedly between two pieces of down to ≈ 15 layers (5 nm) over a millimeter size, to adhesive tape until the surface of the tape is cov- serve as a support membrane for transmission elec- ered by a layer of relatively thin graphite [2.10]. No tron microscopy [2.6]. This result remained relatively specific criterion exists at the time of writing for the unknown outside the electron microscopy commu- optimum degree of such exfoliation. Researchers rely Graphene – Properties and Characterization 2.1 Methods of Production 43 atA Part a) b) c) 0A 9A 13A 300 nm SiO2 2.1

5 µm White light

Fig. 2.3a–c Monolayer graphene flake as seen in (a) an optical microscope [2.7], (b) an atomic force microscope [2.8], and (c) a transmission electron microscope (after [2.9]) on personal experience or historic parameters specific higher degrees of standardization and reproducibility, to their laboratory to carry out this procedure. Once albeit with drawbacks of their own. a satisfactory degree of exfoliation has been accom- It should be noted that the best quality of graphene plished with the adhesive tape, it is pressed against currently available is indisputably that produced by mi- the surface of a desired substrate, such as Si/SiO2 cromechanical cleavage, where the graphene quality is wafer. Again, various scientist-specific parameters ex- defined in terms of crystalline domain size, number ist for this process, such as duration, force, application of defects, carrier mobility, etc. The choice of initial and peeling-off procedure, etc. At the end of this pro- graphite material having large grain size, using freshly cedure, the surface of the graphene as well as the cleaved graphite for further exfoliation, and the cleanli- substrate is often contaminated with adhesive residue ness and quality of the adhesive tape and SiO2 substrate from the tape, which has been shown to limit the are all variables which significantly affect the qual- carrier mobility in the graphene flake. Efforts to re- ity of the final flakes obtained. Flakes of graphene move this residue have included annealing at 200 ◦C hundreds of μm across are routinely produced in labora- in a reducing atmosphere of Ar and H2 [2.11], anneal- tories all over the world by this method, predominantly ing in vacuum at 280 ◦C[2.12], and current-induced for fundamental research purposes. In some cases, Joule heating after graphene device fabrication [2.13]. millimeter-size flakes have also been reported. While Variations such as applying an electric field perpen- this method has not matured for commercial applica- dicular to the substrate during the transfer from the tions, ventures such as Graphene Industries have been adhesive tape have also been explored [2.14]. Covalent established to sell micromechanically cleaved graphene linkers such as perfluorophenylazide between graphene flakes. and SiO2 have been explored to aid in the exfolia- tion process to increase monolayer yield [2.15]. The 2.1.2 Chemical Vapor Deposition (CVD) residue issue can also be completely avoided by evap- orating a thin film of gold on the HOPG surface prior The study of the deposition of thin graphitic layers to the transfer step, to avoid direct contact between on metal substrates by CVD dates back to the late the adhesive and the graphene [2.16]. The gold can 1960s [2.17, 18]. One of the motivations for this was in be subsequently dissolved in a suitable etchant without fact to eliminate the formation of graphitic structures on affecting the graphene. The lack of standardization is in- metals such as platinum which results in degradation of dicative of the early stage of current graphene research, catalytic activity. CVD is currently the preferred route and is perhaps an indication that micromechanical ex- for large-scale fabrication of carbon nanotubes, and foliation is not destined to evolve into a large-scale or therefore has generated substantial excitement as a po- industrial method for graphene production. Competing tential method for large-scale production of graphene. techniques, discussed subsequently, have reached much In general, this involves thermal decomposition of 44 Part A NanoCarbons

atA Part gaseous hydrocarbon sources followed by dissolution arbitrary substrates, and the Cu dewets and evaporates and recrystallization of the cracked carbon on the sur- during the growth process itself, leaving behind the face of metallic substrates. The solubility of carbon in graphene film intact on the substrate [2.27]. The first

2.1 various metals, such as rhodium, ruthenium, iridium, truly large-area production of graphene has recently and rhenium, has been measured [2.19], along with the been reported using a continuous CVD deposition and observation that excess carbon dissolved in such metals transfer process [2.28]. at high temperatures can segregate as graphite on the In general, for substrates with small lattice mis- surface upon cooling. Various metallic substrates and match (< 1%) such as Co(0001) and Ni(111), com- carbon feedstock have been explored in the effort to mensurate superstructures are formed, while substrates grow monolayer graphene, and some of the significant with larger mismatches such as Pt(111), Ir(111), and developments are mentioned next. Ru(0001) yield incommensurate moiré superstructures. As early as 1991, monolayer graphene was The first graphene monolayer on the metal surface has grown on Pt(111) by hydrocarbon decomposition at strong interaction with the substrate, and the spacing be- 800 ◦C[2.20], resulting in islands of 20–30 nm size dis- tween the two is much shorter than between two layers tributed uniformly over the surface. Upon annealing at of graphite (3.35 Å). For the case of Ru it is 1.45 Å, and higher temperatures, the graphene was found to accu- for Ni it is 2.11 Å. mulate into large, regularly shaped islands on terraces and step edges. Recently, a variation in which a beam of 2.1.3 Decomposition of Carbides methane molecules with high kinetic energy (670 meV) impacting a Pt(111) surface at 890 K resulted in large The second substrate-supported route for graphene domains of monolayer graphene covering the entire production involves thermal decomposition of surface Pt surface [2.21]. Ni soon followed, requiring a min- layers of carbides such as SiC. About 250 different imum temperature of 600 ◦C, and monolayer graphite crystal structure of SiC are known, but α-SiC is the on Ni(111) was shown to have an arrangement whereby most commonly encountered polymorph and, until re- one carbon atom in a unit cell of the graphite over- cently, the primary focus of epitaxial graphene growth. layer is located at the on-top site of the topmost Ni α-SiC has a hexagonal crystal structure, similar to atoms, while another carbon atom exists at the fcc hol- wurtzite. The (0001) (Si) and (0001)¯ (C) faces of 6H-(α- low site [2.22]. Carbon has been shown to segregate on SiC) [2.29–31] and 4C-(α-SiC) [2.32] have been shown the surface of Ru(0001) as monolayer graphene [2.23], suitable for the growth of epitaxial graphene. when annealed at 1400 K. STM reveals a (11 × 11) 6H-(α-SiC) is first cleaned for 20 min at 850 ◦C structure with good rotational alignment and structural under a Si flux to prevent Si sublimation during the perfection, a well-defined periodicity of ≈ 30 Å, and cleaning step. At higher annealing temperatures in ul- large domain sizes exceeding 100 μm. Graphene has trahigh vacuum (UHV), the surface of SiC undergoes also been grown by thermal decomposition of benzene various reconstructions, until the graphitization tem- on Ir(111) [2.24]. perature when the surface graphene layers form. The At present, the two predominant metallic substrates Si surface of the hexagonal SiC undergoes the fol- ◦ for CVD graphene growth are Ni and Cu. Large- √lowing√ reconstructions: 3 × 3 at√ 850 C√ under Si flux, size (cm2) films of monolayer and few-layer graphene 3× 3R30◦ below 1000 ◦C, 6 3×6 3R30◦ (6R3) have been grown on Ni [2.25], with monolayer re- at 1150 ◦C, and graphitization at 1350 ◦C. On this face, gions as large as 20 μm in size. Cu does even better, the C atoms are in epitaxy with the SiC underneath af- and predominantly monolayer graphene covering many ter graphitization. The surface is passivated by the first cm2 is grown in various laboratories using methane C layer, the interface extends to two C layers, and sub- CVD [2.26]. The solubility of C in Cu appears to make sequent C layers are decoupled from the substrate and the process self-limiting, and at most 5% of the sur- exhibit the properties of graphene. Graphitization oc- face is covered by small islands of bilayer graphene. curs at 1150 ◦C on the C face of the hexagonal SiC, Most importantly, methods have been developed to de- and the C layer occurs on a SiC 2 × 2 native reconstruc- tach these films from the metallic substrate and transfer tion. This reconstruction saturates the dangling bond them intact onto dielectric substrate (Fig. 2.4), where states, so that the first C layer already exhibits graphene they can be lithographically patterned and processed properties. However, this C layer is no longer epitax- for electronic or optical applications [2.25]. In another ial with the underlying SiC, so the long-range order variation, graphene is grown on a thin copper film on of the SiC substrate no longer imposes itself upon the Graphene – Properties and Characterization 2.1 Methods of Production 45 atA Part a) b) c) d) 3L 2L 2.1 1L Glass

1cm

Graphene Graphene

SiO2

Fig. 2.4a–d Graphene films transferred onto (a) aSiO2/Si substrate and (b) a glass plate. (c) Scanning electron micro- graph (SEM) image of graphene transferred onto SiO2/Si (285 nm-thick oxide layer), showing wrinkles as well as two- and three-layer regions. (d) Optical microscope image of the same regions as in (c) (after [2.26])

a) b) c)

G-peak (1592 ± 5) cm–1 2-D-peak (2706 ± 5) cm–1 D-peak (1356 ± 5) cm–1 54 cm–1 Intensity (arb. units) Intensity (arb.

37 cm–1

(2717 ± 5) cm–1 (1596 ± 5) cm–1 8.0 μm 1.400 1.600 1.800 2.600 2.800 Raman shift (cm–1) Fig. 2.5 (a) AFM image of graphene on 6H-SiC(0001) formed by annealing in Ar (p = 900 mbar, T = 1650 ◦C). (b) Low energy electron diffraction (LEED) pattern at 74 eV showing the diffraction spots√ due√ to the SiC(0001) substrate (blue arrows) and the graphene lattice (red arrows). The extra spots are due to the (6 3×6 3) interface layer. (c) Comparison of Raman spectra of Ar-grown (red)andUHV-grown (blue) epitaxial graphene (after [2.33])

C layer. Therefore, it has not been possible to accom- ture, β-SiC would be unsuitable for graphene growth. plish both long-range ordering as well as decoupling However, recently scientists have succeeded in grow- from the surface simultaneously using 6H-(α-SiC). Epi- ing graphene on the Si-rich (100) surface of β-SiC taxial graphene has also been grown on SiC(0001) in Ar by a series of annealing cycles with temperature in- atmosphere [2.33], at close to atmospheric pressure and creasing from 1200 to 1550 K [2.34]. They found a significantly higher annealing temperature of 1650 ◦C, that the strong lattice mismatch between graphene and resulting in morphologically and electronically superior underlying SiC results in very weak coupling simi- graphene compared with vacuum annealing (Fig. 2.5). lar to the (0001)¯ C face of α-SiC. However, it was Perhaps the greatest limitation of SiC as a sub- found that graphene growth on β-SiC was guided along strate for graphene growth is cost. α-SiC wafers are the [110] crystallographic direction despite the lattice relatively expensive, at about USD 300 for a 50 mm mismatch, raising hopes that both substrate–graphene wafer. Cubic 3C-SiC (β-SIC), however, can be grown decoupling as well as substrate-guided large domain directly on the surface of Si wafers of 300 mm and size might be simultaneously achievable after further larger, and is therefore a more commercially viable process optimization and characterization of graphene substrate. It was believed that, due to its cubic struc- on β-SiC. 46 Part A NanoCarbons

atA Part Ribbons of graphene, a few nanometers wide, de- and disperse individual CNTs from a bundle, it has velop an electronic bandgap due to confinement effects, also been used to individualize graphene layers from which is absent in larger dimensions of graphene, which graphite. This process can be facilitated if the inter-

2.1 is a zero-bandgap semimetal. The importance and meth- layer attraction can be compromised by intercalates. ods of inducing a bandgap in graphene are discussed The resultant flakes of graphene in a solvent can be later. Here, we briefly discuss how SiC decomposition stabilized to prevent aggregation, and separated into can be used to grow graphene nanoribbons [2.35]. It is fractions which are enriched in particular graphene known that the (0001) face of both 6H and 4C α-SiC thicknesses. with vicinal miscuts towards 1100¯  displays bunch- ing of parallel steps into (110¯ n) nanofacets up to 4–5 Graphite Intercalation Compounds unit cells in height and oriented at an angle of ≈ 25◦ Due to the nature of hybridization of carbon atoms in to the basal plane. The α-SiC (0001)¯ face generally graphite, it is capable of reactivity involving incorpo- does not show preferential orientation for nanofacets, ration of atoms, ions, or molecules in its lattice while but step-bunched (110¯ n) nanofacets can be induced by leaving its basic structure unchanged. Such graphite in- suitable pretreatment. It has also been observed that tercalation compounds (GIC) [2.40] may be broadly graphene grown on the (0001) and (0001)¯ faces of α- classified into those with homopolar bonding and SiC are continuous over these steps. Controlled facets polar bonding. Graphite oxide (GO) and graphite flu- can be achieved by conventional photolithography and oride (GF) are examples of homopolar bonding, while microfabrication. Few-layer graphene is shown to grow potassium-, rhodium- and cesium-graphite are exam- selectively on these facets. Facets of other crystallo- ples of polar bonding. GICs were well studied as graphic orientations are possible, and it is expected that early as the 1950s and are a staple of chemistry text- the graphene quality, properties, and growth mechanism books. Here, we restrict our discussion to exfoliation will depend significantly on the crystallographic sur- of GICs, in particular graphite oxide, and its reduction face. However, these preliminary results indicate that, to graphene, which has been achieved with varying de- with further research and optimization, the ideal facets grees of success. A family of GICs with interhalogen and growth conditions might be determined for large- compounds offers control over the stage of intercalation scale controlled growth of graphene nanoribbons. and subsequently the layer distribution in the resultant Interestingly, there was significant research into the graphene. growth and characterization of graphene on other metal Graphite can be oxidized to GO in various ways. carbides [2.36–38], such as TiC, TaC, and HfC, as early In the modified Staudenmaier method [2.41] a mixture as the 1980s. Then, it was referred to as monolayer of 97% sulfuric acid and fuming nitric acid is cooled ◦ graphite. Graphene has been grown on the (100) and down to 5 C in an ice bath, graphite in flake or powder (111) faces of these carbides by heating them to 1700 K form is added, followed by repeated additions of potas- in UHV. As with SiC, graphene nanoribbons as narrow sium perchlorate every hour over a period of 3 days. The as 1.3 nm with well-defined edge structure have been resulting solution, sometimes referred to as graphitic grown, for instance, on TiC (755) surface [2.39]. In all acid, is filtered and washed until the pH of the fil- these cases, a significant degree of hybridization be- trate reaches 5 or more. The Brodie method [2.41]is tween the graphene π-electrons and the electronic bands identical, except that only nitric acid is used, and the of the substrate carbide was reported, similar to some of potassium perchlorate is added every hour for 3 h. In the crystallographic faces of SiC. Despite the revelation the Hummers method [2.42, 43], graphite is oxidized that the graphene is significantly decoupled from certain in a mixture of concentrated sulfuric acid, sodium ni- ◦ other SiC faces, similar exploration into other metallic trate, and potassium permanganate at 45 Cfor2h. carbides remains pending. At this stage, the material is often referred to as ex- pandable graphite, reasons for which are explained in 2.1.4 Exfoliation by a Solvent the next section. In the electrochemical method [2.41], a graphite sheet electrode is anodically polarized in Exfoliation of graphene from graphite involves over- perchloric acid with a platinum wire as counterelec- coming the interlayer van der Waals bonds. This is trode. When dried, the above methods result in a powder the same interaction in play between individual CNTs consisting of graphite oxide flakes. Graphite fluoride in a bundle. Just as ultrasonication in a solvent has has also been used as a starting point for graphene been used to overcome this weak force and separate dispersions. Graphite can be fluorinated under fluo- Graphene – Properties and Characterization 2.1 Methods of Production 47 atA Part a) b) c) GO With NaOH With KOH 2.1

Fig. 2.6 (a) HOPG before (top)andafter(bottom) oxidation and expansion (after [2.45]). (b) Deoxygenation of exfoliated GO under alkaline conditions (after [2.46]). (c) AFM image of exfoliated monolayer graphene oxide sheets (after [2.47]) rine pressure of 200 mmHg, in a temperature range of resulting from the oxidation of graphite ensured col- 375–640 ◦C[2.44]. loidal stability in polar solvents [2.50]. Polymers, Interhalogen compounds such as IBr and ICl also surfactants, DNA, etc. can be used to provide additional form GICs, offering control over the stage of intercala- stabilization of the GO flakes in colloidal suspension. tion. Stage I GIC refers to intercalation of every layer Edge-selective diazonium functionalization [2.51]has of graphite, while stage II GICs only have every sec- also been demonstrated as a way to stabilize high- ond layer of graphite intercalated, and stage III GICs concentration graphene solutions without the stabilizing have every third layer intercalated, etc. As discussed agents perturbing the bulk structure of the graphene in subsequent sections, bilayer and trilayer graphene sheets. are electronically distinct from monolayer graphene and Exfoliated GO can be subsequently reduced to yield in certain instances, such as semiconductor electronics, reduced GO. It cannot be referred to as graphene at might prove superior to monolayer graphene. Exfolia- this stage due to the incomplete nature of the reduction tion of graphene from stage II and stage III GICs has process. One route involves reduction in water with hy- been shown to yield solutions of predominantly bilayer drazine hydrate [2.52, 53] or dimethylhydrazine [2.54]. and trilayer graphene [2.48], and is currently the only A reduced GO suspension can also be obtained by large-scale route available to synthesize these multi- heating the exfoliated GO suspension under strongly layer graphenes. alkaline conditions by addition of NaOH at 50–90 ◦C (Fig. 2.6b) [2.46]. Alternately, the flakes can be de- From GIC to Graphene posited in a substrate and reduced by hydrazine vapors Usually, the next step involves expansion of intercalated or hydrogen plasma [2.55]. All these methods result graphite by decomposing and expelling the intercalate. in the formation of unsaturated and conjugated carbon Rapid annealing of expandable graphite to 1050 ◦Cgen- atoms, which results in electrical conductivity and Ra- erates high-pressure gaseous decomposition products man signatures intermediate between those of GO and which force the individual layers apart. This results pristine graphene. in a ≈ 100-fold expansion in the interlayer spacing in GF can be reacted with n-butyl and n-hexyl graphite, and the material is now referred to as expanded lithium reagents in hexane at 0 ◦C. The alkyl lithium graphite (Fig. 2.6a) [2.45, 47]. Similarly, interhalogen reagent replaces the fluorine functionalization dur- GICs can be expanded by expelling the entrapped in- ing this process. The product can then be dispersed tercalants. in ethanol by sonication. The alkyl functionalization, Expanded graphite or graphite oxide is dispersed followed by a subsequent annealing step, partially re- in a solvent by ultrasonication, resulting in graphene stores the pristine graphene structure similar to reduced or GO solutions, respectively. In the case of GO,the GO [2.56]. A one-step electrochemical approach has predominant product in solution is monolayer GO, been demonstrated to form ionic liquid functional- while stage II and III GICs of interhalogen com- ized graphite sheets, which are then exfoliated into pounds yield solutions of predominantly bilayer and functionalized graphene dispersed in polar aprotic sol- trilayer graphenes. Alternatively, GO can be interca- vents [2.57]. lated and exfoliated, for instance, by tributylammonium It is also possible to exfoliate noncovalent GICs to cations [2.49]. The phenol, carbonyl, and groups yield graphene flakes that do not suffer the disadvantage 48 Part A NanoCarbons atA Part a) b) c) Relative frequency 1.0 f4 f4

2.1 f16 f10 0.8 f28 f16 0.6 DGU f22 0.4 f28 0.2

0.0 1.0 2.0 3.0 Mean flake thickness (nm) Fig. 2.7 (a) Schematic illustration of ordered sodium cholate encapsulation of graphene sheets and a photograph of an unsorted aqueous graphene suspension with graphene loading of ≈ 0.1mg/ml. (b) Photograph of a centrifuge tube following DGU marked with the main bands of monodisperse graphene. (c) Mean flake thickness histogram measured by AFM of sorted graphene taken from the locations marked in panel (b) (after [2.59])

of high defect density; for instance, alkali-metal GICs of monolayer and few-layer graphene flakes [2.61]. have been shown to readily and spontaneously exfoliate Other solvents such as DMA, γ-butyrolactone, and 1,3- in N-methyl-pyrrolidone (NMP), yielding a stable so- dimethyl-2-imidazolidinone yield similar results. The lution of negatively charged graphene sheets. Graphene procedure has been adopted successfully for using can also be noncovalently functionalized and exfoliated water as solvent, in the presence of sodium dode- with 1-pyrenecarboxylic acid by continuous sonica- cylbenzene sulfonate or sodium cholate as stabilizing tion in water [2.58]. Interhalogen compounds do not surfactant [2.62]. The predominant drawback of this covalently functionalize the graphite upon intercala- process lies in the fact that the sonication breaks up tion, and therefore the resultant solution is one of pure the graphene into particularly small fragments, with graphene, without any need for further reduction or the monolayer flakes having lateral dimensions of, reconversion. on average, 100 nm. This is similar to the case of CNTs, where sonication appears to cut them down Exfoliation Without Intercalation to ≈ 200 nm [2.63]. The aqueous graphene dispersion In an effort to avoid disruption to the desirable structure can now be processed by density-gradient ultracentrifu- and properties of graphene, efforts have been under- gation (DGU) using iodixanol as density medium to taken to directly exfoliate and disperse graphite in yield fractions enriched in particular graphene thick- a solvent by ultrasonication, as has been successfully nesses (Fig. 2.7)[2.59]. Highly enriched solutions of demonstrated for debundling CNTs. Systematic study monolayer and bilayer graphene with the above size has been undertaken in the case of CNTs to explore their limitation are now available for research purposes from solubility in various solvents without the assistance of commercial sources such as Nanointegris, but not yet in stabilizing agents such as surfactants. It has emerged industrial quantity. that certain solvents such as N-methyl-pyrrolidone Single- and few-layer graphene sheets with sizes and N,N-dimethylamide (DMA) are ideally suited to up to 0.1 mm have been fabricated by quenching hot dissolve CNTs in significant concentration [2.60]. Dis- graphite in ammonium hydrogen carbonate aqueous solution of CNTs in aqueous media is only possible solution [2.64]. Few-layer graphene has also been using stabilizing surfactants; however, these solutions produced by immersing and intercalating graphite have emerged as the premier option among researchers, in supercritical CO2 for 30 min followed by rapidly since the adsorbed surfactants can be easily desorbed or depressurizing the supercritical fluid to expand and ex- disintegrated if and when required. foliate the graphite [2.65]. The expanding CO2 gas Sieved graphite powder was dispersed in NMP containing the graphene flakes was collected directly by bath sonication. The macroscopic particles and in an aqueous solution containing stabilizing surfactant aggregates were sedimented by mild centrifugation to avoid aggregation. Other supercritical fluids, such as (500–2000 rpm), resulting in a homogeneous dark dis- ethanol, NMP,andDMF, can also be used to exfoliate persion which was found to contain a high fraction graphite into graphene [2.66]. Graphene – Properties and Characterization 2.1 Methods of Production 49 atA Part a) b) c) d) e) W-sub 10 nm

Si Wafer – 2.1 [1120]

500 nm 20 nm 3nm 500 nm

Fig. 2.8a–e Graphene nanoribbons formed by various means. (a) Nanoparticle cutting (after [2.67]); (b) synthesized from polyphenylene precursors (after [2.68]); (c) etching of carbon nanotubes embedded in a polymer (after [2.69]); (d) unzipping of carbon nanotubes (after [2.70]); (e) chemical exfoliation in DCE with PmPV (after [2.71])

2.1.5 Synthetic Production Route preferentially along crystallographic directions. The particles can become deflected into proceeding along If the reduction of bulk graphite into graphene is viewed a different direction if they come within close proxim- as a top-down approach, then the chemical synthesis of ity of each other, of defects in the graphene lattice, or graphene from smaller aromatic hydrocarbons will con- of previously formed cuts. The result is a complex pat- stitute the bottom-up approach. If graphene is regarded tern of cuts which border various well-defined shapes of as a polycyclic aromatic hydrocarbon (PAH), one of the the surface graphene layers, including instances where largest of these synthesized involves 222 carbon atoms two parallel cuts result in a narrow ribbon between or 37 benzene units in a hexagonal structure, 3 nm them. These shapes can be transferred onto arbitrary in diameter [2.72], from an oligophenylene precursor substrates using the mechanical exfoliation methods de- which was planarized by oxidative cyclohydrogenation. scribed earlier [2.67]. The graphene can also be cut These structures have also shown a high tendency to into ribbons after they have been transferred onto any self-assemble on surfaces [2.73] and could potentially arbitrary substrate [2.78]. act as precursors for larger synthetic graphene. Expanded graphite was dispersed in a 1,2-dichloro- ethane (DCE) solution containing a polymer poly(m- 2.1.6 Graphene Nanoribbon ( GNR) phenylenevinylene-co-2,5-dioctoxy-p-phenylenevinyl- ene) (PmPV) by sonication for 30 min followed by cen- The techniques described here have been suitably mod- trifugation to remove larger aggregates. The supernatant ified and developed with particular focus on forming after sonication was shown to contain an appreciable very narrow ribbons of graphene with widths of tens fraction of graphene nanoribbons and related morpholo- of nanometers and with well-defined edge structure gies such as ribbons with kinks, bends, and nonparallel and orientation (Fig. 2.8). This is of particular impor- sides [2.71]. The exact mechanism or variation to the tance in electronic applications, since such nanoribbons liquid-phase exfoliation procedures described earlier of graphene are one of the means to engineer an that results in the significant yield of nanoribbons in this electronic bandgap in otherwise gapless graphene, as case is not clearly understood. discussed in detail in Sect. 2.3.6. Once graphene flakes Graphene nanoribbons, 8–12 nm in length and have been deposited onto a substrate, nanoribbons 2–3 nm width, have also been synthesized by surface- can be fabricated on it using standard and noncon- assisted coupling of molecular precursors into linear ventional lithography and etching processes, such as polyphenylenes and their subsequent cyclohydrogena- electron-beam lithography [2.74] and nanowire lithog- tion [2.68]. raphy [2.75], respectively. CNTs have been described as rolled-up graphene Nickel [2.76] and silver [2.77] nanoparticles have sheets, and now graphene nanoribbons have been made been shown to act as a knife for cutting patterns in sur- from unraveling CNTs. Oxidized nanoribbons were ob- face graphite layers of HOPG. The cutting proceeds tained by suspending CNTs in concentrated sulfuric via catalytic hydrogenation of the graphene lattice, and acid followed by treatment with 500 wt.%KMnO4 for 50 Part A NanoCarbons

atA Part 1hat22◦C and 1 h at 55–70 ◦C. The process, described reversible, and the graphane can be reconverted to as CNT unzipping, could occur as a linear longitudinal graphene by annealing at 450 ◦C in Ar atmosphere cut or in a spiral manner depending on the chirality of for 24 h. The reconverted graphene, however, contains

2.2 the CNT [2.70]. CNTs have also been converted to GNR remnant defects just as vacancies and oxygenated or by controlled plasma etching of CNTs that are partially hydrogenated carbon atoms. In the case of substrate- embedded in a polymer film [2.69]. supported graphene, only one side is hydrogenated, while both sides can be hydrogenated in the case of 2.1.7 Derivatives of Graphene suspended graphene. Graphene can also be fluorinated with xenon di- Graphane, a fully saturated hydrocarbon derived from fluoride. When one side is exposed, F coverage graphene, with formula CH, was predicted to be saturates at 25% (C4F), whereas fluorination of both stable based on first-principles total-energy calcu- sides results in perfluorographene [2.81] and fluoro- lations [2.79]. Experimentally, it was later shown graphene [2.82], which are the nonstoichiometric and that graphene can be hydrogenated and converted to stoichiometric variations. Nonstoichiometric and mul- graphane using a low-pressure (0.1 mbar) hydrogen- tilayered fluorographene can also be exfoliated from argon mixture (10% H2) with direct-current (DC) graphite fluoride [2.56, 83]. Hydrazine treatment has plasma for 2 h [2.80]. The hydrogenation is stable but been shown to reverse the fluorination.

2.2 Properties While its very existence as a freestanding two- been realized by any of the graphene production dimensional material is a feather in graphene’s cap, it is routes. the properties of graphene that make it the truly excep- Two-dimensional structures such as graphene have tional material that has stoked feverish research in this been postulated to be intrinsically unstable, and ac- field. The individual properties of pristine graphene are cording to the Mermin–Wagner theorem [2.84], long- discussed first, while the properties of graphene deriva- wavelength fluctuations destroy the long-range order of tives such as GO, graphane, and fluorographene are 2-D crystals. Even 2-D crystals embedded in 3-D space discussed in the final section. have a tendency to crumple. The puzzling stability of suspended 2-D graphene sheets has been attributed to 2.2.1 Structure and Physical Properties intrinsic microscopic undulations in which the surface normal varies by several degrees and the out-of-plane Graphene shares most of its structure and physical prop- deformation reaches 1 nm [2.9,85]. This observation by erties with graphite, its parent material. The carbon TEM is discussed further in Sect. 2.3.2 and also con- atoms are arranged in a two-dimensional hexagonal forms to atomistic Monte Carlo simulations. Similar lattice (Fig. 2.9b), which can also be constructed as corrugation has also been reported on graphene sup- two interpenetrating triangular sublattices, which takes portedonSiO2 substrates, where it is a superposition particular significance in bilayer and other multilayer of intrinsic rippling as well as extrinsic undulations graphenes. The carbon atoms are sp2 hybridized, and imposed by the substrate surface morphology [2.86]. the in-plane carbon–carbon bond length is a = 1.42 Å. Periodic ripples have also been observed on weakly The remaining p-orbital is oriented perpendicular to coupled graphene monolayers on substrates such as the plane of carbon atoms and delocalizes to form Ru(0001) [2.87]. Corrugations in substrate-supported the π (valence) and π∗ (conduction) electronic bands graphene are primarily observed by STM and are dis- which are discussed in detail in Sect. 2.2.3. The car- cussed further in Sect. 2.3.3. In addition to ripples, bon layers are usually stacked in an ABAB (Bernal) substrate-supported graphene also exhibits ubiquitous stacking; however, in certain few-layer graphenes such wrinkles which could be several nanometers in width. as that grown by CVD, the layers are rotated with Scrolling has been occasionally observed at the edges respect to this standard arrangement. The interplane of graphene flakes, both suspended [2.88] and sub- spacing is 3.45 Å. A staggered ABCABC (rhombo- strate supported [2.89], and this appears to rely on the hedral) arrangement is also possible, but has not fabrication method. Scrolling occurs when graphene is Graphene – Properties and Characterization 2.2 Properties 51 atA Part a) b) 1.42 Å 2.2

A B B

A A B

Fig. 2.9 (a) SEM of graphene suspended over a macroscopic hole of a Cu TEM grid (after [2.90]). (b) Structure of graphene (after [2.92]) subjected to liquid-phase processing during microfab- modulus of 22 eV/Å2 from the elastic modulus of bulk rication, while its solid-phase or gas-phase processing graphite [2.95], the lengths of unsupported graphene appears to avoid this [2.90], and it is possible to ob- observed in TEM samples have been 106 times larger tain large free-standing sheets of monolayer graphene than its effective thickness. Suspended graphene can (Fig. 2.9a). The ripples in graphene also result in pertur- gain additional thickness from large-scale corrugations bations in the electronic structure, and many electronic by a factor of (H/a)2, where H is the characteristic and chemical properties of graphene have been at- height of the corrugations. In addition to supporting its tributed to these ripples, rather than being intrinsic to own weight, suspended graphene has been shown to graphene. However, it has been shown that graphene support significant extra load such as copper nanoparti- deposited on atomically flat terraces of cleaved mica cles [2.90], as well as surviving accidental shocks such surfaces is flat down to the atomic scale [2.91]. The as during handing. height variation observed by AFM waslessthan25pm, Direct measurements of the elastic properties of and such ultraflat graphene is expected to permit ex- graphene have been conducted by nanoindentation ploration of various intrinsic physical and chemical of suspended graphene layers in an AFM [2.96, properties of graphene. 97]. Details of the measurement technique are found in Sect. 2.3.3. Measurements conducted on few-layer 2.2.2 Mechanical Properties graphene of less than 8 nm thickness yielded spring constants of 1–5 N/m. A Young’s modulus of 0.5TPa Carbon materials have made it a habit of setting records was extracted by fitting the data to a model for dou- for their intrinsic mechanical properties. Diamond is the bly clamped beams under tension. For measurements on hardest known natural material, and is assigned a grade monolayer graphene, the force–displacement character- of 10 (highest) on the Mohs scale of hard- istics yield second- and third-order elastic stiffness of ness [2.93]. Similarly, the record for tensile strength 340 and −690 N/m, respectively. The breaking strength has been held by CNTs; a Young’s modulus of 1 TPa was found to be 42 N/m, which represents the intrin- and tensile strength of 150 GPa coupled with elonga- sic strength of a defect-free sheet. This corresponds tion to failure as high as 20% have been experimentally to Young’s modulus E = 1.0 TPa, third-order elastic reported [2.94]. stiffness of 2.0 TPa, and intrinsic strength of 130 GPa. The earliest experimental indication for the extraor- These figures mean that graphene is the strongest ma- dinary stiffness of graphene was the observation that terial ever measured. graphene beams supported on only one end do not scroll Nonlinear finite elasticity theory for graphene res- or fold, quite unlike the papery or cloth-like appearance onators for both electrostatic and electrodynamic cases of graphene. If the effective thickness of monolayer has been developed and agrees well with experiments graphene is estimated to be 0.23 Å from elastic the- on graphene resonators [2.98]. The dynamic response of ory, a bending rigidity of 1.1eV[2.85], and a Young’s clamped graphene resonators resembles that of coupled 52 Part A NanoCarbons atA Part a) b) c) Current (nA) 0.6 Au 2.2 I II Au 0.4

SiO2 0.2 Si Q = 125 0

–0.2 2550 75 100 Frequency (MHz) 5 μm

Fig. 2.10 (a) Schematic of graphene resonator, with electrostatic actuation and electrical readout. (b) SEM image of such a resonator. (c) The graphene resonance (I) at 65 MHz. Resonances of metal beams (II) are also visible below 25 MHz. Inset: the graphene resonance at low driving power, and Lorentzian fit (red line) with Q = 125 (after [2.102])

Duffing-type resonators. Similarly, a continuum plate scribe graphite. The valence and conduction bands of model for the vibration of multilayered graphene sheets, graphene are conical valleys that touch at the high- including the van der Waals (vdW) interaction between symmetry K and K points of the Brillouin zone. Near the layers, suggests that the lowest natural frequencies these points, the energy varies linearly with the mag- are identical for various numbers of layered graphenes. nitude of momentum, i. e., follows a linear dispersion Higher resonance frequencies, however, depend on the relation. In neutral graphene, this point of intersection vdW interaction and are different for different layered coincides with the charge neutrality point, and is re- graphenes [2.99]. In general, natural resonance fre- ferred to as the Dirac point. quencies in the THz regime are expected for graphene In every other material known to condensed matter resonators, due to the combination of their extreme physicists, the electrons behave as and can be described thinness and extraordinary stiffness. Experimentally, by the Schrödinger equation. In graphene, on the other the mechanical vibrations in electrostatically actuated hand, electrons have been shown to behave as rela- graphene resonators have been imaged by a special tivistic particles, and should be described by the Dirac modification of atomic force microscopy [2.100]. Res- equation [2.104–106]. The interaction of electrons with onance frequencies in the tens of MHz have been the periodic potential of the graphene hexagonal lattice recorded in graphene resonators (Fig. 2.10), with qual- results in quasiparticles, which can be viewed as elec- ity factors as high as 4000 at room temperature [2.101] trons devoid of their rest mass m0 and therefore called and 10 000 at 5 K [2.102]. massless Dirac fermions. The linear energy dispersion means that the speed of electrons in graphene is a con- 2.2.3 Electronic Properties stant, independent of momentum, as in the case of the speed of photons. The velocity of electrons in graphene Electronically, monolayer, bilayer, and trilayer graph- is ≈ 106 m/s, about 300 times slower than the speed of ene are electronically distinct materials. Beyond three light (photons). layers, graphene’s electronic properties tend towards The electronic states near the Dirac point are those of bulk graphite. In certain aspects, graphene of composed of states belonging to the two graphene up 10 layers might exhibit deviation in electronic prop- sublattices, and as a result the quasiparticles possess erties from bulk graphite and could be referred to as pseudospin, similar to the electron’s spin [2.107, 108]. graphene, but beyond 10 layers all graphenes are indis- As a result, these Dirac fermions are said to be chi- tinguishable from graphite. ral. Another relativistic feature of these quasiparticles is the Klein paradox [2.107], wherein they tunnel Monolayer Graphene through a potential barrier of any height and width The electronic structure of graphene was first described with a transmission probability of 1 or without a re- in 1946 [2.103], as a theoretical building block to de- flected component. As a result, electrons in graphene Graphene – Properties and Characterization 2.2 Properties 53 atA Part a) b) c) 2.2 π* π* π*

E = ED

π π π

Fig. 2.11a–c Electronic structure of (a) monolayer, (b) symmetric bilayer, and (c) asymmetric bilayer of graphene (af- ter [2.109]) can propagate over (relatively) vast distances of the or- angle-resolved photoemission spectroscopy (ARPES) der of microns through the graphene lattice, even in the (Fig. 2.12). The relative potential of the top and bottom presence of lattice defects or other external perturbing graphene layers is varied by changing the doping level potentials [2.4]. by potassium adsorption. An apparent gap at the K point appears in the as-prepared graphene, disappearing and Bilayer Graphene reappearing with increasing level of K doping. If the hexagonal atomic structure of graphene is com- The electronic gap in bilayer graphene can thus be posed of nonidentical elements, such as in boron nitride, controlled by applying an external transverse electric the lateral in-plane symmetry is broken and a large field, such as by a gate bias, making it the only known bandgap is formed between the π and π∗ states. This semiconductor material with a tunable energy gap. Us- is the case in bilayer graphene, where the AB (Bernal) ing a tight-binding model, the value of the gap was stacking between the two graphene renders the two car- extracted as a function of electron density, showing that bon atoms inequivalent and results in two graphene it can be tuned to values larger than 0.2 eV, using fields sublattices. As a result, the unit cell of bilayer graphene of ≈ 1V/m. contains four atoms, and two additional bands result (π The two key semiconductor parameters, the elec- and π∗ states). If the inversion symmetry between the tronic bandgap and carrier doping concentration, can two layers is broken, then an energy gap between the also be independently tuned by using a dual-gate config- low-energy valence and conduction bands forms at the uration. Reliable determination of the bilayer bandgap Dirac point (Fig. 2.11). has been carried out in such a configuration using The first experimental demonstration of this ef- infrared microscopy [2.110]. Figure 2.13 shows the fect was performed on bilayer graphene synthesized gate-modulated bilayer absorption spectra at the charge on SiC (6H, (0001) orientation) [2.109]. The as-grown neutrality point. The two features present in the spectra, graphene is n-doped due to the depletion of the sub- a peak below 300 meV and a dip around 400 meV, arise strate’s dopant carriers. At low temperature, the SiC from different optical transitions between the bilayer dopant electrons are frozen out and the substrate acts electronic bands. Transition I shows pronounced gate as a nearly perfect insulator while the excess electrons tunability up to 250 meV at 3 V/nm, since it accounts in graphene retain their high mobility. In this case, the for the bandgap-induced spectral response. symmetry of the bilayers is broken by the dipole field By examining the electronic band structure of created between the depletion layer of the SiC and the graphene around the K point within a tight-binding accumulation of charge on the graphene layer next to approach, it has been shown that a single graphene the interface. Further n-type doping can be introduced layer is a zero-gap semiconductor with a linear Dirac- by deposition of potassium atoms onto the vacuum like spectrum around the Fermi energy, while graphite side, which donate their lone valence electrons to the shows semimetallic behavior with band overlap of about graphene layer, forming another dipole. The binding 41 meV. Bilayer graphene has a parabolic band struc- energy–momentum dispersion relation of π, π∗,andσ ture around the Fermi energy and is a semimetal like states along high-symmetry directions was measured by graphite; however, the band overlap is only 0.16 meV. 54 Part A NanoCarbons atA Part a)0.005 e– b)0.0125 e– c) 0.0350 e– 0.2 2.2 0.0 –0.2 –0.4 –0.6 –0.8

Binding energy (eV) Binding energy –1.0 –1.2 0.1 Å–1 –1.4 Momentum Fig. 2.12a–c Evolution of gap closing and reopening by changing the doping level by potassium adsorption. Experimental and theoretical bands (solid lines)for(a) as-prepared graphene bilayers and b,c with progressive adsorption of potassium are shown. The number of doping electrons per unit cell, estimated from the relative size of the Fermi surface, is indicated at the top of each panel (after [2.109])

a) b) c) Absorption difference (%) Absorption difference (%)

12 Experiment 12 Theory

− D = 3.0 V/nm Δ =250 MeV 8 8 IV V

EF Δ − I D = 1.9 V/nm Δ =190 MeV 4 4 − D = 1.4 V/nm Δ =145 MeV II III − D = 1.0 V/nm Δ =105 MeV 0 0

200 400 600 200 400 600 Energy (meV) Energy (meV) Fig. 2.13a–c Infrared spectroscopy to probe bilayer energy gap opening at strong electrical gating. (a) Allowed optical transitions between different subbands of a graphene bilayers. (b) Gate-induced absorption spectra at the charge neutrality point for different applied displacement fields D¯. Curves are offset from zero for clarity. (c) Theoretical prediction of gate- induced absorption spectra based on a tight-binding model where the bandgap value is taken as an adjustable parameter. The fit provides an accurate determination of the gate-tunable bandgap at strong electrical gating (after [2.110])

This overlap increases with the number of graphene electron–hole conversion involves electrons from the layers, and for 11 or more layers it is smaller than 10%. conduction band being converted into a hole from the valence band. This interband conversion is associated Superconductivity with specular reflection instead of the retroreflection Andreev reflection at a metal–superconductor junc- found in normal metals where electron–hole conver- tion involving graphene is fundamentally different from sion occurs within the conduction band (Fig. 2.14). The normal metals [2.111]. In weakly doped graphene, Josephson effect has also been experimentally stud- Graphene – Properties and Characterization 2.2 Properties 55 atA Part a) b)l (nA) c) dV/dI (kΩ) Superconductor Superconductor 1.2

30 2.2 y 2Δ/3 e 1.0 e h 20 h x 10 0.8 Andreev retroreflection Specular Andreev reflection Δ 2Δ 0 –100 10–750 –350 0 350 750 B (mT) V (µV) Fig. 2.14 (a) Andreev retroreflection (left) at the interface between a normal metal and a superconductor, and specular Andreev reflection (right) at the interface between undoped graphene and a superconductor. Arrows indicate the direction of the velocity, and solid or dashed lines distinguish whether the particle is a negatively charged electron (e) or a positively charged hole (h) (after [2.111]). (b) Josephson effect in graphene: dV/dI(I, B)atT = 30 mK (yellow-orange is zero, that is, the supercurrent region, and red corresponds to finite dV/dI)(after[2.112]). (c) dV/dI versus V, showing multiple Andreev reflection dips below the superconducting energy gap. The dips in dV/dI occur at values of V = 2Δ/en,where n is an integer number (after [2.112]) ied in macroscopic junctions consisting of a graphene sition from graphene to bulk graphite. The contrast of layer contacted by two closely spaced superconduct- a graphene flake depends not only on the SiO2 thickness ing electrodes (SGS)[2.112, 113]. A supercurrent is but also on the wavelength λ of light used. Figure 2.15 observed, which can be carried either by electrons summarizes the expected contrast as a function of SiO2 in the conduction band or by holes in the valance thickness as well as wavelength of monochromatic il- band, as determined by the gate voltage. A finite su- lumination, derived using Fresnel theory. It was also percurrent is also observed at zero charge density at inferred that the complex refractive index of graphene is the charge neutrality point, indicating phase-coherent the same as that of bulk graphite, n = 2.6 − 1.3i, which electronic transport at the Dirac point. The diffusive is independent of λ. This can be explained by the fact junction model has been shown to yield quantitative that the optical response of graphite with the electric agreement with experiments [2.114], while a ballis- field parallel to graphene planes is dominated by the tic SGS model is inconsistent with the data. This in-plane electromagnetic response. Since changes in the is attributed to potential fluctuations in graphene due light intensity due to graphene are relatively minor, the to the influence of the substrate as well as metal- observed contrast can be used to determine the number lic leads. Crossed Andreev reflection in graphene– of graphene layers. superconductor–graphene junctions [2.115]andAn- The absorbance of light by monolayer and bilayer dreev reflection in graphene nanoribbons [2.116]have graphene has been measured to be 2.3and4.6%, re- been theoretically investigated, but experimental confir- spectively, in the visual regime (450–750 nm), and this mation remains pending. extends linearity up to five layers. The optical trans- parency of noninteracting graphene is solely determined 2.2.4 Optical Properties by the fine structure constant of quantum electrodynam- ics (α = e2/c = 1/137), which describes the coupling Successful exfoliation of monolayer graphene depends between light and relativistic electrons [2.117, 118]. on the recognition of the optical properties of graphene This is because, as discussed in the previous sec- more than the exfoliation procedure [2.7]. The choice of tion, the electrons in graphene behave as relativistic 300 nm-thick SiO2 on Si substrate allowed optical iden- Dirac particles and electron–electron Coulomb interac- tification of the exfoliated monolayer graphene, which tions can be neglected. The high-frequency (dynamic) would otherwise have been invisible and not practi- conductivity G for Dirac fermions in graphene is a uni- cally detectable; for instance, only flakes thicker than versal constant equal to e2/4. The universal G implies ten layers can be found in white light on top of 200 nm that observable quantities such as graphene’s optical SiO2, which also marks the commonly accepted tran- transmittance T and reflectance R are also universal 56 Part A NanoCarbons atA Part λ (nm) 0.15 700 2.2 410 nm 470 nm 530 nm 590 nm 650 nm λ=710 nm (a)300 nm SiO (b) (c) 200 nm SiO 0.10 2 2 600

0.05 500 5 μm White light λ=560 nm White Green Red 0.00 400 0 100 200 300 λ= SiO2 thickness (nm) 410 nm 470 nm 530 nm 590 nm 650 nm 710 nm

Fig. 2.15 Left: Color plot of contrast as a function of wavelength and SiO2 thickness. The color scale on the right shows the expected contrast. Right: Graphene crystallites on 300 nm SiO2 imaged with white light (panel a), green light, and another graphene sample on 200 nm SiO2 imaged with white light (panel c). Single-layer graphene is clearly visible in the left image (panel a), but even three layers are indiscernible on the right (panel c). Image sizes are 25 × 25 μm2. Top and bottom panels show the same flakes as in (panel a)and(panel c), respectively, but illuminated through various narrow bandpass filters with bandwidth of ≈ 10 nm (after [2.7])

and given by T ≡ (1 + 2πG/c)−2 = (1 + 1/2πα)−2 and Umklapp scattering, since more states are available for R ≡ 1/4π2α2T for normal light incidence. This yields scattering owing to the increased number of phonon graphene’s opacity (1 − T) ≈ πα = 2.3%. branches. The thermoelectric power (TEP) is the voltage 2.2.5 Thermal and Thermoelectric developed across a sample when a constant tem- Properties perature gradient is applied. TEP of 80 μV/Kwas recently measured in graphene at room temperature CNTs are known to have very high thermal conduc- (300 K) [2.125]. Similar to the quantum Hall effect in tivity K with the experimentally determined value of electronic transport, quantized TEP has also been ob- K ≈ 3000 W/(m K) at room temperature for an individ- served in graphene at high magnetic fields [2.125]. The ual multi-walled CNT [2.119]andK ≈ 3500 W/(m K) TEP can be tuned in graphene, even to negative values, for an individual single-walled CNT [2.120]. These val- under the application of a gate bias or chemical poten- ues exceed those of the best bulk crystalline thermal tial [2.126]. Very large TEP values have been predicted conductor, diamond, which has thermal conductivity in for graphene nanoribbons, for instance, 4 mV/Kfor the range K = 1000–2200 W/(m K) [2.121]. a1.6 nm-wide ribbon [2.127]. In comparison, the high- The first experimental determination of the ther- est value experimentally reported so far is 850 μV/Kfor mal conductivity of suspended monolayer graphene two-dimensional electron gases in SrTi2O3 heterostruc- pegged the value at 5300 W/(m K) and a phonon mean tures [2.128], while only a few μV/K has been reported free path of 775 nm near room temperature [2.122], for bulk graphite [2.123]. The TEP power of SWNTs which was extracted from the dependence of the Raman has been theoretically and experimentally shown to be G peak frequency on the excitation laser power and in- 60 μV/K[2.129], inferior to that of graphene. A giant dependently measured G peak temperature coefficient. thermoelectric coefficient of 30 mV/K was reported in Interestingly, this value is higher than the bulk graphite a nanostructure consisting of metallic electrodes peri- limit of K ≈ 2000 W/(m K) [2.123]. It has been experi- odically patterned over graphene, deposited on a silicon mentally shown that the room-temperature thermal con- dioxide substrate [2.130]. ductivity decreases from ≈ 2800 to ≈ 1300 W/(m K) as the number of graphene layers in few-layer graphene 2.2.6 Chemical Properties (FLG) increases from two to four [2.124]. The ob- served evolution from two-dimensional graphene to The chemistry of graphene is dominated entirely by bulk graphite is explained by the cross-plane coupling its surface, since every carbon atom is a surface atom of the low-energy phonons and changes in the phonon twice over, forming a part of two surfaces. For nanorib- Graphene – Properties and Characterization 2.2 Properties 57 atA Part

a) b) c) Intensity n =1 Intensity D/G = 0.185 60 000 Pristine15 000 Functionalized L1 10 000 L3 2.2 10 000 5000 30 000 D 5000 D 0 L2 1300 1600 0 0 10 μm 10 μm 1100 2000 2900 1100 2000 2900 Raman shift (cm–1) Raman shift (cm–1) d) Intensity n =2 Intensity D/G = 0.012 e) Intensity n = ∞ Intensity D/G ≈ 0 40 000 Pristine50 000 Functionalized 60 000 Pristine50 000 Functionalized

20 000 25 000 30 000 25 000 D D 0 0 0 0 1100 2000 2900 1100 2000 2900 1100 2000 2900 1100 2000 2900 Raman shift (cm–1) Raman shift (cm–1) Raman shift (cm–1) Raman shift (cm–1) Fig. 2.16 (a) Microscopic images of single-layer (right), bilayer (left), and (b) multilayer (n ≈∞) graphene. (c–e) Raman spectra of pristine (left) and functionalized (right) sheets: (c) spot L1 on single sheet with inset showing expanded 1300–1700 cm−1 region, (d) spot L2 on bilayer, and (e) spot L3 on multilayer (n ≈∞, graphite). There is no D peak for the pristine samples (left spectra). The D/G ratio after reaction of single layer (0.185) is about 15 times higher than that for a bilayer (0.012) and greater for other multilayers (≈ 0). Reactions all performed at 35 ◦C with 17 mM 4-nitrobenzene diazonium water with 1 wt.% sodium dodecyl sulfate (SDS)(after[2.131]) bons of graphene, the edges play an increasing role in Graphene can be readily functionalized through determining their reactivity. diazonium or nitrene [2.136] reactions, which can in- It has been shown that, for electron transfer chem- troduce reactive species covalently linked to graphene. istries, single graphene sheets are almost 10 times more These groups then serve as templates for further chem- reactive than bilayer or multilayer graphene (Fig. 2.16) istry and grafting of functional groups, for instance, according to the relative intensity of the disorder (D) through an azide linker. Chemical functionalization of peak in the Raman spectrum examined before and after graphene can be monitored through its effect on the con- chemical reaction [2.131, 132]. Substrate-induced dop- ductivity of graphene, serving as a means to control the ing of the graphene resulting in electron-rich regions electrical transport properties of graphene [2.137, 138]. has been proposed to explain this trend. The effect of Furthermore, p-doped and n-doped regions in graphene doping is greatest in monolayers because the screening can be generated by suitably functionalizing them, for length in the c-axis in graphite and graphene is only 5 Å, instance, with diazonium salts and polyethylene imine, comparable to the interlayer spacing of 3.5Å [2.133, respectively [2.139]. Such chemical modification can 134]. Similarly, the reactivity of edges is at least two also be performed on different parts of a single sheet times higher than the reactivity of the bulk graphene to form p–n junctions in graphene [2.140]. sheet [2.131]. Predictions based on Gerischer–Marcus electron transfer theory and tight-binding approxi- 2.2.7 Properties of Graphene Derivatives mations predict that armchair and zigzag graphene nanoribbons (GNRs) have opposite trends in reactiv- Graphane was theoretically predicted to take one of two ity, with the former increasing with width and the latter configurations: a chair conformer with the hydrogen decreasing. In zigzag ribbons the major reactivity con- atoms alternating on both sides of the plane for the two tribution comes from edge states [2.135]. This reactivity graphene sublattices, and a boat conformer with the hy- trend for zigzag GNR is reversed for very narrow rib- drogen atoms alternating in pairs [2.79]. These chair bons due to the presence of large semiconducting gaps and boat conformers have a direct electronic bandgap with correspondingly low reactivities. of 3.5and3.7 eV, respectively. Graphane is a com- 58 Part A NanoCarbons

atA Part pletely insulating material; its resistivity changes by modulus of fluorographene was measured to be 0.3TPa, two orders of magnitude with decreasing temperature which is about 30% the stiffness of graphene. Simi- from 300 to 4 K, and its carrier mobility decreases to larly, fluorination reduces graphene’s intrinsic breaking

2.3 ≈ 10 cm2/(V s) at liquid-helium temperatures for typi- strength by 2.5 times. However, fluorographene is cal carrier concentrations of 1012 cm−2 [2.80]. able to sustain the same ultimate strain of 15% as Similarly, fluorographene [2.82] is a high-quality graphene. Fluorographene is also strongly hydrophobic, insulator with large optical bandgap of > 3eV and and can be considered the two-dimensional equivalent room-temperature resistivity of > 1012 Ω. The Young’s of Teflon.

2.3 Characterization Each of the properties discussed in the previous sec- 2.3.2 Transmission Electron Microscopy tion has to be measured and correlated using multiple characterization techniques, which are discussed in this Transmission electron microscopy (TEM) is one of section; for instance, electronic properties of graphene the most direct observation techniques to elucidate the have to be independently verified by ARPES, optical structure of graphene. High-resolution TEM can resolve spectroscopy, and electronic transport. This is essen- individual carbon atoms as well as adatoms, defects, tial, since just one measurement, for instance, electronic and other anomalies in graphene (Fig. 2.17)[2.88]. The transport, might not be able to sufficiently distinguish high-energy electrons in a TEM can also be used to between an electronic bandgap and a mobility gap. Sim- engineer defects such as vacancies, cause edge recon- ilarly, mechanical properties have to be confirmed by structions and graphene sublimation, as well as observe a combination of tensile testing, electromechanical res- them in situ [2.10]. Various techniques have been devel- onance, and Raman spectroscopy. oped to transfer micromechanically cleaved graphene flakes onto TEM grids. If folds occur in the trans- 2.3.1 Optical Characterization ferred graphene flake, observation of the folded edge can yield information about the number of layers in Based on the optical properties of graphene discussed the graphene flake [2.9]; a monolayer fold edge turns in an earlier section, and the fact that green light is most up as a single dark line, while a bilayer fold edge comfortable for the eyes, optimal SiO2 thicknesses of appears as two dark lines and so on, in analogy to 90 and 280 nm can be recommended [2.7]. Similarly, single-walled and multi-walled carbon nanotubes. In it has been shown that graphene can be observed on addition, nanobeam electron diffraction (Fig. 2.17) can 50 nm Si3N4 using blue light and on 90 nm poly-methyl also be used to quantify the layering in graphene [2.9]. methacrylate (PMMA) using white light [2.7]. Optical Monolayer graphene can be distinguished from higher- contrast can similarly be used to identify graphene ox- layered graphenes by the anomalous intensity ratio of ¯ ide on Si/SiO2 substrates, as well as to visualize its the diffraction peaks; its 0110 peaks being more intense conversion to reduced GO upon annealing, since both than the 1210¯ peaks. When measured as a function of in- the effective index of refraction and the effective extinc- cidence angle, it probes the whole 3-D reciprocal space. tion coefficient increase [2.141]. The total (integrated) intensity of the 0110¯ and 1210¯ Rayleigh scattering can identify the number of peaks of monolayer graphene varies weakly with tilt an- graphene layers as well as probe their dielectric con- gle and no minima in intensity are observed, since the stant [2.142]. Rayleigh imaging relies on elastically intensities in reciprocal space for monolayer graphene scattered incident photons, while Raman spectroscopy, are continuous rods. In contrast, the total intensity of bi- which is discussed later, collects inelastically scattered layer graphene diffraction peaks varies strongly with tilt photons. For graphene on Si/SiO2 substrate, under angle, including minima at certain angles where some white-light illumination combined with interferomet- peaks vanish [2.143]. However, while the total intensity ric detection, the contrast can be tailored by adjusting in monolayer graphene only decreases slightly, signif- the SiO2 thickness and the light modulations depend icant peak broadening is observed with increasing tilt strongly on the graphene thickness. Up to six layers, the angle. This effect is most pronounced in monolayers, graphene behaves as a superposition of single sheets and and decreases with increasing thickness of the graphene the monochromatic contrast increases linearly. flake. This is attributed to nanoscale corrugations in 2-D Graphene – Properties and Characterization 2.3 Characterization 59 atA Part a) b) –– – 0°– 1120 0° 1210 1210 –

– – 2110 2.3 0110 – 1010 – –– 0110 1010 1120 – – – 1100 2110 1210

Tilt axis

c) Intensity (arb. units) d) Intensity (arb. units) 600 150

400 100

50 200

0 0 00.2 0.4 0.4 0.8 1 1.2 1.4 00.2 0.4 0.4 0.8 1 1.2 1.4 Distance (Å–1) Distance (Å–1) e) Intensity (arb. units) f) Intensity (arb. units) – 0–110 –1100 15 000 1010 4000

10 000

2000 – –0110 1010– 5000 0100

0 0 –30–20 –10 0 10 20 30 05010 20 30 40 Tilt angle (deg) Tilt angle (deg) Fig. 2.17 (a) Atomic-resolution TEM of graphene (after [2.88]). (b) Nanobeam electron diffraction patterns of monolayer and bilayer graphene. Relative intensities of 1100¯ and 1210¯ peaks in (c) monolayer and (d) bilayer graphene. Variation of intensity of the 1100¯ peaks with tilt angle for (e) monolayer and (f) bilayer graphene (after [2.143]) graphene, with the surface normal deviating on average beam in order to yield a smooth Gaussian shape of the by ±5◦ in monolayers and ±2◦ in bilayers. Consider- diffraction peak, it is estimated that the corrugations ing that the spatial extent of these corrugations cannot occur on length scales of 10–25 nm. This nanoscale cor- be drastically smaller than the coherence length of the rugation extending into the third dimension squares the diffracted electrons and that a large number of orienta- existence of 2-D graphene with the theoretical predic- tions should be included within the submicron electron tion that perfect 2-D atomic crystals cannot exist. 60 Part A NanoCarbons

atA Part process needs to be repeated at least for every differ- a) b) ent substrate and processing conditions involved in the I graphene preparation.

2.3 In addition, AFM is used to measure the flatness of II graphene on various substrates, and it is revealed that ul- traflat graphene can be obtained on mica surface [2.91] with standard deviation of height and height correla- tion length of 24.1 pm and 2 nm, respectively, compared with 154 pm and 22 nm, respectively, for SiO2 substrate. III AFM can also be used in nanoindentation mode to probe the stiffness of suspended graphene (Fig. 2.18)[2.96]. c) d) This technique has the advantage that the sample geom- etry can be precisely defined and the sheet is clamped around the entire hole circumference; a Young’s modu- lus of 1 TPa and intrinsic breaking strength of 42 N/m 0.5 μm have been measured. EFM has been used to confirm that the surface potential of few-layer graphene increases 1 μm 1.5 μm with film thickness, approaching bulk graphite values for five or more layers [2.145]. This is a measure of the extent of the electrostatic interaction between graphene Fig. 2.18a–d Images of suspended graphene membranes. (a) Scan- and the substrate, and the screening of these perturba- ning electron micrograph of a large graphene flake spanning an . μ tions by underlying graphene layers. array of circular holes (1 and 1 5 m in diameter). Area I shows STM can image graphene with atomic resolution, a hole partially covered by graphene, area II is fully covered, and and the correlation of the graphene hexagonal lattice to area III is fractured from indentation. Scale bar 3 μm. (b) Noncon- . μ the direction of the edge of an exfoliated flake reveals tact AFM image of one membrane, 1 5 m in diameter. The solid the orientation of the edge as being either armchair or blue line is a height profile along the dashed line. The step height at . zigzag [2.146]. It has been shown that, in mechanically the edge of the membrane is 2 5nm.(c) Schematic of nanoindenta- exfoliated graphene flakes, a majority of edges follow tion on suspended graphene membrane. (d) AFM image of fractured either of these orientations and intersect at angles that membrane (after [2.96]) are multiples of 30◦. STM imaging of graphene grown epitaxially on SiC [2.147] or metallic substrates [2.20, 2.3.3 Scanning Probe Techniques 23,87,148] reveals the superlattice structure and the ex- tent of coupling between the graphene and substrate. Scanning probe techniques discussed here in the context STM can also be used to locate and characterize point of graphene characterization include atomic force mi- defects in the graphene lattice [2.149, 150]. STS can be croscopy (AFM), electrostatic force microscopy (EFM), used to probe the atomically resolved local electronic and scanning tunneling microscopy (STM) and spec- structure of graphene [2.147,151,152]. A prominent gap troscopy (STS). in the tunneling spectrum unique to graphene has been AFM in tapping mode is commonly used to measure observed and attributed to a phonon-mediated inelastic the thickness of graphene flakes on substrates; how- tunneling process. ever, the correlation between measured thickness and actual thickness as well as number of layers is chal- 2.3.4 Angle-Resolved Photoemission lenging [2.144]. Electrostatic interactions between the Spectroscopy (ARPES) tip and the graphene, adsorbed moisture, and incorrect choice of AFM parameters such as free amplitude val- ARPES is a direct experimental technique that has ues can all influence the final measured thickness of measured the electronic density of states in graphene a graphene flake. Therefore, while using AFM to char- with both energy and momentum information. The ∗ acterize graphene flakes, an internal reference such as shapeoftheπ and π bands near EF at the K-point a fold in the flake, or a second characterization tool from ARPES reveals the transition from 2-D to bulk such as Raman spectroscopy, is often used to correlate character from one to four layers of graphene [2.153]. the measured thickness and number of layers, and this Fermi velocities and effective masses of the electrons Graphene – Properties and Characterization 2.3 Characterization 61 atA Part a) Intensity (arb. units) d) Monolayer q = Exchanged 514 nm

phonon momentum 2.3 50 000 q π* 40 000 Graphite b c ε L 30 000 a Electron energy π ε 20 000 L = Laser energy Fermi level Bilayer Graphene q1B 10 000 q1A

0 ε 1500 2000 2500 3000 L Raman shift (cm–1) b) Intensity (arb. units) c) Intensity (arb. units) 2500 514 nm 514.5 nm q Edge 2A Graphite graphite q 2000 2B ε 10 layers L 1500 D1 D2 5 layers 1000

2 layers Edge 500 1 layer 1 layer D 0 2600 2700 2800 1300 1350 1400 Γ KMK' Raman shift (cm–1) Raman shift (cm–1) Fig. 2.19 (a) Comparison between Raman spectra of graphene and graphite. (b) Evolution of 2D peak shape with number of layers of AB-stacked graphene. (c) Comparison of D peak between graphene and graphite. (d) Scattering processing causing the 2D peak components in monolayer and bilayer graphene (after [2.156]) can also be measured. ARPES on epitaxial AB-stacked graphene has been able to reproduce key experimental bilayer graphene on SiC has revealed that the magni- observations including the indication of a mismatch be- tude of the gap between the valance and conduction tween the upper and lower halves of the Dirac cone. bands can be varied by controlling the carrier density, for instance, with a transverse electric field. On the other 2.3.5 Raman Spectroscopy hand, APRES also reveals that individual graphene lay- ers of multilayer graphene grown on SiC(0001)¯ behave The three significant Raman spectral features in as decoupled monolayers with independent linearly dis- graphene are the G peak at ≈ 1580 cm−1, the D peak persed bands at the K-point [2.154]. ARPES also al- at ≈ 1350 cm−1, and the 2D peak at ≈ 2700 cm−1,as lows studies of electron–electron, electron–phonon, and seen in Fig. 2.19 [2.156]. The G peak is due to the E2g electron–plasmon interactions, and indicates that all mode, i. e., in-plane vibrations of the carbon atoms. The three must be considered on an equal footing in under- D peak and 2D peak are strongly dispersive, with exci- standing the quasiparticle dynamics in graphene [2.155]. tation energy due to the Kohn anomaly at the K-point, Ab initio simulations of the ARPES intensity spectra of while the G peak is not. 62 Part A NanoCarbons atA Part a) Pos (2D) (cm–1) b) FWHM (G) (cm–1) As-deposited graphene As-deposited graphene p-doping UV-disordered graphene

2.3 Gated graphene [12] 2700 Disordered graphene [19] Suspended graphene [19] 80 Gated graphene [11] Disordered graphite [17] Disorder = 514 nm 60 Suspended graphene [19] 2694 40 2688

20 2682

2676 n-doping 8 Doping 6 2670 1580 1590 1600 1610 1575 15801585 1590 1595 1600 Pos(G) (cm–1) Pos(G) (cm–1) Fig. 2.20 (a) Pos(G) as a function of Pos(2D) for as-deposited graphene and gated graphene. (b) Pos(G) as a function of FWHM(G) for as-deposited graphene, compared with disordered graphene, disordered graphite, and gated graphene. The dotted lines are only a guide to the eye (after [2.159])

The 2D peak is the second order of the zone- from monolayer graphene by its full-width at half- boundary phonons and therefore does not require maximum (FWHM) of 50 cm−1, which is twice that of defects. For monolayer and few-layer graphene, the 2D monolayer graphene. peak serves as a fingerprint for identification [2.156]. A similar observation can also be made for the In general, the 2D peak of graphite has four compo- D peak [2.156]. The D peak of monolayer graphene nents: 2-D1B, 2-D1A, 2-D2A,and2-D2B. Monolayer is a single sharp peak, while in bulk graphite it can graphene has a single sharp 2D peak, dominated by the be resolved into two peaks, D1 and D2. The D peak 2-D1A component. In bilayer graphene, the 2-D1A and is observed in defective graphene, and prominently at 2-D2A peaks have higher relative intensity compared the edges of graphene flakes. In carbon nanotubes, con- with the other two, and the 2D peak appears up-shifted finement and curvature split the two degenerate modes and broader compared with monolayer graphene. In of the G peak into G+ and G− components, whereas monolayer graphene, there is only one phonon sat- only one G peak is observed in graphene. The D peak isfying the double-resonance conditions for the 2D arises from a double-resonance process involving elec- Raman peak. In bilayer graphene, the interaction be- tron scattering by zone-boundary phonons as well as tween graphene layers causes the π and π∗ electronic defects in graphene. Since these do not satisfy the Ra- bands to split into four bands. According to density man selection rule, they do not occur in the Raman functional theory dipole matrix elements, the incident spectra of defect-free graphene. A similar process in- light couples more strongly to two among four possi- volving intravalley scattering gives rise to a D peak at ble optical transitions (Fig. 2.19). The excited electrons ≈ 1620 cm−1 in defective graphene. can be scattered by phonons with momenta q1B, q1A, The Raman spectrum of graphene also responds q2A,andq2B. The corresponding processes for holes to doping, i. e., changes in the Fermi surface of are associated to identical phonon momenta, resulting graphene [2.160, 161]. Graphene can be doped inten- in four components to the 2D peak of bilayer graphene. tionally or unintentionally, by electron transfer from As the number of layers further increases, the 2D1 adsorbed chemical species, and by modulation of the peaks reduce in intensity, and beyond five layers, it electronic band structure by a gate voltage or interac- resembles the 2D peak of bulk graphite. It is also im- tion with the substrate. The G peak upshifts for both portant to note here that non-AB stacked graphene, such hole and electron doping. The position of the 2D peak, as multilayer CVD graphene, also shows a single 2D however, decreases monotonically with increasing elec- peak [2.157, 158]. However, this can be distinguished tron concentration or decreasing hole concentration, Graphene – Properties and Characterization 2.3 Characterization 63

Fig. 2.21a–d Field effect in few-layer A Part Š σ –1 a) (kΩ) b) (mΩ ) graphene. (a) Typical dependences of graphene’s resistivity ρ on gate volt-

age for different temperatures (T = 5, 2.3 8 70, and 300 K for top to bottom curves, respectively). (b) Example of changes to the film’s conductivity 6 3 σ = 1/ρ (Vg) obtained by inverting the70Kcurve(dots). (c) Hall co- efficient RH versus Vg for the same 4 film at T = 5K.(d) Temperature de- 0 pendence of carrier concentration n0 –100 0 100 in the mixed state for the film in (a) V (V) 2 g (open circles), a thicker FLG film (squares), and multilayer graphene d) n (T)/n (4K) ≈ 0 0 0 (d 5nm,solid circles). Red curves –100 –50 0 50 100 in b–d are the dependences calculated Vg (V) 6 from the model of a 2-D semimetal il- c) R H (kΩ/T) lustrated by insets in (c) (after [2.4]) ε F 4 0.5 δε εF

0 2

εF

–100 –50 0 50 100 0 100 300 Vg (V) T (K) and the 2D peak can be used to distinguish between creasing influence of doping. Highly doped samples electron and hole doping. The changes in the G peak yield a slope of < 1. In another approach, if the plot position are related to the nonadiabatic Kohn anomaly of FWHM versus position of the G peak is monoton- at the Γ point, while the shift in the 2D peak posi- ically decreasing, it is related to doping, while if it is tion is due to electron–electron scattering in addition monotonically increasing, it is caused by strain or dis- to electron–phonon scattering. Taken together, a plot order (Fig. 2.20b) [2.161]. Under uniaxial strain, the of 2D versus G peak positions can be used to distin- G peak splits into two bands, G+ and G−, analogous guish between electron and hole doping in graphene to the effect of curvature on the G peak of carbon nano- (Fig. 2.20a) [2.159]. Doping trends can also be ob- tubes [2.165]. served in the FWHM and intensities of Raman peaks. Raman peaks in graphene can also be shifted by bi- 2.3.6 Electrical Characterization axial strain induced, for example, by interaction with the substrate [2.162, 163]. It has been proposed that The first electrical characterization of graphene was the correlation between normalized shift of 2D and carried out on micromechanically cleaved few-layer G peak positions, i. e., Δ2D/2D0 and ΔG/G0, where graphene flakes (Fig. 2.21)[2.4]. The sheet resistivity 2-D0 and G0 are the peak positions for undoped ρ of FLG flakes varies with applied gate voltage Vg, graphene, indicates whether the shift arises from dop- exhibiting a sharp peak of several kiloohms and decay- ing or strain [2.164]. When Δ2D/2D0 versus ΔG/G0 ing to ≈ 100 Ω at high Vg. The conductivity σ = 1/ρ is plotted from Raman spectra obtained at a num- increases linearly with Vg on either side of the resis- ber of points on a graphene samples, a linear fit with tivity peak (conductivity valley). The Hall coefficient slope close to 1.58 ± 0.18 indicates that strain plays RH reverses sign at the same VG as the resistivity a predominant role, while a smaller slope indicates in- peak. This resembles an ambipolar semiconducting 64 Part A NanoCarbons atA Part Š Š a) xx (kΩ) xy (kΩ) b) BF (T) Δ (Ω) 1.5 80 2.3 –100 V at 9 T 4

10 60 +80 V 1.0 +80 V 3 0 50 100 40 T (K) +100 V 2 0.5 –25 V 20 1

–100 V 0 0 2 4 6810 –100 –50 050100 B (T) Vg (V) Fig. 2.22 (a) Examples of ShdH oscillations in a graphene device for different gate voltages; T = 3K,and B is the magnetic field. (b) Dependence of the frequency of ShdH oscillations BF on gate voltage. Solid and open symbols are for samples with δε ≈ 6 and 20 meV, respectively. Solid lines are guides to the eye. The inset shows an example of the temperature dependence of amplitude % of ShdH oscillations (circles)(after[2.4])

field-effect transistor (FET), except that there is no zero- mobility and minimum conductivity also decrease as conductance region since there is no bandgap. This a result of defects, which can be induced in a controlled tunability of conductivity is achieved by transforming fashion to study this effect, for instance, by ion irradia- the graphene by electric-field doping from a completely tion [2.168] or exposure to atomic hydrogen [2.169]. electron to completely hole conductor, passing through The earliest determination of carrier mobility by a mixed state where both electrons and holes con- field-effect and magnetoresistance measurements in tribute equally. This behavior holds true for monolayer few-layer graphene yielded ≈ 10 000 cm2/(V s), which graphene as well as undoped bilayer graphene; how- was independent of the absolute temperature, indicating ever, doped bilayer graphene has an intrinsic bandgap that it was limited by scattering defects. For multilayer and will be discussed subsequently. In both the elec- graphene, the mobility reached 15 000 cm2/(V s) at 2 tron and hole regions, RH decreases with increasing 300 K and 60 000 cm /(V s) at 4 K. Substrate-induced carrier concentration as 1/ne as expected, and the resis- charge puddles are significantly reduced in suspended tivity follows the standard ρ = 1/neμ relation, where graphene, and low-temperature mobility approaching μ is the carrier mobility. Also significant is the mini- 200 000 cm2/(V s) has been reported [2.170, 171]. In mum conductivity of graphene at the charge neutrality such devices, the conductivity of suspended graphene 2 point (σmin), which has been shown to be about 4e /h. at the Dirac point is strongly temperature dependent This σmin is not related to the physics of the Dirac and approaches ballistic values at liquid-helium temper- point singularity, but instead related to charge-density atures [2.171]. inhomogeneities (electron–hole puddles) induced by Graphene flakes also exhibit pronounced Shub- the substrate or charged impurities [2.166, 167]. In- nikov–de Haas (ShdH) oscillations in both longitudinal variably, the position of the charge neutrality point (ρ resistivity ρxx and Hall resistivity ρxy (Fig. 2.22). The peak) is shifted to large positive VG, leaving the ungated oscillations depend only on the perpendicular compo- graphene as a hole metal. This large shift is attributed nent of the magnetic field B cos Θ, where Θ is the angle to unintentional doping of the graphene by adsorbed between the magnetic field and the graphene. The fre- species such as water. The peak position can also be quency of the SsdH oscillations BF depends linearly on shifted by intentional doping [2.166] or removal of VG, indicating that the Fermi energies εF of holes and dopants by thermal or current-induced annealing [2.13], electrons are proportional to their concentration n. This but such charged impurities do not affect σmin. The is different from the 3-D dependence εF proportional Graphene – Properties and Characterization 2.3 Characterization 65 atA Part σ 2 a) xy(4e /h) b) mc/me 8 0.09

B =12T 2.3 6

4 Doped

2 0.06

0

–2

Pristine –4 Screened 0.03 T =4K Unscreened –6 –100 –50 050100 –8 –4 048 12 –2 Vg (V) n (10 cm )

Fig. 2.23 (a) Measured Hall conductivity of pristine (undoped) and chemically doped bilayer graphene (n0 ≈ 5.4×1012 cm−2), showing a comparison of the QHE in both systems. (b) Cyclotron mass versus n, normalized to the free electron mass me. Experimental data are shown as open circles.Theinset shows an electron micrograph (in false color) of the Hall bar device with a graphene ribbon width of 1 μm(after[2.172]) to n2/3, proving the 2-D nature of charge carriers in micromechanically cleaved graphene on an oxidized graphene. silicon wafer (300 nm SiO2)[2.172]. The silicon sub- QHE has been observed in graphene even at room strate was used as a back gate to modulate the carrier temperature, since the electrons suffer little scattering density n, while doping from adsorbed NH3 on the ex- due to their relativistic nature and have a large cyclotron posed graphene surface was used to mimic a top gate gap which exceeds the thermal energy kBT by a factor and open a bandgap corresponding to a fixed electron of 10. density n0. Under applied magnetic field, a plateau at Under the influence of strong magnetic field, elec- zero Hall conductivity σxy = 0 occurs in biased bilayer trons in a two-dimensional system such as graphene graphene, as a result of the gap opened between the va- 2 develop strong Coulomb interactions between them, lence and conduction bands. Plateaus at σxy = 4Ne /h leading to correlated states of matter such as a frac- occur as expected, including at N = 0(Fig.2.23a). This tional quantum Hall liquid. This collective behavior in is the standard integer QHE that is expected for an am- graphene was predicted to yield the fractional quantum bipolar semiconductor with energy gap larger than the Hall effect (FQHE); however, due to prevalent disor- cyclotron energy. This plateau is absent in monolayer der effects, observation of this remained elusive in early and unbiased bilayer graphene, which show an anoma- measurements. The FQHE was eventually reported in lous double step of 8e2/h at n = 0, indicative of the suspended graphene devices when a plateau at filling metallic state at the charge neutrality point. A huge peak factor ν = 1/3 was observed above a magnetic field as in the longitudinal resistivity ρxx at n = 0 was also ob- low as 2 T [2.173, 174]. An insulating state was also served, exceeding 150 kΩ at 4 K compared with 6 kΩ observed at magnetic fields B > 5 T and filling factors for the unbiased case. From Shubnikov–de Haas mea- ν<0.15, which has been attributed to symmetry break- surements, the cyclotron mass, mc, in biased bilayer ing of the zeroth Landau level by electron–electron graphene is found to be an asymmetric function of the interaction. carrier density, which is a clear signature of a bandgap The first electron transport measurement of the tun- (Fig. 2.23b). Measurements with dual-gated bilayer able bandgap in bilayer graphene was conducted on graphene have also confirmed these results [2.175]. 66 Part A NanoCarbons

atA Part Fig. 2.24 a) E (meV) b) g (a) SEM im- E (meV) g age of a set 2.3 of graphene 100 nanoribbon de- 100 vices of varying 10 width. (b) De- pendence of 1 0306090 θ (deg) energy gap on 10 P1 ribbon width P2 and orientation P3 P4 (inset). Dashed D1 D2 line shows the 1 predicted em- 0306090 pirical scaling W (nm) (after [2.177])

Due to the linear dispersion relation in graphene, As discussed previously, another route to opening intrinsic electron scattering by acoustic phonons is in- a bandgap in graphene is to exploit lateral confine- dependent of carrier density and only contributed 30 Ω ment of charge carriers in a graphene nanoribbon, to the room-temperature resistivity of graphene. This which creates an energy gap near the charge neutrality would yield an intrinsic mobility of 2 × 105 cm2/(V s) at point [2.177]. Graphene nanoribbons of varying widths carrier density of 1 × 102/cm2, which would be make it and different crystallographic orientations have been the highest known mobility, superior to those of InSb fabricated by lithographic patterning of monolayer ex- and semiconducting single-walled carbon nanotubes. foliated graphene (Fig. 2.24). An energy gap is observed However, a strong temperature dependence of mobil- for narrow ribbons, which scales inversely with ribbon ity is observed in substrate-supported graphene devices, width. Energy gaps in excess of 100 meV were observed suggestive of extrinsic scattering, which limits the mo- for widths less than 20 nm, which could have poten- bility to about 4 × 104 cm2/(V s) [2.176]. tial technological relevance. It has also been shown that Various approaches have been proposed to increase edge states do not contribute to the dominant electrical the performance of graphene devices, in particular noise at low frequencies for nanoribbons as narrow as to reduce impurity scattering and enhance mobility. 20 nm [2.181]. However, the lack of a well-defined crys- Ultrahigh current density-induced removal of adsor- tallographic structure of lithographically etched edges bents, photoresist, or e-beam resist residue can be used means that the effect of orientation is not observed in to clean graphene in situ during transport measure- the bandgap produced in such nanoribbons. ments [2.13]. A parylene-coated SiO2 substrate used It is also possible to fabricate quantum dot (QD) as a dielectric stack for back-gating yields a stable devices entirely out of graphene using a similar litho- charge neutrality point and low hysteresis [2.178], since graphic procedure [2.182–184]. Such a device consists the hydrophobic nature of the parylene surface sup- of a graphene island connected to the source and drain presses moisture-related doping and charge-injection via two narrow graphene constrictions and three fully effects and yields mobilities of up to 10 000 cm2/(V s). tunable graphene lateral gates (Fig. 2.25). Larger QDs Similar results have been achieved by utilizing an (> 100 nm) show conventional single-electron transis- organic polymer buffer between graphene and conven- tor characteristics, with periodic Coulomb blockade tional top-gate dielectrics [2.179]. It was demonstrated peaks. For smaller QDs, the peaks become strongly that merely changing the dielectric to a high-k dielec- nonperiodic, indicating a strong contribution from tric or media does not increase the carrier mobility quantum confinement. The narrow constrictions remain beyond ≈ 10 000 cm2/(V s), suggesting that Coulomb conductive and show a confinement gap of ≈ 0.5eV. scattering is not the dominant limitation beyond this This can be extended to a double QD system (Fig. 2.25) regime [2.180]. Phonon scattering or resonant scatterers where the coupling of the dots to the leads and between with energy close to the Dirac point have been proposed the dots is tuned by graphene in-plane gates [2.185]. as alternate mechanisms. This structure has been proposed for the realization of Graphene – Properties and Characterization 2.3 Characterization 67 atA Part a) c) G (e2/h) d) Graphene Ti/Au G (e2/h) PG 2.3

S 0.150 D 1.0 B1 B2 0.125 300 nm b) 0.100

GC 0 100 200 GL δVg (mV) GR 0.5 e) VB (V) LR 5 CL GR 0 SD –5 0.0 100 nm –20–10 0 10 20 0 50 100 δV (mV) Vg (V) g Fig. 2.25 (a) SEM of an all-graphene single quantum dot device (after [2.184]). (b) SEM of an all-graphene double quantum dot device (after [2.185]). (c) Conductance of a graphene single QD device over a wide range of gate voltages at T = 4K.(d) Zoom in to low-conductance region showing Coulomb blockade oscillations (after [2.182]). (e) Coulomb diamonds showing differential conductance as a function of gate voltage and drains–source bias spin qubits from graphene QDs[2.186]. It has been has been explored as an alternative to pristine mono- shown that, in an array of many qubits, it is possible layer graphene. However, measurements in individual to couple any two of them via Heisenberg exchange monolayer RGO flakes have yielded conductivities while the others are decoupled by detuning. This unique ranging between 0.05 and 2 S/cm and field-effect feature is a direct consequence of the quasirelativistic mobilities of 2–200 cm2/(V s) at room temperature. nature of carriers in graphene. Conductivity decreases by up to three orders of mag- One of the important considerations in electronic nitude when measured down to 4 K, following a T −1/3 device performance is the signal-to-noise ratio, where dependence, suggesting variable-range hopping con- usually the low-frequency 1/ f noise dominates. In duction between regions of highly reduced (nearly monolayer graphene, the 1/ f noise follows Hooge’s pristine) graphene islands separated by defective or empirical relation with a noise level comparable to poorly reduced regions. carbon nanotube and bulk semiconductor devices. How- ever, in bilayer graphene, the 1/ f noise is strongly Spintronics suppressed and obeys a unique dependence on carrier When graphene devices are fabricated with ferromag- density, due to effective screening of carrier scattering netic electrodes, such as the soft magnetic NiFe or by external impurities [2.134]. In monolayer graphene, Co, it is possible to inject spin-polarized current into the noise amplitude is minimum at the Dirac point and the graphene. A thin Al2O3 or MgO tunnel barrier increases with increasing carrier density. However, in is used at the ferromagnet–graphene interface. High- bilayer graphene, the noise amplitude achieves a max- quality graphene enjoys ballistic transport with spin imum at the Dirac point and decreases with increasing relaxation lengths between 1.5and2μm even at room carrier density. In both cases, the noise is independent temperature, which is only weakly dependent on charge of carrier type. density [2.187]. The switching fields of the elec- In an effort to realize commercial viability of trodes can be controlled by in-plane shape anisotropy. graphene electronics, reduced graphene oxide (RGO) Graphene spin valves have been constructed using ei-