Polymer Composites with Carbonaceous Nanofillers E-Book

Sie Chin Tjong

• Herausgeber: Wiley-VCH
• Kategorie: Fachliteratur
• Sprache: Englisch

Seitenzahl: 519

Veröffentlichungsjahr: 2012

Beschreibung

Written by an expert in the field of nanomaterials, composites, and polymers, this book provides up-to-date information on recent advances in various aspects of polymer composites reinforced by carbonaceous nanofillers, including their fabrication and their electrical, thermal, and mechanical properties. It also extensively covers applications of these nanocomposites in fuel cells, sensors, electromagnetic interference shielding, human implants and scaffolds.

Leseprobe

Table of Contents

Related Titles

Title Page

Copyright

Preface

Abbreviations

Chapter 1: Introduction

1.1 Graphene-Based Nanomaterials

1.2 Carbon Nanotubes

1.3 Carbon Nanofibers (CNFs)

1.4 Physical Properties of Graphene

1.5 Properties of Carbon Nanotubes

1.6 Properties of Carbon Nanofibers

1.7 Current Availability of Carbonaceous Nanomaterials

1.8 Multifunctional Composite Materials

Nomenclature

References

Chapter 2: Preparation of Polymer Nanocomposites

2.1 Overview

2.2 Dispersion of Nanofillers

2.3 Solution Mixing

2.4 Melt Mixing

2.5 In situ Polymerization

2.6 Patent Processes

References

Chapter 3: Thermal Properties of Polymer Nanocomposites

3.1 Crystallization

3.2 Characterization Techniques for Crystallization

3.3 Thermal Stability

3.4 Thermal Conductivity

Nomenclature

References

Chapter 4: Mechanical Properties of Polymer Nanocomposites

4.1 Background

4.2 General Mechanical Behavior

4.3 Fracture Toughness

4.4 Strengthening and Toughening Mechanisms

4.5 Nanocomposites with Graphene Fillers

4.6 Nanocomposites with EG and GNP Fillers

4.7 Nanocomposites with CNT and CNF Fillers

4.8 Composites with Hybrid Fillers

Nomenclature

References

Chapter 5: Electrical Properties of Polymer Nanocomposites

5.1 Background

5.2 Percolation Concentration

5.3 Electrical Conductivity and Permittivity

5.4 Current-Voltage Relationship

5.5 Positive Temperature Coefficient Effect

5.6 Hybrid Nanocomposites

References

Chapter 6: Carbonaceous Nanomaterials and Polymer Nanocomposites for Fuel Cell Applications

6.1 Overview

6.2 Polymer Exchange Membrane Fuel Cell

6.3 Conventional Bipolar Plates

6.4 Polymer Nanocomposite Bipolar Plates

6.5 Electrical Characteristics of Nanocomposite Bipolar Plates

6.6 Mechanical Properties of Nanocomposite Bipolar Plates

6.7 Electrocatalyst Supports

Nomenclature

References

Chapter 7: Polymer Nanocomposites for Biomedical Applications

7.1 Overview

7.2 Bone Implants

7.3 Biocompatibility of Carbon Nanotubes

7.4 CNT/Polymer Nanocomposites for Load-Bearing Implants

7.5 CNT/Polymer Nanocomposite Scaffolds

7.6 Nervous System Remedial Applications

7.7 Biocompatibility of Graphene Oxide and Its Nanocomposites

References

Chapter 8: Polymer Nanocomposites for Electromagnetic Interference (EMI) Shielding

8.1 Introduction

8.2 EMI Shielding Efficiency

8.3 Nanocomposites with Graphene Fillers

8.4 Nanocomposites with GNPs

8.5 Nanocomposites with CNTs and CNFs

8.6 Foamed Nanocomposites for EMI Applications

Nomenclature

References

Chapter 9: Polymer Nanocomposites for Sensor Applications

9.1 Introduction

9.2 Pressure/Strain Sensors

9.3 Gas and Humidity Sensors

9.4 Organic Vapor Sensors

Nomenclature

References

Index

Related Titles

Tjong, S. C.

Carbon Nanotube Reinforced Composites

Metal and Ceramic Matrices

2009

ISBN: 978-3-527-40892-4

Thomas, S., Stephen, R.

Rubber Nanocomposites

Preparation, Properties and Applications

2010

ISBN: 978-0-470-82345-3

Hierold, C. (ed.)

Carbon Nanotube Devices

Properties, Modeling, Integration and Applications

2008

ISBN: 978-3-527-31720-2

Haley, M. M., Tykwinski, R. R. (eds.)

Carbon-Rich Compounds

From Molecules to Materials

2006

ISBN: 978-3-527-31224-5

Sperling, L. H.

Introduction to Physical Polymer Science

2006

ISBN: 978-0-471-70606-9

Reich, S., Thomsen, C., Maultzsch, J.

Carbon Nanotubes

Basic Concepts and Physical Properties

2004

ISBN: 978-3-527-40386-8

The Editor

Prof. Sie Chin Tjong

City University of Hong Kong

Department of Physics and Materials Science

Hongkong, PR China

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Library of Congress Card No.: applied for

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at <http://dnb.d-nb.de>.

© 2012 Wiley-VCH Verlag & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany

All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

Print ISBN: 978-3-527-41080-4

ePDF ISBN: 978-3-527-64875-7

ePub ISBN: 978-3-527-64874-0

mobi ISBN: 978-3-527-64873-3

oBook ISBN: 978-3-527-64872-6

Preface

Carbon nanotubes and graphene sheets exhibit unique and extraordinary electrical, mechanical, and thermal properties rendering them attractive fillers for reinforcing polymers to form functional and structural composite materials of high performance. The performance of the polymer nanocomposites relies on the inherent properties of carbonaceous nanofillers, and on optimizing the dispersion, interfacial interaction, and nanoscale exfoliation of those fillers within the polymer matrix. Designing smart polymer nanocomposite materials with the appropriate processing-structure-property relationships for biomedical, electronic, electromagnetic interference shielding, and chemical sensing as well as structural engineering applications is challenging. In recent years, one-dimensional carbon nanotubes have been incorporated into various types of polymeric materials for achieving these purposes. However, the high cost, tedious purification and high tendency of agglomeration of carbon nanotubes hurdle the development of nanotube/polymer composites in engineering applications. The recent successful synthesis of two-dimensional graphene layers from graphite oxide via chemical and thermal reduction techniques has sparked enormous interest in their properties, functions, and applications. The low cost and ease of fabrication of graphene offer tremendous opportunities for chemists and materials scientists to explore and develop novel graphene/polymer nanocomposites with excellent biological, mechanical, and physical properties. This book focuses exclusively on the latest research related to the synthesis and property characterization of one- and two dimensional carbonaceous nanomaterials and their polymer nanocomposites, and addresses potential applications of these materials to bipolar plates of fuel cells, electrocatalysts, human orthopedic implants and scaffolds, electromagnetic interference shielding materials, and gas-, pressure- and temperature sensors. This book serves as a valuable and informative reference to scientists, engineers, medical technologists, and practitioners engaged in the teaching, research, development, and use of functional polymer composites with carbonaceous nanofillers.

Sie Chin Tjong

CEng CSci FIMMM

City University of Hong Kong

Abbreviations

AC

alternating current

AFM

atomic force microsopy

AIBN

2,2

′

-azobisisobutyronitrile

ARTP

atom transfer radical polymerization

BIB

α-bromoisobutyryl bromide

BP

benzoyl peroxide

CB

carbon black

CF

carbon fiber

CMG

chemically modified graphene

CNF

carbon nanofiber

CNT

carbon nanotube

CS

chitosan

CTAB

hexadecyltrimethylammonium bromide

CTE

coefficient of thermal expansion

CVD

chemical vapor deposition

DBP

dibutyl phthalate

DC

direct current

DGEBA

diglycidyl ether bisphenol-A

DENT

double-edge-notched tension

DMA

dynamic mechanical analysis

DMAc

N,N

dimethylacetamide

DMF

dimethylformamide

DMSO

dimethyl sulfoxide

DSC

differential scanning calorimetry

DWNT

double-walled carbon nanotube

EG

expanded graphite

ECM

extracellular matrix

ECSA

electrochemically active surface area

EDS

energy dispersive spectroscopy

EMI

electromagnetic interference

EVA

ethylene vinyl acetate

EWF

essential work of fracture

FGS

functional graphene sheet

FMWNT

functionalized multiwalled carbon nanotube

GIC

graphite intercalation compound

GDL

gas diffusion layer

GNP

graphite nanoplatelet

GO

graphene oxide

HA

hydroxyapatite

HDPE

high-density polyethylene

HDT

heat deflection temperature

HEK

human epidermal keratinocyte

HiPCo

High-pressure carbon oxide disproportionation

HOPG

highly oriented pyrolytic graphite

HOR

hydrogen oxidation reaction

iGO

isocyanate-treated graphene oxide

LDPE

low-density polyethylene

LEFM

linear elastic fracture mechanics

LLDPE

linear low-density polyethylene

MA-g-PP

maleic anhydride-grafted polypropylene

MD

molecular dynamics

MEA

membrane electrode assembly

MMT

montmorillonite

MTS

3-(4,5-dimethylthiazol-2-yl)-5(3-carboxymethonyphenol)-2-(4-sulfophenyl)-2H-tetrazolium

MTT

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

MWNT

multiwalled carbon nanotube

ORR

oxygen reduction reaction

PA

polyamide

PAA

poly(acrylic acid); polyallylamine; polyamic acid

PC

polycarbonate

PCL

polycaprolactone

PDMS

poly(dimethyl siloxane)

PE

polyethylene

PECVD

plasma-enhanced chemical vapor deposition

PEDOT

poly(3,4-ethylenedioxythiophene)

PEEK

poly(etheretherketone)

PEMFC

proton exchange membrane fuel cell

PEN

poly(ethylene-2,6-naphthalate)

PEO

poly(ethylene oxide)

PET

polyethylene terephthalate

PGMA

poly(glycidyl methacrylate)

PI

polyimide

PLA

polylactic acid

PmPV

poly(

m

-phenylene vinylene)

PMMA

poly(methyl methacrylate)

PS

polystyrene

PSF

polysulfone

PTC

positive temperature coefficient

PTT

polytrimethylene terephthalate

PU

polyurethane

PVA

poly(vinyl alcohol)

PVC

polyvinyl chloride

PVD

physical vapor deposition

PVDF

polyvinylidene fluoride

PVP

polyvinyl pyrrolidone

P3HT

poly(3-hexylthiophene)

rGO

reduced graphene oxide

RBM

radial breathing mode

SAED

selected-area electron diffraction

SAN

styrene-acrylonitrile

SBR

styrene-butadiene rubber

SDBS

sodium dodecylbenzenesulfonate

SDS

sodium dodecyl surfate

SE

shielding efficiency

SEM

scanning electron microscopy

SENB

single-edge-notched bending

SGF

short carbon fiber

SIP

surface-initiated polymerization

sPS

syndiotactic polystyrene

SR

silicone rubber

SWNT

single-walled carbon nanotube

TEA

triethylamine

TEGO

thermally expanded graphene oxide

TEM

transmission electron microscopy

TETA

triethylenetetramine

T

g

glass-transition temperature

TGA

thermogravimetric analysis

THF

tetrahydofuran

TLP

tissue culture plate

TPU

thermoplastic polyurethane

TRG

thermally reduced graphene

VGCNF

vapor-grown carbon nanofiber

VLS

vapor-liquid-solid

WST-1

2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium

XRD

X-ray diffraction

Chapter 1

Introduction

1.1 Graphene-Based Nanomaterials

Carbon exists in many forms including buckyballs, diamond, nanotubes, and graphite. It is naturally abundant as coal and natural graphite. Two-dimensional (2D) graphene, a new class of carbon nanostructure, has attracted tremendous attention in recent years since the successful isolation of graphene by micromechanical cleavage of highly oriented pyrolytic graphite (HOPG) [1, 2]. Graphene is a single atomic layer of sp2 hybridized carbon atoms covalently bonded in a honeycomb lattice. It is a building block for carbon materials of different dimensionalities, including 0D buckyballs, 1D nanotubes, and 3D graphite (Figure 1.1). It shows great potential for technological applications in several areas such as electronics, optoelectronics, nanocomposites, sensors, batteries, and so on [3–7]. Graphene sheets stack together to form graphite with an interlayer spacing of 0.34 nm, showing strong in-plane bonding but weak van der Waals interaction between layers. By virtue of this layered structure, large efforts have been tempted to exfoliate graphite into individual atomic layers. It is difficult to obtain a fully separated sheet layer of graphene because freestanding atomic layer is widely considered to be thermodynamically unstable. A lack of an effective approach to exfoliate graphite into individual, pure graphene sheet in large quantities remains a major obstacle to exploiting its full potential applications.

Figure 1.1 Graphene is a 2D building material for carbon materials of different dimensionalities. It can be wrapped up into 0D buckyballs, rolled into 1D nanotubes, or stacked into 3D graphite.

(Source: Reproduced with permission from Ref. [3], Nature Publishing Group (2007).)

In 2004, Geim and coworkers of the Manchester University (United Kingdom) prepared single layer of graphene using the cohesive tape method through repeated peeling of graphite and deposited onto a Si/SiO2 substrate [1, 2]. This is often referred to as a scotch tape or drawing method. Optical microscopy was initially used to distinguish individual graphene layers followed by their identification in an atomic force microscope (AFM). Geim and Novoselov received the Nobel Prize in Physics for 2010 for their pioneering work in the fabrication and physical characterization of graphene. Such novel preparation of graphene has opened up a new era in nanotechnology and materials science and prompted much excitement in these fields. This technique can only produce low-yield, high-purity graphene for research purposes, and insufficient for practical applications. Moreover, it is hard to control the number of layers for peeled off pieces.

As an alternative, graphene can be grown directly on solid substrates using two different approaches. The first involves graphitization of single-crystal silicon carbide substrate through thermal desorption of silicon in ultrahigh vacuum at high temperatures (circa above 1300 °C). Consequently, excess carbon is left behind on the surface. The carbon-enriched surface then undergoes reorganization and graphitization to form graphene under proper control sublimation conditions. This process yields epitaxial graphene with dimensions dependant on the size of SiC substrate [8, 9]. The shortcomings of this process are the use of high processing temperature, the formation of atomic scale defects in the graphene lattice and the difficulty of achieving large graphite domains with uniform thickness. The second approach involves epitaxial growth of graphene on metal carbide (e.g., TaC, TiC) or metallic substrates (e.g., Ni, Cu) via chemical vapor deposition (CVD) of hydrocarbons at high temperatures. This is commonly followed by chemical etching and transfer printing to arbitrary substrates [10–14]. For example, Kim et al. [11] prepared patterned graphene film on thin nickel layer using a gas mixture of CH4, H2, and Ar, followed by transferring the printing film onto target substrates. The growth of graphene on nickel with higher carbon solubility (>0.1 at%) occurs by the diffusion of the carbon species into the metal surface before segregating and precipitating to the surface on fast cooling. Ni can dissolve more carbon atoms and thus it is difficult to obtain uniform graphene films due to precipitation of extra C during fast cooling. In contrast, the graphene growth on low carbon solubility Cu substrates occurs by means of surface adsorption process [13]. CVD graphene generally exhibits lower electron mobility than mechanically exfoliated graphene because of its higher concentration of point defects, smaller grain sizes, and residual impurities from the transfer or growth processes [14]. The transfer-printing process is also difficult to scale up for industrial applications. Accordingly, wet chemical processing through oxidation of graphite into graphene oxide (GO) followed by reduction appears to be a cost-effective method for mass-producing graphenelike materials.

1.1.1 Graphite Intercalation Compound

Apparently, high-yield production processes for graphene sheets are necessary for practical applications as conductive films and nanofillers for composite materials. Hence, chemical conversion from graphite offers significant advantages over physical approaches and the CVD process for preparing graphene for large-scale applications. This approach converts natural graphite into graphite intercalation compound (GIC) by reacting with electron-donor agents such as alkali metals and electron-acceptor agents such as halogens and acids [15]. Because of its layered structure, acid molecules and alkali metal can penetrate within the gallery spaces of graphite. The layers of graphite interact with the guest molecules through charge transfer process. For example, potassium can be inserted into graphite galleries to yield both first stage and higher stages of intercalation. Stage implies the number of graphite host layers divided by the number of guest layers that occur periodically in the galleries. In the case where every carbon layer in graphite is intercalated, a stage I compound forms, while intercalating on average every other layer yields a stage II compound [16a]. The first-stage intercalation compound, KC8, has a larger d-spacing (0.541 nm) compared to that of graphite. The second-stage compound, KC24, and the third-stage material, KC36, have a spacing of 0.872 and 1.2 nm, respectively (Figure 1.2). KC8 generally forms by heating graphite with potassium under vacuum at 200 °C [16b]. The KC8 compound then reacts with ethanol to yield potassium ethoxide and hydrogen gas, which aid in separating the graphitic sheets to form exfoliated graphite. The reaction takes place as follows:

1.1

Apart from potassium, the intercalate species normally employed for forming GICs include sulfuric acid, perchloric acid, and selenic acid. In the former case, sulfuric species intercalates into the gallery spaces of graphite in the presence of oxidizing nitric acid. The chemical reaction that takes place between graphite and concentrated sulfuric acid is given by Chen et al. [17a,b]

1.2

where O is the oxidant and graphite · HSO4 is the GIC.

Figure 1.2 Schematic diagram of graphite host layers intercalated with different numbers of potassium guest layers.

(Source: Reproduced with permission from [16a], Elsevier (2007).)

Expanded graphite (EG) is an industrial term for exfoliated graphite obtained from sulfuric acid-based GIC precursor [18]. Rapid heating or microwave irradiation causes a large expansion of the graphite flakes along their c-axis to produce EGs of ∼50–400 nm thickness. In general, microwave heating is more effective to exfoliate GIC than conventional thermal treatment because of its high-energy density and fast heating process. Microwave heating vaporizes the acids within the layers of graphite, producing a significant and rapid expansion of the graphite gallery [19]. Figure 1.3a,b shows the low- and high-magnification scanning electron micrographs showing typical porous, vermicular, or wormlike morphology of EGs. The EGs can be further exfoliated to graphite nanoplatelets (GNPs) of 1–15 µm in diameter and a thickness of <10 nm under sonication in the solvents [17b, 20]. The morphology of individual GNP particle is shown in Figure 1.4a,b. The term nanoplatelet describes the formation of several layers of graphene rather than a single graphene layer.

Figure 1.3 (a) SEM image of EGs showing wormlike morphology. (b) High-magnification SEM image showing pores intersperse with EG platelets.

(Source: Reproduced with permission from Ref. [18], Elsevier (2005).)

Figure 1.4 TEM images of (a) surface and (b) cross-sectional view of an individual graphite nanoplatelet particle.

(Source: Reproduced with permission from Ref. [20], Elsevier (2011).)

1.1.2 Graphene Oxide

Graphite oxide can be obtained by reacting graphite with strong oxidizers such as sulfuric acid, nitric acid, potassium chlorate, and potassium permanganate. The typical Staudenmaier [21] oxidation method involves a mixture of sulfuric acid, nitric acid and potassium chlorate. The Hummers process involves chemical oxidation of graphite with KMnO4 and NaNO3 in concentrated H2SO4 [22]. Till present, this is the most commonly used process. Graphite oxide is decorated with hydrophilic oxygenated graphene sheets bearing oxygen functional groups on their basal planes and edges [23, 24]. In other words, functional groups such as epoxide, hydroxyl, carbonyl, and carboxyl groups are formed in the basal planes (Figure 1.5). Thus graphite oxide exhibits an increased interlayer spacing from original 3.4 Å of graphite to 6.0–10 Å nm depending on the water content [25]. Such functional groups make graphite oxide hydrophilic and weaken the van der Waals forces between layers. Thus graphite oxides can be dispersed in aqueous media readily to form colloidal suspensions [26]. This facilitates exfoliation of graphite oxide into GO sheets via sonication [27]. Figure 1.6 is an AFM image of GO exfoliated in water via sonication showing the presence of sheets with uniform thickness of ∼1 nm. This thickness is somewhat larger than the theoretical value of 0.34 nm found in graphite. This is attributed to the presence of covalently bound oxygen in the GO. It appears that large-scale production of graphene sheets can be achieved through chemical oxidation and exfoliation of graphite flakes in the liquid phase, and the subsequent deoxygenation reduction owing to its simplicity, reliability, and low material cost.

Figure 1.5 Chemical structure of graphite oxide. Carboxylic groups at the edges are not shown.

(Source: Reproduced with permission from [24a], Elsevier (1998).)

Figure 1.6 (a) AFM image of exfoliated graphene oxide sheets. The sheets are ∼1 nm thick. The horizontal lines in the image indicate the sections (in order from top to bottom) corresponding to the height profiles shown (b).

(Source: Reproduced with permission from Ref. [27], Elsevier (2007).)

GO is electrically insulating because the functional groups distort intrinsic network of the sp2 carbon–carbon bonds in the graphene sheets. To recover electrical conductivity, chemical reducing agents such as hydrazine and its derivatives have been used to eliminate oxygen functionalities. For example, Ruoff and coworkers added hydrazine hydrate directly to aqueous dispersions of GO to remove epoxide complexes, producing reduced graphene oxide (rGO), and often referred to as chemically modified graphene (CMG) [25, 28]. A possible reaction pathway for epoxide reduction is given by Stankovich et al. [27]

1.3

It is noted that hydrazine (N2H4) is highly toxic and the treatment causes the formation of unsaturated and conjugated carbon atoms, which in turn degrades electrical conductivity. Residual carbonyl and carboxyl groups still can be found in the C1s X-ray photoelectron spectroscopy (XPS) spectrum because of incomplete chemical reduction by hydrazine (Figure 1.7). Further, C–N groups are also incorporated during chemical reduction. The residual oxygen forms sp3 bonds with carbon atoms in the basal plane such that the carbon sp2 bonding fraction in fully reduced GO is ∼0.8 [29]. Very recently, Shin et al. [30] reported that sodium borohydride (NaBH4) is more effective to remove oxygen moieties in GO than hydrazine. Nevertheless, rGO shows promise for technological applications since it can be processed in liquid phase in large quantities, thus facilitating the fabrication of thin films and composites using low-cost solution processing techniques [28, 31]. Figure 1.8 outlines the process scheme for fabricating rGO-based films for polymer composite and graphene-related electronics applications.

Figure 1.7 C1s spectra of (a) GO and (b) reduced GO.

(Source: Reproduced with permission from Ref. [27], Elsevier (2007).)

Figure 1.8 Process scheme for fabricating rGO-based thin films.

(Source: Reproduced with permission from Ref. [28], Wiley-VCH (2010).)

In addition to chemical reduction, large quantities of graphene can be obtained by reducing GO thermally. This involves rapidly heating GO in an inert atmosphere to form thermally reduced graphene oxide (TRG) or thermally expanded graphene oxide (TEGO). Aksay and coworkers [32a,b] reduced GO by rapid heating ( > 2000 °C min−1) to 1050 °C, resulting in the evolution of carbon dioxide due to the decomposition of hydroxyl and epoxide groups. The evolved gas pressure then increases, forcing the sheets apart and producing exfoliation of graphene sheets. TEGO is also known as functionalized graphene sheet (FGS) having a wrinkled morphology (Figure 1.9a). Some functional groups are still retained despite high-temperature annealing. High-resolution transmission electron microscopy (HRTEM) image of TEGO reveals the presence of approximately three to four individual graphene layers within the platelet [33] (Figure 1.9b). Erickson et al. [34] investigated the local chemical structures rGO and TEGO films using a transmission electron microscopy (TEM) corrected with monochromatic aberration. GO was produced using a modified Hummers method and drop cast into TEM carbon grids. GO-containing grids were reduced in a hydrazine atmosphere and then slowly heated to 550 °C under flowing nitrogen to form TEGO. The TEM image of GO obviously shows the presence of the oxidized areas (A and B) and graphene region (C) (Figure 1.10). The TEM image of TEGO reveals a high amount of restored graphene area (B) with little oxidized area (A) associated with oxygenated functional groups (Figure 1.11).

Figure 1.9 (a) SEM image of dry FGS powder. (Source: Reproduced with permission from Ref. [32b], the American Chemical Society (2007).) (b) HRTEM image of TEGO. The inset is selected-electron diffraction pattern of TEGO.

(Source: Reproduced with permission from Ref. [33], the American Chemical Society (2009).)

Figure 1.10 Aberration-corrected TEM image of a single sheet of suspended GO. The scale bar is 2 nm. Expansion (A) shows, from left to right, a 1 nm2 enlarged oxidized region of the material, then a proposed possible atomic structure of this region with carbon atoms in gray and oxygen atoms in dark gray, and finally the average of a simulated TEM image of the proposed structure and a simulated TEM image of another structure where the position of oxidative functionalities has been changed. Expansion (B) focuses on the white spot on the graphitic region. This spot moved along the graphitic region but stayed stationary for three frames (6 s) at a hydroxyl position (left portion of expansion (B)) and for seven frames (14 s) at a (1,2) epoxy position (right portion of expansion (B)). The ball-and-stick figures below the microscopy images represent the proposed atomic structure for such functionalities. Expansion (C) shows a 1 nm2 graphitic portion from the exit plane wave reconstruction of a focal series of GO and the atomic structure of this region.

(Source: Reproduced with permission from Ref. [34], Wiley-VCH (2010).)

Figure 1.11 Aberration-corrected TEM image of a monolayer of TEGO. The scale bar is 1 nm. Expansion (A) shows, from left to right, an enlarged region of the micrograph, then a proposed possible structure for the region and finally a simulated TEM image for this proposed structure. Expansion (B) shows the structure of a graphitic region.

(Source: Reproduced with permission from Ref. [34], Wiley-VCH (2010).)

1.2 Carbon Nanotubes

Figure 1.12 Structural variety of CNTs. (a) Orientation of the carbon network in armchair (n,n) and zigzag (n,0) CNTs. (b) Single-, double-, and multiwalled CNTs.

(Source: Reproduced with permission from Ref. [36], the American Chemical Society (2011).)

1.2.1 Synthesis of Carbon Nanotubes

1.2.1.1 Physical Vapor Deposition

CNTs have been synthesized by a variety of physical vapor deposition (PVD) and CVD processes. Both processes have their advantages and disadvantages for synthesizing nanotubes. PVD can be classified into direct current (DC) arc discharge and the laser ablation. Those involve condensation of hot carbon vapor generated by evaporating solid graphite. In the former method, an arc is formed between two high-purity graphite electrodes under a protective atmosphere of inert gases. The carbon vapor then condenses on the cold cathode forming a cigarlike deposit with a hard outer shell and a softer inner core [37]. CNTs are then deposited in the weblike soot attached on the chamber walls or electrodes [38]. This technique generally favors the formation of CNTs with a higher degree of crystallinity and structural integrity because of the high temperature of arc plasma. However, by-products such as amorphous carbon and other carbonaceous species are also generated. The quality and yield of nanotubes depend on the processing conditions such as efficient cooling of cathode, the gap between electrodes, reaction chamber pressure, uniformity of the plasma arc, plasma temperature, and so on. [39]. Comparedto laser ablation, electric arc discharge technique is cheaper and easier to implement but has lower output yield.

SWNTs can also be synthesized by the arc discharge evaporation of a carbon electrode with the aid of transition metal catalyst in hydrogen-containing environments such as H2–inert gas (Ne, Ar, Kr, and Xe) or H2–N2 [40]. The as-prepared SWNT soot generally contains a large amount of impurities including transition metal catalysts, amorphous carbon, and carbonaceous particle impurities, rendering purification of the arc-SWNTs a big challenge [41] (Figure 1.13). Ando and coworkers [42, 43] demonstrated that the arc discharge of pure graphite in pure hydrogen results in the formation of MWNTs of high crystallinity in the cathode deposit.

Figure 1.13 TEM images of (a) arc-grown SWNTs in H2 and (b) purified SWNTs.

(Source: Reproduced with permission from Ref. [41], Elsevier (2008).)

Laser ablation refers to removal of substantial amount of material from the target by an intense laser pulse. In the process, a graphite target placed inside a tube furnace is irradiated with a focused laser beam. A stream of inert gas is admitted into the specimen chamber for carrying vaporized species downstream to a cold finger [44] (Figure 1.14). The Nd:YAG and CO2 laser sources operated in a continuous wave mode are typically used for generating carbon vapor species. The quality and yield of CNTs can be manipulated by several experimental parameters including the composition of the target, gas atmosphere, pressure and its flow rate, as well as the laser energy, peak power, and repetition rate [38, 45–48]. SWNTs synthesized from this process often assemble into bundles because of the van der Waals attractions between them [49]. Figure 1.15a,b shows the respective TEM images of SWNTs produced by vaporizing a graphite target containing Pt, Rh, and Re catalysts at 1450 °C in nitrogen or helium gas atmospheres with a laser source [48]. SWNT bundles and amorphous carbon can be readily seen in these micrographs. However, the amount of amorphous carbon is relatively higher in the SWNTs synthesized in helium.

Figure 1.14 Laser ablation facility for synthesizing CNTs.

(Source: Reproduced with permission from Ref. [44], Elsevier (2004).)

Figure 1.15 TEM images of SWNTs synthesized from a graphite target containing Pt, Rh, and Re catalysts by a Nd:YAG laser irradiation at 1450 °C in (a) nitrogen and (b) helium.

(Source: Reproduced with permission from Ref. [48], the American Chemical Society (2007).)

1.2.1.2 Chemical Vapor Deposition

CVD method is a versatile and effective technique for massive synthesis of CNTs at low production cost using a wide variety of hydrocarbon gases such as methane, ethylene, propylene, acetylene, and so on. It allows the manufacture of CNTs into various forms including thin films, aligned, or entangled tubes as well as free-standing nanotubes. The process involves catalytic dissociation of hydrocarbon gases over metal nanoparticle catalysts in a high-temperature reactor. Comparing with the arc-grown CNTs, CVD-synthesized nanotubes generally have higher density of structural defects. The types of CNTs produced depend mainly on the synthesis temperatures. MWNTs are generally formed at ∼600–850 °C, while SWNTs are produced at higher temperatures of 900–1200 °C [44]. CVD can be classified into two categories, that is, thermal- and plasma-enhanced processes depending on the heating sources employed. Thermal CVD decomposes hydrocarbon gases using thermal energy. The hydrocarbon precursor is usually diluted with H2, normally acting as carrier gases [50, 51].

The catalytic decomposition of hydrocarbons can be further enhanced by using plasmas generated from DC, hot filament aided with DC, microwave, radio frequency (rf), electron cyclotron resonance (ECR), and inductively coupled plasma sources. These sources ionize the gas precursors, producing plasmas, electrons, ions, and excited radical species. The setup of DC plasma reactor is relatively simple and consists of a couple of electrodes with one grounded and the other connecting to a power supply. The negative DC bias applied to the cathode causes the decomposition of hydrocarbon gases. To enhance deposition efficiency, a metallic wire (e.g., tungsten) is added to the system as resistively heated hot filament [52]. This is known as the DC-plasma-enhanced hot filament CVD [53], typically used for depositing diamond and diamondlike films. However, the DC plasma is less effective in producing reactive species since the majority of energy consumed in DC plasmas is lost in accelerating ions in the sheath, and this leads to a substantial substrate heating [52]. The plasma instability of DC reactors has led to the adoption of high-frequency plasma reactors in the semiconductor industry. In high-frequency plasma reactors, the gas molecules are activated by electron impact. For example, a microwave source operated at a frequency at 2.45 GHz at 1.5–2 kW oscillates electrons effectively, leading to an increase in their density (Figure 1.16a). These electrons then collide with the feed gas to form radicals and ions. In general, the ECR source is capable of producing higher fluxes of lowenergy ions than other sources (Figure 1.16b). It is well known that electrons travel in a circular path with a cyclotron frequency under the influence of a magnetic field. The cyclotron frequency is proportional to the strength of magnetic field. Electrons that move in a circular path in a magnetic field can absorb energy from an AC electric field provided that the frequency of the field matches the cyclotron frequency. In the ECR reactor, microwaves are introduced into a volume of the reactor at a frequency matches closely with ECR. The absorbed energy then increases the velocity of electrons, leading to enhanced ionization of the feed gas. Thus ECR source has the advantages of obtaining high dissociation levels of the precursor gas and high uniformity of plasma energy distribution.

Figure 1.16 Schematic diagrams of (a) microwave PECVD and (b) ECR PECVD reactors.

(Source: Reproduced with permission from Refs. [54] and [55], Elsevier (2006) and (2007), respectively.)

Plasma-enhanced chemical vapor deposition (PECVD) allows nanotubes to be synthesized at lower temperatures [54–56]. It also offers another advantage by forming vertically aligned nanotubes over a large area with superior uniformity in diameter and length by means of the electric field present in the plasma sheath [57–60]. PECVD process is more complicated than thermal CVD since several plasma parameters such as ion density, ion energy, radical species, radical densities, and applied substrate bias are involved [61]. Since the dissociation of hydrocarbon feedstock creates many reactive radicals, leading to the deposition of amorphous carbon on the substrate. It is necessary to dilute the hydrocarbon with hydrogen, argon, or ammonia. The chemistry of gas precursors is considered of primarily importance for growing aligned nanotubes. The hydrogen gas generally assists the growth of aligned nanotubes by etching the substrate. Atomic hydrogen generated within the plasma can remove amorphous carbon deposit on the substrate [62]. Figure 1.17a,b shows the respective low- and high-magnification scanning electron micrographs of aligned CNT film grown on a cobalt-catalyzed Si substrate using microwave PECVD [60]. However, Wong et al. [63] demonstrated that ammonia gas is even more effective to form aligned CNTs from microwave PECVD process. This is because ammonia can inhibit the formation of amorphous carbon during the initial synthesis stage [64].

Figure 1.17 (a) Low- and (b) high-magnification scanning electron micrographs of aligned CNT film grown on a Co-catalyzed Si substrate using mixed CH4-H2 gases.

(Source: Reproduced with permission from Ref. [60], Elsevier (2007).)

The growth of CNTs during CVD process involves two main step sequences, that is, dissolution of the gas precursors into carbon atoms on the metal catalyst surface and surface diffusion of carbon atoms to the growth site to form the nanotube. At an earlier stage of the process, catalytic dissociation of hydrocarbon molecules occurs near metal nanoparticles. Carbon atoms then absorb and diffuse on the metal surfaces, forming liquid alloy droplets. The droplets act as preferential sites for further adsorption of the carbon atoms, forming metal carbide clusters [65]. When the clusters reach supersaturation, nucleation, and preferential growth of 1D carbon nanostructure takes place [66]. This growth behavior is commonly known as the vapor-liquid-solid (VLS) mechanism [67]. Further, the growth can initiate either below or above the metal catalyst, regarding as the “base” or “tip” growth models (Figure 1.18). The former growth mode occurs assuming the presence of a strong metal catalyst-substrate interaction. Thus the nanotube grows up in 1D manner with the catalyst particle pinned on the substrate surface. In the case of a weak catalyst-substrate interaction, the catalyst particle is lifted up by the growing nanotube and encapsulated at the nanotube tip eventually. There is a speculation that MWNTs follow the “top” growth mode, while SWNTs adopt the “base” growth fashion [61]. Figure 1.19 is a TEM image showing the presence of nickel-aluminum composite catalysts at the tips of CVD MWNTs [68].

Figure 1.18 (a,b) “Base” and “tip” growth models for CNTs.

(Source: Reproduced with permission from Ref. [44], Elsevier (2004).)

Figure 1.19 TEM image of CVD MWNTs synthesized using Ni-Al catalyst at 550 °C.

(Source: Reproduced with permission from Ref. [68], Elsevier (2008).)

Park and Lee [69] synthesized CNTs using thermal CVD on a Fe-catalyzed silica substrate using C2H2 gas from 550 to 950 °C. They reported that CNTs grown at 950, 850, 750, and 600 °C exhibit a bamboolike morphology. There are no encapsulated Fe particles at the closed tips (Figure 1.20). The CNTs grown at 550 °C possess encapsulated Fe particles at the closed tips. Accordingly, they proposed propose a base growth model for the bambootype CNTs (Figure 1.21a–e). Carbons produced from the decomposition of C2H2 adsorb on the catalytic metal particle. They then diffuse on the metal particle surface to form the graphitic sheet cap (Figure 1.21a). As the cap grows up from the catalytic particle, a closed tip with a hollow tube is produced (Figure 1.18b). The compartment graphite sheets are then formed on the inner surface of the catalyst due to carbon accumulation as a result of bulk diffusion (Figure 1.21c). The growth process continues progressively, forming a bambootype CNT (Figure 1.21d,e). Similarly, Chen et al. also reported that the CNTs deposited on Ni-catalyzed Si substrate via hot-filament CVD also follow the base growth mode. Figure 1.22a clearly shows the presence of encapsulated Ni particles at the nanotube end-substrate interface. The nanotubes also exhibit a bamboolike morphology. A high-resolution transmission electron micrograph reveals that the Ni particle exhibits a conical morphology (Figure 1.22b). The bamboolike feature of the nanotubes is also apparent.

Figure 1.20 TEM image of CNTs grown at 850 °C. CNTs exhibit a bamboolike structure. The arrow 1 indicates the closed tip with no encapsulated Fe particle. The arrow 2 corresponds to the compartment layers whose curvature is directed to the tip.

(Source: Reproduced with permission from Ref. [69], the American Chemical Society (2001).)

Figure 1.21 Schematic diagrams of a base growth model.

(Source: Reproduced with permission from Ref. [69], the American Chemical Society (2001).)

Figure 1.22 (a) SEM and (b) HRTEM images of CNTs grown on the Ni film/silicon substrate in a C2H4/NH3 atmosphere using hot-filament CVD process. Conical-shape nickel nanoparticles are marked with arrows in (a).

(Source: Reproduced with permission from Ref. [70], Elsevier (2004).)

Apart from the hydrocarbon feedstock, carbon monoxide is an alternative carbon source gas for synthesizing CNTs. The research group at Rice University developed and commercialized high-pressure carbon oxide disproportionation (HiPCo) process for synthesizing SWNTs using carbon monoxide and iron pentacarbonyl (Fe(CO)5) catalyst [71]. This process enables the production of a relatively high yield of SWNTs. The nanotubes are synthesized in a flowing CO reactor at temperatures of 800–1200 °C under high pressures of 1–10 atm. On heating, the Fe(CO)5 decomposes into Fe atoms that condense into larger nanoparticles. SWNTs then nucleate and grow on these particles in the gas phase via CO disproportionation (decomposition into C and CO2) reaction

1.4

1.2.2 Purification of Carbon Nanotubes

The as-synthesized CNTs generally contain a large amount of impurities including metal catalyst particles, amorphous carbon, fullerenes, and multishell graphitic carbon. For example, arc discharge nanotubes possess a considerable content of amorphous carbon, fullerenes, and carbon enclosing metal catalyst particles. The SWNTs synthesized from the CVD process contain metallic catalyst and amorphous carbon. These impurities affect electrical, mechanical, and biological properties of the CNTs markedly. Thus they must be removed before their practical applications in biomedical engineering and industrial sectors. Because of the diversity of the nanotube synthesis techniques, the as-prepared CNTs may have different morphologies, structures, and impurity levels. Thus the purification techniques must be properly tailored and selected to obtain CNTs with desired purity. It is a challenging task to effectively purify CNTs without damaging their structures. The purification routes can be classified into chemical- and physical-based techniques. The former includes gas- and liquid-phase oxidation, while the latter includes filtration, centrifugation, high-temperature annealing, and chromatography. The most widely adopted purification process is the liquid-phase oxidation because of their simplicity and capability for the removal of metal impurities to a certain concentration level. However, such process often results in structural damages to the nanotubes because of their vulnerability to chemical oxidation.

1.2.2.1 Chemical Techniques

Gas-phase oxidation is the simplest technique for purifying CNTs by removing carbonaceous impurities. It is ineffective to remove metal catalyst particles [41]. Accordingly, this method is suitable for purifying arc-grown MWNTs containing no metal catalysts. CNTs can be oxidized in air, pure oxygen, or chlorine atmosphere at a temperature range of 300–600 °C. The oxidants breach the carbon shell and then oxidize metal catalysts into metal oxides. As the oxidation proceeds, the volume of nanotube increases, and metal oxides crack open the carbon shell surroundings accordingly [72]. Carbonaceous impurities are oxidized at a faster rate than the nanotube material via selective oxidation [73]. The main disadvantage of thermal oxidation is that the process can burn off more than 95% of the nanotube material. This leads to an extremely low purification yield [74]. Alternatively, microwave heating has been found to be an effective method to purify arc-grown SWNTs because of its short processing time. The microwave induces rapid local heating of the catalyst particles, causing both the oxidation and rupture of the carbon layer surrounding metal catalyst particles [74].

In general, thermal oxidation is ineffective for the purification of SWNTs having large tube curvatures and metallic impurities. Additional step procedure is necessary for removing metal catalysts. Smalley and coworkers [75] developed the oxidation and deactivation of metal oxides for purifying raw HiPCo SWNTs. The metal oxides formed in the SWNTs by oxygen oxidation were deactivated into metal fluorides through reacting with C2H2F4, SF4, and other fluorine-containing gases [75]. As a result, the iron content was significantly decreased from ∼30 to ∼1 wt% with ∼70% SWNT yield. This purification process is mainly designed for the HiPCo SWNTs with predominant Fe particles. The toxicity of the reagents is another issue that must be considered from using this process.

Liquid-phase oxidation involves the use of oxidizing agents such as concentrated HNO3, mixed HNO3/H2SO4 (1 : 3 by volume), HCl, KMnO4, HClO4, and H2O2 for purifying the nanotubes, followed by the filtering and drying procedures. Strong oxidants such as KMnO4 and HClO4 are mainly used for the purification of MWNTs with higher resistance to oxidation. Nitric acid is found to be fairly effective in removing metal catalysts and amorphous carbon in the arc-grown SWNTs [76]. In general, a mixed HNO3/H2SO4 (1 : 3 by volume) solution is more effective than concentrated nitric acid in removing impurities [77]. Thus this solution is widely used for the liquid-phase oxidation of CNTs today. As nitric acid is a mild reducing agent, a prolonged oxidation time of up to 50 h is needed to eliminate metallic impurities of the arc-grown SWNTs to a level below 0.2% [76] (Table 1.1). This produces significant wall damages, length reduction, and losses of the nanotube materials. In this regard, a two-step (e.g., gas-phase thermal oxidation followed by dipping in acid) process is adopted for further eliminating metallic component of the arc-grown SWNTs. For a combined thermal and acid oxidation treatment, the acid can easily dissolve the metal oxides formed from the gaseous oxidation. Table 1.1 compares metal content of the arc-grown SWNTs containing Ni and Y catalysts before and after purifying by nitric acid, air oxidation/HNO3 reflux, and microwave heating/HCl reflux treatments. Apparently, an initial oxidation of the SWNTs in air at 400 °C for 30 min can reduce the refluxing time in a nitric acid to 6 h to yield a residual metal content <1%. In contrast, microwave heating in air at 500 °C for 20 min followed by refluxing in HCl for 6 h can further remove residual metal catalyst content to a level below 0.2%. Hou et al. [41] used a reverse sequence strategy, that is, an initial sonication in ethanol followed by air oxidation to purify hydrogen arc-grown SWNTs. Sonication is recognized as one of the effective ways to eliminate amorphous impurities in CNTs using suitable solvents and acid [78, 79]. Using such a two-step procedure, a purity of about 96% with a 41% SWNT yield can be achieved (Figure 1.23b). However, sonication can also induce structural defects in CNTs including buckling and bending [80].

Table 1.1 Residual Metal Contents in Arc-Grown SWNTs Containing Ni and Y Impurities Purified by Different Approaches

Figure 1.23 TEM images and selected-electron diffraction patterns of (a) CVD MWNT and (b) MWNT annealed at 2250 °C.

(Source: Reproduced with permission from Ref. [81], Elsevier (2001).)

It is noteworthy that the acid treatment induces structural defects and shortens the lengths of both the MWNTs and SWNTs (Table 1.1). The treatment also disrupts and opens the ends of CNTs, thus introducing oxygenated functional groups (hydroxyl and carboxyl) on the nanotube surfaces. These functional groups degrade electrical conductivity of the nanotubes markedly. However, these are beneficial in improving the dispersion of nanotubes in the polymer matrix.

1.2.2.2 Physical Techniques

Physical-based purification is an alternative route to reduce the damage of CNTs caused by chemical oxidation. It can retain the intrinsic structure of CNTs that is highly desirable for scientific research purposes and technological applications, particularly for the device and sensor applications. Physical techniques are mainly based on the dispersion of nanotubes in a stable colloidal suspension by the size separation through filtration, centrifugation, or chromatography. CNTs generally exhibit poor dispersibility in polar media such as water and organic solvents. SWNTs often agglomerate into bundles and ropes [49]; thus surfactants are widely used to stabilize the nanotube suspensions in water. Insoluble CNTs of large aspect ratios are extracted from the suspensions through filtration. Bonard et al. [82] dispersed pristine MWNTs in water with sodium dodecyl sulfate (SDS), forming a stable colloidal suspension. The suspension was filtered through a funnel large enough to allow the insertion of an ultrasonic probe. In order to increase the separation yield, successive filtrations were carried out until the attainment of desired purity. Smalley and coworkers employed both the filtration and the microfiltration to purify SWNTs produced by pulsed laser ablation [83]. The process separates carbon nanospheres, metal nanoparticles, polyaromatic carbons and fullerenes from the SWNTs. Purity of SWNTs in excess of 90 wt% can be achieved. The disadvantage of this technique is the use of a number of successive filtration steps to achieve desired purity.

Centrifugation is based on the gravity effects to separate carbon nanoparticles, amorphous carbon from the nanotube suspension. Thus it involves the use of centrifugal force for sedimentation of the suspensions. Amorphous carbon can be removed from the nitric-acid-treated SWNTs by low-speed centrifugation (2000g) in an acid solution of pH 2, leaving the SWNTs in the sediment in an acid [84]. Carbon nanoparticles in the SWNTs can be eliminated by high-speed centrifugation (20 000g) of several cycles in a neutral pH. This leads to the sedimentation of carbon nanoparticles, leaving the SWNTs suspended in aqueous media [85]. In another study, centrifugation in the SDS aqueous solutions is found to be effective for the removal of carbonaceous impurities in the SWNTs [86]. In general, filtration and centrifugation can only remove carbonaceous impurities and ineffective for the elimination of metallic impurities.

High-temperature annealing is the only one physical technique that does not involve the use of colloidal solutions. This process can effectively remove residual metal impurities in the CNTs when compared with other purification techniques such as acid treatment, filtration, and centrifugation. As recognized, a trace amount of metallic impurities in the CNTs can induce inflammation and cell apoptosis in mammals. These impurities must be completely removed for biomedical engineering applications. The annealing process involves heat treatment of nanotubes at high temperatures ( > 1400 °C) under inert atmosphere or high vacuum [81, 87–90]. Andrews et al. [81] annealed MWNTs at temperatures between 1600 and 3000 °C. They reported that heat treatment of MWNTs at temperatures above 1800 °C is very effective to remove residual metals (Figure 1.23a,b). Furthermore, high-temperature annealing can induce graphitization, resulting in the appearance of high-order diffraction spots such as (004) and (006) reflections in the selected-area electron diffraction (SAED) pattern. High-temperature annealing converts disordered structure of the walls of MWNTs into a more perfect graphitic structure (Figure 1.24a–f). In general, high-temperature heating can induce structural changes in the CNTs [88]. A twofold increase in the diameter of SWNTs is observed by heating at 1500 °C under argon and hydrogen atmospheres [89]. Similarly, Yudasaka et al. [90] reported that the diameters of HiPCo SWNTs can be increased via heat treatments at 1000–2000 °C.

Figure 1.24 TEM images of the walls of (a) noncatalytic CVD-grown MWNTs in a porous alumina template and those annealed at (b) 1200 °C, (c) 1500 °C, (d) 1750 °C, (e) 1850 °C, and (f) 2000 °C. The initial, disordered carbon structure is converted into a more perfect graphitic structure with increasing annealing temperature. Scale bar is 5 nm.

(Source: Reproduced with permission from Ref. [87], the American Chemical Society (2006).)

1.2.3 Characterization of Purified Carbon Nanotubes

The final quality and purity of purified CNTs can be examined by several analytical techniques including scanning electron microscopy (SEM), TEM, energy dispersive spectroscopy (EDS), Raman spectroscopy, AFM, X-ray diffraction (XRD), ultraviolet-visible-near infrared (UV-vis-NIR), absorption spectroscopy, NIR spectroscopy, and thermogravimetric analysis (TGA). These have been recommended by the International Team in Nanosafety (TITNT) and NASA-Johnson Space Center (United States) as the characterization techniques for both the pristine and purified SWNTs [91, 92]. In general, SAED and TEM offer useful information relating the chemical identity of impurities of both the pristine and purified nanotubes. However, these techniques are unsuitable for quantitative evaluation of the purity of CNTs. Further, the sample preparation procedures for TEM examination are tedious and time consuming. The EDS attached to either SEM or TEM can provide characteristic X-ray peaks of various elements, particularly useful for qualitative assessment of the metallic impurities. The EDS/SEM is routinely used in the research institutions and industrial laboratories world wide for qualitative analysis of inorganic materials. The EDS/SEM can provide quantitative analysis if suitable standards are available. However, it is difficult to analyze and quantify light elements in the EDS spectra because of the absorption of their X-ray radiation by the window material (beryllium) of the spectrometer. This effect can be minimized by using an SEM-windowless EDS facility.

At present, there appears to be a need of reliable techniques for fast and accurate assessment of the nanotube purity. In this regard, Raman spectroscopy is particularly useful for qualitative analysis of purified nanotubes by examining vibrational frequency responses of different carbon species. The position, width and relative intensity of Raman peaks of various carbonaceous species (e.g., amorphous carbon, fullerene, diamond, and SWNT) are related to their sp3 and sp2 configurations [93–95]. Raman spectra of SWNTs are well characterized for the radial breathing mode (RBM) at 150–200 cm−1 and the tangential G-band at ∼1550–1605 cm−1 [94]. The tangential-mode G-band involves out-of-phase intralayer displacement in the graphene structure of the nanotubes. It is a measure of the presence of ordered carbon. Figure 1.25 shows the Raman profile of the SWNT. A band located at ∼1350 cm−1 is attributed to the disorder-induced band (D-band) and related to the presence of nanoparticles and amorphous carbon. Furthermore, a second-order mode at ∼2600 cm−1 is referred to the D* mode, commonly known as the G′ mode. The RBM is associated with the collective in-phase radial displacement of carbon atoms, and only can be found in the SWNTs. It gives direct information for the tube diameter since its frequency is inversely proportional to the tube diameter. The D-peak is absence in graphite and only found in the presence of disorder defects. The D- and G-peak ratio characterizes the disorder degree of the materials studied. Thus the purity level of SWNTs can be qualitatively determined from the D/G intensity ratio and the width of the D-band [95, 96]. The full width at half maximum (FWHM) of the D-band for the carbonaceous impurities is much broader than that of the SWNT. An asymmetry on the right-hand side of G-band is a peak characteristic of MWNTs and normally appears at 1620 cm−1 [97].

Figure 1.25 Raman profile of SWNT showing the presence of radial breathing mode and tangential G-band.

Figure 1.26 Raman spectra of arc-grown SWNT (PS0) after each step of the purification procedure. (1) PS0, (2) PS1, (3) PS2, and (4) PS3a.

(Source: Reproduced with permission from Ref. [98], Elsevier (2010).)

Figure 1.27XRD patterns of PS0, PS1, PS2, and PS3b specimens.

(Source: Reproduced with permission from Ref. [98], Elsevier (2010).)

Figure 1.28 TEM images of (a) PS0, (b) PS1, (c) PS2, and (d) PS3a.

(Source: Reproduced with permission from Ref. [98], Elsevier (2010).)

TGA is a simple quantitative method for measuring the specimen mass in relation to changes in temperatures using a thermobalance. During the tests, the temperature is raised and the weight of the sample is recorded continuously, which permits the monitoring of weight losses with time. It can be performed in air, oxygen, or inert gas (or vacuum). TGA with air/oxygen atmosphere allows the determination of temperatures at which nanotubes and impurities oxidize. Since amorphous carbon, carbon nanoparticle, multishell carbon, fullerene and SWNT have different affinity toward oxygen, thus those impurities and CNTs oxidize at different temperatures as expected [99, 100]. Other quantitative method such as UV-vis-NIR spectroscopy has also been employed for purity evaluation of SWNTs [101, 102].

1.3 Carbon Nanofibers (CNFs)

Vapor-grown carbon nanofibers (VGCNFs) have received considerable attention in recent years because of their relatively low cost when compared to CNTs, especially SWNTs. Thus VGCNFs are an attractive alternative for CNTs for research purposes and technological applications. VGCNFs are also widely known as carbon nanofibers (CNFs). A CNF consists of stacked curved graphene layers forming cones or cups with an angle alpha with respect to the longitudinal axis of the fiber [103–107]. The stacked cone graphene sheets yield the so-called “herringbone” morphology, while the stacked-cup structure is referred to as a bamboo feature (Figure 1.29a–d). The diameters of CNFs range from 50 to 500 nm. CNFs generally have more crystalline defects, rendering them exhibit poorer mechanical properties than MWNTs. Graphitization of MWNTs by high-temperature heat treatment can improve their electrical conductivity [87]. Analogously, graphitization can improve the perfection of grapheme planes in the walls of the VGCNFs. Tibbetts and coworkers [108] reported that heat treatment of CNFs at temperatures above 1500 °C results in significant rearrangement of the core morphology. Heat treatment at temperatures of 1800–3000 °C reduces the structural disorder and increases the graphitic content of the fiber [109, 110]. In other words, increasing graphitization temperature leads to a reduction of the turbostratic layer and an increase of the order in the graphene planes [106].

Figure 1.29 (a) STEM (scanning transmission electron microscopy) micrograph of a herringbone nanofiber and (b) TEM image of a bambootype CNF. (c) Schematic illustration of stacked cone (herringbone) nanofiber. (Source: Reproduced with permission from Ref. [103], American Institute of Physics (2005).) (d) TEM image of the VGCNF wall with canted graphene planes comprising of stacked-cup morphology.

(Source: Reproduced with permission from Ref. [105], Elsevier (2007).)

CNFs were produced commercially by the Applied Sciences, Inc. using high-temperature decomposition of natural gas in the presence of the Fe(CO)5 catalyst (Figure 1.30). When the catalyst particles are properly dispersed and activated with sulfur, VGCNFs are abundantly produced in a reactor at 1100 °C [105]. This manufacturer has produced Pyrograf® III nanofibers of two types (PR-19 and PR-24) in four different grades, that is, as-grown (AS), pyrolytically stripped (PS), LHT and HHT [111]. Pyrograf®-III is available in diameters ranging from 70 to 200 nm and a length estimated to be 50–100 µm. The diameters of PR-19 and PR-24 fibers are about 150 and 100 nm, respectively. The AS fibers have polyaromatic hydrocarbons on their surfaces. The PS-grade fiber is pyrolytically stripped at 600 °C to remove polyaromatic hydrocarbons from the fiber surface. The LHT-grade nanofiber is treated at 1500 °C in order to carbonize and chemically vaporize deposited carbon present on the fiber surface, and the HHT-grade nanofiber is treated to temperatures up to 3000 °C, leading to graphitization of the fiber. On the other hand, vertically aligned CNFs designed for electronic device applications can be synthesized by means of PECVD process [103].

Figure 1.30 Apparatus for manufacturing VGCNFs.

(Source: Reproduced with permission from Ref. [105], Elsevier (2007).)

1.4 Physical Properties of Graphene

1.4.1 Mechanical Behavior

Atomistic simulation can reveal the physical nature of graphene mechanical deformation at atomistic scale using relevant interatomic potential energy models. In this regard, molecular dynamics (MD) is typically used to simulate the physical motion of atoms and molecules numerically. The atoms are allowed to interact for a period of time, giving rise to the movement and force field among them. The trajectories of atoms in the system under consideration can be estimated from the Newtonian equation of motion. The potential energy of the system is a function of the positions of atoms and thus can be described in terms of force field functions and parameters. Several interatomic potentials have been developed for use with a select group of materials including Lennard-Jones, Morse and Tersoff-Brenner potentials [115]. Lennard-Jones and Morse potentials introduced in the 1920s are empirical isotopic pair potentials and particularly suitable for atoms with no valence electrons. The Tersoff-Brenner potential is widely used for simulations in the carbon-based materials such as diamond, graphene, and nanotubes.