Graphene Oxide E-Book

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 1007

Veröffentlichungsjahr: 2016

Beschreibung

Due to its unique properties, graphene oxide has become one of the most studied materials of the last decade and a great variety of applications have been reported in areas such as sensors, catalysis and biomedical applications.

This comprehensive volume systematically describes the fundamental aspects and applications of graphene oxide. The book is designed as an introduction to the topic, so each chapter begins with a discussion on fundamental concepts, then proceeds to review and summarize recent advances in the field. Divided into two parts, the first part covers fundamental aspects of graphene oxide and includes chapters on formation and chemical structure, characterization methods, reduction methods, rheology and optical properties of graphene oxide solutions. Part Two covers numerous graphene oxide applications including field effect transistors, transparent conductive films, sensors, energy harvesting and storage, membranes, composite materials, catalysis and biomedical applications. In each case the differences and advantages of graphene oxide over its non-oxidised counterpart are discussed. The book concludes with a chapter on the challenges of industrial-scale graphene oxide production.

Graphene Oxide: Fundamentals and Applications is a valuable reference for academic researchers, and industry scientists interested in graphene oxide, graphene and other carbon materials.

Leseprobe

Table of Contents

Cover

Title Page

About the Editors

List of Contributors

Foreword

Preface

Part I: Fundamentals

1 Graphite Oxide Story – From the Beginning Till the Graphene Hype

1.1 Introduction

1.2 Preparation of Graphite Oxide

1.3 Discovery of Essential Functional O‐Containing Groups and its Relation to the Development of Structural Models

1.4 Properties of Graphite Oxide

1.5 Epilogue

References

2 Mechanism of Formation and Chemical Structure of Graphene Oxide

2.1 Introduction

2.2 Basic Concepts of Structure

2.3 Preparation Methods

2.4 Mechanism of Formation

2.5 Transformation of Pristine Graphite Oxide Chemical Structure Upon Exposure to Water

2.6 Chemical Structure and Origin of Acidity

2.7 Density of Defects and Introduction of Oxo‐Functionalized Graphene

2.8 Addressing the Challenges of the Two‐Component Structural Model

2.9 Structure of Bulk Graphite Oxide

2.10 Concluding Remarks

References

3 Characterization Techniques

3.1 Nuclear Magnetic Resonance Spectroscopy of Graphene Oxide

3.2 Infrared Spectroscopy

3.3 X‐ray Photoelectron Spectroscopy

3.4 Raman Spectroscopy

3.5 Microscopy Methods

References

4 Rheology of Graphene Oxide Dispersions

4.1 Liquid Crystalline Behaviour of Graphene Oxide Dispersions

4.2 Rheological Behaviour of Aqueous Dispersions of LC‐GO

4.3 Comparison with Other Systems

4.4 Summary and Perspectives

References

5 Optical Properties of Graphene Oxide

5.1 Introduction

5.2 Absorption

5.3 Raman Scattering

5.4 Photoluminescence

5.5 Graphene Oxide Quantum Dots

5.6 Applications

References

6 Functionalization and Reduction of Graphene Oxide

6.1 Introduction

6.2 Diverse Structure of Graphene Oxide

6.3 Stability of Graphene Oxide

6.4 Non‐Covalent Chemistry

6.5 Covalent Chemistry

6.6 Reduction and Disproportionation of Graphene Oxide

6.7 Reactions with Reduced Form of Graphene Oxide

6.8 Controlled Chemistry with Graphene Oxide

6.9 Discussion

References

Part II: Applications

7 Field‐Effect Transistors, Sensors and Transparent Conductive Films

7.1 Field‐Effect Transistors

7.2 Sensors

7.3 RGO Transparent Conductive Films

7.4 Memristors Based on Graphene Oxide

References

8 Energy Harvesting and Storage

8.1 Solar Cells

8.2 Lithium‐Ion Batteries

8.3 Supercapacitors

8.4 Outlook and Future Development Opportunities

References

9 Graphene Oxide Membrane for Molecular Separation

9.1 Rise of Graphene‐Based Membranes: Two Approaches

9.2 GO Membrane: Structural Point of View

9.3 GO Membrane for Gas Separation

9.4 GO Membrane for Water Purification and Desalination

9.5 Other Membrane Applications

9.6 Conclusions and Future Prospects

References

10 Graphene Oxide‐Based Composite Materials

10.1 Introduction

10.2 Why Mix Graphene Oxide and Polymers?

10.3 Graphene Oxide or Graphene Oxides?

10.4 Conclusion

References

11 Toxicity Studies and Biomedical Applications of Graphene Oxide

11.1 Introduction

11.2 Toxicity of Graphene Oxide

11.3 On the Toxicity Mechanism

11.4 Biomedical Applications of Graphene Oxide

11.5 Bioanalytical Applications

Acknowledgments

References

12 Catalysis

12.1 Introduction

12.2 Graphene Oxide Properties

12.3 Oxidative Activity

12.4 Polymerization

12.5 Oxygen Reduction Reaction

12.6 Friedel–Crafts and Michael Additions

12.7 Photocatalysis

12.8 Catalytic Activity of Other Layered Carbon‐Based Materials and Hybrid Materials of GO

12.9 Outlook

References

13 Challenges of Industrial‐Scale Graphene Oxide Production

13.1 Introduction

13.2 Scope and Scale of the Graphene Market

13.3 Overview of Graphene Oxide Synthesis

13.4 Challenges of Graphene Oxide Production

13.5 Concluding Remarks and Future Directions

References

Vocabulary

Index

End User License Agreement

List of Tables

Chapter 06

Table 6.1

Effectiveness of reducing agents toward GO

Chapter 12

Table 12.1

Oxidative reactions catalyzed by GO. The efficiency is presented either in conversion [c] or yield [y], as described in the literature

Table 12.2

Reaction pathways of ORR in aqueous electrolytes

Chapter 13

Table 13.1

Synthesis method, product, production capacity and main application products of several large graphene manufacturers. Reproduced from [4] with permission from Nature Publishing Group, Copyright 2014

List of Illustrations

Chapter 01

Figure 1.1

The structural models of Thiele [14] (left) and Nakajima et al. [44, 45] (right)

.

Figure 1.2

The structural model of Hofmann et al. [15]

.

Figure 1.3

The structural GO model according to Ruess [19]. However, this figure is taken from Ref. [48] because it shows clearly the double bonds left unoxidized

.

Figure 1.4

Possible keto–enol tautomerism in graphite oxide (left) according to Clauss et al. [48], and the new structural model (right) of Scholz and Boehm [16]

.

Figure 1.5

Lerf–Klinowski

model of the as‐prepared GO structure [55]

.

Figure 1.6

(a)

13

C MAS NMR spectra of GO modified by thermal treatment and reduction with KI [20, 55]. (b) Proposed reorganization of the double bond system after liberation of oxygen [20, 55]. (c) Modified structure of GO containing only phenols and double bonds after complete removal of epoxide groups [55]

.

Figure 1.7

Defective graphite layer [46]

.

Figure 1.8

Titration curves of Hamdi [29] (left) and Szabó et al. [102] (right)

.

Chapter 02

Figure 2.1

Graphene oxide in its true 2D single‐layer form. (a) Photograph of GO aqueous solution; the solution color may vary from yellow to brown. (b) Scanning electron microscope image of GO flakes on a Si/SiO

2

wafer. The number of layers can be distinguished by their opacity. All the flakes on this image are single‐layered. The image is darker where the flakes are folded or overlapped, making double‐layered structures

Figure 2.2

The conversion of bulk graphite to GO is a black box. Researchers studying the mechanism of this conversion must reveal and fully describe all the steps and underlying mechanisms that lead to this transformation

Figure 2.3

A graphite flake in the course of its transition from stage‐1 GIC to PGO. (a) Map of the flake showing spots of acquisition; the circles labeled “b”, “c”, “d” and “e” indicate the four typical areas on the flake surface where the spectra were acquired. (b)–(e) Typical Raman spectra acquired from the corresponding spots labeled in (a). The insets represent the

x

‐axis expansion in the G‐band area. The spectra taken from the blue‐colored areas in the middle of the flake (b and e) show the presence of the stage‐1 GIC only. The spectra taken from the light‐yellow‐colored areas (c) on the flake’s edge represent GO. The spectrum acquired from the dark‐looking bordering area between the blue and yellow areas (d) shows the presence of two phases: stage‐1 GIC and PGO. A 514 nm laser was used for excitation.

Figure 2.4

X‐ray diffraction data. (a) X‐ray diffraction patterns for stage‐1 GIC (black), CGO (brown line) and four transition forms (red, blue, pink and dark cyan lines). The labels “TF‐1” to “TF‐4” represent the four consecutive transition forms obtained by consumption of 1, 2, 3 and 4 wt equiv of KMnO

4

, respectively. (b,c) The

x

‐axis expansions of (a) in the 20° to 24° and 8° to 13° 2

θ

diffraction angle regions.

Figure 2.5

Schematics of conversion of bulk graphite into GO with corresponding micrographic images of sample appearances at each phase. The three steps signify formation of the two intermediate products (stage‐1 GIC and PGO) and the final GO product. The solid black lines represent graphene layers; dotted black lines represent single layers of GO; wide blue lines represent H

2

SO

4

/HSO

4

−

intercalant; wide purple lines represent a layer of the mixture of H

2

SO

4

/HSO

4

−

intercalant with reduced form of oxidizing agent.

Figure 2.6

FTIR spectra of the four different GO samples. The samples are obtained by purification of as‐synthesized PGO samples with acetic acid (AA‐GO), methanol (M‐GO), 33% hydrochloric acid (HCl‐GO) and water (cGO). The two absorption bands at 1417 and 1221 cm

−1

are assigned to the symmetric and asymmetric stretching of the S ═ O bonds in covalent sulfates. The intensity of these two bands increases with the sulfur content in GO, and decreases with increasing aqueous/nucleophilic character of the solvent used for purification.

Scheme 2.1

The sulfate ester (

2

) formed in the first step is the intermediate product. The sulfate ester can attack the neighboring epoxide group, resulting in a 1,2‐cyclic sulfate (

3

). 1,3‐Cyclic sulfates (not shown) can also be formed by the reaction. Note that one hydroxyl group is formed for every sulfate ester, and two hydroxyl groups form for every cyclic sulfate

Scheme 2.2

The hydrolysis of cyclic sulfate (

3

) occurs in two steps. The first step occurs with the C–O bond cleavage and results in the formation of the monosulfate (

4

). The second step (hydrolysis of the monosulfate) results in the formation of a 1,2‐diol (

5

)

Figure 2.7

Simplified version of the Lerf–Klinowski GO structural model. GO flakes contain epoxides and tertiary alcohols on basal planes; carboxyl and hydroxyl groups terminate flake edges

Figure 2.8

Simplified version of the Szabó–Dékány GO structural model. GO flakes contain tertiary alcohols and 1,3‐ethers on basal planes. Ketones are formed where C–C bonds are cleaved

Figure 2.9

Aberration‐corrected transmission electron microscope images of (A) graphene, (B) GO and (C) RGO. The scale bar, denoting 2 nm, is valid for all images. On the right are the same images as on the left, where different areas are color‐coded: graphitic domains in yellow‐green, holes in blue and oxidized domains in red‐purple.

Figure 2.10

Typical direct‐pulse

13

C SSNMR spectrum for GO. The spectrum is taken from the author’s personal research files. The signals at 60, 70 and 134 ppm are assigned to epoxides, tertiary alcohols and aromatic carbons of graphitic domains, respectively. The signals at 101, 167 and 193 ppm might be assigned, respectively, to lactols/gem‐diols, carboxylic groups and ketones

Figure 2.11

Acidic properties of GO solutions. (a–c) Forward (black squares) and reverse (red circles) titration curves. In the forward titration, the GO solution (1.0 mg ml

−1

) was titrated with 0.100 M NaOH; and in the reverse titration, the conjugate base

of GO was titrated with 0.100 M HCl. (a) The reverse titration

was performed immediately after the forward titration. The reverse titration curve begins at almost the same pH. (b) Reverse titration was performed 6 h after the forward titration. The reverse titration curve begins at lower pH values due to the acidification during the 6 h time period. (c) The solution was heated at 60 °C for 15 h after the forward titration (III), and before performing the reverse titration (IV). The reverse titration begins at significantly lower pH values. (d) The change in pH of the GO solution with time after the addition of the 0.100 M NaOH solution. Zero time corresponds to the addition of NaOH. The increase in pH during the first 3 min 47 s after NaOH addition is not shown on the graph. All experiments are performed under nitrogen.

Scheme 2.3

Origin of GO acidity. The structure (

6

) is a GO fragment containing a vicinal diol

. The C–C bond to be cleaved is shown in red. Nucleophilic attack of a water molecule on the hydrogen atom of the hydroxyl group results in C–C bond cleavage and in the formation of a ketone

and an enol

at newly formed edges (

7

). The reaction generates a hydronium cation; this explains the acidification of GO aqueous solutions. The hydrogen atom of the as‐formed enol is highly acidic; this enol can further ionize, making the GO solution even more acidic

Figure 2.12

The simplified version of the GO dynamic structural model

by Dimiev

et al

. The carbon framework is shown in black. Neutral oxygen functionalities are shown in blue. Functionalities with acidic properties are shown in red and purple. The enolic group shown in red is acidic due to conjugation with the graphitic domain. The enolic group shown in purple is acidic due to the formation of a vinylogous acid in conjugation with ketone. Organic sulfates are always adjacent to at least one tertiary alcohol. The density of organic sulfates depends on the washing procedures; for thoroughly washed GO samples, there is about one organic sulfate per 90 carbon atoms

Figure 2.13

HRTEM images of GO. (A) Magnification of the oxidized domain area: left is the actual structure, right is the simulated model. (B) Magnification of the intact graphitic domain. The oxidized domain contains multiple points of C–C bond cleavage.

Figure 2.14

The complete version of the Dimiev–Tour GO structural model (DT model), featuring all the proposed functionalities. A GO fragment contains a hole situated on the border between a graphenic domain (lower right corner) and an oxidized domain (upper left corner). Different structural features are represented by different numbers and colors.

1

Ketone and enol groups are formed at the point of C–C bond cleavage.

2

By hydration, ketones can turn into

gem

‐diols, and further into hemiacetals.

3

Here a

gem

‐diol is in

ɑ

‐position to a ketone; this favors the stability of

gem

‐diols in aqueous solutions.

4–6

These are the vinylogous carboxylic acids. Conjugation in the carboxylic acid

4

is limited by two oxygen atoms. Conjugation of the vinylogous acids

5

and

6

extends to the entire graphitic domain; acids

5

and

6

are stronger acids than acid

4

.

Organic sulfates are present in GO samples prepared in sulfuric acid medium

Figure 2.15

Transformation of as‐synthesized metastable GO into a different form upon hydrothermal treatment (“annealing” in the authors’ terminology) in aqueous solution. The top two images are the Auger electron spectroscopy (AES

) oxygen mapping of as‐synthesized (left) and hydrothermally treated (right) GO films. The white spots indicate oxygen‐rich regions, and the black spots indicate oxygen‐deficient regions. The scale bar is 2

 µ

m. On the bottom is the schematic depicting the proposed phase separation process. The as‐prepared GO exhibits a uniform distribution of oxygen functionalities. This metastable GO has the potential to separate into two distinct oxidized and non‐oxidized domains through diffusion of oxygen atoms on the graphene basal plane under the influence of an external stimulus, i.e. high temperature.

Figure 2.16

(A) Synthesis of GO with an almost intact carbon framework (oxo‐functionalized graphene, oxo‐G

1

) starting from graphite in sulfuric acid with potassium permanganate as the oxidant. (B) Reflected light microscope image of natural graphite (flake size 600 µm). (C) Aqueous dispersion of oxo‐G

1

at 0.1 mg ml

−1

. (D) Scanning electron microscope image of graphene on SiO

2

/Si. Raman spectra of (E) graphite, (F) oxo‐G

1

and (G) a flake of graphene obtained from oxo‐G

1

by chemical reduction.

Figure 2.17

Electrical properties of the wet‐chemically prepared graphene derived from oxo‐G

1

. (A) Raman spectrum of a graphene flake (

Γ

2D

 = 35 cm

−1

) used for the measurements shown in (B) and (C). (B) Four‐point resistance

R

xx

at different gate voltages with visible Shubnikov–de Haas oscillations (maxima and minima indicated by arrows). The frequency of the oscillations varies with the charge carrier density. Inset: Linear dependence of the charge carrier density on the gate voltage

V

G

. The red line represents the fit

n

∝

αV

G

with

α

 = −3 × 10

10

 cm

−2

. (C) Analysis of the Landau‐level indices for the Shubnikov–de Haas oscillations for exemplary gate voltages. The Landau‐level indices for a single gate voltage depend linearly on the inverse magnetic field. Inset: Dependence of the oscillation frequency

B

F

shows a linear dependence on the external electric field also. This is an important indication of a pure 2D material (D) Raman spectrum of graphene (

I

D

/

I

G

 = 4.0). (E) Magneto‐transport measurement of the sample characterized in (D). (F) Statistical Raman analysis of individual spectra of a film of flakes of graphene.

Figure 2.18

Oxo‐functionalized graphene from graphite sulfate. (A) Synthesis of graphene and few‐layer graphene from graphite sulfate, followed by the reaction with water to yield oxo‐functionalized graphene and graphene after reduction. (B) Atomic force microscope image, ultraviolet–visible spectrum and Raman spectrum of oxo‐functionalized graphene. (C) Thermogravimetric analysis of oxo‐functionalized graphene coupled with mass spectrometry. (D) Statistical Raman analysis of a film of flakes of graphene. Inset: Histogram of

Γ

2D

. (E) Atomic force microscope image of graphene derived from G

1

‐(OH)

4%

. (F) Raman spectrum of graphene obtained from G

1

‐(OH)

4%

after reduction, with

Γ

D

,

Γ

G

and

Γ

2D

given in italic numbers.

Figure 2.19

Schematic representation of the two‐component GO structural model, as presented by the model developers. OD, which allegedly pre‐exists in aGO samples, is an oxygen‐rich complex polycyclic molecule strongly physisorbed on the surface of slightly oxidized bwGO. OD detaches from bwGO upon treatment in hot basic solutions.

Scheme 2.4

Disintegration of GO in basic conditions via C–C bond cleavage. The structure (

1

) is a GO fragment containing three vicinal diols. The C–C bonds to be cleaved are shown in red. The structure (

2

) shows the same GO fragment after the three C–C bonds have been cleaved. The three different transformations initiated by the attack of hydroxide ions on tertiary alcohols are represented by three separate sets of curly arrows shown in different colors

Figure 2.20

Schematics representing the gradual disintegration of GO in basic conditions. Structure

1

is a fragment of a regular as‐prepared GO. The blue‐colored areas represent intact graphitic domains. The continuous yellow‐colored network represents oxidized domains. White islands represent holes. Structures

2

and

3

represent progressive stages of GO disintegration. The gradual color change from yellow in

1

to orange in

3

represents the change in chemical structure of the oxidized domains. Dotted lines represent invisible cuts formed on GO as a result of the C–C bond cleavage caused by reaction. The small orange‐colored shapes in dashed red ovals represent carbon fragments removed from the respective GO platform during base treatment. The size of the removed carbon fragments varies from subnanometer through several nanometers.

Figure 2.21

Various macro‐forms of GO. (a) A wet‐spun 14 m long continuous fiber with diameter 10 µm and (b,c) SEM images of the cross‐section of the fiber. (d) A paper‐like film made by the filtration method and (e,f) SEM images of the paper edges. (g) Ultra‐light‐weight GO aerogel with a density of 2 mg cm

−3

and (h,i) SEM images of the aerogel showing the voluminous morphology. Scale bars: 3 cm (a), 1 µm (b), 500 nm (c), 1 cm (d), 3 µm (e), 400 nm (f), 2 cm (g), 30 µm (h) and 2 µm (i).

Figure 2.22

XRD patterns (

λ

 = 0.7092 Å) recorded upon compression from a GO–methanol sample (top) and a GO–ethanol sample (bottom). Indexing for the GO–ethanol phases is given on the 0.85 GPa pattern: A, ambient‐pressure phase; H, high‐pressure phase. A solid ethanol phase is indexed on the 2.59 GPa pattern by a P2

1

/c structure with

a

 = 7.543 Å,

b

 = 4.738 Å,

c

 = 7.1904 Å and

β

 = 114.48°, in agreement with literature data.

Figure 2.23

X‐ray diffraction patterns of the hexadecylamine–GO intercalation compound (C

16

/GO) with various C

16

contents, together with that of C

16

. The values in parentheses are the interlayer spacing of a minor phase. The diffraction peaks at 2

θ

 = 1.98°, 2.24°, 4.40°, 6.62°, 8.80°, 10.96° and 13.2° are due to the unreacted C

16

phase.

Chapter 03

Figure 3.1

Example of a basic cross‐polarization (CP) pulse sequence between

1

H nuclei and

13

C nuclei. First, a pulse is applied to polarize the sample in the magnetic field. Next,

1

H is excited by a pulse (π/2) along the

x

′‐axis of the rotating frame. Then, the

1

H spin is locked by another pulse along the

y

′‐axis accompanied by another pulse to

13

C nuclei along the

y

′‐axis. The CP occurs during that time, and the time is called the contact time. Next, the radio frequency applied to the

13

C nuclei is switched off to record the free induction decay (FID) of the

13

C nuclei

.

Figure 3.2

(A) CP MAS

13

C SSNMR spectrum of graphite oxide (top); proton‐decoupled

13

C SSNMR spectrum measured under MAS conditions (bottom). (B) Development of the integrated intensity of the α, β and γ lines under a variable dipolar dephasing time ranging from a few microseconds to 100 µs

.

Figure 3.3

(A) Suggested chemical scheme for the deoxygenation of GO with HI, producing iodine and water, and forming phenol‐like sp

2

carbon atoms.

1

H‐decoupled

13

C MAS SSNMR spectra of (B) graphite oxide treated with KI and (C) as‐prepared GO

.

Figure 3.4

Different thermally treated

1

H MAS NMR spectra of graphite oxide in the temperature range between 20 and 220 °C to reduce the water content

.

Figure 3.5

Direct

13

C pulsed MAS SSNMR spectrum of GO with chemical shifts of signals at 61, 70, 101, 133, 167 and 191 ppm. The minor groups have been assigned to ketones, carboxyl groups and lactols. The signal at 101 ppm may also be assigned to geminal diols (hydrates) or phenol‐like moieties. However, ketones and carboxyl groups may not be directly bound to sp

2

carbon, suggesting that they are not part of the main structure [8].

Figure 3.6

(A) The 2D

13

C double quantum/single quantum correlation SSNMR spectrum of

13

C‐labeled GO, with (B) and (C) predicted correlation spectra assuming the chemical structures shown in (D) and (E), respectively. The results suggest epoxy groups and hydroxyl groups are adjacent to double bonds.

Figure 3.7

Special correlation spectra of

13

C signals of GO allow the identification of motifs of functional groups and the determination of their regiochemistry. Thus, the major motifs are hydroxyl–epoxy–epoxy and hydroxyl–hydroxyl–epoxy, next to sp

2

–epoxy–epoxy and sp

2

–hydroxyl–hydroxyl. Consequently sp

2

–sp

2

–epoxy, sp

2

–sp

2

–hydroxyl and sp

2

–sp

2

–sp

2

motifs are found

.

Figure 3.8

A typical FTIR spectrum for a GO sample can be arbitrarily divided into three characteristic regions: (i) an intense and very broad absorption band in the 3600–2400 cm

−1

region, (ii) the two most recognizable absorption bands at 1723 and 1619 cm

−1

in the middle of the spectrum, and (iii) a bunch of overlapping signals in the fingerprint region

Figure 3.9

FTIR spectra for H

2

O‐GO and D

2

O‐GO. The absorption bands associated with water molecules downshift upon replacing H

2

O molecules by D

2

O molecules. The shift corresponds to the factor 1.373 derived from the difference in masses of the H and D isotopes

.

Figure 3.10

The C 1s XPS spectrum for graphite. The spectrum consists of one non‐symmetric peak centered at 284.8 eV and a low‐intensity peak at 291.4 eV generated by the π → π* shake‐up interactions

Figure 3.11

The C 1 s spectrum for GO. The black line is the actual experimental spectrum. The integral spectrum can be deconvoluted into three major components: at 284.8 eV (blue line), at 286.5 eV (red line) and a shoulder at 289.2 eV (pink line)

Figure 3.12

Deconvolution of the integral C 1s XPS spectra of two carbon‐based materials that includes all the theoretical components. (a) The content of oxygen‐free carbon is 79%. (b) The content of oxygen‐free carbon is 49%. The authors disentangle the C–O–C and C–OH components, as well as graphitic carbon and alkylinic carbon

.

Figure 3.13

Illustration of the Raman process in comparison to infrared excitation. In addition to Rayleigh scattering, Stokes shifts are observed in Raman spectroscopy, as well as anti‐Stoke shifts. The excitation of molecules can occur into either virtual states or excited states, which is called resonant Raman spectroscopy

Figure 3.14

(A) Raman spectra of graphene (top) and graphene with defects (bottom) with the main peaks labelled as D, G and 2D. (B) Illustrations of the excitation and emission in the Raman process responsible for the G, D and 2D peaks. (C) Electronic Brillouin zones of graphene and the first‐phonon Brillouin zone marked in red. Electronic dispersion is visualized by Dirac cones. Phonon wavevectors that connect electronic states of different valleys are shown in red. (D) The black curves represent the dispersion of in‐plane phonon modes of graphene in the frequency range (900–1700 cm

−1

) relevant for Raman scattering and the interpretation of the Raman spectra of graphene.

Figure 3.15

(A) The I

D

/I

G

data points of different graphene samples as a function of the average distance L

D

between defects. Defects generated by Ar

+

ion bombardment. The I

D

/I

G

ratio is shown as a function of L

D

. (B) Activated regions (green) with radius r

a

and structurally disordered regions (red) with radius r

s

. The density of defects increases from panels 1 to 4 (as described in (C)). (C) STM images of the surface of a bulk highly oriented pyrolytic graphite (HOPG) sample subjected to 90 eV Ar

+

ion bombardment. The ion doses were varied between 10

11

(1), 10

12

(2), 10

13

(3) and 10

14

(4) Ar

+

ions per cm

2

and are the basis for the images shown in (B). The defective structure is depicted in the insets in panels 1 and 3.

Figure 3.16

(A) Raman spectra of graphene with a variable amount of defects between 0.005% and 0.77% based on the relation described and illustrated in Figure 3.15. The

I

D

/

I

G

ratio increases with increasing density of defects, and at about 0.3% the

I

D

/

I

G

ratio decreases again. In contrast, the

Γ

D

,

Γ

G

and

Γ

2D

peaks increase with increasing density of defects. (B) Illustration of the carbon framework of graphene with symbolized defects marked as red circles. Here L

D

 = 10 nm and the expected I

D

/I

G

ratio would be 1. (C) Examples of Raman spectra of RGO with a density of defects of about 3% (left) and 0.03% (right).

Figure 3.17

Raman spectra of specially prepared RGO with a variable amount of defects. For the D, G and 2D peaks, the

Γ

values are given in italic numbers. For the displayed spectra, a density of defects between 1% and 0.03% can be estimated

.

Figure 3.18

(A) Raman spectra of GO with degree of functionalization of about 4% and about 40%, respectively (

Γ

values are given in italic numbers). A significant difference between the spectra and the relation to the degree of functionalization are not yet reliably established. (B) Spectra of typical GO and RGO prepared according to standard methods [43], outlined in Chapter 2. Spectral features almost do not change, because, although oxygen functional groups are removed, more than 3% permanent structural defects remain that also cause line broadening and a D peak. Only the

I

D

/

I

G

ratio increases slightly, which is related to a minimal increase of quality, as shown in Figure 3.15(A) for graphene with a density of defects of 0.77% and 0.12%, respectively.

Figure 3.19

Illustration of the chemical structure of GO. (left) GO with structural defects, such as vacancies, on the >3% scale. After chemical reduction, defects can be probed by Raman spectroscopy. Functional groups at defect sites are omitted for clarity. (right) GO with an intact carbon framework is shown, which can be chemically reduced to intact graphene (RGO without defects).

Figure 3.20

(A) Histogram of

I

G

and discrimination of RGO and few layers of RGO as well as edges of flakes of RGO. (B) Plot of

I

G

versus

Γ

G

extracted from a dataset of about 6 × 10

4

Raman spectra from RGO. The highest quality of RGO spectra is indicated by small

Γ

G

values. (C) Plot of I

2D

versus

Γ

2D

; edges (red), monolayer graphene (green) and few‐layer graphene (blue). (D) Plot of the

I

D

/

I

G

ratio and A

D

/A

G

versus

Γ

2D

that allows the quality of RGO to be illustrated.

Figure 3.21

(A) SRM image of a film of 250 × 250 µm

2

of RGO. The

I

G

values are color‐coded according to

I

G

and single layers of RGO are shown in green. (B) Magnification of the gray marked area of (A). Raman information about the thickness of the flakes is illustrated as an overlay of the corresponding AFM image.

Figure 3.22

SEM images of different GO samples on Si/SiO

2

wafers. Images are acquired at different magnifications. (a) The image demonstrates that SEM allows simultaneous imaging of hundreds of flakes on the screen with a size 200 µm by 200 µm. (b) Typical monolayer GO flakes. The flake’s size, edge roughness and other morphological features are easily visible. In the light‐gray areas the flakes are of a single carbon layer. The double‐ and triple‐layer areas, where flakes are folded or overlapped, are easily identifiable by flakes’ opacity. (c) The image demonstrates GO flakes with different morphology; multi‐layered flakes appear as black, non‐transparent bodies. (d) A GO flake with smooth straight edges; the edges intersect at 120°

Figure 3.23

SEM images of GO flakes demonstrating the impact of base treatment on flakes’ morphology. (a,b) Typical GO flakes retain their integrity and have smooth edges. (c,d) GO flakes damaged by base treatment and consecutive sonication for 20 s. These flakes exhibit smaller size and jagged edges.

Figure 3.24

(A, D, G) SEM images and (B, E, H) corresponding AFM images of a GO flake. (C, F, I) Height profiles corresponding to the respective images (B, E, H). The image demonstrates removal of a single GO layer from an original bilayer flake.

Figure 3.25

TEM images of GO–Pd nanoparticle (NP) composite. The tiny dark spots on panels (a)–(d) are Pd NPs. (e) The HRTEM image reveals the crystalline structure of a Pd NP. (f) Histogram of particle size distribution.

Figure 3.26

Atomic‐resolution, aberration‐corrected TEM image of a single‐layer RGO. (a) Original image and (b) the same image color‐coded to highlight different features. The defect‐free crystalline graphene area is displayed in the original light‐gray color. Defective areas are shaded in dark gray. Blue‐colored regions are the disordered single‐layer carbon networks, or extended topological defects, identified as remnants of the oxidation–reduction process. Red‐colored areas highlight individual adatoms or substitutions. Green‐colored areas indicate isolated topological defects, which are single bond rotations or dislocation cores. Holes and their edge reconstructions are colored in yellow. Scale bar (bottom right) is 1 nm.

Figure 3.27

(a,b) Pentagon–heptagon dipole observed at two different times (2 min apart) in the TEM. The separation between the dislocation pairs has increased, indicating mobility of the carbon grid, and a significant amount of strain causing the separation. (c,d) Additional defect clusters, indicated by blue dashed lines. Yellow dashed lines indicate regions with a hexagonal lattice rotated to the dominant orientation (yellow solid lines for comparison). Red dashed lines indicate distortions in the hexagonal lattice. Yellow arrows in (c) indicate strongly elongated carbon polygons. All scale bars are 1 nm.

Chapter 04

Figure 4.1

Schematic representation of the formation of highly ordered structures of 2D anisotropic platelets in liquid media when the particle concentration is increased. Increasing contents of anisotropic particles in liquid media leads to an increasing restriction of their free rotation. As a consequence of this restricted free rotation, the particles tend to organize themselves, forming highly ordered structures, which are known as colloidal liquid crystals (LCs)

Figure 4.2

Polarized light optical micrographs of (a) GO sheets randomly oriented in a very dilute dispersion, and (b) GO sheets forming an LC phase when the concentration is increased, showing the appearance of birefringence under polarized light when, at a higher concentration of particles, they organize themselves, forming a nematic LC phase.

Figure 4.3

Schematic representation of (a) dynamic and (b) steady shear measurements, showing how the samples are deformed (sliding deformation)

Figure 4.4

Typical dynamic strain sweeps on GO dispersions at different concentrations conducted at a constant frequency (1 rad s

−1

) and increasing strain amplitude [42], showing the presence of a LVR (with G′ and G″ independent of strain and G′ > G″) up to a critical strain, which strongly depends on the concentration. Above the critical strain, a liquid‐like behaviour region is entered.

Figure 4.5

(a) Typical dynamic stress sweeps for a 0.83 vol.% GO dispersion, showing the LVR up to a critical stress, above which the two moduli cross and the liquid‐like behaviour region is entered. (b) Variation of the yield stress with concentration of GO, showing a minimum at 0.45 vol.%, which represents an orientation of the sheets [43].

Figure 4.6

Dynamic frequency sweep on aqueous GO dispersions conducted within the LVR (at a constant strain amplitude of 0.1%). Variation of (a)

G

′ and (b)

G

ʺ with increasing frequencies, showing different behaviours depending on the GO concentration. (c) Values of the moduli

G

′ and

G

ʺ within the LVR at a constant frequency of 0.1 rad s

−1

as a function of the concentration.

Figure 4.7

(a) Steady shear rate flow behaviour of the GO dispersions at different concentrations, showing Newtonian or non‐Newtonian behaviour depending on the concentration. (b) The steady shear flow data can be fitted to either the Bingham model (dashed lines) defined by (4.4) or the Herschel–Bulkley model (solid lines) defined by (4.5), as both models incorporate a yield stress.

Figure 4.8

Schematic representation of the elastic network of GO flakes existing within the LVR, which breaks down into flocs by applying increasing rates leaving the LVR. The size of the formed flocs decreases as the shear rates increase

Figure 4.9

Variation of the shear viscosity of aqueous GO dispersions with GO composition, showing a non‐monotonic relationship between them, which demonstrates the isotropic–nematic phase transition.

Figure 4.10

Variation of the ratio of elastic and storage moduli (

G

′/

G

″) with frequency for GO dispersions at different concentrations, showing which processing techniques are suitable to allow the use of GO dispersions for industrial fabrication. For example, when the viscous component (

G

″) dominates, high‐rate processing methods where the dispersion must spread on a substrate are suitable; whereas when the elastic component (

G

′) dominates, extrusion printing and fibre spinning are appropriate fabrication methods.

Figure 4.11

Field‐induced birefringence generated on an aqueous GO dispersion with concentrations of 0.1 and 1.1 vol.% by applying an electric field of 10 Hz, showing the appearance of birefringence only for the 0.1 vol.% dispersion. The disappearance of the birefringence when the electric field is removed can also be seen [74].

Figure 4.12

Rheological properties of GO–PMMA melts with different GO contents (performed at the polymer melting temperature, i.e. 230 °C). Dependence of (a)

G

′ and (b)

G

ʺ with frequency (dynamic frequency behaviour at a constant strain amplitude of 1%), showing the gradual development of a GO percolated network in the matrix with increasing GO content. (c) Dependence of the viscosity with increasing shear rates (steady shear behaviour), showing an initial Newtonian plateau, which corresponds to the presence of the particulate network of GO flakes in the matrix, followed by a shear‐thinning region, which indicates the breakdown of the GO network.

Chapter 05

Figure 5.1

(A) The π orbitals of a C = C bond. (B) The π* orbitals of a C = C bond. (C) Energy diagram of π, π* and n orbitals and the transitions between those

Figure 5.2

Absorption spectrum of GO produced by the modified Hummers method

Figure 5.3

Infrared (IR) spectrum of GO prepared by the modified Hummers method. Labeled features correspond to vibrational bands in functional groups

Figure 5.4

(A) Absorbance spectra of RGO oxidized via controlled ozone treatment for 0 to 25 minutes. (B) Absorbance spectra of GO at different pH values [24]

.

Figure 5.5

Attenuated total reflectance Fourier‐transform infrared (ATR‐FTIR) spectra of solid GO samples isolated from (A) acidic and (B) basic aqueous suspensions. Major bands and proposed assignments: (A) 1040 cm

−1

(C–O stretch), 1164 cm

−1

(C–OH stretch), 1623 cm

−1

(adsorbed water and skeletal vibrations of unoxidized graphitic domains), 1722 cm

−1

(C = O stretch); and (B) 829 cm

−1

(C–H out‐of‐plane wag), 980 cm

−1

(possibly epoxide stretch), 1007 cm

−1

(C–H in‐plane bend), 1309 cm

−1

(C–O stretch), 1367 cm

−1

(COO− symmetric stretch), 1590 cm

−1

(COO− antisymmetric stretch) [24]

.

Figure 5.6

Raman spectra of single‐layer (top) and double‐layer (bottom) pristine and oxidized graphene

.

Figure 5.7

(A) Photoluminescence spectrum of GO produced from graphite by the modified Hummers method [14, 70], 440 nm excitation. (B) Photoluminescence of GO produced by controlled oxidation from RGO; 440 nm excitation

Figure 5.8

(A) General schematics of the bandgap photoluminescence process. (B) Simplistic schematics of electronic transitions considering excitonic effects, where E

gap

is a one‐electron bandgap energy, E

b

is the binding energy of the exciton and E

ts

is the first allowed optical transition energy

Figure 5.9

Structural models of graphene nanodisks

and the scaling characteristics of optical properties. (A) Scaling relations for the exciton binding energy (E

b

) with the size of the graphene nanodisks (D). (B) Transition energy (E

ts

) as a function of the inverse size (1/D)

.

Figure 5.10

(A) PL spectra of aqueous GO suspensions at excitation energies from 1.8 to 2.5 eV indicated by arrows. Inset: A photoluminescence emission intensity map of GO. (B) Change of photoluminescence spectra of GO with hydrazine reduction time

.

Figure 5.11

Schematics of the dual nature of photoluminescence in reduced GO. (A) The predominant emission in GO from disorder‐induced localized states. (B) The predominant emission in thermally reduced GO from confined cluster states

.

Figure 5.12

Photoluminescence spectra of ozone‐treated RGO

Figure 5.13

Photoluminescence spectra of GO, KOH‐ and HNO

3

‐treated GO sheets in deionized water (100 mg ml

−1

) under excitation of 420 nm. Inset: A schematic energy level diagram depicting the multiple photoluminescence emissions due to electron transitions between different molecular orbitals

.

Figure 5.14

(A) Model of graphene sheet fragment with a COO

−

functional group at the edge. (B) Computation of electrostatic potential isosurface for a graphene sheet fragment with a COO

−

functional group at the edge. (C) pH‐dependent photoluminescence spectra of GO [24]

.

Figure 5.15

Implied relaxation of excited carriers within the context of a disordered band model of GO

.

Chapter 06

Figure 6.1

Different proposed structural models of GO and graphite oxide. The Hofmann model can be applied to the structure of graphene epitaxially grown from SiC and exposed to oxygen atoms (scanning tunneling microscopy image displays individual oxygen atoms covalently bound to graphene).

Figure 6.2

(A) Thermogravimetric analysis coupled with mass spectrometry of GO and GO treated with

18

O‐labeled water (

16

OH

2

exchanged by

18

OH

2

). C

18

O

16

O can be detected, evidence for the formation of hydrate species generated from carbonyl or carboxyl groups. (B) Suggested reaction pathways of decarboxylation processes in GO.

Figure 6.3

(A) Scanning electron microscope (SEM) image of flakes of GO with smooth edges. (B) SEM image of a base‐treated GO flake that shows etched edges and disintegration of the flake to form oxidative debris. (C) Schematic model of the disintegration of flakes of GO with holes (white dots), graphitic domains (blue) and oxidized domains (orange). (D) Proposed reaction schemes for the degradation of GO by water and base. Plausible reaction mechanisms for CO

2

generation and the formation of acidic groups are illustrated. Base treatment of

1

can lead to extended π networks with acidic functional groups (

2–7

). Hydrates (

3

,

4

) and C–O–C bonds can be formed and permanent defects (

6

). Hydroxide can react with carbonyl groups to form carboxylic acids after C–C bond breakage (

8–10

). Decarboxylation and subsequent reactions lead to further degradation [1, 28, 29].

Figure 6.4

(A) The carbon framework of oxo‐G

1

is thermally stable up to 100 °C, but it is unstable at higher temperatures [30]. (B) Statistical Raman analysis of graphene derived from oxo‐G

1

after thermal treatment and subsequent chemical reduction [30].

Figure 6.5

(A) Illustration of GO or oxo‐G

1

that is damaged by a laser pulse and the locally thermally reduced material is further reduced chemically. AFM images of (B) GO and (C) RGO after a 5 s laser pulse at 0.06 mW applied to GO. (D) Raman map of the

I

D

/

I

G

ratio of the flake of RGO.

Figure 6.6

Non‐covalent approaches to functionalized GO with π interactions or polar interactions. Example of ssDNA conjugated with a fluorophore (FL, quenched) adsorbed on GO exploiting polar and π interactions. A complementary target leads to desorption and restores fluorescence [33]

Figure 6.7

Non‐covalent polar interaction of PDDA with GO as stabilizing agent, followed by deoxygenation with NaBH

4

to yield stabilized RGO [34]

Figure 6.8

Non‐covalent polar interaction of pyrene with the π system of RGO enables attachment of the RAFT agent for composite formation [35]

Figure 6.9

Chemical sketch of GO accounting for functional groups on both sides of the basal plane and not well‐defined defects

Figure 6.10

Functionalization of GO with α‐bromoisobutyryl bromide to form ester groups and the bromo end‐group that can be exploited for ATRP reactions. The ATRP polymerization with methyl methacrylate is subsequently initiated by CuCl and N,N,N,N,N‐pentamethyl diethylene triamine (PMDETA) [36]

Figure 6.11

Hydroxyl groups in GO exist is arbitrary amounts on the basal plane and at defect sites. Ester or ether formation with hydroxyl groups will proceed at tertiary hydroxyl groups and the more phenol‐like hydroxyl groups located at edges (if the ester and ether formation are sterically possible). Chemical reduction is expected to cleave preferably the C–O bond on the basal plane

Figure 6.12

Illustration of the reaction of hydroxyl groups of GO to form C–C bonds by Johnson–Claisen rearrangement. The yielded ester can be saponified, activated and undergoes amidation with various amines, including propargyl amine, that can be further exploited for alkine–azide reactions. The addition of various groups with ethylene glycol moieties or sulfonate groups was reported. Terminal amines can also be protonated to yield positively charged derivatives suitable for layer‐by‐layer assembly [37]

Figure 6.13

Clemmensen reduction of carbonyl groups to CH

2

groups, proceeding at the edges of flakes and defects as proposed by Sofer et al. [43]. Nascent hydrogen instead leads to defunctionalization of surface bound oxo‐groups [43]

Figure 6.14

(A) GO functionalized with nanoparticles (NPs). Typically, magnetic Fe

3

O

4

NPs are prepared by redox reaction. Fe

2+

ions are adsorbed on GO and the NPs form in the presence of NH

4

OH at pH = 9. (B) Dispersion of GO and (C) images of dispersion after redox reaction.

Figure 6.15

Illustration of the increase of the carbon framework of nano‐graphene with zigzag edges (number of C atoms/number of edge atoms). While benzene bears only edges, coronene with its C

24

unit has 18 edge atoms and six in‐plane atoms. For the C

96

unit there are more in‐plane atoms that edge atoms

Figure 6.16

Illustration of the hexagonal carbon framework of graphene with one missing C atom. Three new edges are produced that may be oxidized to ketone groups. The middle structure illustrates that the space required for three ketones is limited. However, the formation of an enol structure that is forming a semi‐acetal may solve that problem

Figure 6.17

Illustration of the hexagonal carbon framework of graphene that was over‐oxidized to form carboxylic acids. A certain amount of space is required for the formation of a carboxylic acid, and therefore they are more likely to be formed at larger holes. Formation of lactol groups at edges is plausible

Figure 6.18

Functionalization of GO at carboxylic acids, located at the edges of flakes or defect sites. Activation of carboxylic acids by thionyl chloride can also activate surface OH groups and cause elimination or other side reactions. Acid chlorides can subsequently react with alcohols or amines to form esters and amides, respectively

Figure 6.19

Reaction scheme for an example of edge functionalization of RGO, starting from GO with structural defects. Surface functional groups are chemically removed by reduction with hydrazine. Mild oxidation with nitric acid causes oxidation at defect sites to introduce carboxylic acids. Subsequently carboxylic acids are activated by thionyl chloride to enable the reaction with octadecylamine (the defect site is marked as thick black lines and additional functional groups are omitted for clarity) [67]

Figure 6.20

FTIR spectra for GO (black), RGO (red), carboxylated GO (green) and ODC‐GO (blue) with the bands’ assignment as performed by the authors [67].

Figure 6.21

XPS spectra for GO, RGO, carboxylated GO and ODC‐GO with deconvolution of the envelope spectra as performed by the authors [67]. (a) C 1s spectrum for original GO. (b) C 1s spectrum for ODC‐GO. (c) N 1s spectrum for ODC‐GO.

Figure 6.22

Simplified scheme of GO with a flake size of about 5–50 nm with carboxylic acids located at the edges is functionalized by carbodiimide‐mediated amide bond formation to introduce polyethylene glycol groups that provide water solubility [63]

Figure 6.23

Edge functionalization after hydrazine reduction of GO. The reduction of GO with hydrazine can lead to the formation of pyrazole moieties. Epoxy and hydroxyl groups located at the basal plane are removed by this reduction method. The formation of pyrazole at the edges of flakes or defect sites gives evidence for adjacent ketones [68]

Figure 6.24

Possible reaction pathway for the reduction of epoxide groups by hydrazine

Figure 6.25

Morphology of tpGO formed during thermal annealing at 1500 K: (a) original GO contained 20% oxygen; (b) original GO contained 33% oxygen. The structures are obtained by molecular dynamics simulations.

Figure 6.26

The reaction scheme showing deoxygenation of GO during base treatment. Structure (

3

) is a GO fragment containing three alcohol groups. Nucleophilic attack of hydroxide ion on a hydrogen atom results in C–C bond cleavage and in the formation of a ketone. The fragment (

4

) contains one oxygen atom less compared to structure (

3

), signifying deoxygenation

Figure 6.27

Conversion of ketones to carboxyls during base treatment. Structure (

5

) is a GO fragment containing one ketone and two alcohol groups. Nucleophilic attack of a hydroxide ion on a carbonyl group results in C–C bond cleavage and in the transformation of ketone to carboxyl group (

6

)

Figure 6.28

Decarboxylation of GO under strongly alkaline conditions. The reaction mechanism is typical for decarboxylation

Figure 6.29

Three‐step reduction of GO nanoribbons: N

2

H

4

reduction, 300 °C annealing, 900 °C annealing. (A) C 1s XPS and (inset) Raman spectra. (B) Logarithmic I–V curves for the nanoribbons with different degree of reduction; the average data obtained from testing 30 different devices. (C) and (D) Source–drain current (I

sd

), source–drain voltage (V

sd

) and gate voltage (V

g

) dependences for the same device build on an N

2

H

4

‐reduced monolayer RGO NR before (C) and after (D) annealing in H

2

/Ar at 300 °C for 30 min.

Figure 6.30

(A) Statistical Raman analysis of graphene from oxo‐G

1

using different reducing agents. (B) AFM images of graphene from oxo‐G

1

obtained after thermal reduction (at 200 °C) or reduction with vitamin C (ascorbic acid), hydrazine or hydriodic acid and trifluoroacetic acid (HI/TFA) [90].

Figure 6.31

(A) Mechanism of the reduction of oxo‐G

1

by an electron donor, such as iodide and a strong acid. Successive protonation and electron transfer steps remove oxygen functional groups to form graphene on a substrate. (B) Reduced on substrate by vapor of HI/TFA (G

HI/TFA

). (C) Reduced by ascorbic acid (AS) from the reductive subphase (G

AS

). (D) Reduced by the combination of both reduction methods (

) [95].

Figure 6.32

Functionalization of pristine graphene with nitro‐aryl diazonium salts.

Figure 6.33

Scheme showing sodium dodecylbenzene sulfonate (SDBS)‐wrapped GO, its reduction and functionalization of intermediate SDBS‐wrapped RGO with diazonium salts.

Figure 6.34

Schematic of the chemical functionalization of devices with 4‐nitrophenyl groups (left). Electronic devices consist of monolayer RGO nanoribbons contacted with Pt source (S) and drain (D) electrodes. The devices were fabricated on a 200 nm thick thermal SiO

2

over heavily doped p‐type Si that was used as a back gate. The

I

sd

–V

g

curves recorded at

V

sd

 = 

0.1 V after several consecutive grafting experiments (right). The labels on the curves show the total grafting time.

Figure 6.35

(A) Proposed formation of cyclic organosulfate groups in GO and hydrolytic cleavage to organosulfate groups and hydroxyl groups. Acid‐catalyzed epoxide formation of cis‐dihydroxy groups can be assumed [17, 29]. The structure of GO therefore bears epoxy, hydroxyl and organosulfate groups as major functional groups. Defects in GO are omitted and the structure model is also valid for oxo‐G

1

[108]. (B) Thermogravimetric analysis of (left) GO, (middle) GO treated with sodium hydroxide to cleave organosulfate groups and (right) GO treated with sodium hydroxide and mixed with sodium hydrogensulfate [108].

Figure 6.36

Illustration of the reactivity of oxo‐G

1

. Reactivity at 10 °C (top) and 40 °C (bottom), upon treatment with HCl or NaOH is shown. Also the substitution of organosulfate by azide is possible under controlled reaction conditions [110]

Figure 6.37

(A) Infrared spectra of GO‐N

3

and GO‐

15

N

14

N

2

(here GO is oxo‐G

1

). (B) Thermogravimetric analysis of GO and GO‐

15

N

14

N

2

, m/z 29 of GO‐N

3

and GO‐

15

N

14

N

2

. (C) Simplified model of GO‐N

3

with an azide and a hydroxyl group connected to the carbon lattice in trans‐configuration.

15

N NMR shifts are calculated by ab initio methods (107.4 ppm and 225.8 ppm). (D) The

15

N SSNMR MAS (magic angle spinning) spectrum of GO‐

15

N

14

N

2

with two peaks (1 : 1 ratio) [110].

Figure 6.38

(A) Reaction scheme illustrating the reaction of oxo‐G

1

with dodecylamine (oxo‐G

1

/DA), followed by non‐covalent functionalization with a block copolymer of styrene and ethylene oxide (PSEO). Solid‐state NMR spectra of oxo‐G

1

/DA: (B)

13

C NMR, (C)

15

N NMR and (D)

1

H–

13

C correlation NMR. (E) Device structure of a floating‐gate memory device with oxo‐G

1

/DA/PSEO as charge storage material. (F) Device characteristic: write signal at −3 V, erase signal at 2 V and read voltage at −0.5 V. (G) AFM image of flakes of oxo‐G

1

/DA/PSEO composite [111].

Figure 6.39

(A) TGA of oxo‐G

1

/DA between 30 and 700 °C and the temperature profiles of cleaved alkyl, CO

2

and SO

2

formation, identified by FTIR. (B) TGA of oxo‐G

1

/DA/PSEO between 30 and 700 °C with the temperature profile of cleaved aromatic compounds (gray) identified by FTIR [111].

Chapter 07

Figure 7.1

(a) The I–V characteristics of an RGO‐based FET device (gate biases of 15, 0, −5, −10 and −15 V, respectively, from bottom to top). As the gate voltage changes from +15 to −15 V, the conductance of the device increases, indicating a p‐type semiconductor behavior. (b) Temperature‐dependent measurements, confirming the semiconducting behavior of the RGO film

.

Figure 7.2

(a–d) Microfabrication of an all‐carbon graphene FET device. Both GO active channel and electrodes are deposited onto an SiO

2

/Si substrate by drop‐casting or printing. (e–h) Device fabrication using conventional photolithography. Schemes and optical microscopy images of the RGO electrodes (i) and the whole device (j,k)

.

Figure 7.3

Increase of carrier mobility in Au/pentacene organic FET devices due to the presence of a highly reduced GO (HRG) interlayer

.

Figure 7.4

Fabrication of an all‐RGO thin‐film transistor. A GO solution is spin‐coated onto a flexible polymeric substrate, and subsequently scratched to make two separated electrodes. Between the electrodes, another GO film is micro‐patterned using the microfluidics technique. RGO films are then obtained via exposure to hydrazine vapor. A silicone rubber is finally used to insulate the RGO electrodes

.

Figure 7.5

Example of an NO

2

sensor. (a) All‐organic flexible set of interdigitated electrodes generated from highly reduced laser‐scribed graphene (hr‐LSG). (b) The same interdigitated electrodes transferred onto polydimethylsiloxane (PDMS). (c) NO

2

detection using the all‐organic flexible interdigitated electrodes. Here the sensor uses hr‐LSG as the active electrodes and marginally laser‐reduced graphite oxide as the detecting medium. The NO

2

concentration is 20 ppm in dry air

.

Figure 7.6

(a) Layer‐by‐layer fabrication process of a humidity GO nanocomposite film. (b) Optical image of 4 × 6 sensor array on a flexible polyimide substrate

.

Figure 7.7

Three possible mechanisms of hybridization between a probe DNA adsorbed by GO and its cDNA (target DNA): (A) Langmuir − Hinshelwood mechanism; (B) Eley − Rideal mechanism; and (C) displacement mechanism. In all three cases, the probe DNA with a fluorophore label is pre‐adsorbed and the cDNA is added afterwards. The tendency of GO to adsorb double‐stranded ds‐DNA is lower than that to adsorb single‐stranded ss‐DNA

.

Figure 7.8

Example of dependence of different properties of RGO on film thickness. (A) Photograph of films with increasing thickness. (B) Ultraviolet–visible spectra of films with increasing thickness. (C, D) Dependence of the electrical behavior on transparency with different reduction processes

.

Figure 7.9

Preparation process of RGO‐Ag nanowire TCFs

.

Figure 7.10

Schematic representation of a MIM memristive structure based on a GO switching layer, with mechanism based on metal filament formation. (a) Device in pristine state. (b) A positive bias induces the oxidation of the top electrode, originating metal cations that migrate towards the bottom electrode, where the formation of metal filaments is promoted after the reduction. (c) The grown filament reach the top electrode with the generation of a local short‐circuit (device switches from HRS to LRS). (d) On inverting the external bias, the opposite process is achieved

.

Figure 7.11

Typical characteristics of a symmetric MIM device based on GO switching layer, showing switching based on oxygen ion diffusion and bipolar behavior: (a) I–V characteristic curve, (b) endurance and (c) retention tests

.

Chapter 08

Figure 8.1

Schematic of a Li‐ion battery in discharge mode (graphite anode, LiMO

2

cathode).

Figure 8.2

Total capacity of a Li‐ion cell as a function of anode capacity (

C

A

), including masses of other required internal components and case (

Q

M

 = 130 mA h g

−1

). Capacities of cathodes considered were 135, 200 and 400 mA h g

−1

Figure 8.3

SEM images of a chemically reduced graphene oxide (RGO) paper showing the (A) rolling paper‐like electrode surface and (B) non‐uniformity and cracking along a fractured edge.

Figure 8.4

Schematic drawing (left) of the process to introduce in‐plane pores into the GO material and subsequent formation into freestanding holey tpGO paper material. The electrochemical performance (right) of (A) a traditional tpGO paper (no holes) and (B) holey tpGO paper are compared.

Figure 8.5

Schematic showing the nitrogen and boron doping into the graphene plane, and resulting high energy and power performance of doped materials compared to the undoped control samples.

Figure 8.6

Cross‐sectional schematic (not to scale) of the Si–tpGO composite structure (top). Digital image of prepared Si‐GO composite paper (bottom left). Transmission electron microscopy (TEM) image of Si nanoparticles encapsulated between tpGO sheets (bottom right).

Figure 8.7

Cross‐sectional schematic (not to scale) of the holey Si–tpGO composite structure (left). TEM image (right) depicting the highly defective tpGO structure (scale bar: 50 nm). Inset: Zoomed‐in view of the tpGO structure.

Figure 8.8

(a) Fabrication of the RGO‐wrapped metal oxide particles. (b) SEM image demonstrating the Co

3

O

4

particles embedded in the flexible RGO structure. (c) Electrochemical cycling of co‐assembled Co

3

O

4

‐RGO composites (squares), physically mixed Co

3

O

4

‐RGO composite (circles) and bare Co

3

O

4

electrodes (triangles). (d) Cycle performance of the co‐assembled Co

3

O

4

‐RGO electrode for 130 cycles at 74 mA g

−1

.

Figure 8.9

(A) Preparation process and microscale structure of LiFePO

4

–RGO composite. (B) SEM image showing LiFePO

4

primary particles wrapped by the RGO structure. (C) Electrochemical cycling of LiFePO

4

composite electrodes at a 10C–20C asymmetric charge–discharge process.

Figure 8.10

Schematic representation of the synthesis scheme to fabricate the sulfur–PEG–GO architecture (left). Cycling performance of the composite at rates of C/5 and C/2.

Figure 8.11

Schematic of the CTAB‐modified sulfur–GO composite material (top) and accompanying composite electrochemical cycling performance (bottom).

Figure 8.12

Illustrative schematic depicting the fabrication of the freestanding Li

2

S–tpGO composite paper (top). Electrochemical cycling performance of the Li

2

S–tpGO cathode when cycled at 5C (~6 A/g

−1

).

Figure 8.13

Illustrative schematic depicting the fabrication of the “all‐graphene battery” (top). Electrochemical cycling performance of the system when cycled at 500 mA g

−1

(bottom).

Figure 8.14

Synthetic scheme for the preparation of GO functionalized with PMABS to form a GO‐PMABS membrane (top). Electrochemical cycling at 20°C and 80°C demonstrating the reversible charge storage behavior of the thermally responsive membrane (bottom) when cycled at 200 mA g

−1

.

Figure 8.15

Fabrication schematic for the formation of highly porous graphene‐derived carbons with hierarchical pore structures.

Figure 8.16

(A) Schematic illustrating the formation of (B) the crumpled sheet morphology. Illustration (bottom) of the flat sheet (purple) and crumpled ball (golden) morphologies and their respective capacitive performance.

Figure 8.17

(a) Schematic detailing the synthesis method for the combined hydrothermal and chemical activation processes. (b, c) SEM images demonstrate that the material has a sponge‐like and porous structure. (d, e) TEM images show the material is composed of a dense 3D porous structure with a highly curved or wrinkled surface.

Figure 8.18

Illustration of the “holey graphene” material (b, d, f) and reference non‐holey material (a, c, e) as idealized electrodes for supercapacitor applications.

Figure 8.19

Synthetic formation of hierarchical PANI–RGO composites with 3D interconnected pores.

Figure 8.20

Synthetic scheme (top) for the template procedure to form Ni(OH)

2

–tpGO composite framework; and (bottom) TEM images at different magnifications (a–c) and STEM mapping of the elemental components (d–g).

Chapter 09

Figure 9.1

Two distinct approaches to use graphene‐based materials as selective membrane: (a) porous graphene and (b) layered GO membrane

Figure 9.2

Simplified GO membrane structure to be considered for molecular separation

Figure 9.3

(a) Illustration (generalized) of the helium leak rate through GO film under dry and humidified feed conditions [20]. (b) Change of interlayer distance of hydrophilic and hydrophobic domains in GO under wet–dry switching.

Figure 9.4

(a) Gas permeability of thick GO membranes with different GO platelet sizes [22]. (b) H

2

permeance of thin GO‐coated microporous membranes as a function of applied feed pressure [22].

Figure 9.5

(a) Change of gas permeances of GO membrane prepared by method 1 as a function of coating times [22]. (b) Illustration of the H

2

and CO

2

permeances and selectivity of GO as a function of temperature [23].

Figure 9.6

Gas permeance of GO membranes prepared by (a) method 1 and (b) method 2 under dry and humidified feed conditions (based on single gas permeation measurement) [22].

Figure 9.7

(a) Illustration of the permeation through a 5 mm thick GO membrane from the feed compartment with a 0.2 M solution of MgCl

2

. (b) Illustration of the permeation rate through the GO membrane as a function of hydrated ion radius [32]

Figure 9.8

(a) Water flux through a GO thin membrane as a function of number of GO layers. (b) Rejection of monovalent and divalent ions in a GO thin membrane as a function of salt concentration. (c) Rejection of dye (MB, methylene blue; R‐WT, rhodamine‐WT) and salts in a GO thin membrane as a function of number of GO layers.

Figure 9.9

(a) XRD patterns of base‐refluxing reduced GO, hydrazine‐reduced GO and thermally treated bRGO at 220 °C in vacuum. (b) Variation of pure water flux as a function of bRGO loading coated on the membranes.

Figure 9.10

(a) Pressure‐dependent flux and rejection of EB molecules of NSC‐GO membrane under different pressure. The curves marked by black solid squares and red solid circles represent the flux variation during the first and third pressure‐loading processes, respectively. The curve marked by blue solid triangles denotes the rejection rate of EB during the first pressure‐loading process. (b) Simulated changes in the cross‐sectional area of a nanochannel by varying the applied pressure. (c) The response of a half cylindrical GO nanochannel modelled in molecular dynamics simulation.

Figure 9.11

(a) Water desalination performance chart for GOF membranes compared to existing theoretical work on carbon‐based materials. The labels are: CNT, carbon nanotube; nG, nanoporous graphene; and Gy, graphyne. (b) Water desalination performance chart for GOF membranes compared to existing organic and inorganic membrane technologies.

Chapter 10

Figure 10.1

The crystal structure of graphite.

Figure 10.2

Graphite exfoliation. (a) A scanning electron microscope image of thermally expanded graphite, resulting in “accordion‐like” morphology consisting of graphite nanoplatelets with a thickness of a few nanometers (inset). (b) Intercalation of graphite with metal atoms and subsequent chemical reaction result in exfoliation and delamination of graphite into thin nanoplatelets.

(c) A typical setup for electrochemical exfoliation of a graphite electrode.

Figure 10.3

Chemical pathways for functionalization of CNTs and graphene. (a) CNTs can be modified through different interactions. (b) Origin of chemical reactivity of graphene and possible strategies for functionalization of graphene.

Figure 10.4

SWNT helicity map on graphene sheet and examples of (n,m) chiral vectors that give rise to armchair, zigzag and chiral nanotube structures.

Figure 10.5

Analogous concepts for the fabrication of GO‐ and CNT‐based materials. (a) Fabrication of GO‐based transparent conductive films by filtering a GO dispersion through a filter (top) and then wet transfer to a plastic substrate. (b) Thin film of SWNTs using filtration and wet transfer on different substrates with controlled thickness. (c) Fabrication of GO paper by filtration and drying of a GO dispersion through a filter, which results in strong flexible paper‐like material (inset, top). The GO paper possesses a layered structure due to the stacking of GO sheets over each other. (d) Photograph of “buckypaper” (top); the paper (left) is tough and flexible enough to fold into an origami plane (right). Scanning electron microscope images of SWNT‐based buckypaper fabricated by filtration of an aqueous dispersion of SWNTs over a filter.

Figure 10.6

The extraordinary mechanical properties of CNTs raised hopes to achieve a “space elevator”, a conceptual infrastructure that turns out to be unrealistic, at least at the current level of technology.

Figure 10.7

(a) Graphenic sheets with lower thickness induce higher confinement to polymer chains. By keeping the inter‐sheet distance constant, the effective volume fraction of graphene increases on increasing the number of layers. (b) Effective graphene Young’s modulus and maximum graphene volume fraction for different indicated polymer layer thicknesses, as a function of the number of layers in the graphene flakes. (c) Maximum nanocomposite modulus predicted for different indicated polymer layer thicknesses as a function of the number of layers in the graphene flakes.

Figure 10.8

Nacre‐inspired design of GO‐based composites. (a) Red abalone shell. Inset: A cross‐section cut from the shell. (b) Scanning electron micrograph showing the micro/nanostructure of natural nacre (the so‐called “brick and mortar” structure). Scale bar, 1 µm. (c) Scanning electron micrograph of a GO paper resembling the layered structure of nacre. (d) Improved fracture mechanism of GO paper by inserting flexible polymer molecules at interlayer. More complex strain engineering of such composites is presumed to result in tough and strong GO‐based composites.

Figure 10.9

Schematic representation of percolated networks of anisotropic filler. (a) A random network of filler results in a lengthy path for electrons and so a higher resistance is expected. (b) Segregation of filler into wire‐like structure lowers the percolation threshold and enhances the conductivity at lower filler content.

Figure 10.10

Impact of GO aspect ratio on the electrical conductivity of PS–RGO composites. (a–c) Size distribution of GO sheets used for composite fabrication. (d) Schematic of PS–RGO composite thin‐film field‐effect devices employed for conductivity measurements. (e) Device conductivities measured in ambient conditions as a function of channel length for different aspect ratios, clearly demonstrating the high potential for large GO sheets to improve the electrical conductivity of the composites.

Figure 10.11

Schematic illustration for the preparation of PS–RGO composites with an ordered three‐dimensional segregated network, by mixing GO sheets and PS microspheres, followed by GO reduction with vitamin C and hot pressing.

Figure 10.12

Incorporating graphene foam (GF) into a polymer matrix with high conductivity at very low loading. (a,b) CVD growth of graphene films (Ni–G) using a nickel foam as a 3D scaffold template. (c) An as‐grown graphene film after coating a thin polymethyl methacrylate (PMMA) supporting layer (Ni–G–PMMA). (d) A graphene foam coated with PMMA (GF–PMMA) after etching the nickel foam with hot HCl (or FeCl

3