Synthesis of Vanadium Oxide Nanostructures for Functional Applications Haitao Fu

Home , Silver oxide

Synthesis of Vanadium Oxide Nanostructures for Functional Applications

Haitao Fu

A thesis submitted in fulfillment of the requirements for the degree of

Doctor of Philosophy

School of Materials Science and Engineering

Faculty of Science

The University of New South Wales

August 2013

ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ……………………………………………......

Date ……………………………………………......

List of Publications

Peer-Reviewed Journals

1. Fu, H.; Yang, X.; Jiang, X.; Yu, A., Bimetallic Ag–Au Nanowires: Synthesis, Growth Mechanism, and Catalytic Properties. Langmuir 2013, 29, (23), 7134-7142. 2. Fu, H.; Yang, X.; Yu, A.; Jiang, X., Rapid synthesis and growth of silver nanowires induced by vanadium trioxide particles. Particuology 2013, 11, (4), 428-440. 3. Su, D.; Fu, H.; Jiang, X.; Wang, G., ZnO nanocrystals with a high percentage of exposed reactive facets for enhanced gas sensing performance. Sensors and Actuators B: Chemical, 2013, 186, (0), 286-292. 4. Fu, H.; Jiang, X.; Yang, X.; Yu, A.; Su, D.; Wang, G., Glycothermal synthesis of assembled vanadium oxide nanostructures for gas sensing. J. Nanopart. Res. 2012, 14, (6), 1-14.

5. Yang, X.; Fu, H.; Yu, A.; Jiang, X., Large-surface mesoporous TiO2 nanoparticles: Synthesis, growth and photocatalytic performance. Journal of Colloid and Interface Science 2012, 387, (1), 74-83.

6. Yang, X.; Fu, H.; Yu, A.; Jiang, X., Hybrid Ag@TiO2 core-shell nanostructures with highly enhanced photocatalytic performance. Nanotechnology, 2013, accepted.

Book Chapter

7. Yang, X.; Fu, H.; Jiang, X.; Yu, A., (2013) Silver Nanoparticles: Synthesis, Growth Mechanism and Bioapplication, Silver Nanoparticles: Synthesis, Uses and Health Concerns, Chapter ID: 17048, ISBN: 978-1-62808-407-8, NOVA Science.

Conference Proceedings

8. Fu, H.; Yang, X.; Jiang, X.; Yu, A., Glycothermal Synthesis of Urchin-Like Vanadium Pentoxide Nanostructure for Gas Sensing, the 13th IEEE International Conference on Nanotechnology (IEEE-NANO 2013), August 5-8, 2013, Shangri-La Hotel, Beijing, China.

Abstract

Vanadium oxide nanoparticles have displayed excellent properties in the field of clean energy, environment, and catalysis. Nowadays, the applications of vanadium oxides in catalysts, lithium ion batteries (LIB), gas sensors, smart windows, and temperature switches attract increasing more attention. Therefore, the studies of the synthesis and properties of these materials are important for scaling up production and understanding the formation and growth mechanism, and surface behaviours for functional properties and potential applications.

In this thesis, a brief introduction of the relative research and a literature review on the vanadium oxides and their nanocomposites were presented in Chapters 1 and 2, respectively. Chapter 3 systematically described the preparation of various vanadium oxides (V2O5 and V2O3) with different shapes (microspheres, microurchins and nanorods) by a polythermal method, the growth mechanism, and the sensing performance of V2O5 nanoparticles. To enhance the function properties of gas sensing, the nanocomposites of silver vanadium oxides (SVO) and vanadium oxides with Ag nanocomposites (VOx@Ag) were investigated in Chapter 4, in which Ag2V4O11 nanobelts were found to exhibit high sensitivity and selectivity to amines. To enrich the application of V2O3 particles, Chapters 5 and 6 respectively demonstrated wet-chemical methods for induced synthesis of Ag nanowires and Ag-Au bimetallic nanowires by V2O3 particles. In these chapters, the formation mechanisms and catalytic performance for reduction of 4-nitrophenol were discussed. Finally, the conclusions were summarised in Chapter 7.

iii

Table of Contents

List of Publications ...... ii

Abstract ...... iii

Acknowledgements ...... x

Abbreviations and Symbols ...... xi

Captions of Figures ...... xiv

List of Tables...... xxiii

Chapter 1. Introduction ...... 1

1.1 Background ...... 1

1.2 Scope of Research ...... 2

Chapter 2. Literature review ...... 5

2.1 Vanadium oxides ...... 5

2.1.1 Vanadium pentoxide ...... 5

2.1.2 Vanadium dioxide ...... 6

2.1.3 Vanadium trioxide ...... 7

2.1.4 Transformation among three oxides...... 8

2.2 Synthesis Methods ...... 9

2.2.1 Physical methods of vanadium oxides ...... 9

2.2.2 Chemical methods of vanadium oxides ...... 13

2.2.3. Shape controlled synthesis of Ag nanoparticles...... 31

2.2.4 Shape controlled synthesis of Au nanoparticles...... 33

2.3 Surface modifications ...... 34

2.3.1 Polymers ...... 35

2.3.2 Metal dopants ...... 37

2.3.3 Metal Oxide dopants ...... 38

2.3.4 Silica modifier(s) ...... 39

2.4 Properties of vanadium oxide nanoparticles and composites ...... 40

2.4.1 Redox properties ...... 40

2.4.2 Metal insulator transition ...... 41

2.4.3 Electrical properties ...... 42

2.4.4 Magnetic properties ...... 44

2.5 Functional applications ...... 46

2.5.1 Gas sensing materials ...... 46

2.5.2 Catalysts ...... 49

2.5.3 Actuators ...... 49

2.5.4 Optical switches and smart windows ...... 50

2.5.5 Electrode in LIBs ...... 51

2.6 Summary ...... 51

Chapter 3. Glycothermal synthesis of assembled vanadium oxide nanostructures for gas sensing…………………………………………………………………………...…………53

3.1 Abstract ...... 53

3.2 Introduction ...... 53

3.3. Experimental work ...... 55

3.3.1 Materials ...... 55

3.3.2 Synthesis ...... 55

3.3.3 Characterization ...... 56

3.4. Results and discussion ...... 58

3.4.1 Synthesis of vanadium oxide nanoparticles ...... 58

3.4.2 Effects of experimental parameters ...... 62

3.4.3 Formation and growth mechanism ...... 66

3.4.4 Gas sensing detection...... 74

3.5 Summary ...... 77

Chapter 4. Hydrothermal synthesis of Ag-vanadate nanocomposites for amine sensing 78

4.1 Abstract ...... 78

4.2 Introduction ...... 78

4.3 Experimental work ...... 80

4.3.1 Materials ...... 80

4.3.2 Preparation of Ag-vanadate nanocomposites ...... 80

4.3.3 Characterization ...... 82

4.3.4 Gas sensing performance ...... 82

4.4 Results and discussion ...... 82

4.4.1 Synthesis of nanocomposites ...... 82

4.4.2 Effect of experimental parameters ...... 85

4.4.3 Formation mechanism of the nanocomposites...... 99

4.4.4 Gas sensing for amines ...... 101

4.5 Summary ...... 106

Chapter 5. Rapid synthesis and growth of silver nanowires induced by vanadium trioxide particles…………………………………………………………………………………...107

5.1 Abstract ...... 107

5.2 Introduction ...... 107

5.3 Experimental work ...... 110

5.3.1 Materials ...... 110

5.3.2 Synthesis of Ag nanowires ...... 110

5.3.3 Characterizations ...... 110

5.4 Results and discussion ...... 111

5.4. 1 Synthesis of the Ag nanowires ...... 111

5.4. 2 Effect of experimental parameters ...... 113

5.4. 3 Formation mechanism...... 116

vii

5.4. 4 Roles of V2O3 in the formation of Ag nanowires ...... 124

5.4. 5 Optical property of the Ag nanowires...... 129

5.5 Summary ...... 130

Chapter 6. Synthesis of bimetallic Ag-Au nanowires by induction of vanadium trioxide and its catalytic property for reduction of 4-NP...... 132

6.1 Abstract ...... 132

6.2 Introduction ...... 132

6.3 Experimental work ...... 134

6.3.1 Materials ...... 134

6.3.2 Preparation of Ag-Au nanowires ...... 135

6.3.3 Preparation of Au nanoparticles ...... 135

6.3.5 Catalytic reduction of 4-NP ...... 136

6.3.6 Characterizations ...... 136

6.3.7 Molecular dynamics simulation ...... 136

6.4 Results discussion ...... 138

6.4. 1 Synthesis of bimetallic nanostructures ...... 138

6.4. 2 Effect of molar ratio of Au to Ag ...... 143

6.4. 3 Formation of mechanism ...... 146

6.4. 4 Properties of the Ag-Au bimetallic nanowires ...... 152

6.5 Summary ...... 158

viii

Chapter 7 Conclusions ...... 159

References ...... 161

Acknowledgements

I would like to express my sincere gratitude to everyone who has walked along with me on the duration of my research studies and making of this thesis.

I thank A/Prof. Xuchuan Jiang for the constant supervision in project design, experimental conduction, result analysis and manuscripts preparation and modification. He was always available for contact, even after working hours, for questions and comments. I also thank my co-supervisor, Prof. Aibing Yu, for giving constructive advice on improving the quality of my research as well as providing helpful suggestions on my future career development.

I would like to particularly thank China Scholar Councils (CSC) for financial support until the complete of my PhD degree.

I would like to express my appreciation to the technical staff in the electron microscope unit (Analytic Centre), especially Katie Levick, Sean Lim and Sigrid Fraser, for their technical help in TEM, HRTEM, and SEM analysis. I also thank my friends and colleagues of SIMPAS, especially my wife - Xiaohong Yang, Mr. Chuyang Chen, Mr. Shixian Xiong, and Mr. Zhengjie Zhang, for their help in experimental analysis and discussion.

Most importantly, I want to acknowledge the loving support of my family and friends in China, especially my wife and my daughter who is actually the most successful achievement in my PhD study.

Abbreviations and Symbols

4-NP 4-nitrophenol

4-AP 4-aminophenol

AFM Antiferromagnetism

APS Ammonium Peroxydisulfate

AOT Sodium bis(2-ethylhexyl) Sulfosuccinate

BET Brunauer-Emmett-Teller

CTAB Cetyltrimethylammonium Bromide

EDS Energy Dispersive Spectroscopy

EG Ethylene Glycol

EMF Electromotive Force

FM Ferromagnetism

FSP Flame Spray Pyrolysis

FT-IR Fourier Transform Infrared

HDA HexDecyl Amine

HRTEM High Resolution Transmission Electron Microscope

ICD Implantable Cardioverter Defibrillator

ITO Indium Tin Oxide

LIB Lithium Ion Battery

LPG Liquid Petroleum Gas xi

MD Molecular Dynamics

MIT Metal-Insulator Transition

MLE Monolayer Equivalent

MS Mass Spectrum

MWCNT Multi-Walled Carbon Nanotube

PAN Polyaniline

PC Polycarbonate

PEO Poly(ethylene oxide)

PMAA Polymethylacrylic Acid

PLD Pulsed Laser Deposition

PPy Polypyrrole

PVC Physical Vapor Condensation

PVP Polyvinylpyrrolidone

PZC Point of Zero Charge

SAED Selected Area Electron Diffraction

SCE Saturated Calomel Electrode

SDS Sodium Dodecyl Sulfate

SEM Scanning Electron Microscopy

SERS Surface Enhanced Resonance Spectroscopy

SPM Superparamagnetism

xii

SPR Surface Plasmon Resonance

STM Scanning Tunnelling Microscopy

TEM Transmission Electron Microscopy

TEOS Tetraethylorthosilicate

TDAB Tetradecylammonium Bromide

TGA Thermo-Gravimetric Analysis

THF Tetrahydrofuran

TMA Tetramethyl Ammonium

TPO Temperature Programmed Oxidation

TPR Temperature Programmed Reduction

UHV Ultra-High Vacuum

VEG Vanadyl Ethylene Glycolate

VITP Vanadium (V) Triisopropoxy Oxide

XRD X-Ray Diffraction vdW van der Waals

xiii

Captions of Figures

Figure 2-1 A phase diagram of vanadium oxide nanostructures on Rh(111) as a function of the vanadium coverage and the substrate temperature...... 10

Figure 2-2 STM images of a) 0D, b) 1D, c) 2D, and d) 3D vanadium oxide nanostructures prepared by the PVD method...... 11

Figure 2-3 V(V) solute species in aqueous solutions as a function of pH and concentration...... 14

Figure 2-4 Molecular structure of V5+ precursors in the pH range of 2-6...... 17

Figure 2-5 Schematic illustration of the formation path of nanourchins...... 22

Figure 2-6 Possible mechanisms of information of surface area in microemulsion-mediated vanadium pentoxide...... 24

Figure 2-7 SEM images of (a) top view and (b) side view of V2O5 nanotubes; (c) TEM image of isolated V2O5 nanotubes; (d) XRD pattern of the prepared V2O5 nanotubes on Au electrode...... 26

Figure 2-8 Potential-pH diagram for vanadium species in aqueous solution when pH < 4. IV V -2.5 The limits are calculated for a total V or V concentration of 10 M. The H2O/O2 redox system is also indicated in dotted line. Reactions Eq.2-5 – Eq.2-9 are indicated by double arrows...... 28

Figure 2-9 Schematic illustration of the spray pyrolysis process...... 29

Figure 2-10 SEM images of pure V2O5 nanobelts (A) and V2O5@PAN nanostructures (B-F) synthesized under different conditions. B) pH = 2; (C, D) pH = 0; (E) pH = 1, [AN]/[APS] = 2, and (F) pH = 1, [AN]/[APS] = 4. The scale bar in the inset in Figure 2-10 E is 100 nm...... 36

xiv

Figure 2-11 STEM image of Ag-V2O5 nanofibres (scale bar = 100 nm)...... 38

Figure 3-1 Schematic diagram of the gas sensing measurement system...... 57

Figure 3-2 (a) The SEM image of the as-prepared urchin-like precursors; (b) the SEM image of the sample calcined in air at 600 °C; (c) the SEM image of the sample calcined in

N2 at 600 °C...... 59

Figure 3-3 (a) The TEM image of an individual microurchin calcined in air with chain-like structure and (b) the HRTEM image taken from one of the nanorods of the microurchins. 59

Figure 3-6 TEM images showing the effect of the pH on the VEG particle shapes and sizes: (a) pH = 1, (b) pH = 5, (c) pH = 7, and (d) pH = 10...... 63

Figure 3-7 TEM images showing the effect of the temperature on the VEG particle shapes and sizes: (a) 120 °C, and (b) 160 °C...... 63

Figure 3-8 TEM images showing the effect of the concentration of Na3VO4 on the VEG particle shapes and sizes: (a) 0.1 M, (b) 0.15 M, and (c) 0.2 M...... 64

Figure 3-9 SEM images of the effect of surfactants on the morphologies of precursors. (A&B) SDS, (C&D) PVP, (E&F) CTAB...... 65

Figure 3-10 MS spectra of the different stages during the formation of VEG. (a) The initial reaction solution without heating (t = 0 min); (b) the reaction solution heating for 5 min; (c)

The reaction solution heating for 20 min; (d) the reaction solution heating for 30 min, at this moment the precipitates deposit...... 67

Figure 3-11 FT-IR patterns of the precursor, V2O3, and V2O5...... 69

Figure 3-12 XPS pattern of carbon (C1s peak) in the product of V2O3 calcined in N2...... 70

Figure 3-13 TGA (solid line) and DTA (dot line) curves of the precursor calcined in air (a) and N2 (b)...... 72

Figure 3-14 SEM images of the products taken out of the reaction solution at different times. (A) 0.5h (the precipitate just appeared); (B) 1 h; (C) 2 h; and (D) 12 h...... 73

Figure 3-15 Schematically illustrating the growth process of the products...... 73

Figure 3-16 (A) Typical response curves of the V2O5-based sensors during cycling between increasing concentration of acetone and ambient air; and (B) sensitivity of the sensing materials to acetone...... 74

Figure 3-17 (A) Sensitivity of (A) microurchin-based sensing materials and (B) nanorod- based sensing materials to the tested gases: filled square acetone; filled down triangle isopropanol; and filled circle ammonia...... 76

Figure 4-1 (a) The XRD patterns of the products. Pattern a represents Ag2V4O11 and Pattern b corresponds to Ag0.35V2O5; (b) An overview TEM image of the Ag2V4O11 nanobelts; (c)

An overview TEM image of the Ag0.35V2O5 nanobelts...... 83

Figure 4-3 XPS patterns of (a) vanadium species and (b) Ag species in the VOx@Ag nanocomposites. Black line is original curve obtained by measurement. Green line xvi

corresponds to the background, while blue line and red line are the simulated curves corresponding to V4+ and V5+, respectively...... 85

Figure 4-4 TEM images of the products obtained with different molar ratios of Ag to vanadium: (a) 0, (b) 5%, (c) 10%, (d) 15%, (e) 50%, and (f) 100%...... 86

Figure 4-5 XRD patterns of the products obtained with different Ag/V molar ratios: (a) 0, (b) 1%, (c) 5%, (d) 10%, (e) 15%, (f) 50%, and (g) 100%...... 87

Figure 4-6 TEM images (left hand side) and XRD patterns (right hand side) of the products obtained by hydrothermal methods under various reactants. (a) NH4VO3 + SDS; (b)

NH4VO3 + SDS + Ag (10%); (c) V2O5 + Ag (10%); (d) V2O5 + PVP + Ag (10%); (e) V2O5

+ PVP; (f) V2O5 + CTAB + Ag (10%)...... 90

Figure 4-7 The TEM images of the Ag2V4O11 nanobelts obtained at different pH: (a) pH = 13, (b) pH = 10, (c) pH = 6 (original), and (d) pH = 1...... 91

Figure 4-8 XRD patterns of the Ag2V4O11 belt structures produced at different pH: (a) pH = 13, (b) pH = 10, (c) pH = 6 (original), and (d) pH = 1...... 93

Figure 4-9 HRTEM images (left-hand side) and SAED patterns (right-hand side) of the single nanobelts obtained at different pH: (a) pH = 1, (b) pH = 6 (original), and (c) pH = 10...... 94

Figure 4-10 TEM images of some typical products synthesized at different ratios: (a) low magnification of the product obtained at the ratio of 25%, (b) high magnification of (a), (c) 15%, (d) 10%, (e) 5%, (f) 3%, (g) 11.5%, (h) 1.8%, (i) 4.6%, and (j) 4.5%. The above ratios are the Ag/(V2O3+V2O5) ratios as shown in Table 4-1...... 96

Figure 4-11 TEM images of the nanocomposites prepared with different capping agents: (a) PVP and (b) SDS...... 97

xvii

Figure 4-12 TEM images of the nanocomposites prepared at different Ag/(V2O3+V2O5) ratios: (a) 15%, (b) 10%, and (c) 5%. (d) XRD patterns of the nanocomposites shown in (a), (b), and (c), while the XRD patterns a, b, and c correspond to the products shown in Figure (a), (b) and (c), respectively...... 98

Figure 4-13 Sensitivities to 100 ppm 1-butylamine of the sensors based on different materials at different working temperatures. (a) Ag2V4O11 nanobelts obtained at pH 1, (b)

Ag0.35V2O5 obtained at the Ag/V molar ratio of 15%, (c) VOx@Ag nanocomposites obtained at the Ag/total V molar ratio of 15%, and (d) the naked V2O5 nanoparticles. .... 102

Figure 4-14 Sensor response of Ag2V4O11 nanobelts to various concentration of 1- butylamine at optimized working temperature of 260 °C...... 102

Figure 4-15 Sensitivity of the as-prepared materials to various concentration of 1- butylamine testing at working temperature of 260 °C. (a) Ag0.35V2O5 nanobelts and

VOx@Ag obtained with the Ag/V ratios of 5%, 10%, and 15%; (b) Ag2V4O11 nanobelts obtained at different pH...... 104

Figure 4-16 Relative selectivity of the sensors based on the various as-prepared materials to various gases at working temperature of 260 °C, and the gas concentration is fixed at 100 ppm...... 105

Figure 5-1 (a) SEM image of as-prepared Ag nanowires; (b) XRD pattern of the Ag nanowires; (c) EDS spectrum of a single Ag nanowire...... 112

Figure 5-2 (a) A TEM image of the as-prepared Ag nanowires; (b) an enlarged TEM image of the red area in (a); (c) a TEM image of a single nanowire; (d) SAED pattern taken from an individual nanowire; (e) HRTEM image of the nanowires clearly showing the {111} lattice fringes with a spacing distance of ~2.35 Å...... 113

xviii

Figure 5-4 TEM images of silver nanoparticles obtained under different concentrations of silver nitrate: (a) 0.2 mM, (b) 1 mM, (c) 2 mM, and (d) 3 mM. The molar ratio of silver to vanadium was maintained as 5:1. The inset in (d) is a SEM image showing the radial particles...... 115

Figure 5-5 Time-dependent UV–vis spectra for tracking the formation and growth of Ag nanowires, along with blank references of AgNO3 solution and V2O3 suspension (scanning interval step is ~ 3 min)...... 117

Figure 5-6 XPS spectrum of vanadium oxides showing the redox reaction happening in the formation of silver nanowires, in which the peak of V2p3/2 (516.2 eV) could be assigned to 4+ 5+ V , while V2p3/2 (517.5 eV) to V ...... 118

Figure 5-7 TEM images of the effects of various molar ratios of surfactants on the formation of Ag nanowires: (a) CTAB:Ag+ = 2:1; (b) CTAB:Ag+ = 5:1; (c) SDS:Ag+ = 2:1; (d) SDS:Ag+ = 5:1; (e) PVP:Ag+ = 2:1 and (f) PVP:Ag+ = 5:1...... 120

Figure 5-9 (a) TEM image of tiny Ag nanoparticles showing the aggregation and formation of nanowires; (b) a magnified TEM image showing the chain-like structure of nanowires with rough surface; and (c) HRTEM image of the joint section showing the lattice matched {111} planes...... 123

xix

Figure 5-10 Morphology change of V2O3 particles before (a) and after (b) reaction (the molar ratio of silver to vanadium is 100:1)...... 125

Figure 5-11 TEM images of the nanowires prepared at different pH: (A) pH = 4 (original pH); (B) pH = 7; (C) pH = 10; and (D) XRD patterns of the nanowires prepared at different pH: (a) pH = 10; (b) pH = 7 and (c) pH = 4 (original). ▲represents Ag, while ● corresponds to Ag(VO3)H2O...... 126

Figure 5-12 (A) UV-vis spectra of as-synthesized Ag nanowires (a), as compared to Ag nanowires obtained by glycol-thermal approach (b);35 and (B) TEM image of Ag nanowires prepared by glycol-thermal approach...... 129

Figure 6-1 (a, b) TEM images of the Ag nanowires prepared in recent work; (c, d) TEM images of the Au nanoparticles (20-50 nm) attached on the V2O3 particles (200 nm × 1 μm)...... 139

Figure 6-3 (a) STEM image of Ag-Au nanowires with the Au/Ag molar ratio of 1%; (b) EDS elemental mapping of the Ag-Au bimetallic nanowires; green (c) and red (d) corresponding to elemental Au and Ag, respectively...... 141

Figure 6-4 (a, b) TEM images of Ag-Pd bimetallic nanowires; (c) Pd nanoparticles (20 nm) attached on V2O3 particles (1μm × 200 nm)...... 141

Figure 6-5 (a) A STEM image of such Ag-Pd bimetallic nanowires; (b) EDS elemental mapping of the Ag-Pd bimetallic nanowires; as shown on the left-down corner of Figure 6-5 b, green (c) and red (d) correspond to Pd and Ag, respectively...... 142

Figure 6-6 TEM images of the Ag-Au bimetallic nanostructures with different ratios of Au to Ag: (a) 0.1%, (b) 1%, (c) 5%, and (d) 10%...... 143 xx

Figure 6-7 XRD patterns of the Ag-Au bimetallic nanostructures with different molar ratios of Au to Ag: (a) 0.1%, (b) 1%, (c) 5%, and (d) 10%. ○ represents the peaks of Ag/Au; and

▲ corresponds to AgCl...... 144

Figure 6-8 EDS spectra of the Ag-Au bimetals: (a) 0.1% Au/Ag product, (b) 1% Au/Ag product, (c) 5% Au/Ag product, (d) 10% Au/Ag product. The main peak of Cl overlaps with one of Ag peaks...... 145

Figure 6-10 HRTEM images of the joint/fused part of Ag-Au nanostructures...... 149

Figure 6-11 The snapshots of the Au nanoparticles deposition simulation onto V2O3 (104) surface and Ag (111) surface at the final stage (500 ps). Yellow particles correspond to Au atoms; red particles correspond to O atoms; gray particles represent V atoms; light blue particles represent Ag atoms. The black lines show the lattice changes of Au nanoparticles after deposition on Ag surface...... 150

Figure 6-12 Simulated XRD patterns showing the structural crystallization of Au nanoparticles deposited on the Ag (111) surface (black) and the V2O3 (104) surface (red)...... 151

Figure 6-13 UV-Vis spectra of the Ag-Au bimetallic nanostructures with different molar ratios of Au to Ag: (a) 0.1%, (b) 1%, (c) 5%, (d) 10%, and (e) pure Au NPs...... 152

xxi

Figure 6-15 The relationship of ln(At/A0) with time in the catalytic reduction of 4-NP to 4- AP, corresponding to the catalysts listed in Figure 6-14...... 156

Figure 6-16 Time-dependent UV-Vis absorption spectra of the catalytic reduction of 4-NP over the Ag-Pd bimetallic nanowires with different molar ratios of Pd to Ag: (a) 0.1%, (b) 1%, (c) 5%, (d) 10%, and (e) Pd nanoparticles...... 156

Figure 6-17 The plots of ln(At/A0) versus time for the catalytic reduction of 4-NP corresponding to the nanoparticles in Figure 6-16...... 158

Figure A1-1 JCPDS file of V2O5...... 175

xxii

List of Tables

Table 2-1 Electrode process of vanadium and corresponding standard electrode potentials 0 (EA )...... 40

Table 3-1 The molecular formulas corresponding to the peaks in MS pattern...... 66

Table 4-1 The molar ratios of V2O3 to V2O5, Ag to V2O3, and Ag to total vanadium...... 95

Table 6-1 Atomic ratios of each product shown in Figure 6-8...... 145

Table 6-2 The total reduction time and kapp of the Ag nanowires (Ag NWs), the Au nanoparticles (Au NPs), and the Ag-Au nanowires with different molar ratios...... 155

Table 6-3 The total reduction time and kapp of the Pd nanoparticles, and the Ag-Pd bimetallic nanowires with different molar ratios...... 157

xxiii

Chapter 1. Introduction

1.1 Background

Vanadium is a chemical element with atomic number 23. The element exists naturally in minerals and in fossil fuel deposits. In China and Russia, it is produced from steel smelter slag, while other countries produce it either from the flue dust of heavy oil, or as a by- product of uranium mining. It is mainly used to produce specialty steel alloys such as high speed tool steels. The most important industrial vanadium compound, vanadium pentoxide, is used as a catalyst for the production of sulphuric acid.

Nanoparticles are of great scientific interest as they effectively bridge between bulk materials and atomic or molecular structures. Nanoparticles (ultrafine particles) are sized between 1 and 100 nm in at least one of its dimensions (width, thickness, and length). At nanoscale, size-dependent properties are often observed as the percentage of atoms at the surface of a material becomes significant. Moreover, the energies can be described as being quantized, in which the material behaves in discrete rather than continuous form.1

Vanadium oxide particles in nanosize have found to exhibit new and enhanced functions from the quantization effect, which gives rise to unique chemical and physical properties. These materials can be used in catalysts, sensors, electrodes in lithium ion batteries, and optical switches. Therefore, vanadium oxide nanoparticles are an important area of research to enhance its properties for practical use.

Another research area of significant importance is a nanocomposite, which is a multiphase solid material where one of the phases has one, two or three dimensions of less than 100 nm. These materials can combine two or more components’ properties or improve the performances of the substances compared to its separate form. The vanadium oxides applied in this area will create nanomaterials possessing high selectivity and sensitivity for catalysis and sensing.

1.2 Scope of Research

Nanostructures have attractive intensive research interests because of their unique optical, electronic, magnetic and physic-chemical properties that exhibit potential applications in many areas such as optoelectronics, catalysts, and gas sensors. The challenges in developing applications still remain. Of the metal oxides achieved thus far, vanadium oxides are very attractive due to their outstanding layered structural flexibilities and multivalence compositions (VOx with 1.5 ≤ x ≤ 2.5). Because of this, the vanadium oxides have found many potential applications in gas sensing, catalysts, and lithium ion batteries.2-4 The possibility to use vanadium redox couples in both half-cell, thereby eliminating the problem of cross contamination by diffusion of ions across the membrane is the advantage of vanadium redox rechargeable batteries. Lithium vanadium oxide has been proposed for use as a high energy density anode for lithium ion batteries, at 745 Wh/l when paired with a 2,3 lithium cobalt oxide cathode. Notably, V2O5 nanofibres show much lower detection limit than other oxides (e.g., SnO2, TiO2, and MoO3) to organic amines (e.g., 1-butylamine with limit of detection below 30 ppb) that are important analysis in food industry and medical diagnosis.3,5

The functional properties and advanced applications of V2O5 nanoparticles are strongly shape/size-dependent. Various methods have been used to control the synthesis of V2O5 nanoparticles (e.g., rods, fibres) and nanostructures for unique functional properties, such as 6,7 sol-gel process via “chimie douce” colloidal assembly of VO(OR)3 reverse micelles for 8 V2O5 nanorods, a sol-gel reaction followed by hydrothermal treatment for V2O5 nanotubes,9,10 and nanobelts.11 To achieve high sensitivity and selectivity, chemically tailoring the V2O5 nanostructures by metal doping or surface modifications, has been recently studied. Unfortunately, these methods have shown some success yet limitations.

For example, the V2O5 gel is short-range order structures that are difficult to characterize accurately. The true chemical composition of such a gel is not obvious. The wide size distribution and the random metal doping distribution reduce the performance of V2O5

nanoparticles in functioning such as gas sensing. Therefore, there is an unequivocal need to vanadium oxide nanoparticles with shape and functionality control.

To meet some of the challenges of the vanadium oxides and their composites, the research aims of this project can be demonstrated as below:

1) To understand the growth mechanisms of certain vanadium oxides in order to produce shape-controlled nanoparticles.

2) To study the interactions of Ag ions and vanadium oxides for generating nanocomposite materials with multifucntionalities.

3) To examine functional properties of the nanoparticles and nanocomposites in gas sensors, reducing agents, and catalysts.

In Chapter 2, the recent research of vanadium oxide materials were reviewed, mainly concentrating on the synthesis methods, nanocomposites formation, properties, and applications.

In Chapter 3, the use of the polythermal method was found to produce highly uniform microurchins, nanorods, and microspheres of the vanadium oxide precursors (vanadyl ethylene glycolate). With calcination of the precursors in atmospheres of air and N2, the precursors could be converted to V2O5 and V2O3, respectively. The various shapes could be easily adjusted by the reaction time. In this study, the formation of vanadyl ethylene glycolate was investigated by mass spectra. Furthermore, the growth mechanism of the morphology changes was also studied by time-dependent experiments. Finally, the V2O5 nanoparticles were tested for detection of gases such as acetone, isopropanol, and ammonia. The sensing performances of microurchins and nanorods were compared.

In Chapter 4, the nanocomposites (e.g., Ag@VOx, Ag0.35V2O5, Ag2V4O11 and

Ag@Ag2V4O11) were prepared by the hydrothermal method. The different composites could be obtained by changing the molar ratios of Ag to vanadium and temperature. For this reason, the effects of parameters such as temperatures, Ag/V molar ratios, pH, and 3

various surfactants were systematically investigated. For Ag2V4O11, the pH could lead to the appearance of Ag nanoparticles attached on the surface of Ag2V4O11 nanobelts, while the amount of Ag nanoparticles could increase with pH. Finally, the sensing performances of the three materials with various amount of Ag to organic amines were examined. Results show that the sensitivities of the three materials to organic amines increased with the amount of Ag. Moreover, Ag2V4O11 and its nanocomposites exhibited much higher sensitivities than the other two materials probably due to good conductivity and the coordination of Ag and amine groups.

In Chapter 5, the V2O3 particles were found as reducing agents and catalysts to quickly prepare silver nanowires in aqueous solution at room temperature. The novel approach could synthesize nanowires with diameter of 20 nm and length up to a few micrometers. The possible growth mechanism of the Ag nanowires was proposed as oriented attachment of Ag nanocrystals. In this study, V2O3 rods were used for synthesis of silver nanowires for the first time, playing multiple roles as reducing agent, template, and catalyst. Finally, the Ag nanowires showed high catalytic performance for reduction of 4-nitrophenal.

Chapter 6 focused on the reducing properties of V2O3 which were further developed in the synthesis of Ag-Au bimetallic nanowires. This newly developed wet-chemical method was also carried out in aqueous solution at room temperature. The Ag nanowires and Au nanoparticles were sequentially formed by reduction with V2O3 particles, while the Au nanoparticles with ~ 5 nm in diameters attached on the surface of Ag nanowires. The formation and growth mechanisms of the nanostructures were proposed to be driven by thermodynamics and lattice match. Moreover, the bimetallic nanowires exhibited catalytic properties superior to those of Ag nanowires, Au nanoparticles, and Ag-Pd bimetallic nanostructures prepared under the reported conditions. The final comments were finally concluded in Chapter 7.

Chapter 2. Literature review

2.1 Vanadium oxides

Vanadium oxides belong to the important class of early 3d transition metal compounds12 which have received considerable attention in research and technology over the last decades. In bulk form, vanadium oxides display different oxidation states and V–O coordination numbers and exhibit a broad variety of electronic, magnetic and structural properties13,14 which make these materials attractive for many industrial applications. Prominent examples range from the area of catalysis, where vanadium oxides are used as components of important industrial catalysts for oxidation reactions15 and environment pollution control,16 to optoelectronics, for the construction of electrical switching devices17 and thermochromic smart windows.18

When the dimensionality of oxide materials is reduced, new properties, either local or collective, may appear. Nanometric oxide structures in the form of ultrathin layers, line elements, quantum dots and three-dimensional clusters are of significant importance as passive and active elements in different areas of the emerging nanotechnologies. Examples comprise electronic, optoelectronic and magneto-electronic device technology, sensing devices, heat- and corrosion-resistive coatings or the field of advanced catalysis. In all these areas the controlled fabrication of nanostructure oxide phases is required.

Vanadium oxide mainly contains three species, vanadium trioxide (V2O3), vanadium dioxide (VO2), and vanadium pentoxide (V2O5), which differs in color, molecular structure, and properties. Every oxide phase can be transferred by reduction, oxidation, and calcination environment. This section will briefly introduce the basic properties of vanadium oxides and phase transformation among these oxides.

2.1.1 Vanadium pentoxide

Vanadium pentoxide is the highest oxidation state, which is the most stable oxide in the V- O system. It crystallizes with an orthorhombic unit cell structure belonging to Pmnm space 5

group with lattice parameters of a = 11.510 Å, b = 3.563 Å and c = 4.369 Å. It has a layer- like structure and it is built up from distorted trigonal bipyramidal coordination polyhedral of O atoms around V atoms, that share edges to form (V2O4)n zigzag double chains along the (001) direction and are cross linked along (100) through shared corners. The distorted polyhedral have a short vanadyl bond (1.58 Å) and four O atoms located in the basal plane at distances ranging from 1.78 to 2.02 Å. The sixth O atom in the coordination polyhedral of V lies along the vertical axis opposite to the V–O bond at a distance of 2.79 Å.19

Vanadium pentoxide is famous in the manufacture of sulphuric acid, an important industrial chemical. Vanadium (V) serves the crucial purpose of catalysing the mildly exothermic oxidation of sulphur dioxide to sulphur trioxide by air. Other catalysis including maleic 20 anhydride, phthalic anhydride, as well as NOx (selective catalytic reduction) are also widely reported.

Vanadium pentoxide is also applied in lithium ion batteries due to its layer structure, and gas sensors due to its semiconductor electronic property. However, the performance of

V2O5 is not ideal. The common way to improve the performance is to prepare nanocomposites.

2.1.2 Vanadium dioxide

Vanadium dioxide can be divided into three phases, including monoclinic (M), rutile (R), and B phase. The three phases can be transferred each other by calcination at different temperatures. B phase can be synthesized by hydrothermal method, 21-24 which was the most widely reported phase. VO2 rutile can be produced by calcination of B phase between 400-480 °C. The temperature of transformation between rutile and monoclinic phase is quite low, around 68 °C. This transformation is a very important application of VO2 and has attracted great interest over the last several decades because of their characteristic near- room-temperature insulator–metal phase transition.25,26 This phase transition is accompanied by an ultrafast four-to-five orders of magnitude change in electrical conductivity and a dramatic diminution of optical transmittance across the visible and

infrared regions of the electromagnetic spectrum.25,27,28 A structural phase transition from a monoclinic to a tetragonal rutile polymorph generally accompanies this insulator– metal phase transition.

The origin of the phase transition remains controversial (Peierls distortion vs. Mott– Hubbard),29,30 but the recent demonstration of a thermally decoupled electric field-induced phase transition in VO2 thin films underlines the remarkable potential of this material for integration in nanoscale electronic devices such as Mott field-effect transistors.31

In addition, according to the Peierls electron-lattice distortion picture, the structural phase transition in VO2 occurs at around 67 °C with an ultrafast transformation of the semiconducting monoclinic structure to a metallic rutile structure.32 The tetragonal structure formed above the phase transition temperature is composed of adjacent VO6 octahedra that share edges along the c-axis forming a close-packed hexagonal lattice that brings adjacent vanadium atoms into close proximity.33 Below the phase-transition temperature, a semiconducting monoclinic phase is existent wherein the vanadium atoms form dimmers as part of a more expanded crystal lattice with alternating short and long V– V bonds. Electron hopping between the d1 V sites is thus considerably easier in the high- temperature polymorph.27,29,34

2.1.3 Vanadium trioxide

V2O3 is another metal–insulator transition (MIT) material in family of vanadium, except for

VO2. However, this transformation is single and the temperature is around 160 K, changing from a low-temperature antiferromagnetic insulator to a high-temperature paramagnetic phase.35 The transition temperature may be controlled by changing the proportion of the doped metals such as Cr36, Al37, and Mo.38 This property has been used for some functional 39,40 devices such as temperature sensors and current regulation. Furthermore, V2O3 powder can be used in conductive polymer composites41 and in catalysts.42

In the past years, many methods for preparing V2O3 powder have been studied. Spherical 43 V2O3 particles were prepared by the O2–H2 flame of V2O3 at 2000 °C; spherical and 7

necking V2O3 powder was synthesized by reducing V2O5 obtained by evaporative 44 decomposition of solutions in H2 flow at 850 °C for 6 h. Furthermore, V2O3 powder was prepared by pyrolyzing the hydrazine-containing vanadium salt45 and reducing the sol–gel- 46 synthesized V2O5 in H2 stream. However, only micropowder could be obtained, and these methods are limited by high temperature.

2.1.4 Transformation among three oxides

Theoretically, the three oxides can be transferred one another by simply reduction and oxidation. Since the reduction process much more easily control the phase transformation than the oxidation process by adjusting the concentration of the reducing gas, it is widely used. The reduction process always starts from V2O5, in which vanadium is the highest valence and most stable phase of vanadium oxide in air. For example, Corr et al.47 demonstrated that VO2 (B) nanorods can be solvothermally prepared by reduction of V2O5 via formaldehyde or isopropanol at near room temperature. Furthermore, the metastable

VO2 (B) nanorods can be converted to rutile VO2 by heating in argon, and to corundum

V2O3 by reducing in 5% H2: 95% N2, respectively. Su et al. reported that V2O5 can transfer -7 to V2O3 via VO2 in a specimen chamber of a TEM by heating 600 °C in vacuum at 10

Torr. The initial decomposition of V2O5 to V2O3 is followed by a combination of diffusion, coalescence, and stabilisation processes.

The oxidation process is normally achieved by calcination of low-valence vanadium oxide in air, usually starting from V2O3. In this process, VO2 as an intermediate valence is difficult to be obtained by direct oxidation process. Liu et al.48 demonstrated that the yolk- shell V2O5 microspheres can be prepared by calcination of yolk-shell V2O3 microspheres at

400 °C for 2 h. The V2O3 microspheres were prepared by thermal treatment of mixture of vanadium (IV) acetylaceton and N,N-dimethylformamide in autoclave at 220 °C for 24 h.

Another application of the oxidation process is to avoid the structure decomposition when preparing V2O5 by calcination vanadium organic precursors. The direct calcination of such precursors in air always companies with rapid loss of water/solvent and strong exothermic

reaction which can cause the decomposition of the structure.49 However, the process from organic precursors to V2O3 then to V2O5 is a relevant peaceful exothermic reaction when the precursors were calcined in reducing atmosphere. The gradual loss of water/solvent in reducing atmosphere is essential to retain the structure of V2O3, followed by a simple oxidation transformation from V2O3 to V2O5. Wang and co-workers compared the two ways to prepare V2O5 nanoparticles from precursor. Results show that the way through

V2O3 could retain the spherical nanostructures, whereas the direct calcination in air caused 50 significant structure collapse. Therefore, preparation of V2O5 nanoparticles from precursors through the intermediate product (V2O3) is proved as a better way to retain the structure of final products.

2.2 Synthesis Methods

A variety of methods can be used to synthesize vanadium oxide nanoparticles. By changing the phase, sources of energies, raw chemicals, and other parameters, vanadium oxide nanoparticles can be prepared in different sizes and shapes depending on the method. The method for synthesis of nanoparticles can be classified into two types: physical and chemical. In the synthesis of vanadium oxide, the chemical methods are more widely used to control the size and shapes of nanoparticles, compared to physical methods. The benefit of the chemical method is that the nanoparticles prepared from basic materials are much easily controlled, because there can be control over the amount of chemicals and the energy input to the system. Therefore, the size, shape, and structure of the final products are much more distinct. Besides, the low cost and the ambient synthetic conditions of chemical methods are another attracting reason for widely use.

2.2.1 Physical methods of vanadium oxides

2.2.1.1 Physical vapour deposition (PVD) PVD is one of the most common physical synthetic methods for growing vanadium oxide nanoparticles. Normally, a certain surface of prototypical metal can be as a deposition substrate (e.g. Au(111),51 Cu(100),52 Pt(111),53 Pd(111),54 and Rh(111) surface55). By

balancing the thermodynamic and kinetic forces that drive the chemistry and the mass transport associated with the formation of the oxide layer, the vanadium oxide nanostructures can be formed in various dimensionalities and a high degree of structure order. The thermodynamic forces are given by the chosen materials and include the surface and interface energies of the metal (oxide) surface and the oxide-metal interface, whereas the kinetic effects can be adjusted by varying the experimental parameters, such as substrate temperature, oxygen pressure and evaporation rate.

Schoiswohl et al. reported four types of well-defined vanadium oxide nanostructures prepared by the PVD method on Rh(111) surface.55-59 The nanoparticles included: (i) zero- dimensional (0D) nanostructures, which was the molecular oxide clusters with dimensions < 1 nm; (ii) 1D nanostructure, which was the elongated oxide islands with a width of only a few unit cells and an aspect ratio of ~ 10; (iii) 2D nanostructures, which was the ultrathin oxide layers with a thickness of less than 1 nm; (iv) 3D nanostructures, with an average size of less than 10 nm.

Figure 2-1 A phase diagram of vanadium oxide nanostructures on Rh(111) as a function of the vanadium coverage and the substrate temperature.59

Figure 2-2 STM images of a) 0D, b) 1D, c) 2D, and d) 3D vanadium oxide nanostructures prepared by the PVD method.59 A phase diagram of the vanadium oxide/Rh(111) system in terms of different oxide structures with reduced dimensions is presented in Figure 2-1. The vanadium oxide coverage is given in monolayer equivalents (MLE), where 1 MLE contains the same number of vanadium atoms as one monolayer of Rh(111) atoms. From this diagram, it can be seen that the formation of 0D nanostructures could be formed by deposition of small (< 0.2 ML) amounts of vanadium metal onto Rh(111) surface pre-covered with oxygen, followed by a low-temperature (< 250 °C) flash. With the higher temperatures (250 °C) during the flash annealing, the oxide clusters were elongated along the main azimuthal directions of the substrate, caused by the aggregation into oxide islands. The 1D nanostructures with an aspect ratio of up to 10 were temporarily stable on the Rh(111) surface and quickly transform into 2D islands when annealed to high temperatures. The 2D morphology could be obtained in a relatively broaden range of coverage (0.2-0.8 MLE). At 11

high MLE level (> 1), the oxide growth mode changed from 2D to 3D. Finally, the regular 3D morphology could be obtained at high temperature and high level of vanadium coverage. The scanning tunnelling microscopy (STM) images of different dimensions (0D-3D) are displayed in Figure 2-2.

Except for the substrate of Rh(111), the vanadium oxide nanostructures can also be deposited on other metal substrates. For example, Lewis et al. demonstrated that ordered vanadium oxide films which were formed when an evaporated layer of metallic vanadium was annealed in ultra-high vacuum (UHV) could be grown on an Au(111) substrate. A Cu(100) surface with and without sodium on the surface as a substrate to prepare vanadium oxide overlayers by PVD method was reported by Kishi et al.52 The similar scenario 54 60 61 happened in the substrates of Pd(111), TiO2(110), and Al2O3(0001).

2.2.1.2 Pulsed laser deposition Pulsed laser deposition (PLD) is a thin film deposition technique which utilize a high power pulsed laser beam to strike a target of the material that is to be deposited in a vacuum chamber. The material is vaporized from the target which deposits as a thin film on a substrate (e.g., indium tin oxide (ITO) glass or silicon wafer). The process can occur in ultra high vacuum or in the presence of a background gas, such as oxygen which is commonly used when depositing oxides to fully oxygenate the deposited films, especially for V2O5.

Recently, Lida et al. reported that V2O5 thin films were prepared using pulsed laser 62 deposition. In this study, commercial V2O5 powders were pressed into a pellet of diameter 22.5 mm with thickness of 2 mm at the pressure of 30 kg/cm-2. A KrF excimer laser with a wavelength of 248 nm was used to ablate the V2O5 pellet. The substrate was ITO glass with the distance of 40 mm from the pellet and the temperature of the substrate was varied in the range from 423 to 723 K when deposition. During the ablation, the pellet was rotated to avoid the depletion of the material at same spot. The pulse repletion rate was 10 Hz, and the shot number of ablation was kept as 18000 (30 min) for the deposition film. The deposition chamber was evacuated to a base pressure of 39.9 mPa before the deposition. An oxygen 12

partial pressure was also varied in the range from 6.7 to 26.7 Pa. The thickness of the laser- ablated V2O5 thin film was calculated as about 0.3 μm. The roughness of the thin films was varied with the deposited temperatures and oxygen partial pressures.

The surface structures of the films are largely dependent on growth temperatures, compared with the oxygen partial pressures.62-64 For example, Ramana et al. demonstrated the 63 morphology of V2O5 thin films as a function of growth temperature. For deposited temperature lower than 200 °C, the V2O5 thin film did not show characteristic features and were amorphous. The nanostructure grain growth appeared at 200 °C. At this temperature, the film was composed of spherical particles varying in size and the average grain size was 50 nm. The grain size increased with further increase in temperature and the nearly spherical shape of the grains of V2O5 thin films gradually changed to a long and rectangular shape when the temperature was increased to 500 °C.

2.2.2 Chemical methods of vanadium oxides

2.2.2.1 Hydrothermal methods Hydrothermal methods is one of the effective routes to the fabrication of high quality vanadium oxide nanoparticles, utilizing a solvent under pressures and temperatures close to its critical point to increase the solubility of a solid and to speed up reactions between solids. Particularly, it provides a commonly used method for producing 1D nanoparticles.65 The major advantage of this approach is that most materials can be made soluble in a proper solvent by heating and pressuring the system close to its critical point, and it should be well-suited for use with solid materials. The process mechanism can be summarized as below:

1) Vanadium (V) species in aqueous solutions

The aqueous chemistry of V(V) has been extensively studied and a large variety of molecular species were described in the literature.66 At room temperature, the molecular species mainly depend on vanadium concentration and pH, as shown in Figure 2-3.

3- V(V) is a highly charged cation, so that oxo-anions (VO4) in which vanadium is surrounded by four equivalent oxygen atoms are formed in highly alkaline aqueous solutions (pH > 14). Protonation occurs as the pH decreases, giving rise to hydrolysed 3-n -4 monomeric species (HnVO4) in diluted solutions (c < 10 M). The mean negative charge of these anionic species decreases with the pH, as n increases, down to the point of zero + charge (pH = 2). Below that pH value, monomeric cationic species (VO2) are observed (Figure 2-3).

As demonstrated by Livage, two basic processes, hydrolysis and condensation, occur when vanadium salt is dissolved in water.67

Figure 2-3 V(V) solute species in aqueous solutions as a function of pH and concentration. Hydrolysis

In aqueous solutions, V5+ ions are dissolved as a form of dipolar water molecules giving 5+ hydrated [V(OH2)n] species. However, due to the strong polarizing power of the small

and highly charged V5+ ion, the solution becomes more acidic by the partially deprotonated coordinated water molecules, as described by the following equation:

V OH ])([ 5 hH  VO OHOH ])()([ h)5(   hH O Eq. 2-1 62 2 h 62 h 3

The “partial charge model” based on the electronegativity equalization principle of R.T. Sanderson can be used to describe the relationship between the hydrolysis ratio ‘h’ and pH. The increase of ‘h’ with pH can lead to the formation of aquo, hydroxo and oxo species.68 + Based on this model, the V(V) precursor corresponding to [V(OH)4(OH2)2] can be found at low pH (pH < 2). However, some internal proton transfer occurs between V-OH groups in order to decrease the positive charge of vanadium. This leads to the formation of vanadyl + + cations [VO2(OH2)4] or (VO2) in which two V = O double bonds are formed in cis positions.

The coordination number of V(V) decreases from 6 to 4 above pH = 6. The transformation of the “σ” electron from the coordinated water molecules toward the empty d orbital of V(V) ions increases with deprotonation and the positive partial charge of vanadium decreases. V- O bonds become more covalent. The decrease of vanadium coordination results in four-fold (3-n)- coordinated vanadates species [HnVO4] . This change can be easily seen with naked eyes. The colour of V5+ (3d0) ions is caused by the electron transformation from the oxygen bonding orbital to the empty vanadium “d” orbital. With the crystal field splitting of “d” orbital decreasing, the charge transfer bands move toward UV. Six-fold coordinated V(V) decavandium solutions are typically orange while four-fold coordinated metavanadate are 69 3- (3-n)- colourless. The deprotonated species (VO4) from tetrahedral species (HnVO4) are formed with pH increasing above 12.

Condensation

Condensation happens at high vanadium concentration (Figure 2-3). The process includes two main reactions.

Olation V OH  V OH2  V OH ------ 2OHV Eq. 2-2

Oxolation V HOOH  ------ 2OHVOVV Eq. 2-3

The two reactions both involve the nucleophilic addition of negatively charged OHδ- groups onto positive vanadium cations Vδ+. Hydroxo V-OH groups are needed to initialize the condensation reactions. Olation reactions, in which unstable water molecules are already formed, are kinetically faster than oxolation.

+ At low pH, the main form of the vanadium species is (VO2) , which is difficult to form condensed species. To make precipitation occur, anions are required to be added. An + example is vanadyl phosphates (VOPO4∙nH2O), which was obtained by (VO2) cations surrounded by phosphate anions. Under this circumstance, V-O-V bonds were not 70 formed. With pH increase, V2O5 precipitates at the Point of Zero Charge (PZC) (around 0 pH ≈ 2), resulting from the polycondensation of the neutral precursor [VO(OH)3(OH2)2] . According to the molecular structure, condensation unlikely occurs along the z direction

O=V-OH2, whereas V-OH bonds are presented only within xy plane (Figure 2-4). Fast olation reactions along x direction lead to chains of edge sharing (VO5) square pyramids, and then the chains are linked by oxolation reactions along y direction, causing the formation of vanadium oxide layer structures made of corner sharing double chains

(orthorhombic V2O5). This two-steps mechanism can be used to explain the ribbon-like 71-73 structure of vanadium oxide gels V2O5∙nH2O.

(6-n)- Above pH= 2, negatively charged decavanadate clusters (HnV10O28) , made of 10 edge- sharing (VO6) octahedral, are formed. In this structure, closed clusters in which V=O bonds point radially toward outside are made of small and highly charged V5+ ions with polarized terminal O2- ligands. Due to the strong acidic behaviours, the further condensation of decavanadates does not occur under ambient conditions. Polyvanadate solid phases are precipitated in the presence of cations.74

Figure 2-4 Molecular structure of V5+ precursors in the pH range of 2-6. Vanadium coordination decreases with pH increase to 6, while the main form of vanadate is (3-n)- tetrahedral anionic precursors (HnVO4) . Around pH 6-9, cyclic or chain-like - metavanadates are condensed from difunctional precursors (H2VO4) . Chain metavanadates 4- 51 are formed in the solid state, whereas cyclic species such as (V4O12) evidenced by V and 17 75 O NMR are usually observed in solution. For example, KVO3 is composed of single chains of corner-sharing (VO4) tetrahedral, and KVO3∙H2O contains double chains of edge- (3-n)- sharing (VO5) trigonal bipyramids. Their formation from cyclic (HnV4O12) solute precursors can be described via a ring opening polymerization mechanism. A phase transition from cycles to chains can be observed in the solid state of ter-butylammonium metavanadate. Depending on temperature, both cyclic [(CH3)3CNH3]4[V4O12] and chain 76 [(CH3)3CNH3][VO3] metavanadates can be prepared in aqueous solutions.

2- (HVO4) is the deprotonated precursor above pH 9. The condensation of these 4- monofunctional species only happens in dimeric pyrovanadates (V2O7) made of two 3- corner-sharing tetrahedral. Fully deprotonated oxo-anions (VO4) are observed at very high

pH. There is no functional V-OH group and V-O-V bonds cannot be formed. Only orthovanadates made of isolated (VO4) tetrahedral can be obtained above pH 12.

Evolution of molecular precursors in hydrothermal process

Under hydrothermal conditions, the correlation between the molecular structure of vanadate precursors in solution and vanadate anions in solid precipitate is totally different from the case at room temperature. The molecular structure of V(V) precursors in aqueous solutions does not depend only on pH, but also on temperature.77 The progressive transformation of 4- decavanadate solutions to cyclic metavanadates (V4O12) by heating can be confirmed by 51V NMR experiments. Only metavanadates are observed around 200 °C, probably caused by the dissociation of decavanadates into metavanadates by heating, as described by the following equation:

6  )(54)(2 4 12HOVOHOVH  Eq. 2-4 2 2810 2 4 12

This is a reversible reaction. Decavanadates can be observed again when decreasing temperature. Livage67 suggested that, in dilute solution, the deprotonation of the neutral 0 - precursor [VO(OH)3(OH2)5] led to anionic species such as [VO(OH)4(H2O)] that would (3-n)- gave more or less protonated tetrahedral vanadate anions (HnVO4) by dehydration.

Through hydrothermal treatment, solid phases built of (VO5) polyhedral were formed via - the polycondensation of the intermediate molecular precursor [VO(OH)4(H2O)] . Its molecular structure indicates that oxolation reactions only occur in the xy plane, along the four V-OH bonds leading to 2D layered compounds instead of ribbon-like particles.

Reduction also exist during the thermal transformation of tetramethyl ammonium 78 decavanadates into layered TMA[V4O10] oxide. The transformation cannot be found in inorganic sodium decavanadate, since only stable oxide phases NaVO3 and NaV3O8 can be obtained by directly heating sodium decavanadate. The transformation of

(TMA)4(H2V10O28)∙4H2O into TMA(V4O10) only occurs in the presence of organic cations. Furthermore, it is found that the transformation would not occur until some V5+ are reduced

into V4+. The reduction is contributed to condensation and coordination expansion, because of the partial decomposition of organic species.

2) Vanadium oxides Nanostructures

A large number of vanadium oxide nanostructures have been synthesized by hydrothermal treatment of aqueous solution of V(V) precursors. Their morphologies are usually related to the layered structure of orthorhombic V2O5. Therefore, 1D and 2D, even 3D nanostructures, such as nanowires, nanorods, nanoribbons, nanosheets and microurchins are typically reported in the literature. Temperature and pH are the main parameters to control the 79 morphologies of these V2O5 based nanomaterials.

1D nanostructure

A large variety of 1D nanostructures of vanadium oxide can be prepared over past decade, depending on the experimental conditions (pH and temperature). The morphologies include nanowires,80-82 nanorods,8 nanobelts, 83,84 and nanotubes.85

An example of long belt-like V2O5 nanowires was reported by Pan et al. who demonstrated 86 a hydrothermal route for synthesis of V2O5 nanowires at 180 °C for 5 h. The product was mainly composed of layered V2O5∙0.3H2O with a small amount of V2O5∙xH2O (0.3 < x <

1.7) and V2O5. The well-crystallized nanowires was several tens of micrometers long and tens of nanometres wide, growing along [010] direction.

Nanorods are another 1D nanostructure which is lower aspect ratios compared to nanowires. One example of vanadium oxides was reported by Pavasupree et al. who demonstrated that vanadium oxide nanorods with high crystallinity and high surface area were prepared by the hydrothermal method via laurylamine hydrochloride, metal alkoxide and acetylacetone.

The product was uniform B phase VO2 nanorods with the width of 40-80 nm and lengths of

1 μm. After calcination in air at 400 °C for 4 h, V2O5 rod-like structure with the width of 100-500 nm and the lengths of 1-10 μm were obtained.87

Nanotubes have recently attracted more and more attention due to the novelty of the morphology. Vanadium oxide nanotubes are firstly reported by Nesper and co-workers.88 The nanotubes were composed of multiwall tubular structures with several μm in length. The walls contained up to 30 vanadium oxide layers, giving an outer diameter up to 100 nm.

The preparation of the VOX nanotubes normally utilizes hydrothermal methods (~ 180 °C, a few days) in the presence of long chain alkyl amines like HDA (HexDecyl Amine) as templates. The template organic molecules could be removed without breaking the structure of the nanotubes that remained redox-active allowing the reversible insertion of Li+ ions.89- 91 A number of precursors was used such as V2O5, VO(OR)3, VOCl3, NH4VO3, and gels

V2O5∙n H2O. In all cases, layered vanadium oxides were formed in the aqueous solution, and then rolled up into nanotubes.92 A mixture of vanadium oxide nanotubes and unrolled vanadium oxide sheets could be observed during the hydrothermal treatment, confirming the link between nanotubes and exfoliated lamellar intermediates.93 The number of plate- like particles progressively decreased with time while more nanotubes were formed and only nanotubes were observed after a few days.

Two main chemical processes are involved in the formation of VOx nanotubes, including the intercalation of organic molecules between the oxide layers and the reduction of V5+ ions. Intercalation is an important step during the formation of nanotubes. The size of intercalated organic molecules is also important to decrease interaction between oxide layers. Vanadium oxide nanotubes about 120 nm in diameter were observed in the V2O5- 94 HDA system while only nanorods with diameter of 20 nm were formed with V2O5-EtOH. Due to the existence of organic molecules, the reduction of the oxide also appears, indicated by the colour transformation of samples from orange to green and black. This leads to the formation of a mixed valence compound containing both V5+ and V4+ ions (about 50%).95 Reversely, VOx nanotubes can be prepared by the oxidation of V4+ 96 4+ 5+ solutions. Because of the much bigger radius of V compared to that of V (rV4+ = 0.85

Å, rV5+ = 0.49 Å), a significant curvature of the oxide layers can be formed.

2D nanostructures

The formation mechanism of 2D vanadium oxides has been presented in previous section. A good example of the 2D nanostructure was reported by Pang et al. who synthesised the polyaniline-vanadium oxide nanosheets by in situ intercalation polymerization of aniline 97 with layered V2O5 under hydrothermal conditions. The thickness of the nanosheets was between 10 and 20 nm, and the typical lateral dimensions of the nanosheets were in the range of hundreds of nanometres to several microns. Moreover, the thickness of nanosheets decreased with the increase of the concentration of aniline.

3D nanostructures

3D nanostructures of vanadium oxides such as hollow microspheres, and microurchins are obtained via the self assembly of the 1D structure.

Hollow V2O5 microspheres were obtained via the self-assembly of nanorods with a diameter of 200 nm and a length of 2 μm.98 The assembly into hollow microspheres occurred when vanadium acetylacetonate [V(acac)3] was heated with ethylene glycol (EG) in the presence of PVP. The formation of rod-like vanadium precursors could be explained by the coordination of EG giving a vanadium glycol followed by an oligomerization reaction. PVP acted as a template in which vinyl groups are hydrophobic while carbonyl groups are hydrophilic, leading to the formation of micelles. Vanadium oxide microspheres with closely packed particles and aligned rods with radial structure could be also formed in the presence of oxalic acid.99

Rose-like nanostructure vanadium oxide films were synthesized from water-ethanol i solutions of vanadium tri-isopropoxide VO(OPr )3, with hexadecylamine as a template. The films were obtained by drop-casting of the solution onto Si wafers. Such structure was made of radially packed petal-like spherical with 40 nm in diameter. The films exhibited photo-induced hydrophilic properties which was the reversible transformation between superhydropholic and superhydrophilic under UV irradiation and dark storage respectively.100 Such flower-like nanostructures are not limited to V(V) oxides, also

including reduced vanadium oxides such as VO2 petaloid cluster or VOOH hollow dandelions.101,102

Another interesting nanostructure of vanadium oxides reported during the past few years are the spherical clusters looking like ‘nanourchins’, which were obtained via the hydrothermal treatment of an ethanol solution of vanadium tri-isopropoxide and alkyl amines.103 A layered compound was first formed with alkyl amines intercalated between the oxide layers, then lamina self-organize into a fan-like laminar structure, because of the presence of non-intercalated free amines. Finally, hollow nanotubes with the length of several micrometers were formed by rolling-up of the laminar structure. The tube walls were made of vanadium oxide layers with intercalated organic surfactant molecules. The redial self-organization of VOx nanotubes led to the formation of ‘nanourchins’, as shown in Figure 2-5.

Figure 2-5 Schematic illustration of the formation path of nanourchins.

2.2.2.2 Microemulsion Method Microemulsion systems have been widely used as ideal media to prepare nanoparticles. This method has been extensively used for synthesis of nanoparticles because of its simplicity in preparation and control of particle size and shape. A water-in-oil (w/o) microemulsion is a transparent and isotropic liquid medium with nanosized water pools dispersed in a continuous phase and can be stabilized by surfactant and co-surfactant molecules at the water-oil interface. The water pool provides an ideal microreactor for the formation of nanoparticles. For shape control, a nanoparticle grows randomly under normal conditions, normally leading to spherical particles due to the minimization of its surface area, while the nanoparticle size can be tunable by the size of reverse micelles which can be controlled by adjusting the ratios of surfactant, co-surfactant, oil, water, and raw materials.

By this method, BiVO4 spherical nanoparticle precursors could be prepared using a typical quaternary water-in-oil microemulsion of TX-100/n-hexyl alcohol/ cyclohexane/water.104 In this study, the particle size could be controlled by changing the amount of water. With the molar ratio of water to surfactant increased from 3.5:1, 7:1 and 14:1 (other parameters were fixed), the precursor size increased from 5 nm, 20 nm, to 200 nm. The BiVO4 nanocrystals were obtained by calcination of the precursors.

Although in most case the microemulsion method results in spherical nanoparticles, it also can be used to prepare some 1D or 2D nanostructures under certain conditions, such as ZnO,105,106 FeOOH,107 TiPO,108 and so on. Regarding to vanadium oxides, Cussler et al. demonstrated 2D microporous vanadium pentoxide prepared by this method.109,110 In the study, ribbon-like precursors was prepared by adding vanadyl isopropoxide into microemulsion of Aerosol OT (AOT), toluene, and water. And then, by different drying routes, three samples with different pore size could be obtained. The possible formation mechanism of surface area in microemulsion-mediated vanadium pentoxide is shown in Figure 2-6. The starting material was a suspension of vanadium pentoxide clusters coated with AOT surfactant. In route 1, most of surfactant was removed by washing with acetone. During drying, the clusters reacted upon contact due to the high reaction rate in vanadium

system, finally forming highly porous structure. In route 2, the colloid still contained surfactant during the drying step. The coating of surfactant could lower the effective reaction rate of hydroxyl groups on the vanadyl clusters. The following acetone removed the surfactant within the vanadium pentoxide film, allowing reactions to occur between clusters. This led to the formation of less porous structure. In route 3, washing with water may hydrolyse some of the V-O-V bonds formed within clusters, then collapsed the structures. Simultaneously, excess water stopped condensation reactions between clusters. As a result, capillary forces collapsed the structure when drying, finally resulting in the non-porous material.

Figure 2-6 Possible mechanisms of information of surface area in microemulsion-mediated vanadium pentoxide.

2.2.2.3 Electrodeposition method This process is also known as "electroplating" and is typically restricted to electrically conductive materials. There are basically two technologies for plating: electroplating and electroless plating. In the electroplating process the substrate is placed in a liquid solution (electrolyte). When an electrical potential is applied between a conducting area on the substrate and a counter electrode (usually platinum) in the liquid, a chemical redox process takes place resulting in the formation of a layer of material on the substrate and usually some gas generation at the counter electrode. In the electroless plating process a more complex chemical solution is used, in which deposition happens spontaneously on any surface which forms a sufficiently high electrochemical potential with the solution. This process is desirable since it does not require any external electrical potential and contact to the substrate during processing. Unfortunately, it is also more difficult to control with regards to film thickness and uniformity. Currently, the former one is most widely used in vanadium oxide synthesis.

By this method, some 0D or 1D morphologies of vanadium oxide nanoparticles can be controlled, such as nanospheres,111 nanofibres,112,113 nanorods114 and nanotubes.115 In addition, this is a useful method for preparation of vanadium oxide thin films.116 However, the most exciting highlight of this method is to prepare ordered 1D array of vanadium oxides with assistance of hard templates.

One of good examples for a nanotube array of vanadium pentoxide was reported by Cao’s group who demonstrated that the nanotube array could be prepared by electrodeposition method assisted by the templates of hydrophilic polycarbonate (PC) membrane with pore diameters of 200 nm and thickness of 10 μm.117 The electrolyte was composed of the aqueous solution of 0.1 M VOSO4 and 0.03 M H2SO4. The solution was blue and with a pH of 1.7. The primary vanadium ionic clusters in the solution were VO2+. An aluminum sheet of 9 mm in diameter was used as the working electrode and placed beneath the template. A platinum mesh was used as the counter electrode. The distance between the two electrodes was kept as 25 mm. To make sure a good electrical contact between the template and

electrodes, the back of the membrane template was sputter-coated with Au-Pd alloy before attaching to the working electrode. The complex was attached onto ITO substrate with silver paste. Then electrodeposition was achieved under various voltages (1.5-2 V) and times. The sample was dried at 110 °C for 6 h in air. To remove the hard template, the methylene chloride was used to sink the as-prepared sample to dissolve away PC membrane, resulting in as ensemble of V2O5 nanotubes attached on the ITO substrate. As shown in Figure 2-7, the nanotubes with outer diameter of 200 nm and inner diameter of 100 nm stood apart from one another. The XRD pattern reveals that the nanotubes were amorphous.

Figure 2-7 SEM images of (a) top view and (b) side view of V2O5 nanotubes; (c) TEM image of isolated V2O5 nanotubes; (d) XRD pattern of the prepared V2O5 nanotubes on Au electrode.

A single-crystal V2O5 nanorod arrays were also prepared by this group with the similar method.114 In this study, the concentration of VO2+ varied from 0.02 M to 0.2 M, which suggests no change of the valence state of vanadium ions. The applied voltage ranged from 1.5 V to 2.5 V. A big difference from the method prepared nanotube arrays was the way to remove PC membrane. In this case, the removal of PC membranes was through the calcination of the dried deposited sample in air at 485 °C for 1 h. The V2O5 nanorods with 10 μm in length and 100-200 nm in diameter were single crystal grown along [010] direction.

The theory of the electrodeposition was described by Pourbaix118 via a potential-pH diagram for vanadium species in aqueous solution when pH < 4, as shown in Figure 2-8.119 2+ VO cations existed in acidic medium and precipitation into VO2 occurred when the V + 4- solubility product was attained. For pH < 4, soluble V species were VO2 and H2V10O28 V -2.63 mainly. Insoluble V2O5 appeared at pH 1.8 for a total V concentration of 10 M. If the pH was different, the precipitation of V2O5 required a larger concentration, V2O5 was formed between pH 1.68 and 1.92 when VV was 10-2.5 M.

The electrodeposition of V2O5 can occur either directly, according to:

2  VO 32 2 52  26 eHOVOH Eq. 2-5

Or in two steps:

2 4   VO 1810 2  2 2810  1034 eHOVHOH Eq. 2-6

4  2 2810  354 252 OHOVHOVH Eq. 2-7

2   VO 2OH VO2 2  eH Eq. 2-8

  2VO 22 52  2HOVOH Eq. 2-9

Eq.2-9 Eq.2-8

Eq.2-7 Eq.2-5 Eq.2-6

Figure 2-8 Potential-pH diagram for vanadium species in aqueous solution when pH < 4. The limits IV V -2.5 are calculated for a total V or V concentration of 10 M. The H2O/O2 redox system is also indicated in dotted line. Reactions Eq.2-5 – Eq.2-9 are indicated by double arrows.

2.2.2.4 Sol-gel method The sol-gel process involves conversion of monomers into a colloidal solution (sol) that acts as the precursor for an integrated network (or gel) of either discrete particles or network polymers. Typical precursors are metal alkoxides. The advantages of the methods are that the kinetics of the various chemical reactions can be easily controlled by the low temperature and by the often dilute conditions, and the method can create very fine powder. In addition, this method can also combine with hard templates to prepare 1D nanostructure arrays, such as nanorod arrays, and nanotube arrays.

A representative example was reported by Lakshmi and co-works who used the sol-gel method to prepare TiO2 nanotube and nanorod arrays, and MnO2, V2O5, Co3O4 nanofibres 120 by assistance of PC membrane. For V2O5, vanadium (V) triisopropoxy oxide (VITP) was used as a precursor. A 600-nm-pore-diameter PC membrane was placed onto a Ni wafer, as current collector. In an atmosphere of argon, the liquid VTIP was then dropped onto the surface of the template, and then hydrolysed by exposing to air with ambient humidity at 60 °C for 2 h. Finally, the template was removed by exposure to oxygen plasma. SEM images reveal that monodisperse V2O5 nanofibres were formed within the pores of the template membrane.

Recently, Al Zoubi et al. demonstrated that V2O5 nanoparticles could be synthesized at moderate reaction temperatures (< 100 °C) by hydrolysis of VTIP in air and water-stable ionic liquids with reflux of acetone and isopropanol as solvents and co-solvents.121 The ionic liquids (1-butyl-1-methyl pyrrolidinium bis(triﬂuoromethylsulfonyl)amide

([Py1,4]Tf2N and 1-ethyl-3-methylimidazolium bis(triﬂuoromethylsulfonyl)amide

([EMIM]Tf2N)) were with the same anion and different cations which was proved to affect the crystallinity, morphology and the surface area of the produced nanoparticles.

Specifically, [Py1,4]Tf2N made the products with higher crystallinity especially with acetone as refluxing solvent, while [EMIM]Tf2N gave a clear mesoporous morphology with isopropanol as refluxing solvent.

Additionally, this method was also used to prepare hybrid organic/inorganic nanocomposite without solid-state fusion. This part will illustrate in the following section (Section 2.3).

2.2.2.5 Spray pyrolysis Spray pyrolysis is a process in which a thin film is deposited by spraying a solution on a heated surface, where the constituents react to form a chemical compound. The chemical reactants are selected so that the products other than the desired compound are volatile at the temperature deposition.122 The process is particularly useful for the deposition of oxides. The equipment of this process is displayed in Figure 2-9.

Figure 2-9 Schematic illustration of the spray pyrolysis process.

This method for synthesis of V2O5 nanoparticles or V2O5 thin films is also popular. For example, spherical porous V2O5 was synthesized by spray pyrolysis of the precursor made 29

123 of V2O5, H2C2O4 and citric acid aqueous solution. The precursor was fed into a vertical spray-pyrolysis reactor at a rate of 50 ml/min. An atomizing nozzle was used in combination with compressed air, and the spraying was carried out at a pressure of 2.0 MPa and an atmospheric temperature of 350 °C to produce the oxide precursor. And then, the oxide precursor was heated at high temperature in air to form spherical porous V2O5. SEM images show the morphology difference of the V2O5 spheres before and after heat treatment. It was found that the shape can be kept after heat treatment but shrank and became porous. With the calcination temperature increase, the spherical particles became dense, and finally collapsed when temperature increased to 600 °C. The BET measurement indicated that the surface areas of the samples heated at 420 and 450 °C were 114 and 76 m2/g, confirming the increasing density and less porous.

Another spray pyrolysis process developed from the traditional one is so-called flame spray pyrolysis (FSP). The difference is that the pyrolysis is not achieved by heating plates but flame. An example was presented by Ng and co-workers who demonstrated that V2O5 nanoparticles with size of 30-60 nm were prepared by a one-step and scalable flame spray pyrolysis process.124 In their study, VTIP as the precursor was first dissolved into EG, and then stirred into toluene and tetrahydrofuan (THF). And then, the precursor solution was injected at 3 to 6 ml/min through the reactor nozzle and dispersed with 5.0 ml/min of oxygen into a fine spray while maintaining a constant pressure drop of 1.5 bar across the nozzle tip. A premixed flame fuelled by 1.3 L/min of methane and 3.0 L/min of oxygen was maintained to ignite and support the combustion of the spray. A sheath gas of 5.0 L/min of oxygen surrounding the flame was used to ensure complete combustion. The powder was collected by placing a glass fibre filter (257 mm in diameter) above the flame and drawing the gas streams with a vacuum pump. According to composition analysis, they found that the products contained both V2O5 and VO2. The content of V2O5 (wt %) increased with V concentration in the precursor solution. The particle size and crystallinity were associated with the molar concentration of the precursor and the feed injection rate. In this study, the particle sizes made by this method were no more than 60 nm. The low concentration and low injection rate led to monocrystalline particles with small particle size, 30

confirmed by the equal particle diameters measured by N2 absorption/desorption and XRD, respectively. Particle size and polycrystalline increased with the concentration and injection rate, caused by vanishing of VO2 phase. In addition, the surface areas of the products increased with the concentration and the feed rate.

2.2.3. Shape controlled synthesis of Ag nanoparticles

In this thesis, V2O3 will be used to assist for synthesis of Ag nanoparticles. Therefore, the background of controllable synthesis of Ag nanoparticles will be briefly introduced in this section.

2.2.1 Hydrothermal Method Hydrothermal methods are a sort of popular chemical methods for shape control synthesis of Ag nanoparticles. One good example of this method was reported by Jiang et al. who demonstrated a facile synergetic reduction method to synthesize silver triangular nanoplates under ambient conditions (e.g., aqueous solution, room temperature)125-130. The synergetic reduction approach means the use of a few reducing agents with different reducing abilities

(e.g., citric acid, L-ascorbic acid, and sodium borohydride) toward the reduction of AgNO3 simultaneously. In this approach, the nucleation reaction is initiated by adding trace of

NaBH4 solution, small colloids could be formed at the beginning, and the subsequent growth of silver nanoplates was achieved by the synergetic reduction of both citric acid and L-ascorbic acid in aqueous solution at low temperatures. This synthesis method shows some features such as fast reaction (within 10 minutes), high yield, large quantity, and nearly monodispersed silver nanoplates.

2.2.2 Solvothermal Method Against the hydrothermal method, solvothermal method has been widely investigated in preparation of silver LD nanostructures (e.g., rods, wires, plates, or disks) because of the advantages in desirable shape/size control and diverse choices in solvents (e.g., organic solvent, water, or their mixtures), as well as surface capping agents (surfactants, polymers, or dual systems). For example, Chen et al. 131 presented the CTAB-assisted approach to

generate silver nanodisks in water by ageing at 40 C, and the size ranging from 40 to 300 nm was obtained by adjusting the molar ratio of CTAB to silver, although the reported particles are major in truncated structure. Zhang et al. 132 used PVP as a surfactant to assist the synthesis of silver nanoprisms by heating a water/PVP/n-pentanol ternary system at 95 C for 48 h. This solvothermal reduction method could generate silver nanoparticles, however, diverse morphologies such as triangular, hexagonal, and truncated particles were obtained in the final product, which may affect further applications.

2.2.3 Photochemical Method Photochemical method is performed in solution assisted by UV-Vis or laser light illumination. A good example in this field was first reported by Mirkin et al. 133,134 who demonstrated for the generation of nearly monodispersed silver nanoprisms through the illumination of visible light, in which the conversion of citrate-capped Ag colloids into larger particles with triangular morphology and an edge length of ~100 nm was observed in the presence of stabilizer such as bis(p-sulfonatophenyl) phenylphosphine dehydrate dipotassium salt (BSPP), which played a key role in the growth of silver nanoprisms or nanoplates. They further proposed that the formation of Ag nanoprisms might be caused by the photoinduced fusion or aggregation of pre-formed nanoparticles which may serve as nuclei in the formation of triangular plates.

Furthermore, Mirkin et al. 134 reported in subsequent work that the excitation wavelengths could be chosen to control the prism size, so that the use of longer excitation wavelength could produce larger particles in an edge length of 30-120 nm. They also proposed that a bimodal nanoprism size distribution could be induced by dual-beam illumination, and the so-called secondary beam suppressed the fusion of pre-formed nanoprisms, induced by charge re-distribution on the particle surface resulting from dipole plasmon excitation. Such a fusion of small prisms would be responsible for the formation of a secondary population of larger ones.

There are some other methods for controllable synthesis of Ag nanoparticles, such as template assist methods, electrochemical methods, and so on. Here, the brief introduction 32

regarding the synthesis methods of Ag nanoparticles is just related to the corresponding content in this thesis.

2.2.4 Shape controlled synthesis of Au nanoparticles

Similar to synthesis of Ag nanoparticles, the methods for preparation of Au nanoparticles mainly concentrate on chemical methods which can result in well shape-controlled nanoparticles by adjusting experimental parameters (such as temperature, concentration, stabilizers, etc.). The following part will briefly introduce the background of preparation of Au nanoparticles.

The first scientific article on Au nanoparticles, was reported by Faraday in 1857,135 and in 1908 Mie rationalized their visible absorption using Maxwell’s electromagnetic equations.136 However, a host of ways to prepare Au nanoparticles commenced from the 3+ 0 application of commercial HAuCl4. Later, citrate reduction of Au to Au in water was introduced by Turkevith et al. in 1951.137 This method is still used nowadays.138

For the shape control of nanosize particles, stabilizers play a very important role in their formation and further application. A large variety of stabilizers have been used, such as ligands, surfactants, polymers, dendrimers, biomolecules, etc.138 However, the most stable Au nanoparticles were reported by Giersig and Mulvancy to be stabilized by thiolates via the strong Au-S bond between the soft acid Au and the soft thiolate base.139 Along this way, nowadays the most popular synthetic method using such sulfur coordination for Au nanoparticles stabilization is the Shiffrin-Brust biphasic synthesis using HAuCl4, a thiol, tetraoctylammonium bromide and NaBH4 in water toluene yielding thiolate-Au nanoparticles.140

In addition, besides spherical Au nanoparticles, other shapes of the nanoparticles can be varied using appropriate techniques. For example, Au nanorods with controlled aspect ratio in the range of 2-6 have been synthesized by the micelle-template seed and feed technique developed by Murphy and El Sayed.141 Halas’ group has developed the synthesis of Au nanoshells composed of a silica core (100-200 nm in diameter) surrounded by a thin Au 33

layer (5-20 nm).142 Furthermore, it was found that due to the aggregating effect of salt ions, citrate-capped Au nanoparticles and Au nanoshells as well as CTAB-capped Au nanorods were not stabilized in the presence of a buffer solution, but these Au nanoparticles were readily stabilized by thiol-functionalized PEG ligands.143

Recently, Xia group developed a novel class of nanostructures (Au nanocages) prepared by a template-engaged galvanic replacement reactions. Specifically, Ag nanocubes were used as a template, and then HAuCl4 solution was controllably added to a boiling suspension of Ag nanocubes.144 From the TEM investigation, it can be found that pinholes on one of the six faces of each Ag cube appeared after the nanocube with sharp corners reacted with a small amount of HAuCl4 solution, indicating that the reaction was initiated locally at high- energy sites, instead of over the entire cube surface.145 With the reaction going, the pinhole served as an anode, where Ag was oxidized and electrons were stripped. The released - electrons transferred to the surface of the nanocubes and were captured by AuCl4 , forming Au0 growing on the surface. In later stages, the pinhole closed, presumably through mass diffusion processes or direct deposition of Au near the pinhole.

The galvanic replacement reaction is an effective way to prepare such shell structures for Ag and Au. Meanwhile, this is also a great challenge for preparation of Ag@Au bimetallic nanostructures. In this thesis, we will introduce a new developed method to synthesize Ag@Au bimetallic nanostructures.

2.3 Surface modifications

The surface modifications of nanostructures have attracted much attention due to the as- prepared materials with unique physical-chemical properties. Particularly for materials with surface-related properties, the modified nanostructures can lead to enhance the electronic, mechanical, catalytic, photocatalytic and gas sensing performance.3,146-149 The limitation of preparing such nanoparticles with expected properties is the lack of understanding on particle surfaces. The design of the modification may include factors such as ion energy and ion flux of depositing species, interface volume, crystalline size, coating thickness,

surface and interfacial energy, and so on. All the factors are affected by the selection of materials, deposition method and experimental process parameters.

2.3.1 Polymers

As mentioned before, although various properties have been reported, the main interest for the material remains their electronic properties in LIBs. The polymers with various electronic properties (conductive or insulated) can change the surface properties of V2O5 by coating. Therefore, using polymers as a coating agent is extremely popular. The polymer coated particles can be synthesized by the in situ method, i.e. monomer polymerization in the present of the synthesized nanoparticles,150-152 or the ex situ method, i.e., dispersion of the nanoparticles in a polymeric solution.153

For LIBs, the capacity of the V2O5 nanomaterials can be decreased by several charge- discharge cycles due to solvent exchange, steric hindrance limiting rechargeablity, and irreversible charge in the structure. In order to improve lithium ion diffusion and electrochemical stability, intrinsic conductive polymers (polyaniline (PAN)123,124,127-130 and polypyrrole154-159) have been extensively used to prepared transition metal oxide-based hybrid materials. Another approach to increase lithium ion diffusion is to intercalate non- intrinsic conducting polymers such as poly(ethylene oxide) (PEO)153,160-164 to enhance the ionic conductivity in lamellar and porous electrolytes.

One good example of the polyaniline coated V2O5 nanoparticles was reported by Li et al. who fabricated uniform one-dimensional V2O5@polyaniline core-shell nanobelts by a simple in situ polymerization method in the absence of any surfactant and additional 150 initiator. It is found that the pH was critical for the information of V2O5@polyaniline core-shell nanobelts. With a decrease in the pH value to 0, the original morphology of the

V2O5 nanobelts was destroyed. When ammonium peroxydisulfate (APS) was added, some separated PAN nanofibres were formed, as shown in Figure 2-10.

Moreover, the polymer coating cannot only enhance the stabilization of the electrochemical response of V2O5 in aqueous solution, but also result in electrochromic properties and 36

mechanical flexibility of films. For this purpose, polyvinylchloride (PVC) as the coating polymer was reported by Zampronio et al.149 who prepared the composite film by drying the hydrogel made of PVC resin, V2O5 particles and THF.

2.3.2 Metal dopants

The application of metal/vanadium oxide nanoparticles has been found to be extremely useful in optical, electrical and environmental area. The doping metals include Au,165 Ag,166,167 Cu,148 Pt,168 acting as multiple properties or enhanced performance particles. For gas sensing area, the doping of metal can both increase the sensitivity and selectivity, in which the selectivity is considered as a more important performance than sensitivity.169 In addition, the metal modified V2O5 nanoparticles show better electrochemical performance in lithium ion battery.148

The metal coated vanadium pentoxide using direct chemical reduction of metal salts in aqueous solution in the presence of V2O5 particles is quite difficultly achieved, due to the 5+ 4+ + 0 very high standard electrode potential of the pair (V /V ) in the presence of H (E V5+/V4+ = 0.991). Therefore, the coating methods are always concentrated on the physical methods, such as sputtering coating,168 and pulsed laser deposition.165,166

However, some chemical methods can still be used for synthesizing metal coated V2O5 nanoparticles. One good example was reported by Schlecht et al. who prepared the Ag-

V2O5 nanofibres by adding vanadium-oxytreiisspropoxide and small amount of silver acetate into water, followed by stirring the mixture for 5 min at 70 °C.167 The nanofibres was with the width of 8-15 nm, while the size of Ag nanoparticles was found in the range of 5-15 nm, as shown in Figure 2-11. Another approach is the polythermal method. By this method, metal particles and vanadium precursors were both reduced and precipitated in the solution.148 The vanadium precursors were normally with multivalence (V5+ and V4+). Therefore, heat-treatment like calcination must be followed. However, the calcination may cause the aggregation of the metal nanoparticles, which may reduce the relevant performances.

Figure 2-11 STEM image of Ag-V2O5 nanofibres (scale bar = 100 nm).

2.3.3 Metal Oxide dopants

The coating of metal oxides is very popular for the surface modification of vanadium oxide 147,170-172 173,174 175 nanoparticles. The metal oxides are divers including TiO2, NiO, MoO3, 176-178 179-181 182-184 WO3 , SnO2, and TeO2 etc. The most widely used methods to coat oxides on the surface of vanadium oxide are performed by sol-gel methods, followed by high temperatures treatment,146,171,172,185-188 or simply by physical mixing.173,179,189 The purpose of this coating is to enhance the catalytic, photocatalytic, and electrical properties, or to combine one or more functions of the materials onto another to form a combined effect in one single particle. The properties that are enhanced or combined depend on the coating metal oxide.

TiO2 as a multifunctional material can be used in diverse areas, and shows excellent performances in photodegradation of organic dye, catalyst, and gas sensing. In the early stage, the method to prepare the catalyst was mainly around mechanical mixing,179,189 by which the morphology was difficult to control. Recently, sol-gel method has been widely 171,190 used for the preparation of V2O5/TiO2 nanocomposite, while the focus of the

application has been transferred from catalysis to photodegradation.147 The nanocomposite cannot only increase the degradation efficient but also lead to the photodegradation under sunlight to some extent.185

Electrical and optical properties are important properties of V2O5. TeO2 is found to be a good metal oxide for enhancement of the properties.182-184,191 Pal et al. demonstrated that

V2O5-TeO2 amorphous films were prepared by vacuum deposition of V2O5 and TeO2 with different molar ratios.182 Results show that the conductivity of the films varied from 1.07 × -7 -15 -1 10 to 3.73 × 10 Scm at 373 K. The conductivity increased with annealing and V2O5 content. For the optical property, the transmission of the films was almost 80% from 700- 900 nm wavelengths, but after annealing the transmission decreased to 20%. The optical band gap energy (Eopt) increased with the V2O5 content. The Eopt values of the V2O5-TeO2 amorphous films (2.96-3.11 eV) was larger than that of the films prepared by blowing from bulk glass (Eopt = 2.06-2.13 eV) with the same compositions.

2.3.4 Silica modifier(s)

The preparation of the material is easily achieved by the sol-gel process with the precursor 146,187,188,192-194 of tetraethoxysilane (TEOS). The SiO2 modified V2O5 as catalyst is widely studied. The reactions catalysed by this material are extremely diverse, such as epoxidation 146 of styrene and cyclooctene, oxidative dehydrogenation of isobutane with CO2 to 195 196 isobutene, selective oxidation of H2S to elemental sulphur, selective oxidation of ethanol to acetaldehyde,197 and selective oxidation of o-xylene to phthalic anhydride,198 etc.

Besides the catalytic property, the electrical property of this material has attracted more and more attention. For example, Barbosa et al. demonstrated that the V2O5/SiO2 xerogel showed higher conductivity than pure SiO2 matrix; measurement indicated room temperature conductivity almost 1000 times higher than what is found in literature. And the 194 electrochemical behavior is quite similar to that found in V2O6 xerogel. Furthermore, silica matrix provides an improvement of the electrochemical properties, mainly in relation

to the lithium electroinsertion into the oxide matrix with little decrease in the total charge during successive redox cycles.187

2.4 Properties of vanadium oxide nanoparticles and composites

2.4.1 Redox properties

Due to the multivalence of vanadium, various vanadium oxides show different levels of redox properties. For example, V5+ as highest valence only exhibits strong oxidation property, while V4+ as intermediate valence shows both oxidation property and reduction property, and V3+ normally displays reduction property, as described in Table 2-1.

0 Table 2-1 Electrode process of vanadium and corresponding standard electrode potentials (EA ).

0 Electrode process EA /V   2 0.991 VO2 2 eH VO  2OH   3 0.668 VO2  224 2OHVeH 2  3 0.337 VO 2  2OHVeH  0.7996 Ag e  Ag

V2O5 has been widely known as the catalyst for producing sulphuric acid by utilizing the oxidation property for years. Another widely used application of the oxidation property is to produce maleic anhydride179,180 and phthalic anhydride.198 Kim et al. systematically studied the V2O5 catalytic oxidation process of o-xylene to phtalic anhydride. According to temperature programmed reduction and temperature programmed oxidation (TPR/TPO) studies, it was found that the reduction of V2O5 underwent three redox steps in the catalytic process. The first step was from V2O5 to V6O13, in which the group V=O of V2O5 on (010) plane first vanished. Transformation from V6O13 to V2O4 and further to V2O3 was the second and third steps. The group of V-O-V was removed in these steps. The oxidation proceeded in the reverse order of reduction process, and both the reactions proceeded via quite a stable intermediate. 40

Recently, Li et al. reported that 1D polyaniline nanostructures with controllable surfaces and diameters were prepared by vanadic acid (V2O5/HCl) as the oxidant in the absence of porous templates and structural directing molecules.199 The diameters of the polyaniline nanorods decreased from 150-300 to 40-100 nm with the increase of the HCl concentration, + while the surfaces also became smooth. During this process of polymerization, VO2 could be reduced to VO2+ by aniline. The appropriate redox potential of oxidants and static condition were in favour of the formation of polyaniline nanofibres due to homogeneous nucleation. The synthesis of polyaniline could be also achieved by V2O5 particles, as reported by Li’s group. The synthesis of polyaniline by V2O5 particles as oxidants could be formed a core-shell (V2O5/polyaniline) structure, which is so-called in situ polymerization.150,200

Although VO2 and V2O3 exhibit redox properties to some extent, the utilization of these properties is unpopular for production in recent years, whereas they are extensively studied because of the electrical and metal-insulator transition (MIT) properties. However, the reduction properties of V2O3 can be used for synthesis of Ag nanowires, which will be discussed in the later section.

2.4.2 Metal insulator transition

The MIT process has been found in both VO2 and V2O3. VO2 is a prominent example for a material exhibiting MIT as a function of temperature. It undergoes a first-order transition from a high-temperature metallic phase to a low-temperature insulating phase at almost room temperature (T = 340 K). The resistivity jumps by several orders of magnitude through this transition, and the crystal structure changes from tetragonal phase (R phase) at high temperature to monoclinic phase (M phase) at low temperature. The driving mechanism of the MIT is still debated, but both strong electron correlation (Mott-Hubbard transition) and electron-lattice coupling (Peierls transition) are generally thought to be 34,201 important. Furthermore, for VO2 thin films, the optical properties (transmittance) change along with the obvious changes of electrical properties due to the lattice change of

VO2 during MIT process. Thus, the optical properties of VO2 thin films are associated with 41

electrical properties and temperature. On the basis of these relationships of properties, the 202 203 VO2 thin films can be used as optical switches and smart window.

Vanadium oxide (V2O3) displays a single MIT around 160 K, changing from a low temperature antiferromagnetic insulator to a high-temperature paramagnetic phase. In a semiconductor, the doping level also affects the MIT property. It has been observed that higher dopant concentrations in a semiconductor create internal stresses that increase the free energy of the system, thus reducing the ionization energy. For example, the metallic 204 phase of V2O3 could be stabilized with Ti doping and the insulating phase with Cr 205 doping. For VO2, many metals have been doped in the structure of VO2 for decrease of the MIT temperature, such as Mg,206 Cr,207 Fe, Al, and Ga.208 However, the most efficient metal is W, which can decrease the MIT temperature by ~ 20 K/at.%W for the bulk and by ~ 50 K/at.%W for the nanostructure, and even metallic ground state is realized for W 209 concentration of ≥ 8 at.%. In addition, the metal-insulator transition of VO2 accompanied the change of optical property at the MIT temperature, so-called thermochromic property. That is, the transmittances of the films change with temperature. The thermochromic temperature is equivalent to the MIT temperature. The thermochromic property can be characterized by the widths and slopes of the hysteresis loops (the temperature sensitivity of the transition) in the figure of temperature dependence of optical transmittance.

Recently, Kang et al. demonstrates that thermochromic VO2 films with various optical properties and phase transition parameters were synthesized on fused silica substrates via a simple solution process with inorganic precursors and PVP.210-212 The widths and slopes of the hysteresis loops could be regulated by controlling grain sizes and grain boundary conditions, which were believed to dominate the generation of the elementary hysteresis loop of each gain and the propagation of the MIT.

2.4.3 Electrical properties

The investigation of V2O5 crystal structure has attracted numerous attention on lithium ion batteries (LIBs) because its layer structures perpendicular to the c-axis via van der Waals

interaction can make the guest species (lithium ions) feasibly intercalate into this compound.

Another reason for the attractive of vanadium is the low cost of V2O5 due to the high abundance in the crust of Earth. The investigation of V2O5 as an electrode material has gone a long time, since 1975, because of its high capacity, high output voltage, and low 213 cost. However, bulk V2O5 suffers from poor rate capability and cycleability due to its low diffusion coefficient of lithium ions, slow electron conductivity, and the irreversible phase transitions upon deep discharge.81,214,215

To overcome these limitations, two functionalization routes have been generally used to optimize the properties of V2O5 cathode. One route is the insertion of metallic cations or conductive polymer in the interlamellar space of the oxide, forming the “bronze vanadium oxide” or organic-inorganic hybrid. Normally, the organic-inorganic hybrid has advantages of both organic and inorganic components and exhibit good synergistic effect.216 Therefore, the modified V2O5 is expected to obtain higher electrical conductivity and more stable layer 217 structure. The other route is the production of V2O5 in nanometric scale. By this route, the diffusion path is shorter than that in a macroscopic material and the kinetic of the intercalation is faster due to the high area to mass ratio.218

Recently, effects of various morphologies on the V2O5 as the cathode material have been reported, such as nanowires,215,219 nanorods,220 and thin films,153 etc. However, research shows that the hollow spheres of metal oxides integrate the small size effects of ultrafine particles, such as surface effect, volume effect and quantum effect. The spherical shell material has some unique features, such as ability to hold large amounts of guest molecules or large-sized object, low density, large specific surface, and strong surface penetrability.221 Therefore, the hollow spheres are proposed as a suitable cathode material of LIBs. Wang et al. demonstrated that hollow spheres and its polypyrrole (PPy) hybrid were synthesized by 222 polymerization of pyrrole monomer in the hollow microspherical V2O5 host. It was found that the intercalation of PPy molecules between the layers of V2O5 could cause slight reduction of capacity, but substantially improve cycleability and electrochemical activity compared with the pure hollow structure V2O5 microspheres. Furthermore, the

microurchins with hollow structures used as cathode material in LIBs were reported by Cao et al. who demonstrated that this structure exhibited desirable electrochemical properties such as high capacity and remarkable reversibility.223

In addition, vanadium trioxide (V2O3) has been investigated as the anode material for LIBs.

Xu et al. reported that new-phased metastable V2O3 porous urchin-like microstructures were fabricated on a large scale by pyrolyzing a vanadyl ethylene glycolate (VEG) precursors in the absence of any templates or matrices.224 This anode exhibited improved electrochemical properties with relatively high first discharge capacity and better cycle retention relative to thermodynamically stable V2O3 (R). Such high electrochemical properties were believed to derive from the unique microscopic crystal structure and macroscopic 3D framework with rigid morphology, porous structure, and high specific surface area.

2.4.4 Magnetic properties

As an important physical property of vanadium oxide (VOx, 1.5 ≤ x ≤ 2.5), the magnetic properties of various vanadium oxide based nanomaterials has attracted much attention.225- 228 Generally, magnetic behaviour of a material depends on the electron spin vector or the total magnetic dipole moment. In a paramagnetic material, the unpaired electrons are free to align in any direction. These electrons can be influenced by an external magnetic field, causing them to align in one direction. In a ferromagnetic material, or permanent magnets, the unpaired electrons are aligned in parallel spin even without an external magnetic field induced. Inversely, when the unpaired electron spin direction is permanently fixed, and the spins between neighbouring atoms are aligned in opposite directions such that they have a zero net magnetic moment, such phenomenon is so-called antiferromagnets, similar to ferromagnets. Thus, the contribution of magnetic properties of VOx is from V4+ magnetic ions, instead of V5+ non magnetic ions. In addition, it is noted that other materials may affect the spin states of metal oxides, especially in the preparation of nanocomposites.229 The nature of its spin direction may therefore enhance or reduce the magnetic behaviour due to spin polarizations. 44

An important concept in this field is the Curie-Weiss law that can describe the magnetic susceptibility χ of a ferromagnet in the paramagnetic region above the Curie point, as displayed in Eq.2-10.

C  Eq. 2-10 TT C

Where C is a material-specific Curie constant, T is absolute temperature, and TC is the Curie temperature. The law predicts a singularity in the susceptibility at T = TC. Below this temperature the ferromagnet has a spontaneous magnetization.

The interest in this field of VOx is simulated by the observation of the room-temperature ferromagenetism in iodine doped VOx multiwall nanotubes (VOx NTs).225 However, pure VOx NTs exhibit a strong departure from the Curie-Weiss law at temperature T > 100 K due to the presence of antiferromagnetic dimmers formed by V4+ S = ½ magnetic ions,226 so they are supposed to be antiferromagnets. Some features of magnetization and NMR spectra in VOx NTs have been assigned to the trimmers or other complex V4+ S = ½ spin clusters.226

Another interesting feature of the magnetic properties of VOx NTs is so-called ferromagnetic-antiferromagnetic (FM-AFM) crossover consisting of a transition at T ~ 100 K from the FM Curie-Weiss temperature dependence of dynamic susceptibility to an AFM one with a temperature decrease. This effect is accompanied by a noticeable increase of Curie constant at low temperatures, which may be a consequence of a change in the V4+ magnetic ion concentration.230 However, it is difficult to determine concentrations of various spin species in the VOx NTs. The estimates from the static magnetization were reported by Vavilova226 who demonstrated a discrepancy at high and low temperatures, where the number of spins at low temperature was less than that at high temperature. Meanwhile, dynamic data suggested opposite behaviour.230 A solution of this problem was attempted by Demishev et al.231 but the applied procedure of separation of various terms was not well justified. The above-mentioned difficulties reflect poor understanding of the

fundamental magnetization structure in various VOx nanomaterials. Therefore, the research in this area needs to be further investigated.

2.5 Functional applications

2.5.1 Gas sensing materials

When a chemical reaction of a gas molecule occurs at the surfaces of semiconductors, the absorption/desorption will generate a change in resistance according to the gas concentrations. The electric resistance can be measured to give quantitative results. Therefore, nanoparticles with high surface area enable larger fraction of the constituent atoms for reactive sites which will exhibit advantageous properties, such as faster and more accurate response.

There are a set of parameters used to characterize sensor performance,232 as listed below:

 Sensitivity is a change of measured signal (voltage or resistance) per analyte concentration unit. This parameter is sometimes confused with the detection limit.

 Selectivity refers to characteristics that determine whether a sensor can respond selectively to a group of analytes or even specifically to a single analyte.

 Stability is the ability of a sensor to provide reproducible results for a certain period of time. This includes retaining the sensitivity, selectivity, response, and recovery time.

 Detection limit is the lowest concentration of the analyte that can be detected by the sensor under given conditions, particularly at a given temperature.

 Dynamic range is the analyte concentration range between the detection limit and the highest limiting concentration.

 Linearity is the relative deviation of an experimentally determined calibration graph from an ideal straight line.

 Resolution is the lowest concentration difference that can be distinguished by sensor.

 Response time is the time required for sensor signal to return to its initial value (i.e., 90%) after a step concentration value.

 Working temperature is usually the temperature that corresponds to maximum sensitivity.

 Hysteresis is the maximum difference in output when the value is approached with an increasing and a decreasing analyte concentration range.

 Life cycle is the period of time over which the sensor will continuously operate.

All of these parameters are used to characterize the properties of a particular material or device. An ideal chemical sensor would possess high sensitivity, dynamic range, selectivity and stability; low detection limit; good linearity; small hysteresis and response time; and long life cycle. Normally, only some of these ideal parameters are characterized in research. This is because the task of creating an ideal sensor for some gases is extremely difficult. On the other hand, real applications usually do not require sensors with all perfect characteristics at once. For example, a sensor device monitoring the concentration of a component in industrial process does not need a detection limit at the ppb level, though the response time at range of seconds or less would be desirable. In case of environmental monitoring applications, when the concentrations of pollutants normally change slowly, the detection limit requirements can be much higher, but response time of a few minutes can be acceptable.

Most semiconductors can be used for detection of reducing or oxidizing gases. In addition, although various morphologies (nanorings, nanowires, and porous spheres) are reported to exhibit high sensitivity to ethanol, NOx, CO, NH3, liquid petroleum gas (LPG), and so on, the 1D nanostructure (e.g. nanowires, nanoribbons) is proposed to be more suitable for gas detection due to high surface-to-volume ratios, single crystal, and the size is likely to

produce a complete depletion of carriers inside the structure.233 Dan et al. summarized a variety of nanowire materials as gas sensors to detect various gas species.234

For vanadium oxides, the gas sensing performance has been extensively studied in the past decade. And 1D nanostructure also exhibited higher sensitivity than other morphologies (i.e., thin films, and spheres) so as to be widely reported.3,11,235 In addition, various gases were reported to be detected by V2O5 nanoparticles, including ethanol, H2, and organic amines etc. V2O5 nanostructures showed higher sensitivities to ethanol and organic amines. Liu et al. demonstrated that vanadium pentoxide nanobelts prepared by a hydrothermal method showed very high sensitivity to ethanol with short response time 30-50 s at the concentration range from 10 to 1000 ppm at working temperature of 200 °C. Furthermore, hundreds of cycling tests were employed for stability, suggesting that no appreciable variations were detected and both sensitivity and electrical conductance were reproducible enough during the test. Another successful example was reported by Raible who demonstrated that long V2O5 nanofibres (μm) prepared by reacting ammonium vanadates with ion exchange resin for at least 3 months showed very low detection limit (30 ppb) to 1-butylamine and moderate sensitivity for ammonia. Due to the sufficiently high conductivity of the fibres derived from micrometer length of fibres, the sensors could be operated at room temperature.

The sensing mechanism is considered as surface conduction modulation by adsorbed gas molecules. Specifically, in an ambient environment, n-type V2O5 is expected to adsorb both oxygen and moisture. The adsorbed O and OH groups trap electrons from the conduction band of the V2O5 nanocrystals and increase the resistance. For V2O5, the adsorbed oxygen and lattice oxygen (O) can react with target gas molecules which are chemically adsorbed at the active sites on the surface of V2O5 materials. In this process, electrons will transfer to the surface of V2O5 to lower the number of trapped electrons, which induces a decrease in the resistance. Based on Ohm’s law, the electric resistance of the sensor undergoes a decrease or increase when the gas is injected (released).236 Schilling et al. systematically studied the mechanism for sensing reducing gases with vanadium pentoxide films. It is

found that the mechanism involves a surface redox process and a subsequent oxygen loss by means of lattice oxygen diffusion.237 In addition, for the absorption, Rey et al. measured the oxygen adsorption by pressure change. They found that the predominant oxidation below 300 °C is chemisorptions of oxygen onto the surface, whereas between 300 and 400 °C bulk oxygen vacancies were filled linearly with the square root of time.

2.5.2 Catalysts

The most relevant catalytic application of V2O5 has been extensively used in catalytic oxidation of hydrocarbons and sulphur dioxide.238 The generally accepted working mechanism of V2O5 is the so-called “oxidation-reduction” mechanism, which is similar to that of sensor.239 That is, the oxygen in the catalyst is regenerated as the oxidation reaction proceeds. It is assumed that the V=O bond plays the most important role in catalytic oxidation reactions.240 The catalytic activity ascribed to the V=O groups has two aspects. On one hand, the molecules that take part in the oxidation reaction are adsorbed at these centres.241 On the other hand, it has been often stressed that the vanadyl oxygen vacancy 240 plays an essential role for the electrical and catalytic properties of V2O5.

The catalytic effect can be enhanced by coating metal nanoparticles on the surface.

Research shows that the combination of TiO2 can enhance such catalytic performances, including ammoxidation of 3-Picoline242-244 and toluene,147,172,245 selective oxidation of benzene,246 selective oxidation of ethanol to acetaldehyde,197 and selective oxidation of o- 198 xylene to phthalic anhydride. Besides TiO2, there are various metal oxides which are used to dop on the V2O5 particles to enhance the catalytic property, such as NiO for 173,186 177,247 oxidative dehydrogenation of propane, WO3 for nitric oxide reduction , MoO3 for 248 175 benzene oxidation and ammoxidation of 3-picoline, as well as SnO2 for o-xylene oxidation.180

2.5.3 Actuators

A decade ago, vanadium oxides were proposed for the redox-dependent application: the direct conversion of electrical energy to mechanical energy in actuators (artificial 49

muscles).249 However, electromechanical actuators based on vanadium oxides was realized by Gu et al. who utilized high modulus V2O5 sheets comprising entangled V2O5 250 nanofibres. The high surface area of the V2O5 sheets facilitated electrochemical charge injection and intercalation that causes the eletromechanical actuation.

The eletromechanical actuation was achieved for the V2O5 nanofibre sheet by immersing in an aqueous electrolyte. Actuation results from applying a potential between the working electrode and a carbon fibre counter electrode, while the potential is measured using a saturated calomel electrode (SCE) as the reference. The length of changes of the sheets derived from the cantilever displacements can be measured using an optical sensor. It is found that the sheets can lead to a large actuation strain (0.21%) and 5.9-MPa force- generation capability.82 However, the slow actuation rate is a problem. It is proposed that the actuator mechanical properties could be improved by shear flow or electric or magnetic fields to orient the nanofibres within sheets.

2.5.4 Optical switches and smart windows

The applications of optical switches and smart windows are originated from the MIT process. As aforementioned, the temperature change can cause the changes of electrical properties and optical properties of VO2 thin films. With a band gap energy of 0.6 to 0.7 eV, the semiconductor phase of VO2 is highly transparent in the 3 to 12 μm region of the 251 spectrum. In the metallic phase, VO2 is highly absorbing and reflecting. The semiconductor-to-metal phase transition in VO2 can be brought about by conventional heating at the temperature of 68 °C. Furthermore, switching times of the order of nanoseconds have been observed.252 These provide excellent conditions for optical switches.

These natures of VO2 thin films can also apply in smart windows, which can automatically change infrared radiation of glass so as to automatically adjust the room temperature. Specifically, when temperature is below 68 °C, the window is opaque and allows infrared through to increase the room temperature. When temperature is higher than 68 °C, the window became translucent, partially blocking infrared while maintaining a clear view

through the glass. The high MIT temperature of pure VO2 thin films (68 °C) can be reduced by doping mentals in the films. The most effective metal is tungsten (W) which can modify the MIT temperature to 28 °C.253

2.5.5 Electrode in LIBs

Rechargeable lithium ion batteries (LIBs) have been extensively developed as energy storage devices for electric, hybrid electric vehicles, and intermittent renewable energy sources because of their high energy and power densities and long cycle lifetime. The successful adoption of LIBs for transportation and stationary electric energy storage mainly depends on the factors such as cost, safety, cycleability, and power and energy densities of the electrode materials. Due to low cost, high capacity, and high output voltage, V2O5 is proposed an excellent material as cathodes in LIBs. However, as mentioned before (Section

2.4.3), limitations still exist. For V2O5 nanoparticles, the main problem is the stability

(cycleability). A good method to overcome this problem is to modify V2O5 nanoparticles with conductive polymers such as PPy,222 polyaniline,150 and poly (ethylene oxide).254

2.6 Summary

The progress of vanadium oxides nanoparticles research has been extensively studied in the past decade. Many size and shape control methods have been explored and the growth mechanisms into various morphologies are well explained. Furthermore, some breakthrough experiments have been successful to prepare and control the deposition and doping of these nanoparticles with various materials.

The various highly functional applications of vanadium oxides nanoparticles and nanocomposites have been achieved for practical uses due to their extremely attractive properties. For example, they have been successfully functionalized as selective oxidation and ammoxidation catalysts; they have exhibited highly sensitive and low detection limit in the detection of gases and organic amines under low temperature; they have been doped with tungsten that decrease the MIT temperature to 38 °C which can make the practice use

of the materials for smart windows possible at room temperature; and they have been combined with polymers that are highly conductive and stable to be used as LIBs.

The challenge that researchers currently face is whether the synthesized materials are efficient and practical on a mass production level. The problem of vanadium oxide is often their stability from phase change and multivalence. The vanadium oxides favour to form 1D nanostructure so that other shapes (nanospheres) is difficult to be controllably obtained. The production of widely distributed nanoparticles may significantly reduce the performance of the material. Nevertheless, the broad scientific implications have given insights for possible commercial application.

In this thesis, the methods for synthesis of vanadium oxides and their nanocomposites revolved around hydrothermal methods and polythermal methods. Actually, every methods own their both merits and defects. For example, the hydrothermal method can achieve large scale production to some degree, but the problem regarding multivalence is hard to avoid; while the polythermal method can conduct relative good shape controlled synthesis, and combined with calcination, multivalence is no more a problem, but the cost for the scale-up is rather high. It is difficult to find a perfect method which can overcome all these limitations. Our knowledge had enhanced the efficiency of these materials to provide improved commercial products. Therefore, further development of vanadium oxide nanoparticles and nanocomposites is still an important area of research for our understanding and design of materials.

Chapter 3. Glycothermal synthesis of assembled vanadium oxide nanostructures for gas sensing

3.1 Abstract

This study demonstrates a facile but effective glycothermal method to synthesize vanadium oxide nanostructures for gas sensing detection. In this method, sodium orthovanadate was first dispersed and heated in ethylene glycol at 120-180 C for a few hours, and then the precipitates were collected, rinsed, and sintered at high temperatures (e.g., 600 C) for V2O5 in air and V2O3 in nitrogen, respectively. The as-prepared vanadium oxide particles are nanorods (200 nm × 1 μm) and can assemble into microspheres or urchin-like structures with a diameter of ~ 3 μm. The experimental parameters (temperature, time, surfactants) and the formation mechanisms were investigated by various advanced techniques, such as

TEM, SEM, XRD, FT-IR, and TGA. Finally, the V2O5 nanoparticles were tested for sensing detection of gas species of acetone, isopropanol, and ammonia. The microurchin structures show higher sensing performance than the nanorods.

3.2 Introduction

Vanadium oxide nanoparticles have been widely studied since the discovery of VOx nanotubes by Nesper and co-workers.2,4,255,256 Due to the outstanding structural flexibilities, ranging from one dimension (1D) to three dimension (3D),59 and multi-valence compositions (VOx with 1.5 ≤ x ≤ 2.5), the vanadium oxides have found many potential applications in lithium ion batteries (LIB), gas sensors, and catalysts.2-4,257-259 For example,

V2O5 nanofibres show higher sensitivity than other oxides (e.g., SnO2, TiO2, and MoO3) to organic amines (e.g., 1-butylamine with limit of detection below 30 ppb) that are important 3,5 for analysis in food industry and medical diagnosis. The sensitivity of V2O5 fibres upon interaction with organic amines or NH3 can be significantly enhanced via doping noble metal clusters (gold, silver).167,260

103,113,261- The functional properties of V2O5 nanoparticles are strongly shape/size-dependent. 264 Various methods have been used to control the synthesis of V2O5 nanoparticles (e.g., rods, fibres, tubes) and their assembled nanostructures for unique functional properties, such as chemical vapour deposition or anodic deposition,113,263,264 and wet-chemical synthesis.6,7,10,11,85,265,266 Wet-chemical methods have been widely reported, such as sol-gel 6,7 process via “chimie douce”, colloidal assembly of VO(OR)3 reverse micelles for V2O5 265 10,84,85,266 nanorods, and hydrothermal approach for V2O5 nanotubes and VO2 nanobelts.

Recently, glycol-mediated synthesis has been extensively studied, which could lead to the kinetics or thermodynamics growth. For example, ethylene glycol (EG) can serve as a  + 267-270 reducing agent for metal ions (e.g., AuCl4 , Ag ). Hollow V2O5 microspheres were obtained via the self-organization of nanorods with the average diameter of ~ 200 nm and up to length of 2 μm by heat-treatment of vanadium acetylacetonate [V(acac)3] in the 223 presence of poly(vinylpyrrolidone) (PVP) in EG solution. Urchin-like V2O5 nanostructures have been obtained via such a glycothermal treatment on an ethanolic solution of vanadium tri-isopropoxide and alkyl amines.261,262,271,272 Six-fold rotationally symmetric vanadium oxide nanostructures were obtained from V2O5 xerogels assisted by dodecanethiol.261,272

These methods have shown not only some success but also limitations. The role of EG in the formation of low-dimensional vanadium oxide structures is not properly understood.

The multi-valence structure (VOx with 1.5≤ x ≤2.5) and their heavy pH dependence make it difficult in shape and size control. Vanadium oxides such as VO2 petaloid clusters or VOOH hollow dandelions have also been obtained.100,273 The mixture of multi-valence oxides and the wide size distribution usually reduce the performance of V2O5 nanoparticles in functioning such as gas sensing. Therefore, it is still a challenging task to synthesize vanadium oxide nanoparticles with composition, shape, size, and functionality control.

This study will employ a developed glycothermal approach to synthesize vanadium oxide nanoparticles (V2O5, V2O3) and their assembled structures. The particle properties (morphology, size, and assembly structures) will be characterized by various advanced 54

techniques such as TEM, SEM, TGA, and FT-IR spectroscopy. The possible mechanisms on formation, growth, and assembly will then be discussed. Finally, the sensing capability of such nanostructures toward acetone and ammonia is evaluated and discussed.

3.3. Experimental work

3.3.1 Materials

Sodium orthovanadate (Na3VO4, 99%), EG (99.9%), hydrochloric acid (HCl, 32%), and various surfactants such as PVP (MW = 55000), SDS (A.R. Grade), CTAB (A.R. Grade), were all purchased from Sigma-Aldrich and used as received without further treatment. All solutions were freshly made, and ultrapure water was used in all synthesis process.

3.3.2 Synthesis

The glycothermal approach was used for the synthesis of vanadium oxides. In a typical procedure, 0.55 g Na3VO4 was added into a three-necked flask containing 20 ml EG, and the mixture was stirred until the sodium orthovanadate was dissolved completely. Then, a few droplets of concentrated HCl were gradually added to adjust pH = ~ 1. It was found that the colour of the solution changed from opaque light yellow to brick red with the addition of HCl. The brick-red solution was heated in an oil bath at 120 C; after ~ 20 min, the colour of the solution became greenish black. Continuously heated for another 10 min, blue precipitates were formed. The solution was continuously heated for several hours and then cooled down to room temperature. The precipitates were collected by centrifugation and washed with ethanol for 2-3 times, followed by drying in oven at 60 C. The corresponding parametric optimization was performed in this study, such as temperature, pH, concentration, and surface modifiers. Finally, the blue color powder was calcined at ~ 600 C for 2 h in the atmosphere of nitrogen and air for different phases.

3.3.3 Characterization

Particle characteristics and properties were examined using various techniques, as demonstrated below.

 The shape and microstructure is confirmed by TEM (JEOL 1400), operated at an accelerated voltage of 100 kV. The high-resolution TEM (CM200) was used for inspecting crystalline lattice, which was operated at an accelerated voltage of 200 kV. For TEM observations, the specimen was prepared by dropping solution onto the copper grid and dried in air naturally.

 To observe surface structure, SEM (Hitachi 900) was used and operated at 20 kV.

 TGA was employed to determine phase transformation from precursor to oxides, operated by a thermal analyzer (DuPont 950) under nitrogen gas flow with heating rate of 10 C/min in the range of 25-1,000 C.

 FT-IR spectroscopy (Nicolet 360) was used to investigate the formation and phase transformation mechanism of vanadium oxides. In order to prepare the pill, a mixture of dried KBr and the precursor powders or liquid samples were used.

 To further understand the surface, XPS (conducted on a Physical Electronics PHI 5000 Versaprobe spectrometer with Al Kα radiation, 1486 eV) analysis is used to measure

the C1s VxOy.

 The surface area of the as-synthesized powders and pore size distribution were measured by BET equipment (TriStar 3000) via nitrogen gas adsorption and desorption isotherms.

 To track and confirm the chemical composition of vanadium glycolate particles, electron spray ionization mass spectrometry (ESI-MS) was carried out by mass spectrometer (model QT of Ultima API, Micro mass, Manchester, UK) under

conditions of a capillary voltage of 1.3 kV, a MCP voltage of 2150, and a collision gas of argon.

 The composition analysis of final products was conducted by XRD (Philip MPD diffractometer) with Cu Kα radiation with scanning step of 2= 0.02 o/s.

 Gas sensing detection. According to the schematic diagram of WS-30A gas sensing measurement sensor system (see Error! Reference source not found.), the working

emperature can be controlled by adjusting the heating voltage (Vheating) across a Ni-Cr resistor wire inside the ceramic tube (D).

Figure 3-1 Schematic diagram of the gas sensing measurement system.

A reference resistor is put in series with the sensor to form a complete measurement circuit. In the test process, the change of the resistance of the sensor C in air or in a test

gas can be monitored via the voltage (Voutput), which is the voltage at the two ends of

the reference resistor (Rreference). To prepare the sensor C, the as-prepared V2O5 was

first dispersed in tetraethyl ammonium tetrafluoroborate (Sigma-Aldrich, 99%, as

binder) and ethanol to form slurry. Then deposit the mixture as a thin film on a ceramic 57

tube with Au electrodes and Pt conducting wires. Finally, the ceramic tube was heated at 400 °C for 2 h to evaporate the solvent. The gas sensing measurements were carried out at a working temperature of 150 ºC and 30% relative humidity.

3.4. Results and discussion

3.4.1 Synthesis of vanadium oxide nanoparticles

The morphology and size of the glycothermally prepared particles were characterized. Figure 3-2 shows the SEM images that vanadium oxide precursors obtained at 120 °C were urchin-like microstructures with an average diameter of ~ 3 μm (Figure 3-2 a). With careful inspection, it was found that the urchin-like microspheres were composed of quadrangular- prism-like nanorods with diameter of ~ 200 nm and length of 1-2 μm. In particular, these nanorods align and assemble from the center. The vanadium oxide precursor can be converted into pure oxides with different valences depending on reaction conditions. Calcined in air at 600 °C for 2 h, the urchin-like structures were still remained but shrank to ~ 2 μm in diameter, in which rough surface and chain-like nanorods were formed (Figure 3-2 b). This is probably caused by losing water, ethylene glycol, and other carbon-related groups. The chain-like rods were found to link by small particles with diameter of ~ 50 nm. In contrast, there was no shrinkage and rough-surface nanorods were observed in the calcined particles under the nitrogen (N2) atmosphere (Figure 3-2 c). The distinct difference in the surface structure may be caused by the different calcination atmospheres. In air, the carbon-related materials can be completely oxidized and removed; while in N2, the carbon black in amorphous or other phases originated from carbon-related groups may remain but cannot be oxidized, thus the surface structure is unchanged. This is consistent with the 274 similar scenario observed in the PVP-assisted VO2 film preparation.

Figure 3-2 (a) The SEM image of the as-prepared urchin-like precursors; (b) the SEM image of the sample calcined in air at 600 °C; (c) the SEM image of the sample calcined in N2 at 600 °C.

Figure 3-3 (a) The TEM image of an individual microurchin calcined in air with chain-like structure and (b) the HRTEM image taken from one of the nanorods of the microurchins. The microurchins calcined in air were further confirmed by TEM technique, as shown in Figure 3-3. The TEM image shows the urchin-like spheres with diameter of ~ 2 μm consist of interconnected nanocrystallites (Figure 3-3 a). A high resolution TEM (HRTEM) image (Figure 3-3 b) taken from the edge of a nanorod in the microurchin reveals that the lattice 59

fringes were clearly visible with a spacing of 0.438 nm, which can be assigned to {001} 275 planes of V2O5 (JCPDS 00-041-1426), with growth direction along [010].

Figure 3-4 XRD patterns of the precursor calcined in different atmospheres. From top to bottom, the curves correspond to V2O5, the mixture of V2O5 and VEG, V2O3, and VEG, respectively. ● represents the peaks of V2O5, while △ corresponds to the peak of VEG. The X-ray Diffraction (XRD) patterns were used to confirm the composition of precursors and the calcined samples. It clearly shows that the phase transformation occurred when the precursor was calcined at a high temperature in the atmosphere of N2 and air. The XRD pattern at the bottom of Figure 3-4, corresponding to the precursor, can be indexed as monoclinic phase of VEG (space group: C2/c (15), a=9.271, b=9.735, c=9.918Å; JCPDS 00-049-2497). The sharp diffraction peaks reveal that the VEG precursor was well crystallized. When calcined at 250 °C in air, the pattern shows that V2O5 and VEG coexisted in the product, clearly suggesting that there was a phase transformation from

VEG to V2O5. As calcined at 600 °C in air for 2 h, the orthorhombic phase of V2O5 was formed (JCPDS 00-041-1426),275 identified by the XRD pattern at the top of Figure 3-4.

While calcined in N2, totally different from the XRD pattern of V2O5, the rhombohedra 224 phase of V2O3 (JCPDS 00-034-0187) was formed and shown in the middle of Figure 3-4.

Figure 3-5 The N2 adsorption/desorption isotherms of different shapes of V2O5: (a) microurchins, (b) nanorods, and (c) hollow microurchins. The inset shows the corresponding pore distribution of the sample. The nitrogen gas adsorption and desorption was used to measure the surface area of the calcined particles, as shown in Figure 3-5, presenting the isotherm of the V2O5 microurchins (a), nanorods (b), and hollow microurchins (c). The specific areas calculated 61

by standard multipoint BET method. The surface areas of these three samples show a little difference. No obvious hysteresis loop was found, suggesting the surface area would not be very large. The surface area was displayed in each figure. The insets show the pore size distribution, suggesting that the pore sizes of the microurchins and the hollow microurchins are both larger than 20 nm. In addition, there was a peak around 2-3 nm in the inset of Fig. 3-5 a and c, suggesting that there were mesopore formed after calcination.

3.4.2 Effects of experimental parameters

To understand the formation and growth process of vanadium oxide precursors, various parameters have been investigated in this study, such as pH, temperature, concentration, and surface modifiers, as detailed below.

Effect of pH

The pH of the system was adjusted from 1 to 5, 7, and 10, while other parameters were kept as constants (0.15M of Na3VO4 in EG, 120 °C, 12h). Error! Reference source not found. hows the TEM images that VEG nanoparticles with different shape and size were formed at different pH. From the figures, it can be seen that only the product obtained at pH = 1 is uniform rod-like particles with a diameter of ~ 200 nm and a length of 1 μm (Error! eference source not found. a), which generated from decomposition of microurchins. While other products obtained at other pH are irregular shapes. The H+ and its concentration are critical to the formation of urchin-like and rod-like nanoparticles (Error! eference source not found. b-c), as described in Eq. 3-1 and Eq. 3-2. That is, the pH greatly affects the morphology of the obtained VEG particles, and lower pH can help to form microurchins and nanorods.

Figure 3-6 TEM images showing the effect of the pH on the VEG particle shapes and sizes: (a) pH = 1, (b) pH = 5, (c) pH = 7, and (d) pH = 10.

Effect of temperature

Figure 3-7 TEM images showing the effect of the temperature on the VEG particle shapes and sizes: (a) 120 °C, and (b) 160 °C. Figure 3-7 shows the TEM images that the shape and size of VEG nanoparticles are little affected by the temperature. At 120 and 160 °C (other parameters: 0.15M of Na3VO4, pH = 63

1, 12h), the rod-like nanoparticles can be prepared. The discrepancy is that high temperature (160 °C) can cause the high reaction rate, compared with low temperature (120 °C). The high reaction rate may make the products irregular nut short reaction time. Therefore, to obtain relative uniform particles and save energy, 120 °C was proposed as the optimum temperature.

Effect of concentration of Na3VO4

The effect of different concentrations of Na3VO4 was investigated. Figure 3-8 shows the

TEM images of the products obtained with different concentrations of Na3VO4. Similar with temperature, the concentration affects little on the shape and size of VEG particles.

Figure 3-8 TEM images showing the effect of the concentration of Na3VO4 on the VEG particle shapes and sizes: (a) 0.1 M, (b) 0.15 M, and (c) 0.2 M.

Effect of surface modifiers

We also investigated the effect of surfactants of cationic (CTAB), anionic (SDS) and polymer (PVP). The molar ratios of the surface modifiers to vanadium are from 0.5, 1 to 2. The ratio little affected the morphologies of products. Figure 3-9 shows the hollow microurchins with different morphologies prepared by the existence of different modifiers with the molar ratio of 2. In the present system, the surfactants are proposed to act as a “soft template”, in which the CTAB, PVP and SDS molecules may form micelles in polar EG solution The vanadyl ester nanospheres formed at the early stage are easily adsorbed on the surface of the micelles through the vanadyl groups, and then form urchin-like microstructures.

98 The similar morphology has been observed by Cao et al. who reported the hollow V2O5 276 structure synthesized by assistance of PVP. Shi et al. synthesized V2O5 nanobelts in the present of SDS as a shape controller via a hydrothermal method. Luca et al.277,278 demonstrated the mesostructure of vanadium oxide via the combination of CTAB (CTA+) and vanadate in aqueous solution. Besides vanadium oxides, the hollow microurchins have 279 280 281 also been observed in other oxides, such as α-MnO2, γ-MnO2, and TiO2. That is, the surfactant may be helpful for the formation of hollow urchin-like particles under the reported conditions.

Figure 3-9 SEM images of the effect of surfactants on the morphologies of precursors. (A&B) SDS, (C&D) PVP, (E&F) CTAB.

3.4.3 Formation and growth mechanism

3.4.3.1 Formation of vanadyl ethylene glycolate

Various methods for the synthesis of VEG and V2O5 or V2O3 have been widely reported.223,224,269,275 However, the formation mechanism of VEG through a glycothermal method is little discussed. To properly understand, MS was used to identify the molecular structures at different stages in the VEG formation. The samples were taken from the reaction solution around 120 °C in accordance with color changes of solution. The molecular formulae corresponding to the peaks in MS pattern were speculated and listed in Table 3-1.

Table 3-1 The molecular formulas corresponding to the peaks in MS pattern.

Molecular Molecular/ionic Formula weight of the m/z Error (%) fragment

+ {(HOC2H4O)VO [O(C2H4O)2H]2} 338 336 +0.6

+ [(HOC2H4O)2OV-O-VO(OC2H4OH)2] 394 398 -1

+ {(HOC2H4O)2OV-O-VO[O(C2H4O)2H]2} 438 440 -0.5

+ {[H(OC2H4)2O]2OV-O-VO[O(C2H4O)2H]2} 482 484 -0.4

+ {[H(OC2H4)2O]2OV-O-VO[O(C2H4O)3H]2} 528 524 +0.8

+ [VO (OC2H4OH)2] 189 189 0

*All the fragments shown in Table 3-1 are molecular ions.

Figure 3-10 shows the MS pattern of the brick-red color solution before heating (t = 0 min). + The massive peaks reveal that esterification (between EG, H and Na3VO4) and oligomerization (among EG molecules) occurred simultaneously. After heating for ~ 5 min, the solution became darker, and the corresponding MS pattern of the sample was measured and shown in Figure 3-10 b. It was found that the esterification and oligomerization took place. However, the relative intensity of 339 peak (m/z = 339) was the only one that was 66

found to increase, which means that the mixture produced before heating have already converted to the substance (m/z = 339, Table 3-1). Heating for ~ 20 min, the solution became greenish black, which was probably caused by redox reaction. At this moment, only one intensive peak (m/z = 189) can be found in the MS pattern (Figure 3-10 c). After 30-min heating, some blue precipitates appeared but the MS pattern of the solution (Figure 3-10 d) is similar to that shown in Figure 3-10 c, which indicates that the VEG product was transformed from the species corresponding to peak at m/z = 189.

Figure 3-10 MS spectra of the different stages during the formation of VEG. (a) The initial reaction solution without heating (t = 0 min); (b) the reaction solution heating for 5 min; (c) The reaction solution heating for 20 min; (d) the reaction solution heating for 30 min, at this moment the precipitates deposit. Based on above discussion, the formation of VEG is supposed to undergo a few steps, as expressed in Eq. 3-1 to Eq. 3-5. At the initial stage (20 min), esterification and oligomerization occurred simultaneously (Eq. 3-1 and Eq. 3-2). By continuously heating for another 10 min, the vanadium (V) esters could be reduced by EG, and the other complexes produced by esterification and oligomerization may eventually convert to one substance (m/z = 189) (Eq. 3-3 and Eq. 3-4). Through heating over 30 min, an EG molecule was

probably released from the substance (m/z = 189) via intermolecular condensation (Eq. 3-5) to form the VEG precipitate. Therefore, EG may act as both a reducing agent and catalyst in this system. In addition, the role of H+ as the catalyst and evocating agent in esterification in the formation of VEG could not be ignored (Error! Reference source not ound.), as expressed by the possible reactions (Eq. 3-1 to Eq. 3-5).

3  Esterification HnHOC3 42 OH VO 4 H3   HOC 42  OH VO 42  HOHCO  n m 2 Eq. 3-1   2OH3m2n

3  Esterification  m2n2 HOC H42 OH 2VO 4 H6    2OH5m2n2 Eq. 3-2  HOC 42 n 2  42 m HOHCOOVOVOOH 2

Redox HOC 42 n 2  42 m 2  HmHOC2HOHCOOVOVOOH 42 OH   Eq. 3-3 2VOOC H42 OH 2   2  2H2OH5m2 HO42 OHC  1n CH2COH 

4HO42  1n HOHC

Redox HOC 42 n OH VO 42 m 2 2OH2HOHCO  VOOC H42 OH 2 Eq. 3-4   HO42 OHC  1n CH2  2COH HO42  1m  HHOHC

VOOC H42 OH 2  VO OCH CH22 O  HOC H42 OH Eq. 3-5

*n, m>1; n, m are natural number.

3.4.3.2 Formation of V2O5 and V2O3

The transformation from VEG to V2O5 and V2O3 can also be confirmed by FT-IR spectra, as shown in Figure 3-11. In the curve of precursor, the peak centered at ~ 3416 cm1 corresponded to the OH stretching band, and the peaks at 1573, 2878, and 2940 cm-1 corresponded to CH2O group of vanadium glycolate. Similarly, the peak centered at 992 cm-1 could be assigned to V=O stretching bond, while those centered at 886 and 546 cm-1 originated from VO bond. For V2O3, there were only three distinct absorption peaks observed, in which the peaks at 982, 886, and 466 cm1 could be assigned to V=O stretching bond and VO bending bond, respectively. For V2O5, the V=O stretching bond appeared at 1019 cm-1, whereas the peaks at 513 and 477 cm-1 were assigned to the VOV

stretching and bending bonds, respectively. The peak at 631 cm-1 corresponded to the stretching of V–O bond, while the 837 cm-1 peak was attributed to the coupled vibration between V=O and VO–V.161,282,283

Figure 3-11 FT-IR patterns of the precursor, V2O3, and V2O5.

In comparison with V2O5 and V2O3, many absorption peaks corresponding to VEG disappear during calcination at 600 C. The vibration of V=O bonds trend to shift to high –1 –1 wavenumbers with increasing valence, (e.g., from V2O3 at 982 cm , to VEG at 992 cm , –1 and V2O5 at 1019 cm ). This is probably caused by the increase of the number of electron withdraw group (O) surrounding V=O group. The transformation from VEG to V2O5 is an oxidation process (V4+V5+) during calcination in air, as shown in Eq. 3-6; while the 69

4+ 3+ formation of V2O3 from VEG is a reducing process (V –V ) during calcination in N2. Here the reducing reaction may be caused by pyrolysis process of VEG without oxygen. The pyrolysis process is also a self-redox process (Eq. 3-7). In the self-redox process, the 4+ 3+ + decreasing valences of elements are vanadium (V -V ) and hydrogen (H -H2), while the valence of carbon increases from −1 (carbon-related group in VEG) to +2 (CO) and 0 (C), as shown in XPS pattern in Figure 3-12.

4VOOC 42 OH  11 522 8OVO CO  22 OH8 Eq. 3-6

2VOOC 42  32 3OVOH CO 2  CH4 Eq. 3-7

Figure 3-12 XPS pattern of carbon (C1s peak) in the product of V2O3 calcined in N2.

However, the reduction reaction in the atmosphere of N2 is different from those in the 284 atmosphere of H2, as reported by Eid et al., who demonstrated that Fe3C was prepared by the pyrolysis of FeAc2/PVP in the atmosphere of H2 and argon, since the carbon-related 70

groups remain in the structure during this process. The possible reasons for the difference are addressed as below: two kinds of atmospheres are not exactly same. In their case, the reducing gas H2 was continuously provided to reduce the precursor particles, which may help the carbon pyrolyzed from carbon-related groups to react with iron oxide to form

Fe3C. The formation of Fe3C via Fe2O3 and C needs not only the sufficient carbon but also 285,286 the continuous supply of reduction atmosphere (H2 or acetylene). In our case, however, N2 is not exactly a reduction atmosphere. V2O3 and C are generated from the pyrolysis of VEG. Although there are CO and H2 generated from pyrolysis, the small amount of such gases can be easily blew away by N2 and inadequate to supply the 287 reduction. In addition, the formation temperature of VCx is at least 1000 °C when carbon reacts with V2O3. In our case, however, the low temperature (600°C) was insufficient to form VCx. Another reason may be attributed to nature of vanadium and iron. Actually, carbon in the form of Fe3C proverbially exists in pig iron. The formation of steel is achieved by removing the carbon from pig iron, while VCx is an extremely hard refractory ceramic material.288 The exact reason is still unknown. However, our results show that there is no other phase like VCx but V2O3 confirmed by XRD pattern (Figure 3-4).

As a further confirmation, TGA/DTA technique was employed. Figure 3-13 shows the

TGA/DTA curves that the decomposition processes took place in air (a) and N2 (b), respectively. There was a total weight loss of ~ 39% around 300 °C (Figure 3-13 a), corresponding to the removal of physically adsorbed molecules and the decomposition of

VEG to VO2. By further heating to 600 °C, there was a gain in weight, corresponding to the formation of V2O5. In comparison, the decomposition temperature of VEG was higher than those reported by Weeks and Ragupathy.269,275 In addition, a sharp peak in DTA curve at 295 °C reveals that a violent exothermic reaction took place in a short range of temperature, which was attributed to the oxidation reaction as described in Eq. 3-6. Figure 3-13 b shows the TGA/DTA curves of the precursor calcined in N2. The total weight loss of ~ 44% may correspond to the decomposition of VEG to V2O3. Two exothermic peaks appeared at around 365 and 405 °C, which may correspond to the chemical reaction of weight loss of

CH2 group in VEG (~ 12%). Therefore, the transformation from VEG to V2O3 could be a 71

two-step de-carbonizing process for the –CH2 groups in VEG molecules. The two kinds of exothermic ways may be an alternative reason for the surface difference of the samples 4+ calcined in air and N2. The release of CO and/or H2 (Eq. 3-7) may further reduce V in 3+ VEG to V and hence form V2O3 under the N2 flow.

Figure 3-13 TGA (solid line) and DTA (dot line) curves of the precursor calcined in air (a) and N2 (b). 3.4.3.3 Formation of VEG nanostructures For better understanding, time-dependent experiments were performed to gain insight into the formation process of such vanadium precursors. Precipitate was collected at different times (t = initial stage, 30 min, 60 min and 12 h) through heating at 120 C. Figure 3-14 a shows that the initial precipitate composed of microspheres with rough surface and wide size distribution (0.5-3 μm in diameter). The rough surface suggests that the microspheres were probably formed by aggregation of nanospheres with diameter of ~ 50 nm. The microspheres appeared instantaneously; therefore it is impossible to catch the initial product (nanospheres), probably due to the initially formed particles too small to catch and observe or the very fast aggregation. After 30 min, the surface became rougher (Figure 3-14 b), suggesting that nanospheres grew bigger with time due to Ostwald raping process. The urchin-like microstructure was formed in the following heating around 60 min (Figure 3-14 c), in which the microspheres seem to be composed of assembled nanorods, and the urchin- structure remained in geometry within 12 hours (Figure 3-14 d). Over 12 h, the urchin-like

microspheres would break into randomly dispersed nanorods with diameter of ~ 200 nm and length of ~ 1 μm. Thus, the different shapes (microspheres, microurchins, and nanorods) can be obtained by controlling reaction time. Based on experimental observations (Figure 3-14), the evolution process of VEG microstructures, including aggregation of small particles, growth, splitting and breaking, is schematically illustrated in Figure 3-15. In addition, this process may also help explain the formation of hollow microurchins when the surface modifiers (CTAB, PVP and SDS) exist in the system (Figure 3-9).

Figure 3-14 SEM images of the products taken out of the reaction solution at different times. (A) 0.5h (the precipitate just appeared); (B) 1 h; (C) 2 h; and (D) 12 h.

Figure 3-15 Schematically illustrating the growth process of the products. 73

3.4.4 Gas sensing detection

Figure 3-16 (A) Typical response curves of the V2O5-based sensors during cycling between increasing concentration of acetone and ambient air; and (B) sensitivity of the sensing materials to acetone.

The gas sensing performance of the as-prepared V2O5 microurchins and V2O5 nanorods was investigated via measuring the selectivity to acetone, NH3, and isopropanol. Figure 3-16 a shows the typical isothermal response curves of two kinds of materials toward acetone in the range 10-1000 ppm and at working temperature of 150 °C and 30% relative humidity. The response and recovery times (defined as the time required reaching 90% of the final equilibrium value) of the as-prepared urchin-like and rod-like V2O5 based sensors are of 200-500 s. Based on Ohm’s law, the electric resistance of the sensor undergoes a decrease or increase when the acetone was injected (released), which is the typical sensor behavior of n-type semiconductor sensors. In an ambient environment, n-type V2O5 is expected to adsorb both oxygen and moisture. The adsorbed O and OH groups trap electrons from the conduction band of the V2O5 nanocrystals and increase the resistance. For V2O5, the adsorbed oxygen and lattice oxygen (O) can react with acetone molecules which are chemically adsorbed at the active sites on the surface of V2O5 materials. In this process, electrons will transfer to the surface of V2O5 to lower the number of trapped electrons, which induces a decrease in the resistance.236,289

Figure 3-16 b shows the sensitivity (S) of urchin-like and rod-like V2O5 nanomaterials toward acetone at different concentrations. The gas sensitivity is defined as the ratio of the stationary electrical resistance of the sensing materials in the test gas (Rgas) to the resistance in air (Rair), that is, S = Rair/Rgas. It is found that both vanadium oxide materials show good sensitivity toward acetone, and the V2O5 microurchin structures show higher sensitivity than the nanorods. This could be contributed to the different structures and shapes of two materials. The sensing mechanism is surface conduction modulation by adsorbed gas molecular, and the electrical conductivity depends strongly on surface states produced by molecular adsorption. The adsorbed species are active in producing a surface depletion layer. For the urchin-like structure, the radial but not aggregated rods may allow more gas molecules to contact in sensing detection. In contrast, the rods may pack more densely than those in microurchins, and hence the contact area with gases would be smaller than microurchins.

Figure 3-17 (A) Sensitivity of (A) microurchin-based sensing materials and (B) nanorod-based sensing materials to the tested gases: filled square acetone; filled down triangle isopropanol; and filled circle ammonia.

Other gas species such as ammonia and isopropanol were also investigated by the V2O5 microurchins (Figure 3-17 A) and the nanorods (Figure 3-17 B). Figure 3-17 A shows that the microurchins are highly sensitive to acetone, compared with isopropanol and ammonia.

Similar scenario could be found for those V2O5 nanorods (Figure 3-17 B). In comparison, the microurchins show higher sensitivity to ammonia than other V2O5 sensing materials.3,290,291

In the case of sensing detection of acetone, the microurchin-like V2O5 powder shows good 292 236,289 293 294 selectivity compared with ZnO thin film, Fe2O3, nickel ferrite and SnO2. The relative low sensitivity of the V2O5 powder may be caused by the small surface area (~ 7.7 2 m /g) of the V2O5 microurchins, although the urchin-like structure can allow gas molecules enter and reach the active positions readily. Alternatively, to enhance the sensing 295 296 performance is to modify V2O5 with some noble metals, e.g., Ag, Pt. The detailed investigation is under progress in our group. Nonetheless, the microurchin-like V2O5 materials are promising for gas sensing toward acetone under the reported conditions.

3.5 Summary

Vanadium oxide microurchins with a diameter of ~ 3 μm assembled by nanorods (200 nm × 1 μm) can be synthesized by a facile glycothermal method. The calcination of VEG under the atmospheres of air and nitrogen could result in different products, V2O5 and V2O3, respectively. By this method, various morphologies and sizes of microspheres, microurchins, and nanorods could be obtained by optimization of experimental parameters, such as reaction time, temperature, concentration, pH, and surface modifiers (polymers or surfactants). The formation mechanism of the precursor was supposed on the base of esterification, redox reaction and followed by intramolecular condensation under the considered glycothermal conditions. In our reported system, EG acts as multiple roles such as a reducing agent to form VEG precursor with V(IV), a solvent, and a coordinator. The as-prepared V2O5 microurchins can be used as gas sensing materials toward acetone, isopropanol, and other gas species.

Chapter 4. Hydrothermal synthesis of Ag-vanadate nanocomposites for amine sensing

4.1 Abstract

In this chapter, Ag2V4O11 nanobelts, Ag0.35V2O5 nanobelts and VOx@Ag nanocomposites were prepared by hydrothermal methods and wet chemical methods, respectively. The

Ag0.35V2O5 nanobelts could be synthesized at low Ag/V molar ratio of 15% with assistance of SDS which acted as a reducing agent in the formation of Ag0.35V2O5. Whereas the synthesis of Ag2V4O11 nanobelts did not need any additives, but high molar ratio of Ag to vanadium (V) (> 50%) is essential. Interestingly, the composition and crystallinity of

Ag2V4O11 could be controlled by adjusting solution pH. The effects of the parameters of the hydrothermal method were investigated. Moreover, VOx@Ag nanocomposites were prepared via Ag ions and the mixture of V2O3 and V2O5 particles in aqueous solution at room temperature. The relevant formation and growth mechanisms of these materials were discussed. Finally, the gas sensing performances of the three materials were tested. It was found that Ag2V4O11 nanobelts with high crystallinity (obtained at pH 1) showed much higher sensitivity towards amines than the other materials at optimized working temperature of 260 °C. In addition, low detection limit (5 ppm) and significant high selectivity to amines versus ammonia were observed from the Ag2V4O11 nanobelts. This study may offer a promising sensing material for detection of organic amines in industry and particle use.

4.2 Introduction

Recently, silver vanadium oxides (SVO) have attracted increasingly more attention because of their excellent electrical properties in the use of batteries.297-301 Materials containing various amounts of silver, vanadium, and oxygen can form a number of phases of SVO, depending on reaction conditions and material stoichiometry. Among these phases, 299,300,302 Ag2V4O11 and Ag0.35V2O5 have been extensively studied. Especially, Ag2V4O11 as

a typical transition metal vanadate has been used as a commercial success cathode material in the past decade. Commercially, lithium/Ag2V4O11 has been as a cathode material in a primary (non-rechargeable) lithium anode cell which is used for implantable cardioverter defibrillators (ICDs),303 as the material can provide high discharge capacity, high rate capability and long-term reliability for advanced biomedical devices.

However, only a few have been reported on other properties of Ag2V4O11, especially its sensing properties.302,304 As a sensing material, one representative example was reported by 305 Liang et al. who demonstrated that a gas sensor fabricated by Ag2V4O11 was tested for the detection of ethanol and showed high sensitivity to ethanol with the concentration in the range of 10 to 600 ppm under low working temperature. Due to their toxicology306 and wide use in drugs and gas treatments,307 the detection of organic ammines has recently attracted concerns from scientists. For example, Raible et al. demonstrated that the V2O5 nanofibres with width of ~ 10 nm and several μm show high sensitivity and low detection limit (30 ppb) to 1-butylamine.3 In addition, Ag nanowires with diameter of 150-950 nm and lengths of 100 μm were reported as a high-sensitive-to-amine sensor material.308 Based on these studies, we propose that the combination of the two materials may show high sensitivity to ammines.

Methods for the synthesis of SVO traditionally include solid-state thermal reaction by decomposing silver salts or combining silver oxide with vanadium pentoxide, which suffers high temperature (> 500 °C) and multiphase of products.309-311 Recently, hydrothermal methods, which are proposed to be an effective method for large-scale synthesis of

Ag2V4O11 nanoparticles with high yield at relative low temperature (< 200 °C), have been widely used for the synthesis of Ag2V4O11. For example, Ag2V4O11 particles with high yield were made from the reaction of Ag2O and V2O5 by the hydrothermal method at the 312 temperature of 150 °C for 24 h. Furthermore, Ag2V4O11 nanobelts were synthesized by heating V2O5 and AgNO3 aqueous solution with organic template (hexanediamine) at 180 °C for 2 days.302

This chapter is to demonstrate a developed hydrothermal method for controllable synthesis of Ag2V4O11 nanobelts with various crystallinities and compositions (Ag2V4O11 nanobelts and Ag doped Ag2V4O11 nanobelts) by adjusting pH. By optimizing synthetic parameters, a silver vanadium oxide bronze (Ag0.35V2O5) has also been prepared. The possible formation mechanisms of the materials are then discussed. In addition, other nanocomposites

(VOx@Ag) prepared by mixing Ag ions, V2O3 and V2O5 in aqueous solution at room temperature are used as a comparison material for the detection of amines. To characterize the structures of the materials, various advanced technique were used, such as TEM, HRTEM, XRD, and XPS. Finally, the gas sensing performance of the materials was tested by detecting organic amines.

4.3 Experimental work

4.3.1 Materials

Silver nitrate (AgNO3, analytical reagent) were purchased from Ajax Finechem Pty. Ltd.

Vanadium trioxide (V2O3) and vanadium pentoxide particles were prepared by the polythermal method as described in previous section (Section 3.3.2). Other materials such as PVP (MW = 55,000), SDS (A. R. Grade), CTAB (A.R. Grade), were all purchased from Sigma-Aldrich and used as received without further treatment. All solutions were freshly made, and ultrapure water was used in all synthesis process.

4.3.2 Preparation of Ag-vanadate nanocomposites

4.4.1.1 Preparation of Ag2V4O11 nanobelts

The Ag2V4O11 nanobelts were synthesized by a hydrothermal method. In a typical procedure, a mixture of vanadium pentoxide powders (1.5 mmol) and silver nitrate (1.5 mmol) was dissolved in 15 ml pure water. The solution mixture was then placed into a Teflon-lined stainless steel autoclave, sealed and maintained at 180 °C for 2 days. Various pH for this system were considered to investigate the effect of pH on the formation of the materials. Cooled down to room temperature, the greenish gray precipitate was separated 80

by centrifugation at a rate of 4000 rpm. It was then washed with pure water and ethanol several times. Finally, it was naturally dried in air at room temperature.

4.4.1.2 Preparation of Ag0.35V2O5 nanobelts

The Ag0.35V2O5 nanobelts were also synthesized by a hydrothermal method. Similar to the previous preparation method, a mixture of 1.5 mmol vanadium pentoxide powder and 0.6 mmol SDS was dissolved in 15 ml pure water. Afterward, various molar ratios of Ag to vanadium were added in the mixture in order to investigate the effect of the ratio on the formation of the nanobelts. To investigate the effect of additives, CTAB and PVP were used. The mixture was then placed into a Teflon-lined stainless steel autoclave. After heating at 180 °C for 2 days, greenish gray precipitate was formed. Finally, the precipitate was centrifuged and washed by pure water and ethanol several times after being cooled down to room temperature, then dried at room temperature.

4.4.1.3 VOx@Ag nanocomposites

VOx@Ag nanocomposites were prepared by adding various amounts of Ag ions into the mixtures containing various ratios of V2O3 to V2O5 in aqueous solution at room temperature.

The initial purpose of this part is to prepare Ag doped V2O5 nanocomposites. One of the most used methods for synthesis of metal doped metal oxide revolves around liquid phase deposition (the reduction of metal precursors in the present of metal oxide particles)313 or co-precipitation314 which may generate SVO in our case. For the liquid phase deposition, it + is impossible to reduce Ag in the present of V2O5 particles as the standard electrode + potential of V5+/V4+ (EV5+/V4+ = 0.991 V) is higher than that of Ag /Ag (EAg+/Ag = 0.7996 V).

Therefore, little has reported on the preparation of Ag doped V2O5 nanocomposites by this method. In accordance with our precious studies, Ag nanoparticles can be reduced on the surface of V2O3 particles in aqueous solution at room temperature. Here, we propose to prepare Ag doped vanadium oxides by adding Ag ions to the mixture of V2O3 and V2O5 suspension. Then the nanocomposites will be aged at a certain temperature (~ 300 °C) for 24 h in air to oxidize all the vanadium to V5+. In addition, high molar ratios of Ag to

vanadium (> 30%) can lead to the formation of Ag nanowires, which is difficult to evenly attach on the surface of V2O5 nanoparticles. To avoid the formation of Ag nanowires, PVP was used as the capping agent. The separation of the nanocomposites was achieved by centrifugation. The nanocomposite was then washed by pure water and ethanol several times.

4.3.3 Characterization

The morphology size and structure of the samples were investigated with a JEOL 1400 microscope (TEM) and a Philips CM200 FEG microscope (HRTEM), operated at an accelerated voltage of 100 kV and 200 kV, respectively. The surface morphology was studied by scanning electron microscope (SEM, Hitachi 900), operated at 20 kV. To characterize the composition of the materials, powder X-ray diffraction (XRD) pattern was recorded on a Philip MPD diffractometer with Cu-Kα radiation. To understand the composition of VOx@Ag, the X-ray photoelectron spectra (XPS, conducted on a Physical Electrinics PHI 5000 Versaprobe spectrometer with Al Kα radiation, 1,486 eV) analysis was used to measure the vanadium in the nanocomposites.

4.3.4 Gas sensing performance

The gas sensing performance of the materials was conducted by WS-30A gas sensing measurement sensor system. The procedure is similar to that in previous section (Section 3.3.3). It is noted that, in order to make the material stable and obtain a stable single, all the materials were aged at 300 °C in air until the base line became stable (~ 24 h). Therefore, for VOx@Ag nanocomposites, the composition of the materials would not change and all the VOx oxidized to V2O5.

4.4 Results and discussion

4.4.1 Synthesis of nanocomposites

4.4.1.1 Ag2V4O11 and Ag0.35V2O5 nanobelts

The compositions of the products were investigated by XRD, as shown in Figure 4-1 a.

Curve a is a typical pattern assigned to Ag2V4O11 which was produced with original pH as described in Section 4.4.1.1, while Curve b shows a pattern corresponding to Ag0.35V2O5 prepared by the assistance of SDS at the Ag/V molar ratio of 15%. The sharp diffraction peaks reveal that the two materials were well crystallized. The morphology and size of the nanostructures were investigated by TEM, as shown in Figure 4-1 b and c, corresponding to

Ag2V4O11 and Ag0.35V2O5, respectively. It is found that the Ag2V4O11 product was composed of belt-like structures with the width of 100-200 nm and the length of ~ 5 μm.

The Ag0.35V2O5 product also contained belt-like structure with uniform size of 100 nm × 5 μm.

Figure 4-1 (a) The XRD patterns of the products. Pattern a represents Ag2V4O11 and Pattern b corresponds to Ag0.35V2O5; (b) An overview TEM image of the Ag2V4O11 nanobelts; (c) An overview TEM image of the Ag0.35V2O5 nanobelts.

4.4.1.2 VOx@Ag nanocomposites

The morphology, size and composition of the VOx@Ag prepared with the assistance of PVP at the Ag/V molar ratio of 15% were investigated by TEM and XRD, as show in 83

Figure 4-2. In Figure 4-2 a, the small particles with dark contrast correspond to Ag nanoparticles, while the big particles with light contrast represent V2O5. It is found that the Ag nanoparticles with 20 nm in diameters aggregated, instead of attaching on the surface of + V2O5 particles. The possible reason is that the Ag was reduced on the surface of V2O3 via heterogeneous nucleation to form Ag nanoparticles. Therefore, the Ag nanoparticles only physically adsorbed on the V2O5 particles. The composition of the nanocomposites was confirmed by XRD, as shown in Figure 4-2 b. It clear shows that the nanocomposites mainly contained V2O5 and a small amount of metal Ag.

The use of PVP in this system is to avoid the formation of Ag nanowires. Without PVP, high molar ratio of Ag/V can lead to the formation of Ag nanowires. Compared to Ag nanoparticles, it is more difficult for Ag nanowires to coat on the surface of V2O5 particles. Thus, a capping agent to stop the growth of Ag nanowires is needed. According to our research, PVP can effectively stop the formation of Ag nanowires (Section 5.4.3.4). Other surfactant like SDS was also considered to investigate the effect, which will be discussed in the later section.

Figure 4-2 (a) TEM image of the nanocomposites prepared with the Ag/V molar ratio of 15%. The small particles with dark contrast correspond to Ag, while the big particles with light contrast present V2O5. (b) The XRD pattern of the products, suggesting that the products comprise the mixture of V2O5 and Ag nanoparticles. The V3+ compounds were not found in the XRD patterns. To find out reasons, XPS was used to further reveal the composition of the nanocomposites, as shown in Figure 4-3 a. 84

The XPS pattern shows the coexistence of two vanadium species (V4+ and V5+) in the nanocomposites, suggesting that V3+ was oxidized to V4+. This is also the reason why the nanocomposites are named as VOx@Ag, instead of V2O5@Ag. Figure 4-3 b shows two strong peaks at the Ag region of 368.5 and 374.5 eV, assigning to Ag 3d(5/2) and Ag 3d(3/2), respectively, in good agreement with the previous study.302 It is noted that the multivalence of vanadium would not affect the sensing performance. According to the testing procedure, the materials need to be aged at high temperature (~ 300 °C) in the air for long time until the base line stable. This means that all the V4+ will be oxidized to V5+ so that the resistance of the materials will not change.

Figure 4-3 XPS patterns of (a) vanadium species and (b) Ag species in the VOx@Ag nanocomposites. Black line is original curve obtained by measurement. Green line corresponds to the background, while blue line and red line are the simulated curves corresponding to V4+ and V5+, respectively. 4.4.2 Effect of experimental parameters

4.4.2.1 Preparation of Ag0.35V2O5 and Ag2V4O11 nanobelts

a. Effect of molar ratios of Ag to vanadium

The effect of the molar ratios on the formation of the SVO was investigated with the assistance of SDS at the different Ag/V molar ratios (0, 5%, 10%, 15%, 50%, and 100%). The corresponding morphologies and sizes of the materials were observed by TEM, as shown in Figure 4-4. It was found that the structure obtained without Ag (Ag/V molar ratio

equals to 0) is soft belts with wide distribution (Figure 4-4 a). With the ratios increase to 5%, the morphology of the nanostructure changed little (Figure 4-4 b). With the further increase of the molar ratio to 10%, some uniform nanobelts with 50 nm in width appeared. However, some wide soft belts still existed (Figure 4-4 c). Continuously increasing the ratio (15%, Figure 4-4 d), the products mainly contained the uniform nanobelts with 50 nm in width. When the ratio increased to 50% and 100% (Figure 4-4 e and f), the morphologies were similar to the products obtained at the ratio of 15%.

Figure 4-4 TEM images of the products obtained with different molar ratios of Ag to vanadium: (a) 0, (b) 5%, (c) 10%, (d) 15%, (e) 50%, and (f) 100%. 86

Figure 4-5 XRD patterns of the products obtained with different Ag/V molar ratios: (a) 0, (b) 1%, (c) 5%, (d) 10%, (e) 15%, (f) 50%, and (g) 100%. The compositions of the products synthesized with various Ag/V molar ratios were confirmed by XRD, as shown in Figure 4-5. Obviously, the different ratios resulted in the change of the composition of the products. It is found that the crystallinity of the nanoparticles increased with the molar ratios. When the ratio was lower than 1%, the products were close to amorphous, and it was difficult to identify its composition. When the ratios were between 5% and 15%, the typical patterns corresponding to Ag0.35V2O5 appeared. With the ratios further increase, the compositions of the products changed to

Ag2V4O11. That means two materials can be prepared by adjusting the Ag/V molar ratios. The Ag/V molar ratio affected not only the morphology of the products but also their composites and crystallinity.

b. Effect of additives on the formation of Ag0.35V2O5 nanobelts

The effects of vanadium precursors, the additives and the existence of Ag ions on the formation of the Ag0.35V2O5 nanobelts were studied in this section. Sodium vanadates

(Na3VO4 or NaVO3) as precursors in the system can lead to impurity by the formation of other sodium vanadates (like NaV6O15 in Figure 4-6 a) which can co-precipitate with silver vanadates. To avoid the impurity, NH4VO3 and V2O5 were selected as the vanadium precursor. For the additives, besides SDS, PVP and CTAB were also considered. Although CTAB can cause the formation of AgBr (precipitates) in the presence of Ag+, previous studies suggested that Ag nanoparticles will be eventually formed and the AgBr only affected the route of the formation of the final product, instead of introducing impurity.315,316 Therefore, in this work, CTAB was also considered as a surfactant to investigate the effect on the formation of the nanocomposites.

The belt-like structured products prepared with different reactants at 180 °C for 2 days are shown in Figure 4-6, in which the figures on the left-hand side are the TEM images and those located on right-hand side are the XRD patterns corresponding to left-hand figures.

Figure 4-6 a shows the TEM image and XRD pattern of the product synthesized byNH4VO3 and SDS, while Figure 4-6 b shows the TEM image and XRD pattern of the product obtained by NH4VO3, SDS and 10% (molar ratio) of Ag. In comparison, it is found that the structure of the products synthesized with Ag was “harder” than that synthesized without Ag as some twisted nanobelts appeared in the TEM image (Figure 4-6 a), as marked by red arrows. The discrepancy of the structures is probably attributed to the change of the compositions. From XRD patterns, the product synthesized without Ag mainly contained

VO2 B phase. However, due to the existence of sodium (from SDS), an impurity corresponding to NaV6O15 can also be observed. According to Figure 4-6 b, the product was mainly composed of Ag0.35V2O5 and Ag metal. Compared with the products in Figure

4-1 a (Curve b) and c, it is found that NH4VO3 can increase the yields of both Ag0.35V2O5 and Ag (based on the intensity of corresponding peaks). However, this vanadium precursor can cause the impurity of the product (Ag particles). This requires further investigation.

Figure 4-6 c, d, and e show that the products obtained with various sets of reactants. The TEM images show that the belt-like structures of the products were retained. On the other hand, the XRD reveals the differences of the compositions of the products. Without additives (SDS or PVP) (Figure 4-6 c), no reaction occurred under the reported conditions since there was only V2O5 pattern in the figure. However, the morphology significantly changed from the plate-like structure, (as shown in Figure 4-2 a) to the belt-like layered structure. Figure 4-6 d shows the morphology and composition of the product produced by

V2O5 and PVP. It can be found that the nanobelts were VO2 B phase. Figure 4-6 e reveals that the product comprised VO2 B phase nanobelts and small amount of Ag nanoparticles.

Figure 4-6 f corresponding to the material prepared by V2O5, CTAB, and 10% (Ag/V molar + ratios) Ag , suggesting that the nanobelts were amorphous, and no Ag and Ag0.35V2O5 appeared. This may be attributed to the formation of poorly crystalline CTAV which were formed by the cation [CTA]+ (from CTAB) and vanadium, as reported by Luca et al.277,278

On the basis of the above discussion, it can be concluded that: (i) Ag0.35V2O5 cannot be prepared without SDS; (ii) both SDS and PVP can reduce V5+ in the absence of Ag+, however, SDS will introduce small amount of sodium; (iii) the reducing property of PVP is + higher than that of SDS, because in the present of Ag , PVP can totally reduce V2O5 and + + Ag to VO2 and Ag while SDS can partially reduce V2O5 and Ag to Ag0.35V2O5; (iv) compared to NH4VO3, V2O5 as the vanadium precursor can lead to purer products; (v) CTAB in this system can lead to an amorphous belt-like nanostructure, instead of

Ag0.35V2O5; (vi) the appropriate reactants for synthesis of Ag0.35V2O5 are V2O5, SDS, and Ag+.

Figure 4-6 TEM images (left hand side) and XRD patterns (right hand side) of the products obtained by hydrothermal methods under various reactants. (a) NH4VO3 + SDS; (b) NH4VO3 + SDS + Ag (10%); (c) V2O5 + Ag (10%); (d) V2O5 + PVP + Ag (10%); (e) V2O5 + PVP; (f) V2O5 + CTAB + Ag (10%). 90

c. Effect of pH on the formation of Ag2V4O11 nanobelts

The production of Ag2V4O11 was achieved with high Ag/V ratio (50-100%) in the absence of any additives (Section 4.4.1.1). With such high ratios, the pH probably affects the + morphology as both V2O5 and Ag are sensitive to pH. In this section, the investigation of pH was conducted at the Ag/V ratio of 100%. The original pH of the reaction system was 6, while the pH was adjusted by 1 M HNO3 and 1 M NaOH solution.

Figure 4-7 The TEM images of the Ag2V4O11 nanobelts obtained at different pH: (a) pH = 13, (b) pH = 10, (c) pH = 6 (original), and (d) pH = 1. Figure 4-7 shows the TEM images of the products obtained at different pH. The products synthesized at different pH were mainly composed of the belt-like structure. At pH 13, the uniform nanobelts with width of ~ 100 nm can be obtained (Figure 4-7 a). However, some spherical nanoparticles with diameter of 50-100 nm appeared at this pH, suggesting the pH can significantly affect the formation of the products. Furthermore, by careful inspection, the difference among particles exists in length and width, consisting with experimental conditions. 91

The XRD technique was employed to confirm the compositions of the products prepared at different pH. Figure 4-8 shows the XRD patterns of the belt-like structures. It can be found that the increase of pH from 1 to 13 resulted in the decrease of intensity of the peaks corresponding to Ag2V4O11, especially for the (111) plane. In addition, most of the peaks (except for (020)) corresponding to the material obtained at pH 13 became weak, suggesting that this material were close to amorphous with the increase of pH when the concentration of vanadium and molar ratio of Ag to vanadium were fixed. In addition, Ag (111) peak can also be found in these patterns, and the intensity decreased with the increase of pH. The crystalline planes like (111) and (003) and others may increase or decrease in intensity, mainly dependent on the crystallinity, morphology and size of nanoparticles. Here why (111) plane change significantly but (020) insignificantly may be caused by structure change, but the nature needs further investigations in the future. Here, combined with Figure 4-7, it can be concluded that pH gives little effect on the morphology but composition.

The differences of the morphology and crystallinity in terms of pH can be further confirmed by HRTEM images and the corresponding SAED patterns. Figure 4-9 a shows the HRTEM images of the single nanobelt synthesized at pH 1 (left-hand side) while the figure located in the right-hand side is the corresponding SAED pattern. Two sets of typical crystalline planes could be indexed as {20 1 } and {210} for the Ag2V4O11 nanobelts, respectively.302 The alternately dark and bright rings located at the centre of SAED pattern indicates that some parts of the nanobelts were polycrystalline. Figure 4-9 b displays the single nanobelt obtained at pH 6 which is the original pH of the system, without adding any acid and basis. From the HRTEM image, it was found that the contrast of the nanobelt was dark, probably revealing that the nanobelt was composed of double layer. The corresponding SAED pattern located in the right-hand side confirmed the double layer structure of the nanobelts. With careful observation of the SAED pattern, it could be found that the strong bright points indexed as {210} were actually composed of two bright points which were closed to each other. The alternately dark and bright rings were not observed from this pattern, indicating that the nanobelt was single crystalline. The HRTEM image of 92

the product prepared at pH 13 is shown in the left-hand side of Figure 4-9 c, suggesting that the Ag spherical nanoparticles with dark contrast in 20-50 nm diameters attached on the surface of the nanobelt. On the other hand, it can be observed from the corresponding SAED pattern (right-hand side) that several diffraction spots irregularly distributed in the dark area and the diffused rings located at the centre. The irregular diffraction spots should generated from the Ag nanoparticles attached on the surface of the nanobelt whereas the diffused rings indicated that the nanobelt was more close to amorphous. The observation from the HRTEM images and corresponding SAED patterns is good agreement with that from XRD patterns. That is, the pH affects the crystallinity and composition of Ag2V4O11 nanobelts, while with the increase of pH, the crystallinity decreases and the amount of Ag nanoparticles increases. This may be caused by the various formation mechanisms of the nanobelts under different pH. The mechanism will be discussed in the later section.

Figure 4-8 XRD patterns of the Ag2V4O11 belt structures produced at different pH: (a) pH = 13, (b) pH = 10, (c) pH = 6 (original), and (d) pH = 1.

Figure 4-9 HRTEM images (left-hand side) and SAED patterns (right-hand side) of the single nanobelts obtained at different pH: (a) pH = 1, (b) pH = 6 (original), and (c) pH = 10. 4.4.2.2 Wet chemical method for synthesis of VOx@Ag nanocomposites

a. Effect of Molar ratios of Ag to V

+ The mixture of Ag , V2O3 particles and V2O5 particles was used to prepare VOx@Ag nanocomposites in aqueous solution at room temperature. Due to the diverse reactants, the molar ratios of the reactants are complicated, including the molar ratio of Ag to V2O3, V2O3 to V2O5, and Ag to total vanadium (V2O5 + V2O3). Among these ratios, the molar ratio of Ag to total vanadium can be calculated by the other two. In addition, according our previous study, the high ratio of Ag to V2O3 (≥ 30%) can cause the formation of Ag nanowires which preferentially attached on the surface of V2O3 so that it may be more difficult to adsorb on the surface of V2O5 particles (Figure 4-10 a). To avoid the formation 94

of Ag nanowires, two strategies can be applied: i) to keep the molar ratio of Ag to V2O3 lower than 30%; ii) addition of capping agents into the system to stop the growth of the nanowires.

For Strategy i), it meets the problem that the total Ag/V ratio may be too low to obtain enough Ag nanoparticles on the surface of V2O5. Table 4-1 shows the different molar ratios as mentioned before. V2O5 is the main component for gas sensing. For this purpose, the ratios of V2O3 to V2O5 need to be no more than 50%, which leads to insufficient

Ag/(V2O3+V2O5) ratio if the Ag/V2O3 ratio lower than 30%. Therefore, an appropriate capping agent is probably needed.

Table 4-1 The molar ratios of V2O3 to V2O5, Ag to V2O3, and Ag to total vanadium.

V2O3:V2O5/% 50 30 10

Ag:V2O3/% 75 50 30 20 10 30 20 10 50 30 20 10

Ag:(V2O3+V2O5)/% 25 15 10 5 3 7 4.6 2.3 4.5 2.7 1.8 0.9

To practise the first strategy, the investigation of the series of molar ratios was conducted. The morphologies of the typical products synthesized at the different ratios (Table 4-1) are shown in Figure 4-10. Figure 4-10 a shows the low magnification TEM image of the products obtained at high Ag/V2O3 ratio (75%). It can be clear observed that both Ag nanowires and V2O5 nanoparticles appeared. With a closer look of the V2O5 nanoparticles (Figure 4-10 b), no Ag particles was found to be attached on the surface. Other figures (Figure 4-10 c-j) show the different of the ratios in which the total Ag/V ratios gradually decrease. It is found that the surface of the V2O5 nanoparticles was naked in most case, whereas only Figure 4-10 g shows some big aggregated Ag nanoparticles around V2O5. Therefore, it seems that in order to ensure adequate amount of Ag nanoparticles, the ratio of

Ag to total vanadium must be high. For this purpose, the ratio of Ag to V2O3 needs increase because the ratio of V2O3 to V2O5 should be no more than 50%. On the other hand, when the ratio of Ag to V2O3 is higher than 30%, Ag nanowires will appear. Therefore, capping agents (the second strategy) need be used. 95

b. Effect of capping agents

The effect of capping agents mentioned in Strategy ii) were investigated in this section. Figure 4-11 shows the TEM images of the nanocomposites prepared by the assistance of different capping agents with the additive/V molar ratio of 300%. The other conditions are similar to those preparing the products in Figure 4-10 d (Ag/(V2O3+V2O5) = 10%). According to our study, the Ag nanowires grow by the mechanism of oriented attachment. That is, the Ag nanoparticles generated first, and then the nanoparticles fused together along one direction so as to form Ag nanowires. The capping agents would absorb on the surface of newly formed Ag nanoparticles so as to stop the fusion. The detail will be discussed in the later section (Section 5.4.3.1). From the TEM images, it can be found that PVP could successfully stop the fusion of Ag nanoparticles, whereas the Ag nanowires still existed in the presence of SDS. Thus, with the assistance of PVP, Ag nanoparticles with diameters of ~ 20 nm could attach on the surface of V2O5. However, the Ag nanoparticles still slightly aggregated and were difficult to evenly attach on the surface.

Figure 4-11 TEM images of the nanocomposites prepared with different capping agents: (a) PVP and (b) SDS. Based on the results above, we investigated the effect of different ratios on the formation of the nanocomposites in the presence of 300% (the PVP/V molar ratio) PVP, as shown in Figure 4-12 a-c. From the TEM images, it can be found that the amount of the Ag 97

nanoparticles significantly decreased with the total Ag/V ratio from 15% to 5%. It is unfortunate that the Ag nanoparticles still aggregated and did not evenly attach on the surface of V2O5 nanoparticles. This reason may be that the interaction between Ag nanoparticles and V2O5 is very weak as the most exposure surface is the V2O5 (010) planes where the oxygen atoms point outside.317 Figure 4-12 d shows the XRD patterns corresponding to the various products shown in Figure 4-12 a-c. It can be seen that the nanocomposites were composed of V2O5 and small amount of Ag. With the total Ag/V ratio decreased, the amount of Ag nanoparticles decreased. It is noted that the XRD pattern can show only crystal V2O5 in the product, however, there are still other vanadium species which may be amorphous such as V4+, as shown in Figure 4-3 a.

4.4.3 Formation mechanism of the nanocomposites

4.4.3.1 The formation of Ag2V4O11 nanobelts

Ag2V4O11 is a monoclinic crystal structure with the space group C2/m. The crystal structure of Ag2V4O11 consists of [V4O16] units that made of VO6 distorted octahedral sharing their apexes, which build infinite [V4O12] quadruple strings. These quadruple strings are further linked by the corner-shared oxygen atoms to provide the continuous [V4O11]n layers 300,304 separated by AgO5 trigonal bipyramid layers. Thus, this material trends to form 1D nanostructure. By hydrothermal method, V2O5 is dissolved and hydrolysed to form various - 3- - species such as VO3 , VO4 , and H2VO4 in both acidic system and basic system with such high concentration as the temperature increases. When the reaction proceeds to a certain extent, these species condense and polymerize to form a distorted VO6 octahedron, further 2- linking up through sharing the edges or the corners to form V4O11 framework structure. The silver ions can be accommodated in these framework tunnels.302

The formation of the material can be achieved by hydrothermal methods in acidic,304 neutral,318 and basic system.319 It is interesting that although the products are similar, the routes are different. The formation of Ag2V4O11 can be expressed by the following equations:

In the neutral and acidic system,

  Ag 32 OV 52  Ag 42 OV 11  2VO2 Eq. 4-1

In the basic system,

  Ag OH  AgOH Eq. 4-2

2AgOH Ag 2  2OHO Eq. 4-3

Ag 2 2 OVO 52  Ag 42 OV 11 Eq. 4-4

Ag 2O 42 Ag  O2 Eq. 4-5

These equations can be well explained the formation of naked Ag2V4O11 nanobelts (Figure

4-9 a and b) in acidic and neutral system and Ag doped Ag2V4O11 nanobelts in the basic system (Figure 4-9 c). However, it is still difficult to explain why the crystalline of (111) plane decreased when the nanobelts prepared in the basic system. The reason is studied in progress by our group.

4.4.3.2 The formation of Ag0.35V2O5 nanobelts

The precise formula of Ag0.35V2O5 can be represented as Ag(I)0.35V(IV)0.35V(V)1.65O5, which is a non-stoichiometric solid solution of silver in V2O5. The silver atoms in the solid solution are singly ionised.299 Therefore, the formation mechanism of the 1D nanostructure is similar to that of V2O5, as stated in Section 2.2.2.1.

In addition, a reducing agent with suitable reducing properties is significant for the formation of Ag0.35V2O5 by hydrothermal methods. This kind of reducing agents must be 5+ 4+ weak enough to reduce a small part of V in the V2O5 to V , instead of totally reducing 5+ the V so as to form VO2, especially under hydrothermal conditions (high temperature and + long reaction time). As compared in Figure 4-6, only V2O5 and Ag cannot lead to the + formation of Ag0.35V2O5 (Figure 4-6 c), and PVP can totally reduce V2O5 and Ag to VO2 and Ag (Figure 4-6 d and e), while Ag0.35V2O5 cannot be prepared without SDS although the different reactants can cause the impurity (Figure 4-6 a and b). That is, SDS owns a very weak reducing property at high temperature in aqueous solution, even weaker than that of PVP.

The reducing property of SDS is likely to generate from the decomposition. Prolonged heating at 40 °C or higher in aqueous solution can cause hydrolysis of SDS into fatty alcohols and sodium sulphate.320 As known, fatty alcohols own weak reducing property. Therefore, the true reducing agents in this case might be the fatty alcohols decomposed from SDS.

100

4.4.4 Gas sensing for amines

In this section, the gas sensing performances of the as-prepared nanocomposites were investigated via measuring the selectivity to organic ammines, ammonia, and ethanol. All the three materials (e.g., Ag2V4O11, Ag0.35V2O5, and VOx@Ag) are proposed as n-type semiconductors.49,305 Therefore, the gas sensing mechanism for the reducing gases (i.e., ammonia and ethanol) is similar to the discussion in the previous section (Section 4.2). However, the sensing mechanism for organic ammines is slightly different. Raible et al. suggests that the change of V2O5 nanofibres conductance which is induced by organic ammine adsorption possibly result from not only the adsorption on the fibres surface but also the intercalation of amine molecules into the layered structures, since previous studies reveals that the intercalation of alkyl ammonium ions can significantly increase the distance between the oxide layers.321 Thus, the increase of conductance may be attributed to electron 322 transfer from the analyte to V2O5 structure. Due to the similar layered and molecular 304 structure of V2O5 and SVO in this study, the sensing mechanism of detection of amines + should be similar to V2O5. In addition, the interaction between Ag and amine groups may enhance the absorption of amine.323 Therefore, the sensing mechanism of these materials to amines is attributed to the synergistic effect of the intercalation and absorption.

Working temperature is an important characteristic for a sensor, and significantly affects the sensitivity.324 Figure 4-13 shows the sensitivities to 100 ppm 1-butylamine of the sensors based on various materials at different working temperatures. The sensitivity is defined the same as that in previous chapter (Section 3.4.4). It can be found that the as- prepared Ag2V4O11 nanobelts obtained at pH 1 show much higher sensitivity than the other materials at the temperature range from 200 to 340 °C. The lowest sensitivity can be seen from the naked V2O5 nanoparticles, while Ag0.35V2O5 nanobelts and VOx@Ag nanocomposite show similar sensitivities which were between Ag2V4O11 nanobelts and

V2O5 nanoparticles. Furthermore, the optimized working temperatures for the three materials were around 260 °C, which may be attributed to the similar composition and molecular structures.

101

Figure 4-13 Sensitivities to 100 ppm 1-butylamine of the sensors based on different materials at different working temperatures. (a) Ag2V4O11 nanobelts obtained at pH 1, (b) Ag0.35V2O5 obtained at the Ag/V molar ratio of 15%, (c) VOx@Ag nanocomposites obtained at the Ag/total V molar ratio of 15%, and (d) the naked V2O5 nanoparticles.

Figure 4-14 Sensor response of Ag2V4O11 nanobelts to various concentration of 1-butylamine at optimized working temperature of 260 °C. 102

Figure 4-14 shows the typical isothermal response curve of Ag2V4O11 nanobelts towards 1- butylamine in the range 5-500 ppm at working temperature of 260 °C. It is manifest that the response increased with the increase of the gas concentrations. In addition, low detection limit (5 ppm) to the target gas can also be observed from this figure.

The sensitivities of the as-prepared nanocomposites towards 1-butylamine at different concentration are shown in Figure 4-15. The sensing sensitivities of the Ag0.35V2O5 nanobelts and the VOx@Ag nanocomposites obtained at the Ag/V ratios of 5%, 10%, and

15% are displayed in Figure 4-15 a, while those of the Ag2V4O11 nanobelts obtained at pH 1, pH 6 (original), and pH 13 are shown in Figure 4-15 b. It can be found that the sensitivity increased with the Ag/V ratio for VOx@Ag nanocomposites. The reason can be explained by the spillover process. That is, the induction of Ag nanoparticles can increase the total surface area, and may lead to the formation of new donor or acceptor energy states.325,326 In addition, the Ag0.35V2O5 nanobelts showed a similar sensitivity to VOx@10%Ag nanocomposites. Figure 4-15 b shows the curves of sensitivities to the gas concentration for the Ag2V4O11 nanobelts obtained at different pH. Obviously, the sensitivities increased with the decrease of pH, and Ag2V4O11 nanobelts obtained at pH = 1 showed fairly high sensitivity towards 1-butylamine, especially at high concentrations. Normally, the metal doped materials will show higher sensitivity than the naked materials.326 However, it is interesting that although Ag doped Ag2V4O11 nanobelts were obtained at pH 13, the sensitivity of this material were much lower than the naked Ag2V4O11 nanobelts which were obtained at pH = 1. This may be attributed to the different crystallinities. The worse crystallinity maybe results in the worse conductivity, which may further affect the sensing performance. In accordance with the sensing mechanism, the insufficient electron transformation property of materials will lead to low sensing signals. In addition, another reason affecting the sensitivity may be attributed to the adsorption of H+ or OH- ions on the + AgVOx surfaces. That is, more H would be more readily adsorb organic amine at pH=1; while at pH = 13, the adsorption of OH- ions would slightly repel organic amine molecules. Details need more work to conduct to investigate.

103

Figure 4-15 Sensitivity of the as-prepared materials to various concentration of 1-butylamine testing at working temperature of 260 °C. (a) Ag0.35V2O5 nanobelts and VOx@Ag obtained with the Ag/V ratios of 5%, 10%, and 15%; (b) Ag2V4O11 nanobelts obtained at different pH.

104

Figure 4-16 Relative selectivity of the sensors based on the various as-prepared materials to various gases at working temperature of 260 °C, and the gas concentration is fixed at 100 ppm.

Selectivity is another important characteristic of gas sensors. Theoretically, sensors should have high sensitivity to some gases and little or no sensitivity to other gases in the same surroundings. The sensitivities of the sensors based on the Ag2V4O11 nanobelts obtained at pH = 1, the Ag0.35V2O5 nanobelts, and the VOx@Ag nanocomposites obtained with the Ag/V ratio of 15% to several amines, ammonia, and ethanol were measured at the working temperature of 260 °C with the gas concentration of 100 ppm. The results are presented in Figure 4-16. For each material, the sensitivities towards the amines were found to be similar but slightly different. The order of the sensitivities to the ammines is as follows: 1- butylamine > t-butylamine > hexylamine. Because of high boiling point (~ 130 °C), the vapour of hexylamine easily condenses in the chamber so as to affect the diffusion of the gas. Therefore, among the three amines, all the materials showed lowest sensitivity to hexylamine. The lower sensitivity to t-butylamine than that to 1-butylamine (the

105

discrepancy < 0.2) may be attributed to the steric hindrance. That is, due to the branch chain of t-butylamine molecules, the adsorption amount of t-butylamine is slightly smaller than that of 1-butylamine. Furthermore, the Ag2V4O11 nanobelts exhibited quite high selectivity to organic amines versus ammonia, suggesting that organic amines could be easily detected in the mixture of amines and ammonia which own similar odour at low concentration. This amines-to-ammonia selectivity of Ag2V4O11 nanobelts was much higher than that of the other two materials. In addition, Ag2V4O11 nanobelts showed a modest sensitivity to ethanol.

4.5 Summary

By the hydrothermal method, Ag2V4O11 nanobelts and Ag0.35V2O5 nanobelts with width of

100-200 nm and the length of 5 μm can be prepared with V2O5 and AgNO3 at the molar ratio of Ag to V of 100% and 15%, respectively. There is no need for any additives to assist the synthesis of Ag2V4O11 nanobelts, while SDS as a weak reducing agent plays a key role in the synthesis of Ag0.35V2O5 nanobelts. In addition, the composition and crystallinity of

Ag2V4O11 are significantly influenced by pH, which has been proved to further affect the sensing properties of the materials. By a simple wet chemical method, VOx@Ag nanocomposites are prepared via Ag ions and the mixture of V2O3 and V2O5 particles at room temperature in aqueous solution. The sensing performance testing reveals that

Ag2V4O11 nanobelts obtained at pH = 1 are much more sensitive to amines than the nanobelts obtained at other pH and the other materials (Ag0.35V2O5 nanobelts and VOx@Ag nanocomposites) at optimum working temperature of 260 °C. The Ag2V4O11 nanobelts also show low detection limit and a significant high selectivity to organic amines versus ammonia. This study offers a promising sensing material for the detection of organic amines.

106

Chapter 5. Rapid synthesis and growth of silver nanowires induced by vanadium trioxide particles

5.1 Abstract

This study demonstrates a novel approach for rapid synthesis of silver (Ag) nanowires induced by vanadium trioxide (V2O3) particles in aqueous solution at room temperature. Silver nanowires have an average diameter of 20 nm and length up to a few micrometers by parametric optimization. The micro-structure of the silver nanowires was characterized by TEM, HRTEM, SEM, and XRD techniques. The optical property of the as-prepared product was measured by ultraviolet-visible (UV-Vis) spectroscopy. The possible growth mechanism of Ag nanowires via oriented attachment of Ag nanocrystals was discussed. The present approach shows several unique features such as rapid (a few minutes), reproducible and high-yield reaction with no need of any modifiers. V2O3 rods were reported for the first time to be used for synthesis of silver nanowires, playing multiple roles as reducing agent, template, and catalyst. The silver nanowires produced are promising for optical applications (e.g., SERS) due to their rough surface.

5.2 Introduction

Many efforts have been devoted to controlled synthesis and assembly of metal nanowires because of their potential function as interconnects or active components in fabricating nanodevices.65,327-329 In particular, one-dimensional (1D) silver nanostructures have become one of the most important members of these functional metal nanowires. Silver nanowires attract increasing attention due to their unique optical, electronic, and physicochemical properties. With proper shape and size control, they can be used in diverse applications including optical devices,330 SERS331,332, and sensors.333

Various methods have been developed to synthesize Ag nanowires to make use of their beneficial properties for large-scale application. These methods can be divided into two main categories, physical methods, e.g., electron beam induced lithography,334-336 and thermal vapour deposition,337 and chemical methods, e.g., photochemical,338,339

107

electrochemical,340-342 and wet-chemical methods,343 which have been proved to be powerful in shape control. However, limitations exist in the above methods. Many of these methods are restricted by critical experimental conditions such as high reaction temperature (> 100°C), long reaction time (tens of minutes or longer), and requirement for specified surface modifiers (surfactants or polymers). For example, in the electron beam induced lithography method,334-336 energetic beams of photons, ions, and electrons have to be used, while thermal vapor deposition method337 needs both high vacuum and high temperature (as high as 1173 K). Similarly, surface modifiers are needed in chemical methods, such as citric acid as capping agent in photochemical method,338,339 CTAB and TDAB as surfactants in electrolyte in electrochemical method,340-342 and PVP as template in wet- chemical method.267,343,344 These limitations may impede the development of efficient and cost-saving strategies for making silver nanowires efficiently, and with high yield and reproducibility. Therefore, a facile, efficient, and low-cost method needs to be explored to meet these challenges.

Recently, wet-chemical synthesis was extensively studied, showing that silver nanoparticles can be prepared at low temperatures in aqueous solutions with low cost, especially in our group. For example, Jiang et al.125,345 demonstrated that triangular and circular silver nanoplates (2.3 nm in thickness) could be successfully made by a self-seeding co-reduction method in aqueous solution. This method does not need external seeds, organic solvents, and can be carried out at room temperature. Other examples of synthesis of silver nanoparticles can be found in the literature. Yuan et al.346 suggested that silver nano/microparticles with different morphologies could be prepared by electrosynthesis with an H2O–oleic acid or an H2O–glycerol mixed solvent (volume ratio 1:1) as the electrolytic medium and AgNO3 as the supporting electrolyte. The silver nanoparticles exhibited good catalytic property in the reduction of methyl orange and methylene blue. Zhang et al.347 reported that Ag nanowire thin films can be synthesized by a wet chemical method at room temperature in aqueous solution. The formation of products was influenced by the concentration of the polymer (PMAA), pH and glass wall of the reactor. Jana et al.348 obtained silver nanorods with different aspect ratios from nearly spherical 4 nm silver nanoparticles by a seed-mediated growth approach in a rodlike micelle medium. Another

108

example of the wet chemical method is presented by Caswell and co-workers349 who synthesised silver nanowires in an aqueous solution, in the absence of seeds and surfactant. However, these methods could hardly avoid temperatures above 100°C, requirement of surface modifiers, and long reaction time.

The orientated attachment mechanisms of nanowires during synthesis are commonly reported in metal oxides. For instance, Halder et al.350 demonstrated that ultrathin single crystalline Au nanowires could be obtained by adding ascorbic acid to a mixture containing gold nanoparticles and aging at room temperature for an extended period. It was demonstrated that the formation of Au wires was due to the preferential removal of the capping agent of amine from the {111} planes of the nanoparticles followed by fusion.

SnO2 nanowires were obtained from an ensemble of 0D quantum dots through assistance of oleylamine in alcohols. They found that {010} is the most common attachment plane in the 351 352 nanowires. Similar growth mechanism also took place in CeO2 nanoflowers, and 353 MnO2 nanowires, etc. For silver nanoparticles, some low-dimensional structures formed by oriented attachment such as nanorods354 and dendrites355 were discussed. However, there have been but few reports on the formation and growth mechanism of Ag nanowires via orientated attachment. Understanding the mechanism will be useful for shape-controlled synthesis of nanowires.

In this paper we demonstrate a rapid wet-chemical approach, in which V2O3 particles are introduced for the first time to generate silver nanowires in aqueous solution at room temperature. This method shows several advantages, such as simple synthesis system, room-temperature and rapid (within a few minutes), and no need of seeds and external additives. By this approach, silver ions can be quickly reduced by V2O3 (in a few minutes), followed by nucleation and growth on the surface of V2O3 particles. The morphology and composition of V2O3 particles is shown in Figure 3-2(a) and Figure 3-4 (600 °C in N2). In the whole process, V2O3 particles play multiple roles, including acting as reducing agent, template and catalyst. The particle characteristics (morphology, size, and composition) are characterized by various advanced techniques such as TEM, HRTEM, EDS, SEM, XRD, and XPS. The possible formation and growth mechanism of silver nanowires through

109

oriented attachment is also discussed. Finally, the unique optical property of such Ag nanowires will be evaluated.

5.3 Experimental work

5.3.1 Materials

Ethylene glycol (EG, reagent plus, ≥ 99%), sodium orthovanadate (Na3VO4, ≥ 90%), HCl

(32%), SDS, CTAB, PVP (Mr = 55000), AgNO3 (analytical reagent), HNO3 (70%), and NaOH (analytical reagent) were purchased from Sigma Aldrich and Ajax Finechem Pty.

Ltd. V2O3 particles were prepared by the method described in Section 3.3.2. Pure water was purchased from Rowater, Australia. All the chemicals were directly used without any further treatment.

5.3.2 Synthesis of Ag nanowires

0.5 ml 0.01 M AgNO3 solution was rapidly dropped into 10 ml transparent suspension containing 0.2 mmol of the as-prepared vanadium trioxide (V2O3). The transparent suspension then became slightly brown though still transparent. Within 1 min, black cotton- like precipitate was quickly formed and deposited naturally on the bottom. This precipitate could be redispersed into water by sonication and the suspension was transparent again. Finally, the product was separated by centrifugation, and then rinsed several times using pure water.

5.3.3 Characterizations

The morphology of the samples was investigated by JEOL 1400 microscope (TEM) and Philips CM200 FEG microscope (HRTEM), operated at an accelerating voltage of 100 kV and 200 kV respectively. Scanning electron microscopy (SEM) was performed for microstructure on a Hitachi 900 microscope operated at 20 kV. Powder X-ray diffraction (XRD) pattern for sample composition was recorded on a Philip MPD diffractometer with Cu-Kα radiation. UV-Vis absorption spectra were obtained for tracking the formation and growth of silver nanowires on a Varian Cary 5 UV-VIS-NIR spectrophotometer. XPS analysis was conducted for the confirmation of redox with a Physical Electronics PHI 5000 110

Versaprobe spectrometer with Al Kα radiation (1486 eV). The angle between the crystalline plane and the analyzer was 45°. Analysis of the spectra was performed using the Physical Electronics Multipak software package. The solution pH was obtained on a Sartorius PB-10 pH Metre.

5.4 Results and discussion

5.4. 1 Synthesis of the Ag nanowires

To characterize the microstructure of the as-produced samples, TEM (HRTEM), SEM, XRD, and EDS techniques were employed in this work. Figure 5-1 a shows a SEM image of the as-prepared silver nanowires, which suggests that the product mainly contained relatively uniform nanowires with diameter of ~ 20 nm and length up to a few micrometers. The composition of the product was identified by XRD technique. Figure 5-1 b shows that all silver nanowires were metallic elemental silver (JCPDS file NO. 04-783). Notably, the intensity ratios between Ag (200) and (111), and between (220) and (111) peaks were smaller than the conventional values (0.14 vs. 0.4), and (0.12 vs. 0.25) based on the standard JCPDS card. These abnormal intensity ratios indicated that silver nanowires had abundant {111} facets. The small diffraction peak centered at 35.9º might be attributed to 4H silver rather than face-centered-cubic (fcc) silver structure.356 To confirm the purity of the nanowires, EDS for a single nanowire was carried out as shown in Figure 5-1 c that no other elements and phases appeared (except for copper element originated from TEM copper grid), indicating that the nanowires were composed of pure silver. In addition, to remove VOx, various methods will be used, e.g., centrifugation, membrane separation and pH adjusting. Their effectiveness needs more work to be performed in the future.

111

Figure 5-1 (a) SEM image of as-prepared Ag nanowires; (b) XRD pattern of the Ag nanowires; (c) EDS spectrum of a single Ag nanowire. Detailed TEM images of the as-prepared Ag nanowires are presented in Figure 5-2. Figure 5-2 a shows that the Ag nanowires were straight in shape, while some were formed with a bent structure similar to those generated via the glycothermal method, as observed in our recent study.357,358 Figure 5-2 b shows a high-magnification image of the thicker nanowires, indicating that the thicker nanowire was composed of several single nanowires aggregated side by side along the longitudinal direction. The results further prove the uniform diameters of the nanowires. Figure 5-2 c shows a single silver nanowire with rough surface, which might be formed by a mechanism to be described in a later section. The corresponding SAED pattern is shown in Figure 5-2 d, in which three sets of crystalline planes could be indexed: {220}, {311} and {2 2 2} respectively. Weak diffraction spots were found among the dominant spots, indicating that the nanowires probably contained some stacking faults during their formation.347,356,359 HRTEM technique was used to further identify the crystal structure, indicating that the domain crystalline plane was Ag {111}

112

with lattice space of 2.35 Å (Figure 5-2 e). The growth direction of the silver nanowires was clearly perpendicular to the {111} planes.

Since the synthesis rapidly occurred in a simple system at room temperature, the morphology of the product was mainly affected by two parameters: molar ratio of silver to vanadium oxide and concentration of AgNO3. To obtain shape and size controlled silver nanowires, the experimental parameters were optimized in this work as demonstrated below.

113

5.4.2.1 Molar ratio of silver to vanadium oxide

Figure 5-3 TEM images of silver products obtained under different molar ratios of silver to vanadium: (a) 0.05:1, (b) 0.3:1, (c) 0.5:1, (d) 5:1, (e) 10:1, and (f) 100:1. All scale bars are 2 μm. This molar ratio is one of the critical influencing factors in the shape and size controlled synthesis. The molar ratio was adjusted from 0.05:1, through 0.3:1, 0.5:1, 5:1, 10:1, to

100:1, while other parameters were kept constant ([V2O3] = 0.2 mM). The TEM images in Figure 5-3 show silver nanoparticles with different shapes and sizes formed at different molar ratios. At low molar ratio (0.05:1, Figure 5-3 a), instead of wires, spherical particles were formed on the surface of V2O3 (marked by arrows). Nanowires appeared at the ratio of 0.3:1 (Figure 5-3 b), though the nanowires were so soft that they could intertwine on the rod-like V2O3. The inset in Figure 5-3 b shows an enlarged TEM image of the V2O3 rod intertwined by Ag nanowires. With increasing molar ratio, relatively uniform silver nanowires with diameter of 20 nm could be collected through a wide range, from 0.5:1 to 5:1 (Figure 5-3 c-d). When the ratio increased to 10:1, the branch-like particles were

114

formed (Figure 5-3 e). Through careful observation, the formation of branch-like particles was caused by the aggregation among the individual nanowires. With the ratio further increasing to 100:1, the appearance of diverse morphologies including thin film structure was observed, which might result from the rapid nucleation and growth of the nanowires (Figure 5-3 f) due to such high ratio. That is, the molar ratio of silver to vanadium oxide greatly affects the morphology and size of the obtained silver nanoparticles. The suitable ratios of silver to vanadium oxide for generation silver nanowires are 0.5:1 to 5:1.

5.4.2.2 Concentration of AgNO3

The concentration of reactant(s) can also affect the morphology and size of the Ag nanowires. The TEM images in Figure 5-4 illustrate that the shape and size of silver nanoparticles were heavily affected by the concentration of silver nitrate. At low concentration (0.2 mM of AgNO3), relatively uniform nanowires could be obtained (Figure 5-4 a). The lower concentration (such as [Ag+] = 0.1 mM) could be found in previous section (Section 5.4.2.1, the Ag/V ratio =0.5, [V] = 0.2 mM, Figure 5-3c). To achieve such 115

+ low concentration of Ag , the corresponding amount of V2O3 particles must be as low as 0.00003 g which is impossible to accurately scale, if the reaction occurred in 10 ml. If using diluted suspension (for example 0.0030 g/10 ml dilute 100 times), the exact concentration of V2O3 for each sample may be very nonuniform. Therefore the investigation of the concentration (< 0.2 mM) cannot be achieved by directly decreasing the concentration of Ag+. That is why the investigation was put in previous section in order to avoid repetition. When concentration increased to 1 mM, the nanowires were rougher and softer (Figure 5-4 b). Considering the formation mechanism of the nanowires (see the following section), the rough and soft nanowires were probably caused by orientated attachment of non-uniform silver nanoparticles generated at an early stage. High concentration of Ag+ could accelerate the rate of nucleation and growth of the initial Ag nanoparticles, and further make them non-uniform. Higher concentrations of 2−3 mM would evidently not be suitable for generating 1D silver nanostructure. The particles were of radial morphology, as shown in Figure 5-4 c and d, especially in the inset SEM image of Figure 5-4 d. The concentration of silver nitrate, with its important role in the formation and growth of silver nanowires, is another crucial impact factor in shape and size control.

5.4. 3 Formation mechanism

The formation of the nanowires was fairly rapid. The UV-Vis spectra shown in Figure 5-5 could be used to prove the rapid formation and to track the growth process. The spectra show that there was a small absorption peak at ~ 620 nm for V2O3 suspension, and another peak at ~ 300 nm for blank AgNO3 solution. After adding AgNO3 solution to V2O3 suspension, the plasmon resonance immediately appeared around 350 nm, which was assigned to the Ag nanowires. In particular, the plasmon resonance band was nearly kept the same at 350 nm (tested for ~ 30 min), suggesting that the formation of the silver nanowires was completed within a short time. We propose that the rapid formation is inextricably linked with the formation mechanism. The reaction of the formation of the nanowires is obviously a typical solid-liquid surface reducing reaction. Via our study, it was found that the formation process of the nanowires should be divided into four parts: absorption, reduction, desorption, and fusion.

116

The formation of Ag nanowires achieve by surface reaction. The pH of the V2O3 suspension would reflect the surface property of V2O3. It was found that the pH was lower + than that of Di-water. The pH (~ 4.30) of V2O3 suspension before adding Ag solution reveals that some H+ ions may have been released from water (see Eq. 5-1 below). The surface of V2O3 would be negative due to the weak bonding with the hydroxyl group. The + negatively charged sites on the V2O3 particles would adsorb Ag ions, so as to enhance the surface redox reaction, which has positive influence on nucleation and growth of silver nanocrystals.

5.4.3.2 Reduction

0 0 + V2O3 is a strong reducing agent (E V+3/V+4 = ‒0.337 V and E V+4/V+5 = −0.991 V) to Ag 0 (E Ag+/Ag0 = 0.799 V). The XPS result (Figure 5-6) reveals that the product mainly contained pentavalent vanadium accompanied by a small amount of tetravalent vanadium indicating

117

that most of the trivalent vanadium was transformed to pentavalent vanadium.360 Such high redox potential provides huge energy drop to accelerate the reduction reaction.

Figure 5-6 XPS spectrum of vanadium oxides showing the redox reaction happening in the 4+ formation of silver nanowires, in which the peak of V2p3/2 (516.2 eV) could be assigned to V , 5+ while V2p3/2 (517.5 eV) to V .

5.4.3.3 Desorption Desorption of the Ag nanoparticles is probably due to the following reason. The high + + reaction rate between Ag and V2O3 and the low ionization potential of Ag (EAg=7.6 eV)361 may easily make Ag particles grow larger,362 and then the neutral Ag particles without electrostatic attraction moved away from the surface of V2O3 particles and entered in water. That is also the reason why the Ag nanowires can separate from V2O3 particles, instead of forming VOx@Ag core-shell structures. However, some of the nanoparticles may still attach onto the surface of V2O3 particles by van der Waals forces. The nanowires formed on the basis of such Ag nanoparticles will wrap around V2O3 rods (Figure 5-3 b) or form a radial structure.

5.4.3.4 Fusion Due to the rapid reaction, it is hard to capture the growth process of the Ag nanowires. Therefore, a growth inhibitor is required to track the rapid formation process of silver 118

nanowires. In this study, several surfactants (CTAB and SDS) and polymer (PVP) were tested as capping agents. 0.1 ml and 0.5 ml 0.1 M additives (molar ratios of additives to Ag+ were 2:1 and 10:1, respectively) were added into the suspension, followed by adding + Ag . Due to the negatively charged surface of V2O3 particles (proved in the previous section), cationic surfactant cations ([CTA]+) may first absorb on the surface to impede the absorption of Ag+. Moreover, different amounts of CTAB lead to different formation process. For low molar ratio of CTAB (0.1 M), no precipitate appeared, though some precipitate could still be found after centrifugation. The corresponding TEM image is shown in Figure 5-7 a, indicating that no nanowires were formed other than rather spherical nanostructures. In this system, cation with large molecular weight would be preferentially + absorbed on the surface of V2O3 particles, which could prevent the absorption of Ag . The white particles easily disappeared when exposed to sunlight. Therefore, it is proposed that the particles in Figure 5-7 a might be AgBr which probably came from the reaction between Br- and Ag+. For high molar ratio of CTAB (0.1 M), white precipitate AgBr immediately appeared after adding Ag+ into the suspension due to high concentration of Br- (Figure 5-7 b). That is, the CTAB is not suitable for shape control of silver nanowires in our proposed system, though others used it for shape/size control of silver nanoparticles.363

For SDS, due to the anionic group in the molecule, the function of stopping or slowing down the reaction is not obvious, since the negatively charged molecules are difficult to absorb on the negatively charged surface of V2O3 particles. Figure 5-7 c shows the Ag nanowires formed when 0.1 ml 0.1 M SDS (SDS:Ag+ = 2) was added, which is similar to Figure 5-4 c, indicating that the low molar ratio of SDS has little affect on the formation of Ag nanowires. However, for high molar ratio of SDS, SDS molecules can absorb on the freshly formed Ag nanoparticles, due to their long molecular chains. Therefore, the chain- like structure can be observed from Figure 5-7 c and d.

119

The effects of PVP are different from those of CTAB and SDS. Furthermore, differences also appeared in high and low molar ratios of PVP (Figure 5-7 e and f). At a low molar ratio of PVP, the nanowires and nanospheres coexisted. With increasing molar ratio, only nanospheres were observed, indicating that PVP had successfully stopped the growth of silver nanowires. The neutral molecule PVP would adsorb on the surfaces of both V2O3 particles and the fresh Ag nanoparticles, due to the influence of charge. The effect of stopping the growth of Ag nanowire was obvious for high molar ratio of PVP. Therefore, high molar ratio of PVP (PVP:Ag+ = 5:1) was used to stop the growth of Ag nanowires.

PVP molecules can be adsorbed not only on the surface of V2O3 to slow down the reaction rate, but also on the surface of the freshly-formed Ag nanoparticles to stop the further growth of Ag nanowires. Figure 5-8 a to d shows the TEM images taken from the samples prepared at different time when PVP (PVP:Ag+ = 5:1) was added into the reaction system.

120

At the beginning of the reaction, spherical nanoparticles with size of ~ 20 nm were formed (Figure 5-8 a). With increased time, the adsorption of PVP molecules on the surface of the nanoparticles would somehow stop their further growth (Figure 5-8 b). After a few seconds, the length of the nanowires increased obviously, while the diameter kept constant (Figure 5-8 c). Over ~ 1 min, almost all nanowires were formed (Figure 5-8 d). Figure 5-8 e shows the UV-vis spectra of the Ag nanoparticles stopped by PVP during the formation of Ag nanowires. It can be seen that the maxim absorption blue shifted and the intensity of the peak increased with time. Figure 5-8 f displays the change of position of maxim absorption with growth time, clearly suggesting that the peak gradually shifted from 470 to 350 nm within 90 s, also indicating that the shape of the Ag nanoparticles transformed from sphere to wires.

In fact, PVP has recently been used for shape and size control in some anisotropic growth of silver nanostructures. For example, Sun and Xia267 demonstrated that PVP, as a coordination reagent, can help to obtain uniform silver nanowires in a polyol-thermal system. They proposed that the PVP macromolecules might coil around the nanowires as they grow along their longitudinal directions. It is generally accepted that this coordination reagent kinetically controls the growth rates of various faces of a metal through selective adsorption and desorption on these planes. A similar situation can also be found in our previous study.357,358 Here, the role of PVP is different from the previous study due to the different growth mechanism of nanowires. In fact, in both systems, the PVP molecules will coil around the initial nanoparticles (the seeds in polyol-thermal system), thus leading to the failure of fusion among Ag nanospheres in our case, and kinetical growth in poly-thermal system. This role of PVP in our case also confirms to some degree the growth mechanism of Ag nanowires.

121

Figure 5-8 The effect of PVP on the growth of silver nanowires at different growth times: (a) at the initial stage, (b) 20 s, (c) 40 s, and (d) 90 s; while (e) UV spectra of the nanoparticles stopped by PVP at from 0 s (corresponding to Figure 5-8 a) to 90 s (corresponding to Figure 5-8 d); (f) relationship between peak position of maximum absorption and growth time. Figure 5-9 shows TEM images of a forming nanowire with the assistance of PVP. At the initial stage, some small silver nanoparticles (10−20 nm in diameter) were formed (Figure 5-9 a), which then aggregated and fused together to form a chain-like nanowire (Figure 5-9 b). To further identify the chain-like structure, the HRTEM technique was employed. Figure 5-9 c shows the joint portion of two nanoparticles, revealing that the chain was formed by a head-to-tail oriented growth through lattice match of the Ag {111} planes. These nanoparticles comprised nanowires were actually single-crystal rods which were made of well-ordered parallel {111} planes. Such aggregation of nanoparticles may lower the surface energy in the whole system. In addition, the concavo-convex surface of the nanowires also confirmed the head-to-tail oriented growth of the individual nanoparticles.364 Jiang et al.365 suggested that particle shape and the force among particles are the two factors which substantially affect the self-assembly of particles. In our case, the single-crystal rods were able to fulfil the condition of particle shape, though the driving force is yet unknown. Normally, fusion happens between the planes which own high surface energy, thus lowering the whole system energy. For fcc metals, the {111} planes are most stable, which means the lowest surface energy, then followed by the {100} and 122

{110} planes. Here, we do believe that the fusion by Ag {111} planes plays a key role in the formation of silver nanowires, though the exact mechanism calls for further investigation.

Similar scenario of oriented attachment has appeared in other silver nanostructures such as nanorods and dendrites. Sun et al.354 reported that Ag nanorods could be obtained in a magnetic field, in which the formation of nanorods was driven by twinned defects. Wen and co-workers355 synthesized Ag nanodendrites with the aid of Zn microparticle suspension, demonstrating that diffusion and oriented attachment of silver clusters along different directions (e.g. <100> and <111>) in alternation led to the formation of Ag nanodendrites. However, both above methods hardly led the particles to forming longer Ag nanowires. Formation by oriented attachment can also be found in other nanostructures, e.g.,

123

350 351 352 353 Au nanowires, SnO2 nanowires, CeO2 nanoflowers, and MnO2 nanowires, although different conditions were employed.

In summary, the mechanism can be described that: the positive charged Ag+ first absorbed + 0 on the negative charged V2O3 surface in aqueous solution; then Ag were reduced to Ag nanoparticles by V2O3 particles; the initial Ag nanoparticles then desorbed from the surface of V2O3; the final step is the fusion of (111) planes among the initial Ag nanoparticles.

5.4. 4 Roles of V2O3 in the formation of Ag nanowires

In the proposed synthesis strategy, V2O3 particles obviously play important roles in the formation of Ag nanowires, as will be discussed in detail in this section.

5.4.4.1 As a reducing agent As well-known, the reducing agents used to synthesize Ag nanowires, could be roughly classified into strong and weak. For the strong reducing agents (e.g. NaBH4, ascorbic acid), they lead to rapid growth at low temperature but not to uniform size. Therefore, they are always used with capping agent (trisodium citrate, PVP, or other surfactants) to slow down the growth rate, to further terminate kinetic growth. For example, Jana et al.348 used ascorbic acid and NaBH4 as reducing agents, CTAB as micelle template, and sodium citrate as stabilizer, to prepare Ag nanorods and nanowires in water at room temperature. NaOH was also used to adjust the pH of the system. By using this method, aspect ratio could be controlled by the adjustment of pH. Another typical example of preparation by a strong reducing agent was reported by Zhang.347 In their study, Ag nanowires were prepared by a wet chemical method in aqueous solution at room temperature, using ascorbic acid as a reducing agent, and PMAA as a shape controller. From these examples, we may conclude that the preparation of Ag nanowires by a strong reducing agent is always carried out in complicated systems in which capping agents need to be added to slow down the growth rate. In contrast, the low reaction rate driven by a weak reducing agent (e.g., PVP and EG) can lead to uniform growth, but accompanied by long reaction time and high temperature. Sun and Xia267 demonstrated that Ag nanowires could be obtained by a glycol-thermal method under assistance of PVP at 160ºC for 60 min. In their study, ethylene glycol acted as a reducing agent and a solvent, while PVP acted as a coordinating agent. In fact, some of 124

the weak reducing agents also act as capping agents such as trisodium citrate and PVP. For instance, Caswell et al.349 synthesized Ag nanowires in trisodium citrate solution with proper pH. In addition, Washio et al.366 reported that Ag nanoplates could be obtained via reduction of Ag+ by PVP at 60ºC for 21 hours. In both cases, trisodium citrate and PVP acted respectively as reducing agent and capping agent. In a word, rapid reduction is not favored to obtain well shape-controlled nanoparticles, and either strong reducing agents need addictives to slow down reaction rate or weak reducing agents need long reaction time and higher temperatures.

Using the reducing property of V2O3 to make Ag nanowires is an excellent idea. In our case, + although V2O3 is actually a strong reducing agent to Ag , both uniform morphology and rapid growth can be achieved simultaneously, which is attributed to the formation mechanism of the Ag nanowires. The initial Ag nanorods were rapidly formed by strong reducing agent V2O3, followed by fusion between their {111} planes. In addition, the surface of V2O3 rods changed significantly after reduction (Figure 5-10). That is, the high ratio of silver ions to vanadium oxide could lead to the smooth surface destroy or more defects after redox happens.

Figure 5-10 Morphology change of V2O3 particles before (a) and after (b) reaction (the molar ratio of silver to vanadium is 100:1).

5.4.4.2 As a template Based on the previous discussion, the formation of the small Ag nanorods undergoes three steps: adsorption, reduction, and desorption. Therefore, the surface property of V2O3

125

particles needs to be investigated. To investigate the effects of pH on the formation of Ag nanowires and verify the surface property of the V2O3 particles, the pH of the V2O3 + suspension was adjusted to 1, 7 and 10 by adding HNO3 and NaOH solutions, and then Ag was added into the suspension. Unfortunately, no precipitate was found in the system at pH

1, since the V2O3 particles were dissolved in such acidic system. Evidently the appearance time of the precipitates was extended with the decrease of pH, indicating that a proper basic system favored the rapid formation of Ag nanowires. The corresponding HRTEM images of the as-prepared Ag nanowires are displayed in Figure 5-11, clearly showing the difference of the nanowires at pH = 4 (original pH) (Figure 5-11A), 7 (Figure 5-11B) and 10 (Figure 5-11C). With increasing pH, some small nanoparticles appeared together with Ag nanowires. XRD patterns shown in Figure 5-11D confirmed that the small particles was

Ag(VO3)H2O.

126

+ The redox reaction between V2O3 and Ag proceeds rapidly, as shown by the instantaneous decrease of pH from 4.30 to 3.83 after the addition of Ag+. The pH drop was probably caused by further releasing H+ during the redox process, as might be speculated by Eq. 5-2 – below, in which the relative species [VO3(H2O)] was supposed from the XRD pattern (Figure 5-11D) and consistent with the reference.6 When OH– was added into the system, Eq. 5-1 would move to the right due to the reaction between H+ and OH–, which would lead – to the increase of (V2O3)x(OH )y. Eq. 5-2 would then move to the right as well, resulting in – the increase of [VO3(H2O)] . The peaks of AgVO3(H2O) (Figure 5-11D) became more obvious with pH. That is also why the reaction rate increased with pH.

y  yH2O xV 32  OVO 32 x OH y  yH Eq. 5-1

 y 5x  2  OV6xAgOHy 32  OH y  x Eq. 5-2 0  4xAg  23 OH2xAgVO  6x  Hy

Apparently, the template (V2O3 particles) in our case is different from those used to assist the growth of nanowires. The templates helping growth of nanowires belong mainly to two types: “hard” and “soft”. Soft templates mean that their structure is flexible and tunable in solution, such as micelles, reverse micelles, and DNA molecule networks. A good example is presented by Pileni and co-workers367 who demonstrated that Ag and Cu nanoparticles could be prepared via reverse micelles (AOT/isooctane/water reverse micelles) as a “soft” template. They found that the Ag nanodisk size could be tuned between 30 nm and 100 nm by changing the ratio of reducing agent (hydrazine) to Ag+. Another example of such template was presented by Braun et al.,368 who used DNA as a “soft” template to prepare Ag nanowires, 12 mm long and 100 nm wide. Different from the “soft” template, “hard” templates (mesoporous materials, alumina membranes, carbon nanotubes etc.) can normally give uniform size, and are stable and can restrict the formation and growth of metallic nanoparticles under different conditions. Yu et al.369 demonstrated that long and continuous silver nanowires within multi-walled carbon nanotubes (MWCNTs) were formed after irradiating the suspension of MWCNTs-sliver nitrate solution in the presence of 2-propanal due to radiation-induced reduction of silver ions. The Ag nanowires could be as long as 2 mm, with well-oriented structure. Another example of the “hard” template is presented by

127

Zong et al.,370 who reported that transparent Ag nanowires arrays embedded in anodic alumina membrane could be prepared by a template-based approach combining ac electrodeposition and subsequent etching of substrate. After the ac electrodeposition, the anodic alumina channels were filled with Ag nanowires. However, in our case, the role of template for V2O3 is not associated with shape formation, but nucleation. With V2O3 in solution, the rough surfaces of V2O3 particles (Figure 5-10) provide not only absorption sites but also nucleation sites, which leads to heterogeneous nucleation that has been proved to more easily obtain precipitate (Ag nanoparticles) than in homogeneous system. This is probably another reason for the rapid formation of Ag nanowires.

5.4.4.3 As a catalyst or “microelectrode”

Besides the two basic roles of V2O3 particles discussed above, we propose that V2O3 particles also play a role on the formation of the wire-like structure, and on the high reaction rate (the reaction completes within a few minutes). Actually, the formation of nano-wires is a typical catalytic process for V2O3 (absorption, reaction, and desorption). In 371 addition, Dupuis et al. reported that V2O3 (0001) film can be prepared on Au {111} planes by evaporation of metallic vanadium in an oxygen atmosphere, followed by annealing at 700 K. Due to the similar lattice structure of Au and Ag, we propose that the

{0001} surface of our V2O3 particles could also induce the generation of Ag {111} to result in the initial rods composed of large area {111} plane with large surface energy. To reduce such surface energy, the fusion between the Ag {111} planes occurred, as confirmed by the HRTEM observation (Figure 5-9 c). Moreover, Michaelis et al.372 demonstrated that the silver could be immediately formed and deposited on Pt particle surface, but not when the core particles were absent. A similar situation may occur in our case. That is, the V2O3 particles can also act as “microelectrodes”, which immediately reduce silver and make the

Ag nanorods deposit on V2O3 particles surface.

Combining the two examples, we deduce that the quick reaction rate and the formation of the regular shape by simple redox reaction without any surface modifier are associated with

V2O3 particles in this system. However, this role still needs further confirmation.

Nonetheless, the V2O3 particles are of significant importance and are first reported in the formation of silver nanowires. 128

373 In addition, V2O3 is antiferromagnetic, with a Néel Temperature of 168 K. For antiferromagnetic materials, when temperature is higher than Néel Temperature, the antiferromagnet shows the same magnetic behavior as paramagnet. Without external magnetic field, paramagnet does not show any magnetism in macroscopic perspective. On the other hand, Ag is diamagnetic when external magnetic field exists. In our case, no external magnetic field was used, and thus the magnetism of V2O3 has little effect on the formation of Ag nanowires.

5.4. 5 Optical property of the Ag nanowires

The optical property of silver nanoparticles is structure-dependent. Figure 5-12 A shows the UV–Vis absorption spectrum of the chain-like silver nanowires (Curve a), where only one broad plasmon resonance peak centered at ~ 353 nm was observed. In comparison, the Ag nanowires generated via a high-temperature glycol-thermal method357 displayed two plasmon resonance peaks. An intense peak centered at ~ 383 nm (Curve b) could be assigned to the transverse plasmon resonance, while a shoulder located at ~ 353 nm is the same as the peak position in Curve a. In both cases the longitudinal plasmon resonance was not significant in the UV-Vis spectra, as is consistent with observations in previous studies.349

Figure 5-12 (A) UV-vis spectra of as-synthesized Ag nanowires (a), as compared to Ag nanowires obtained by glycol-thermal approach (b);357 and (B) TEM image of Ag nanowires prepared by glycol-thermal approach.

129

Generally speaking, gold and silver nanorods and nanowires have two principal plasmon bands. The band at short wavelengths corresponds to absorption and scattering of light along the short axis of the nanorod/nanowire (transverse plasmon band), and the band at long wavelengths corresponds to absorption and scattering of light along the long axis of the nanorod (longitudinal plasmon band). The plasmon bands are tunable from the visible to the near-IR with the nanorod aspect ratio.374 However, the band usually disappeared for silver nanowires with large aspect ratios.349,375,376 This result may be rationalized by considering that long silver nanowires usually do not conform to any uniform aspect ratio since they can easily bend or twist in solution, thus significantly decreasing the intensity of the longitudinal plasmon band leading to the disappearance of the band. In some cases, transverse plasmon band may appear with a shoulder (the peak located at ~ 350 nm in Curve b of Figure 5-12 A). The shoulder should be assigned to bulk Ag.375,376

The discrepancy of the UV spectra is probably due to the structures of two types of nanowires. Figure 5-12 B displays the morphology of the nanowires prepared by high- temperature method. Compared to the nanowires generated by our method, the nanowires formed via oriented attachment show rough surface, imperfect crystallization (Figure 5-2 c and e), and most importantly “soft” structure due to high aspect ratio; while the Ag nanowires generated by the high-temperature glycol-thermal approach have smooth surface, twinning structures, and “tough” structure.357 The “soft” structure made the nanowires bend easily and coil together, which might lead to similar optical property with bulk Ag (one broaden peak at 353 nm). For the Ag nanowires generated by the high-temperature method, the “tough” structure always made the nanowires “straight”. Therefore, these nanowires show typical UV-vis absorption spectrum of Ag nanowires (two peaks at 353 and 383 nm). In addition, besides the bend structure the aggregation of several nanowires along the longitudinal direction (Figure 5-2 b) might also lead to varied “apparent sizes” further leading to the broadened peak for Curve a in Figure 5-12 A.

5.5 Summary

We have demonstrated a facile but effective method for the synthesis of Ag nanowires with uniform diameter and reproducibility under ambient conditions. By this method, Ag

130

nanowires can be rapidly prepared in aqueous solution at room temperature. The formation of the Ag nanowires undergoes four steps: absorption, reduction, desorption, and fusion, assisted by V2O3 particles which are used for the first time in the synthesis of silver nanowires, displaying multiple functions such as a reducing agent, a template and a catalyst. The silver nanowires show optical properties different from those prepared by the glycol- thermal method, probably due to the rough surfaces. The proposed method shows many advantages, e.g., cost saving and rapid synthesis. However, we need to point that the removal of V2O3 particles from product seems a little difficult, and more work needs to be performed in the future. In addition, without considering separation, we propose that the product of VxOy-Ag nanocomposites can be obtained under appropriate control, and would be useful for gas sensors167 or catalysts.377

131

Chapter 6. Synthesis of bimetallic Ag-Au nanowires by induction of vanadium trioxide and its catalytic property for reduction of 4-NP

6.1 Abstract

Silver-gold (Ag-Au) bimetallic nanowires have been controllably synthesized by a newly developed wet-chemical method at room temperature. The Ag nanowires and Au nanoparticles could be sequentially formed by the reduction by vanadium oxide (V2O3) particles so as to form Ag-Au bimetal, in which the Ag nanowires show a diameter of ~ 20 nm and length up to 10 μm. A few unique features could be found in our developed approach, such as rapid (within a few minutes), controllable in shape and size, reproducible, and no need of any surface modifiers. The formation and growth mechanisms of such Ag-

Au bimetallic nanostructures driven by lattice match and unique reducing agent (V2O3) have been proposed in this study. Molecular dynamic (MD) simulation was used to confirm the mechanism by investigating the interaction between Ag/Au and Au/V2O3. Moreover, the application of such bimetallic nanoparticles for catalytic reduction of 4-nitrophenol (4- NP) to 4-aminophenol (4-AP) was performed, and they exhibit superior catalytic property to the Ag nanowires, Au nanoparticles, and Ag-Pd bimetallic nanostructures prepared under the reported conditions. These Ag-Au bimetallic nanoparticles could be explored to be a highly efficient catalyst in reduction of 4-nitrophenol. This study may open a new path for generation of other bimetallic nanostructures with excellent catalytic efficiency.

6.2 Introduction

Bimetallic nanoparticles have attracted increasing attention because of their specific optical, electronic, magnetic, and catalytic properties greatly distinct from the corresponding monometallic particles,378-380 especially for those noble metals (e.g. Ag-Pt381, Ag-Pd382, Pt- Pd,383 Au-Pt381,383 and Au-Pd384). For example, catalytic activities of Au core structured bimetallic nanoparticles (Au-Pt, Au-Pd, Au-Rh) for hydrogenation and water reduction are higher than that of Pt, Pd, and Rh monometallic nanoparticles.385 On the other hand, bimetallic nanoparticles can also save cost through replacing noble metals by relative cheap

132

metals. A good example was reported by Lu,386 who demonstrated that the Cu/Pd (4/1, mol/mol) alloy show the similar performance of SERS as pure Pd colloids.

Of the achieved so far, Ag-Au bimetallic nanostructures have been extensively investigated because of their excellent performance on SERS386,387 and catalysis.388 To obtain such particles, many efforts have been made in the past. The Ag-Au bimetallic colloids could be prepared by simultaneous co-reduction of two kinds of metal ions with or without protective agents, or by successive reduction of one metal over the nuclei of another to form Ag-Au alloy and Au@Ag core-shell structure,389,390 such as alcohol citrate reduction,391 seeding growth method,392 poly process,390 and alcohol reduction method.393 Different reducing agents for generating such Ag-Au nanoparticles have been used, such as sodium citrate,394 sodium borohydride,395 ascorbic acid,350 and hydroxylamine 396 hydrochloride (NH2OH∙HCl). Among them, NH2OH∙HCl was considered well- acceptable to successfully prepare Ag@Au core-shell structure.392,397 Ivana et al.392 demonstrated that the layer core-shell Ag@Au structure could be prepared by a seed 397 growth method using NH2OH∙HCl. Another example was reported by Shahjamali et al. who demonstrated that Ag@Au core-shell structure was prepared by injecting both

NH2OH∙HCl and HAuCl4 solution into Ag nanoparticle suspension.

Despite some successes, limitations still exist, such as long reaction time (> 3 hours),398 high temperature (> 100°C),391 and carefully chosen surfactant(s).380,389,398 Specifically, the synthesis of Ag@Au core-shell nanostructure remains challenging because of significant etching of Ag cores by Au ions, known as galvanic replacement which usually leads to the generation of nanoframes399 or hollow nanoboxes400 or destroy the original silver structures. In addition, the preparation of Ag@Au core-shell structures was commonly limited in spheres, triangles or rods,392 but little reported about the Ag-Au nanowires bimetallic structure. The mechanism how to keep silver with less- or non-destroyed structure needs further to be understood.

Catalytic applications of Ag and Au nanoparticles have attracted considerable attention in recent years.401-404 Moreover, Ag-Au bimetallic structures are of excellent catalysts. The main active component for catalysis is proposed as Au rather than silver.401,405-407 The

133

catalytic activities are largely affected by the size of Au nanoparticles, the smaller the better.408,409 However, the smaller Au nanoparticles could easily aggregate and hence result in a considerable decrease in the catalytic activities. To avoid aggregation, a suitable substrate is required, such as crystalline cellulose single nanofibres,409 porous carbon spheres,410 and poly(methyl methacrylate).411 Further investigation shows that the catalytic performance of Au nanoparticles adsorbed on these substrates, however, is not as good as those pure Au nanoparticles.407-412

To solve the problems, Ag nanowires were proposed to be a fantastic/ideal substrate because of good catalytic activity,401 excellent electron transfer property,413,414 similar lattice parameter with Au (a = 0.409 nm for Ag and 0.408 nm for Au),392 and relatively low cost.401 In this study, we first report a facile and efficient wet-chemical method to prepare

Ag-Au bimetallic nanowires, assisted by V2O3 particles at room temperature. By this method, the Ag-Au wire-like nanostructure can be simply and rapidly prepared by immediately adding Au3+ into the system which formed Ag nanowires. Compared to the methods mentioned above, a few unique features could be found in our developed approach, such as rapid (within a few minutes), controllable in shape and size, reproducible, and no need of any surface modifiers. Moreover, the bimetallic material show high catalytic performance for reduction of 4-nitrophenol, which is similar with that of pure Au nanoparticles, but the cost is largely reduced. To characterize the structure, composition and optical property, various advanced techniques will be used, such as TEM, HRTEM, STEM, EDS elemental mapping and UV-Vis spectroscopy. The possible formation and growth mechanisms of such Ag-Au bimetallic nanostructures will be discussed. Finally, the catalytic property of the Ag-Au bimetallic nanostructures will be evaluated by reduction of 4-NP as a case study.

6.3 Experimental work

6.3.1 Materials

Silver nitrate (AgNO3, Analytical Reagent) and chloroautic acid (HAuCl4·7H2O) were purchased from Ajax Finechem Pty. Ltd., respectively. Vanadium oxide (V2O3) particles were prepared by the procedures as described in our previous section (Section 3.3.2). Pure 134

water was purchased from Rowater, Australia. All the chemicals were directly used without any further treatment.

6.3.2 Preparation of Ag-Au nanowires

0.5 ml of 0.01 M AgNO3 solution was rapidly dropped into 10 ml transparent suspension containing 0.2 mmol of the as-prepared V2O3 particles. The suspension became slightly brown colour though still transparent. Within 1 min, some black cotton-like precipitates were quickly formed and deposited to the bottom. Based on the previous section (Section 3.3.2),415 silver nanowires were formed as shown in Error! Reference source not found. a nd b. These precipitates could be redispersed into water by sonication and the suspension became transparent in light grey. To prepare Ag-Au nanostructure with different ratios of

Au to Ag, different amounts of HAuCl4·7H2O aqueous solution with certain concentration of HAuCl4·7H2O was gradually added into the transparent grey silver suspension with rate of 0.6 ml per min, then stirring for 1 min. The color of the suspension was darkened with the increase of the ratio of Au to Ag. Before separation, the product was sonicated for 30 min to make the Ag-Au nanowires out of the surface of V2O3 particles. Then, the product was centrifuged at a low speed of 1000 rpm for 3 min. The larger particles (V2O3 particles: 200 nm × 1 μm) deposited faster at the bottom, while the Ag-Au nanowires would suspend in the liquid phase. After taking the liquid phase out, the as-prepared Ag-Au nanostructure was separated by centrifugation at a high speed of 4000 rpm for 15 min, and then rinsed using pure water for several times for further characterizations. This method can remove most of V2O3 particles. Several chemical methods were also employed to remove the V2O3 particles, but not ideal.

6.3.3 Preparation of Au nanoparticles

0.5 ml 0.01 M HAuCl4·7H2O solution was rapidly dropped into 10 ml transparent suspension containing 0.2 mmol of the as-prepared vanadium oxide (V2O3). Within 1 min, the transparent suspension then became purplish gray and cloudy. And then, the product was separated as the procedure in Section 6.3.2 .

6.3.4 Preparation of Ag-Pd bimetallic nanoparticles

135

The procedure and amount of (NH4)2PdCl6 for preparation of Pd nanoparticles were same with those of preparation of Au nanoparticles. The procedure and amount of (NH4)2PdCl6 for preparation of Ag-Pd bimetallic nanowires were same with those of preparation of Au- Ag bimetallic nanowires.

6.3.5 Catalytic reduction of 4-NP

Due to easily monitoring on the colour change by UV-vis absorption spectroscopy, the reaction of 4-NP to 4-AP by NaBH4 was employed as a probe reaction. In a typical process, 9 ml of distilled water was mixed with 0.5 ml of 4.0 mM 4-NP in a 10 ml glass vial. When

0.5 ml of 0.2 M NaBH4 solution was added, the colour of the solution changed from light yellow to dark yellow immediately, and then the mixture was stirred for 10 min at room temperature. The conversion from 4-NP to 4-nitrophenolate anion happened at this stage. After that, 0.5 ml of 0.5 g/L as-prepared nanoparticles was added to the system. Finally, ~ 2.5 ml of the nitrophenolate anion solution was quickly poured into a quartz cuvette for UV-Vis absorption spectra measurement at a time increment of 1 min.

6.3.6 Characterizations

The morphology size and structure of the samples were investigated by JEOL 1400 microscope (TEM) and Philips CM200 FEG microscope (HR-TEM), being operated at an accelerated voltage of 100 kV and 200 kV, respectively. EDS elemental mapping and line scanning were also obtained by Philips CM200 FEG microscope. XRD pattern for identification of the sample composition was recorded on a Philip MPD diffractometer with Cu-Kα radiation. UV-Vis absorption spectra were obtained for tracking the formation and growth of silver nanowires on a Varian Cary 5 UV-VIS-NIR spectrophotometer.

6.3.7 Molecular dynamics simulation

The simulation conditions were the similar to the reference.416

The interaction energy between Au nanoparticles and Ag and V2O3 surface was calculated by the equation

136

eractionint total  surface  EEEE nanoparticle  Eq. 6-1

The energy of a system could be expressed as a sum of the bonded and nonbanded interactions:

total bonded  EEE nonbonded Eq. 6-2

The energy of bonded interactions included bond stretching (bond), valence angle bending (angle), dihedral angle torsion (torsion), and out-of-plane interactions (oop):

bonded bond angle torsion  EEEEE oop Eq. 6-3

The energy of interactions between nonbonded atoms includes van der Waals (vdW), electrostatic (Coulomb), and hydrogen bond (hbond):

nonbond vdW coulomb  EEEE hbond Eq. 6-4

Diffraction analysis was used to simulate the X-ray pattern of the molecular system. The atom pair separations were calculated and averaged over all atoms and frames. In radiation scattering experiments, the objective was to obtain characteristics of the intermolecular structure. In such cases, the main interest was the variation of scattered intensity with changing angle, whereas the absolute intensity was of no concern. Thus, ignoring all intensity scale factors and correction factors, the amplitude and intensity of radiation scattered coherently from an arbitrary set of n atoms may be written as

)(   fQF i Qrj )exp( Eq. 6-5 j1 and

 )(*)()(    kj iQrffQFQFQI jk )exp( Eq. 6-6 kj 1,

Where Q was the scattering vector, fj and fk are the atomic scattering factors for the radiation used, and rjk denotes the vector connecting atoms j and k. This equation was applicable, in principle, to scatter from a gas, liquid, amorphous solid, or crystalline solid.

137

The normalized point intensity was then plotted against 2θ to give the predicted diffractogram.

6.4 Results discussion

6.4. 1 Synthesis of bimetallic nanostructures

6.4.1.1 Au nanoparticles

For comparison with our product, the pure Ag nanowires prepared by the method was shown in Figure 6-1 a and b. Au3+ could also be rapidly reduced under the similar condition although the shape was irregular (Error! Reference source not found. c and d). It can be een that Ag nanowires, Au nanoparticles and V2O3 particles can be easily identified by their particle sizes.

138

Figure 6-1 (a, b) TEM images of the Ag nanowires prepared in recent work; (c, d) TEM images of the Au nanoparticles (20-50 nm) attached on the V2O3 particles (200 nm × 1 μm).

6.4.1.2 Ag-Au bimetallic nanostructures

The morphology and size of the Ag-Au bimetallic structure were investigated by TEM and HRTEM. Figure 6-2 a shows an overview image of the Ag-Au bimetallic nanowires with the Au/Ag molar ratio of 1%. The length of the nanowires was up to tens of micrometers, similar to the Ag nanowires in Error! Reference source not found. a. As well-known, the g nanowires are easily destroyed by Au3+ via electrochemical etching. However, in our case, the structure of Ag nanowires remained after introducing Au3+ ions. To observe clearly, Figure 6-2 b shows an enlarged TEM image of a single Ag-Au bimetallic nanowire, in which the Au nanoparticles showed a diameter of ~ 5 nm and attached on the surface of Ag nanowires. The particles with light contrast might be the fragments of V2O3 particles, which decomposed due to the reduction reaction. The corresponding selected area electron diffraction (SAED) pattern was recorded and shown in Figure 6-2 c. Three sets of typical crystalline planes (circled) could be indexed as {220}, {311} and {2 2 2} for Ag nanowires,415 respectively.

139

Figure 6-2 (a) An overview TEM image of the Au-Ag bimetallic nanowires; (b) An enlarged TEM image of a single Ag-Au bimetallic nanowire; and (c) The select electron diffraction scattering pattern of the area framed in the nanowire in (b). The composition of the bimetallic nanostructure was characterized by EDS elemental mapping. This is because the similar lattice structure of Ag (a = 0.409 nm) and Au (a = 0.408nm), and the XRD technique is difficult to identify Ag from Au. Figure 6-3 a shows the STEM image of Ag-Au bimetallic nanostructure with the Au/Ag molar ratio of 1%. Two elements of Ag (red color) and Au (green color) were scanned as shown in the EDS mapping profile of Figure 6-3 b. Figure 6-3 c and d show independent elemental Au and Ag distribution, respectively. That is, the Ag-Au bimetallic structure is formed under the considered conditions.

140

6.4.1.3 Ag-Pd bimetallic nanostructures

Figure 6-4 (a, b) TEM images of Ag-Pd bimetallic nanowires; (c) Pd nanoparticles (20 nm) attached on V2O3 particles (1μm × 200 nm). 141

Figure 6-4 a shows an overview image of the Ag-Pd bimetallic nanowires with the Pd/Ag molar ratio of 1%. The dimension of the bimetallic nanowires was similar to the nanowires in Figure 6-2 a. Figure 6-2 b shows an enlarged TEM image of a single Ag-Pd bimetallic nanowire, in which the Pd nanoparticles with a diameter of ~ 5 nm attached on the surface of Ag nanowires. The uniform dark contrast suggested that the attachment of Pd nanoparticles was more evenly than that of Au nanoparticles. Pd nanoparticles attached on the surface of V2O3 particles is shown in Figure 6-4 c, in which the Pd spherical nanoparticles with 5 nm in diameter can be clear observed.

Figure 6-5 a shows the STEM image of a single Ag-Pd bimetallic nanowire with the Pd/Ag molar ratio of 1%. Similarly, two elements of Ag (red color) and Pd (green color) were scanned as shown in the EDS mapping profile of Figure 6-5 b. Figure 6-5 c and d show independent elemental Pd and Ag distribution, respectively. That is, the Pd-Au bimetallic structure is formed under the considered conditions.

142

6.4. 2 Effect of molar ratio of Au to Ag

The effect on the formation of the bimetallic structure was investigated with the different Au/Ag molar ratios varying from 0.1%, 1%, 5% to 10%. The corresponding morphology and size of such bimetallic materials were checked by TEM. It was found that the structure could be kept while the Au/Ag molar ratio is 0.1% (Figure 6-6 a). The Au nanoparticles attached on the surface of Ag nanowires which remain the structure. With increasing the molar ratio to 1% (Figure 6-6 b), 5% (Figure 6-6 c), the size of Au particles nearly remained as 5-15 nm in diameter, whereas the Ag nanowires was gradually chemically etched, particularly for 5% (Figure 6-6 c). When the ratio increased to 10%, the Ag nanowires were almost destroyed due to the galvanic reaction between Ag and Au3+. The size of Au nanoparticles slightly increased to 5-20 nm and the aggregation occurred (Figure 6-6 d).

Figure 6-6 TEM images of the Ag-Au bimetallic nanostructures with different ratios of Au to Ag: (a) 0.1%, (b) 1%, (c) 5%, and (d) 10%.

143

The XRD technique was employed to confirm the composition of the product. Figure 6-7 shows the XRD patterns a-d of the bimetallic nanostructures. The different Au/Ag molar ratios from 0.1% to 10% could result in changes of composition reflected by peak intensity and position. When the molar ratio is lower than 1%, the product only contains metallic Au and Ag (Patterns a and b). When the molar ratio is higher than 5%, some new peaks (marked as triangles) emerged in the XRD pattern (Figure 6-7 d) could be attributed to AgCl. However, it seems difficult to quantify the composition of AgCl from the XRD pattern. To confirm the ratios of Ag, Au and Cl quantitatively, the EDS technique was used and the results are shown in Figure 6-8 and Table 6-1. The AgCl might be generated from the three ways: 1) the reaction between Cl− ions the excessive Ag+ in the synthesis Ag − nanowires; 2) galvanic reaction between Ag nanowires and AuCl4 ions; and 3) the active surface silver atoms of Ag nanowires reacting with Cl− ions.417 Despite of the existence of AgCl at high Au/Ag ratios, it may not deposit on the surface of Ag-Au nanowires and thus lead to less impact on the catalytic performance. The details need further investigation in the future. Similarly, vanadium species was not detected by XRD due to the small amount of the fragments of V2O3 particles, implying that the catalytic performance of the Ag-Au nanostructures might not be affected.

Figure 6-7 XRD patterns of the Ag-Au bimetallic nanostructures with different molar ratios of Au to Ag: (a) 0.1%, (b) 1%, (c) 5%, and (d) 10%. ○ represents the peaks of Ag/Au; and ▲ corresponds to AgCl. 144

Table 6-1 Atomic ratios of each product shown in Figure 6-8.

Figure 6-8 EDS spectra of the Ag-Au bimetals: (a) 0.1% Au/Ag product, (b) 1% Au/Ag product, (c) 5% Au/Ag product, (d) 10% Au/Ag product. The main peak of Cl overlaps with one of Ag peaks.

145

6.4. 3 Formation of mechanism

- 6.4.3.1 Rapid reduction of AuCl4 by V2O3 particles

V2O3 particles play a key role in the formation of the Ag-Au bimetallic nanowires. This can be evidenced from the shape and size change without addition of V2O3 particles as shown in Figure 6-9. The Ag nanowires were completely destroyed by adding 1% (molar ratio of − AuCl4 to Ag) HAuCl4 into the suspension of Ag nanowires. In the absence of the reducing − agent of V2O3, Ag nanowires were easily etched by a small amount of AuCl4 because of the galvanic process. In contrast, the Ag wires could be remained in the whole structure if

V2O3 was added in this reaction system.

Figure 6-9 TEM image of the particles prepared by adding 1% (Au/Ag molar ratio) HAuCl4 solution into the suspension of Ag nanowires without V2O3 particles. The wire-like structure was totally destroyed. In this study, Au nanoparticles could be formed by two possible ways. One is the galvanic − reaction between Ag and AuCl4 , as described in Eq. 6-7. The products may include Au and AgCl particles, confirmed by XRD (Pattern d in Figure 6-7). Some studies reported that such a galvanic reaction can be used to etch Ag nanocubes to form cages.412 For example, 407 − Xia et al. demonstrated that Au nanocages and nanoboxes could be prepared via AuCl4 and Ag nanocubes which served as a sacrificial template. The other one may be the 146

− reduction between AuCl4 and V2O3, similar to the reaction happened between Ag and 415 V2O3, as described in Eq. 6-8 and Eq. 6-9. The product of vanadium species in Eq. 6-9 was estimated according to the solution pH and the concentration of vanadium species.6,415

  3Ag  4 (AuCl)s( aq) Au )s(AgCl3)s(  Cl (aq) Eq. 6-7

y  yH2O(aq)  xV 32  OV)s(O 32 x OH y )s(  yH (aq) Eq. 6-8

 y 4xAuCl4  2  yx32 (s)(OH))O3(VO(aq)3y)H(15x(aq)  Eq. 6-9 0     6x[VO(s)4xAu (H23   (aq)16xCl(aq)y)H(18x(aq)O)]

The above three reactions (Eq. 6-7 to Eq. 6-9) are thermodynamics controlled according to the Nernst equation (Eq. 6-10 and Eq. 6-11). The final products in the reaction were mainly dependent upon the electromotive force (EMF) of Eq. 6-7 and Eq. 6-9, closely related to the concentration of reactant(s) and temperature. At low Au/Ag ratios (< 5%), the EMF of Eq.

6-9 (E3) would be larger than that of Eq. 6-7 (E1), and result in the formation of Au nanoparticles that could attach on the Ag nanowires as a nucleation center. With the Au/Ag molar ratio increase to ~ 5%, the EMF of Eq. 6-7 (E1) may increase fast and be close to that − of Eq. 6-9 (E3). Under such conditions, Ag and V2O3 could co-reduce AuCl4 , and the formation of Ag-Au bimetal may be consistent with the study reported by Shahjamali,397 namely, i) initial deposition of Au atoms on Ag nanowire surface; ii) galvanic etching of nanowires; iii) backfilling of the etched pinholes with Ag-Au bimetal; and iv) further deposition of Au atoms/clusters on Ag nanowires. With further increase the molar ratio of − AuCl4 to Ag (e.g. 10%), Eq. 6-7 would become the domain reaction and the products could be Au and AgCl particles. Unfortunately, the Ag nanowires were totally destroyed by formation Ag-Au bimetal (Figure 6-6 d). Here we need to point that the EMF situation is difficult to be quantified at this moment due to the complicated reactions.

 a    RT  Cl  EE 11  ln Eq. 6-10 F3  a    AuCl4 

147

 x6 18  y3x 16x  H aaa Cl  RT  VO 23 OH   EE 33  ln x4 Eq. 6-11 4xF  a    AuCl4 

  Where, E1 and E3 are the standard EMFs of Eq. 6-7 and Eq. 6-9, respectively. a is the chemical activity for the relevant species in solution. x and y are the constants in Eq. 6-9. R is the universal gas constant. F is the Faraday constant.

6.4.3.2 Lattice match

Lattice match was considered as the second reason for the formation of Ag-Au bimetallic nanostructures because of the similar lattice parameters of Au (a = 0.408 nm) and Ag (a = 0.409 nm).392 The lattice match induced fusion between Au and Ag is easily achieved. To understand the crystallization and the lattice matching between Au and Ag, HRTEM was employed to observe the joint part between the two metals (Figure 6-10 ). Figure 6-10 a shows the Ag nanowires grew along [111] direction, while Figure 6-10 b displays an enlarged image of the joint part. It can be seen that the d-space of Au (002) plane was 2.00 Å, while the d-space of Ag (002) was 2.03 Å. The lattice mach were achieved by (002) plane of Ag and Au. Based on the previous studies, Ag and Au can be distinguished by the HRTEM technique according to their different contrast in the HRTEM image. This is because Ag and Au have different atomic weight: Ag 108 and Au 197. For example, Mukherjee et al. demonstrated that synthesis and characterization of bimetallic Au-Ag core- shell nanostructures, and they considered a darker area corresponds to the Au and the light area could be attributed to the Ag in one HRTEM image,418 conducted on JEOL, 2010EX at an acceleration voltage of 200 kV. In addition, they assigned the Miller planes of the Au- Ag core-shell structure to Au(111) 2.35 Å and Ag(101) 2.42 Å, respectively. Similarly, Li et al. demonstrated the preparation of Au-Ag alloy nanoparticles, in which the (111) lattice fringes with an accuracy of 0.001 Å have been confirmed by HRTEM (TecnaiG F20) at an operation voltage of 200 kV. They provided lattice fringes of such Au-Ag alloys as below: 419 d(111)Au1Ag2 = 2.356 Å, d(111)Au2Ag1 = 2.372 Å, and d(111)Au1Ag1 = 2.369 Å, respectively. Therefore, based on the previous studies on the Au-Ag nanostructure, it is possible to distinguish Au and Ag, and obtain a 0.03 Å-difference in the lattice through the HRTEM

148

technique (Note that in our experiments, HRTEM (Philips, CM200) was also operated at an acceleration voltage of 200 kV).

In addition, the Au nanoparticles formed on V2O3 surface trend to absorb on the surface of Ag nanowires due to the strong metal-metal band and lattice match, which can be confirmed by molecular dynamics (MD) simulation. The much higher interaction energy of

Ag-Au (-1032.41 kcal/mol) than that of Au-V2O3 (-334.04 kcal/mol) reveals that Au-Ag system is much more stable. In another words, when the Au nanoparticles formed on the surface of V2O3, they are likely to move to the Ag surface due to such strong driven forces due to the metal-metal bond and lattice match between Au and Ag. Thus, Au nanoparticles more easily absorb on the surface of Ag nanowires. Figure 6-11 shows the snapshots of the

Au nanoparticles deposition simulation onto V2O3 (104) surface and Ag (111) surface at the final stage (500 ps). The interaction area of Au and Ag is obvious larger than that of V2O3 and Au, implying that the Ag-Au system is more stable than V2O3-Au system. Moreover, the lattice of Au nanoparticles changed after deposition on Ag surface, in good agreement with the case reported by Yue et al.416

Figure 6-10 HRTEM images of the joint/fused part of Ag-Au nanostructures.

149

(111) peak in Ag-Au system is much higher than that in Au-V2O3 system, which means the

Ag-Au system is more crystalline than Au-V2O3 system, further confirmed the stability of the Ag-Au system.

150

Figure 6-12 Simulated XRD patterns showing the structural crystallization of Au nanoparticles deposited on the Ag (111) surface (black) and the V2O3 (104) surface (red). 6.4.3.3 Surface defects

The third reason for the formation of bimetallic structure may be attributed to large amounts of surface defects due to the rough surface of such Ag nanowires, generated by the oriented attachment of Ag nanoparticles as reported in the previous section (Section 5.4. 3).415 The rough surface provides many surface defects that can facilitate the deposition of Au nanoparticles. Yue et al.416 reported a theoretical study using molecular dynamics (MD) on the deposition mechanism of precious metal nanoparticles (e.g. Au, Pt, Ag, and Pd) onto iron oxides. The MD simulation results show that the total potential energy of the metal nanoparticles decreases significantly when the metal particles deposited onto porous or defected sites. The interaction is heavily dependent on nanoparticles size, elemental species, and surface structure of the substrates. For Au cluster(s), the size close to the pore or defect size will lead to strong interactions. In this study, the rough surface of Ag nanowires with many defects would benefit for Au nanoparticles nucleation, growth, deposition and fusion as Ag-Au bimetallic structure.

151

6.4. 4 Properties of the Ag-Au bimetallic nanowires

6.4.4.1 Optical property

Figure 6-13 shows the UV-Vis spectra of the Ag-Au bimetallic nanostructures showing 300-700 nm with magnification in Intensity. With small ratio ≤ 1%, the surface plasmon resonance (SPR) peaks in Curves a and b are mainly dominated by that of pure Ag nanowires, as described in the previous section (Section 5.4. 5).415 When the ratio was increased to 5-10%, the SPR band red shifted to around 560 nm, probably attributed to the formation of Au-Ag nanoparticles.412,420 The formation of AgCl particles plays a slight impact on the optical properties.421 The formation of the bimetallic structure can also be evidenced by the fact that the optical absorption spectra show only one composition- sensitive plasmon band but not separate peaks for Ag and Au nanoparticles.422 In addition, the SPR band of the Ag-Au bimetallic nanowires became increasing broader and weaker with a long tail band, similar to those observed in the Ag@Au nanocrystals.397,423

Figure 6-13 UV-Vis spectra of the Ag-Au bimetallic nanostructures with different molar ratios of Au to Ag: (a) 0.1%, (b) 1%, (c) 5%, (d) 10%, and (e) pure Au NPs.

152

6.4.4.2 Catalytic property

Figure 6-14 Time-dependent UV-Vis absorption spectra of the catalytic reduction of 4-NP to 4-AP with Ag nanowires (a), Au nanoparticles (b), Ag-Au bimetallic nanowires with different molar ratios of Ag to Au: (c) 0.1%, (d) 1%, (e) 5%, and (f) 10%. To evaluate the catalytic activity of the as-obtained bimetallic structures, the reduction of 4- 407,412 NP to 4-AP by NaBH4, as a case study, was employed as a probe reaction. As well known, 4-AP is very useful in a great amount of applications including analgesic and antipyretic drugs, photographic developer, corrosion inhibitor, and anticorrosion- lubricant.424-426 The reaction could be easily monitored by UV-Vis spectroscopy because the reactant (4-nitrophenolate anion) and the product can show different absorption position for the maximum absorption peak, i.e. λmax = 400 nm (reactant) and 300 nm for product. In comparison, individual Ag nanowires or Au nanoparticles prepared by the same method were tested as a blank. 153

In the absence of the catalyst (Ag/Au bimetallic structure), the absorption peak centered at 400 nm (4-nitrophenolate anion) remained even over several days, suggesting that the reduction did not occur. When a small amount of bimetallic structures was added, the absorption peak (λmax = 400 nm) gradually decreased at room temperature. Meanwhile, a new absorption peak centered at ~ 300 nm corresponding to the product (4-AP) appeared.

After a while, the reactant peak (λmax = 400 nm) disappeared, and the color of the solution faded from dark yellow (the color of 4-nitrophenolate anion) to transparent. In the whole − process, the bimetallic structures acted as a catalyst to transfer electrons from BH4 to 4-NP enabling its reduction. The time for the reaction completion varied with the molar ratio of Au to Ag in the bimetallic structure. Generally, the reaction time shortened with an increasing amount of Au, and the details were listed in Table 6-2. Figure 6-14 shows the time-dependent UV-Vis absorption spectra of the reduction process for different bimetallic structures. In comparison, the Ag nanowires and the Au nanoparticles prepared by HAuCl4 and V2O3 were also tested, respectively. The reactions were completed within 30 min and 18 min, respectively, as shown in Figure 6-14 a and b. It was noted that the Ag nanowires for catalytic reduction is better than those previous reported,401 due to the rough surface with more defects. Moreover, with the molar ratio of Au to Ag increased, the reduction time decreased from 8 min (1%), 5 min (5%) to 4 min (10%). Herein the sample with ~ 5% Au bimetallic nanostructures shows similar performance to that of the pure Au nanoparticles in the literature.407,412,420 However, the cost of such a catalyst could be significantly reduced.

Figure 6-15 shows the relationship between ln(At/A0) and reaction time t, where A is the absorbance corresponding to the concentrations of reactants and products. In this study, the − − concentration of BH4 was so high compared with that of 4-NP (molar ratio of BH4 to 4- NP is ~ 50). The first-order approximation may be suitable for fitting the kinetics 427,428 calculation. The apparent rate constants (kapp) for the reduction by different catalysts were calculated and listed in Table 6-2, suggesting that the kapp increased with the Au/Ag molar ratios. The kapp of 1% is similar to the Au nanoparticles prepared by our method. However, it is still lower than the pure Au nanoparticles as reported in literature,412,424 whereas the kapp of 5% is close to the activities of the reported Au nanoparticles, and much

154

higher than other substrate-supported Au nanoparticles.409-411 Although the irregular or broken nanoparticles (10% Au-Ag sample) show a slight higher catalytic activity than the Ag-Au wire-like ones (5% Au-Ag sample), we still prefer to consider the 5% Au/Ag sample as a good one in this study, because of two main reasons: 1) the cost is lower than 10% Au/Ag if for large scale production, and 2) ≥ 10% Au may cause the damage of wire- like structures to irregular particles which have been widely studied. Moreover, the wire- like structures also show a few unique advantages: easy separation, no aggregation for Au nanoparticles and low cost compared to the pure Au nanoparticles. In comparison, the catalytic reduction of 4-NP for Ag-Pd bimetallic nanowires was also conducted. The time- dependent UV-Vis spectra of such structures and the relationship between ln(At/A0) and time t was shown in Figure 6-16 and Figure 6-17, and the total reduction time and reaction rate constant kapp were listed in Table 6-3. Note that the concentration (C) is proportional to the absorbance (A) according to Beer-Lambert law. The kapp of the Ag-Pd bimetallic structures was smaller than that of the Ag-Au ones under the reported conditions. In our case, doping Pd does not enhance the catalytic activity. The reason is unknown and requires further investigation. However, this phenomenon does not only happen in our case. For example, Oh et al. demonstrated that the catalytic performance for reduction of 4-NP of Pd monometal nanoparticles is higher than that of its bimetal Pd-M (M including Ag).429 This may be attributed to nonuniform particle size or its intrinsic characteristics.

Table 6-2 The total reduction time and kapp of the Ag nanowires (Ag NWs), the Au nanoparticles (Au NPs), and the Ag-Au nanowires with different molar ratios.

0.1% 1% 5% 10% Products Ag NWs Au NPs* Au/Ag Au/Ag Au/Ag Au/Ag Reduction time 30 18 12 8 5 4 (min)

-3 -1 kapp(10 s ) 0.47 1.3 0.87 1.4 3.4 3.8

* The Au NPS shown in Table 6-2 is the product synthesized by our designed method in Section 6.3.3.

155

Figure 6-15 The relationship of ln(At/A0) with time in the catalytic reduction of 4-NP to 4-AP, corresponding to the catalysts listed in Figure 6-14.

156

The high catalytic performance may be attributed to the large surface area of Ag-Au nanoparticles and the rapid electron transfer of Ag nanowires.413,414 It was well-accepted that the apparent kinetic rate constant kapp was proportional to the total surface (S) of metal 420 nanoparticles. Hence, the apparent kinetic rate constants kapp and k1 can be defined as:

−dCt/dt = kappCt = k1SCt, where Ct is the concentration of 4-NP at time t, and k1 is the rate constant normalized to S which could be normalized to the unit volume of the system. In this study, Au nanoparticles are ~ 5 nm in diameter with high kinetic rate compared to those with larger size (> 10 nm).412 Another reason is probably caused the rapid electron transfer on Ag nanowires. As well known, metal Ag owns the fastest electron conductivity compared to others. This may benefit for the catalytic reduction reaction of 4-NP using − BH4 to form 4-AP at the surface of Au nanoparticles and/or interface of Ag-Au bimetal. The slight increase of the catalytic activities from 3.4 to 3.8 is probably caused by the morphology change from wire to sphere, due to the enlarged surface area. The main contribution of catalysis derived from Au NPs, however, this does not mean that the Ag NPs have no catalytic property (Table 6-2). The broken Ag nanowires lead to increase the surface area of Ag NPs, slightly increase the catalytic activity of the product. Even though the Au catalyst was replaced 90% with Ag, the catalytic performance could be kept, even better than that of pure Au nanoparticles on other substrates (e.g. silica nanotubes).420 Interestingly, the cost would be significantly reduced while used Ag-Au as the proposed catalysts in practice.

Table 6-3 The total reduction time and kapp of the Pd nanoparticles, and the Ag-Pd bimetallic nanowires with different molar ratios.

0.1% 1% 5% 10% Products Pd NPs Pd/Ag Pd/Ag Pd/Ag Pd/Ag Reduction time 28 40 50 80 100 (min) -3 -1 Kapp(10 s ) 0.41 0.30 0.23 0.17 0.053

157

Figure 6-17 The plots of ln(At/A0) versus time for the catalytic reduction of 4-NP corresponding to the nanoparticles in Figure 6-16. 6.5 Summary

We have demonstrated a facile and effective wet-chemical method for synthesis of Ag-Au nanostructures under ambient conditions. The morphology and size of the bimetallic nanostructures could be controlled by adjusting the amounts of gold ions (molar ratio of Au to Ag ≤ 5%) in the suspension containing silver nanowires and vanadium oxide nanoparticles. In the formation process, the galvanic reaction of Au with Ag and the rapid − reduction of AuCl4 ions by V2O3 played a key role, particularly for the generation of the Ag-Au wire-like nanostructure. The successful synthesis of such Ag-Au bimetallic structures is probably contributed by three aspects: i) the rapid reduction of Au by V2O3; ii) lattice match between Ag and Au; and iii) more defects on Ag nanowires surface. The catalytic performance of these Au-Ag bimetallic structures was evaluated by catalytically reducing 4-nitrophenol to 4-aminophenol under the assistance of NaBH4 at room temperature. The findings show that Ag-Au nanostructures (e.g., Au/Ag = 5%) exhibit a good catalytic performance similar to the pure Au nanoparticles reported in the literature, however, the cost would be greatly reduced. Moreover, such Ag-Au nanowires (up to tens of micrometer) could be readily recovered. This study would open a new strategy for preparation of other bimetallic catalysts for diverse uses. 158

Chapter 7 Conclusions

Various shapes (microurchins, microspheres, and nanorods) of V2O5 and V2O3 nanoparticles have been prepared by a polythermal method followed by calcination at different atmospheres. The shape control can be achieved by simply adjusting reaction time. The formation mechanism of VEG has been investigated by MS, showing that the process undergoes esterification, redox reaction, and intramolecular condensation. In addition, the growth of the VEG nanorods follows such steps: the formation of nanospheres, assemble into microspheres, the formation of microurchins by splitting of microspheres, and finally decomposition to nanorods. EG in this system plays an important role, such as solvent, reducing agent, and coordinator.

The as-prepared V2O5 particles can be used as a good sensing material for detection of acetone. For the target gases (acetone, isopropanol, and ammonia), V2O5 mcirourchins show higher sensitivities than V2O5 nanorods. Among these gases, the microurchins are highly sensitive to acetone. In addition, the sensitivity of the microurchins towards ammonia is higher than that of other V2O5 sensing materials. This may be because the less packing density of urchin-like structures than that of rod-like structures.

Silver vanadate (Ag2V4O11 and Ag0.35V2O5) nanobelts have been successfully synthesized by a hydrothermal method at 180 °C for 2 days. Ag2V4O11 can be prepared with V2O5 and

AgNO3 without any additives, while the preparation of Ag0.35V2O5 is essential to be assisted by SDS as a reducing agent except for V2O5 and Ag ions. For Ag2V4O11, the composition and crystallinity are significantly affected by solution pH. It was found that at optimum working temperature of 260 °C, the Ag2V4O11 nanobelts (obtained at pH 1) show high sensitivity to amines and high selectivity to amines versus ammonia, as well as low detection limit (5 ppm).

For V2O3 particles, this thesis, for the first time, applies them to induce synthesis of Ag nanowires and Ag-Au bimetallic nanowires. This novel method shows several unique features, such as simple operation, rapid and room-temperature reaction, and no need for

159

any additives. The Ag nanowires are formed by oriented attachment. Especially for synthesis of the Ag-Au bimetallic nanowires, the use of V2O3 particles can avoid the galvanic reaction between Au ions and Ag metal so as to reduce Au ions by V2O3 particles in the present of Ag metal and make the newly formed Au nanoparticles attached on the surface of Ag nanowires. V2O3 particles play key roles in the formation of both Ag nanowires and Ag-Au nanowires. Experimental results show that the Ag-Au bimetallic nanowires exhibit high catalytic performance for reduction of 4-nitrophenol. The performance of the bimetallic nanowires with Au/Ag of 5% can be comparable with pure Au nanoparticles, however, the cost is largely reduced. This may be potential for industrial applications in the future.

160

References

(1) Shipway, A. N.; Katz, E.; Willner, I. Chemphyschem 2000, 1, 18. (2) Gopinath, R.; Patel, B. K. Org Lett 2000, 2, 577. (3) Raible, I.; Burghard, M.; Schlecht, U.; Yasuda, A.; Vossmeyer, T. Sensor Actuat B-Chem 2005, 106, 730. (4) Bachmann, H. G.; Ahmed, F. R.; Barnes, W. H. Zeitschrift für Kristallographie - Crystalline Materials 1961, 115, 110. (5) Eranna, G.; Joshi, B. C.; Runthala, D. P.; Gupta, R. P. Critical Reviews in Solid State and Materials Sciences 2004, 29, 111. (6) Livage, J. Chem Mater 1991, 3, 578. (7) Gu, G.; Schmid, M.; Chiu, P. W.; Minett, A.; Fraysse, J.; Kim, G. T.; Roth, S.; Kozlov, M.; Munoz, E.; Baughman, R. H. Nat Mater 2003, 2, 316. (8) Pinna, N.; Wild, U.; Urban, J.; Schlögl, R. Advanced Materials 2003, 15, 329. (9) F. Krumeich, H.-J. M., M. Niederberger, F. Bieri, B. Schnyder, and R. Nesper J. Am. Chem. Soc. 1999, 1999, 8324. (10) Niederberger, M.; Muhr, H. J.; Krumeich, F.; Bieri, F.; Gunther, D.; Nesper, R. Chem Mater 2000, 12, 1995. (11) Liu, J. F.; Wang, X.; Peng, Q.; Li, Y. D. Adv Mater 2005, 17, 764. (12) Hervieu, M. Adv. Mater. 1995, 7, 91. (13) Hermann, K.; Witko, M. In The Chemical Physics of Solid Surfaces; Woodruff, D. P., Ed.; Elsevier: 2001; Vol. Volume 9, p 136. (14) Surnev, S.; Ramsey, M. G.; Netzer, F. P. Progress in Surface Science 2003, 73, 117. (15) Kung, H. H. Transition Metal Oxides, Surface Chemistry and Catalysis; Elsevier, 1989. (16) Weckhuysen, B. M.; Keller, D. E. Catalysis Today 2003, 78, 25. (17) Stefanovich, G.; Pergament, A.; Stefanovich, D. J. Phys.: Condens. Matter 2000, 12, 8837. (18) Babulanam, S. M.; Eriksson, T. S.; Niklasson, G. A.; Granqvist, C. G. Solar Energy Materials 1987, 16, 347. (19) Julien, C.; Haro-Poniatowski, E.; Camacho-Lopez, M. A.; Escobar-Alarcon, L.; Jimenez- Jarquin, J. Mat Sci Eng B-Solid 1999, 65, 170. (20) Bauerle, G. L.; Wu, S. C.; Nobe, K. Ind Eng Chem Prod Rd 1978, 17, 117. (21) Li, X. L.; Chen, X. J.; Chen, X. Y.; Han, C. L.; Shi, C. W. J Cryst Growth 2007, 309, 43. (22) Wu, X. C.; Tao, Y. R.; Lin, D.; Wang, Z. H.; Zheng, H. Mater Res Bull 2005, 40, 315. (23) Whittaker, L.; Zhang, H. S.; Banerjee, S. J Mater Chem 2009, 19, 2968. (24) Park, J.; Oh, I. H.; Lee, E.; Lee, K. W.; Lee, C. E.; Song, K.; Kim, Y.-J. APPLIED PHYSICS LETTERS 2007, 91, 153112. (25) Ruzmetov, D.; Zawilski, K. T.; Narayanamurti, V.; Ramanathan, S. J Appl Phys 2007, 102, 113715. (26) Morin, F. J. Phys Rev Lett 1959, 3, 34. (27) Stefanovich, G. B.; Pergament, A. L.; Kazakova, E. L. Tech Phys Lett+ 2000, 26, 478. (28) Lopez, R.; Haynes, T. E.; Boatner, L. A.; Feldman, L. C.; Haglund, R. F., Jr. Phys Rev B 2002, 65, 224113. (29) Cavalleri, A.; Dekorsy, T.; Chong, H. H. W.; Kieffer, J. C.; Schoenlein, R. W. Phys Rev B 2004, 70, 161102. (30) Kim, B.-J.; Lee, Y. W.; Choi, S.; Lim, J.-W.; Yun, S. J.; Kim, H.-T.; Shin, T.-J.; Yun, H.-S. Phys Rev B 2008, 77, 235401.

161

(31) Kim, H.-T.; Chae, B.-G.; Youn, D.-H.; Kim, G.; Kang, K.-Y.; Lee, S.-J.; Kim, K.; Lim, Y.- S. APPLIED PHYSICS LETTERS 2005, 86, 242101. (32) Goodenough, J. B. Annual Review of Materials Science 1971, 1, 101. (33) Goodenough, J. B. Journal of Solid State Chemistry 1971, 3, 490. (34) Wentzcovitch, R. M.; Schulz, W. W.; Allen, P. B. Phys Rev Lett 1994, 72, 3389. (35) Toledano, D. S.; Metcalf, P.; Henrich, V. E. Surface Science 2000, 449, 19. (36) Kokabi, H. R.; Studer, F.; Toulemonde, M. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 1996, 111, 75. (37) Joshi, M. G.; Honig, M. J. Rev. Chim. Miner. 1982, 19, 251. (38) Tenailleau, C.; Suard, E.; Rodriguez-Carvajal, J.; Crosnier-Lopez, M. P.; Lacorre, P. Chem. Mater. 2002, 14, 3569. (39) Andrich, E. Electron. Appl. 1996, 26, 123. (40) Perkins, S. R.; Riiegg, A.; Fisher, M. Adv. Ceram. 1983, 7, 166. (41) Pan, Y.; Wu, G.; Yi, X. Journal of Materials Science 1994, 29, 5757. (42) van der Lee, G.; Schuller, B.; Post, H.; Favre, T. L. F.; Ponec, V. Journal of Catalysis 1986, 98, 522. (43) Kittaka, S.; Sasaki, S.; Morimoto, T. Journal of Materials Science 1987, 22, 557. (44) Sullivan, R. J.; Srinivasan, T. T.; Newnham, R. E. Journal of the American Ceramic Society 1990, 73, 3715. (45) Wu, J.; Liu, H.; Zhu, S.; Ma, Z. Chin. Chem. Lett. 1991, 2, 901. (46) Piao, J.; Takahashi, S.; Kohiki, S. Jpn. J. Appl. Phys. 1998, 37, 6519. (47) Corr, S. A.; Grossman, M.; Shi, Y.; Heier, K. R.; Stucky, G. D.; Seshadri, R. Journal of Materials Chemistry 2009, 19, 4362. (48) Liu, J.; Zhou, Y.; Wang, J.; Pan, Y.; Xue, D. Chemical Communications 2011, 47, 10380. (49) Fu, H.; Jiang, X.; Yang, X.; Yu, A.; Su, D.; Wang, G. Journal of Nanoparticle Research 2012, 14, 1. (50) Wang, S.; Lu, Z.; Wang, D.; Li, C.; Chen, C.; Yin, Y. Journal of Materials Chemistry 2011, 21, 6365. (51) Lewis, K. B.; Oyama, S. T.; Somorjai, G. A. Surface Science 1990, 233, 75. (52) Kishi, K.; Hirai, K.; Yamamoto, T. Surface Science 1993, 290, 309. (53) Surnev, S.; Sock, M.; Kresse, G.; Andersen, J. N.; Ramsey, M. G.; Netzer, F. P. The Journal of Physical Chemistry B 2003, 107, 4777. (54) Surnev, S.; Kresse, G.; Sock, M.; Ramsey, M. G.; Netzer, F. P. Surface Science 2001, 495, 91. (55) Schoiswohl, J.; Surnev, S.; Sock, M.; Eck, S.; Ramsey, M. G.; Netzer, F. P.; Kresse, G. Physical Review B 2005, 71, 165437. (56) Schoiswohl, J.; Surnev, S.; Sock, M.; Ramsey, M. G.; Kresse, G.; Netzer, F. P. Angewandte Chemie International Edition 2004, 43, 5546. (57) Schoiswohl, J.; Kresse, G.; Surnev, S.; Sock, M.; Ramsey, M. G.; Netzer, F. P. Physical Review Letters 2004, 92, 206103. (58) Schoiswohl, J.; Sock, M.; Eck, S.; Surnev, S.; Ramsey, M. G.; Netzer, F. P.; Kresse, G. Physical Review B 2004, 69, 155403. (59) Schoiswohl, J.; Surnev, S.; Netzer, F. P.; Kresse, G. Journal of Physics: Condensed Matter 2006, 18, R1. (60) Zhang, Z.; Henrich, V. E. Surface Science 1992, 277, 263. (61) Schuler, H.; Klimm, S.; Weissmann, G.; Renner, C.; Horn, S. Thin Solid Films 1997, 299, 119. (62) Lida, Y.; Kaneko, Y.; Kanno, Y. J Mater Process Tech 2008, 197, 261. 162

(63) Ramana, C. V.; Smith, R. J.; Hussain, O. M.; Massot, M.; Julien, C. M. Surf Interface Anal 2005, 37, 406. (64) McGraw, J. M.; Perkins, J. D.; Hasoon, F.; Parilla, P. A.; Warmsingh, C.; Ginley, D. S.; Mateeva, E.; Readey, D. W. Journal of Materials Research 2000, 15, 2249. (65) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.; Kim, F.; Yan, H. Advanced Materials 2003, 15, 353. (66) Baes, C. F.; Mesmer, R. E. The hydrolysis of cations; Wiley, 1976. (67) Livage, J. Materials 2010, 3, 4175. (68) Henry, M.; Jolivet, J.; Livage, J. In Chemistry, Spectroscopy and Applications of Sol-Gel Glasses; Reisfeld, R., Jjørgensen, C. K., Eds.; Springer Berlin Heidelberg: 1992; Vol. 77, p 153. (69) Pettersson, L.; Hedman, B.; Andersson, I.; Ingri, N. Chemica scripta 1983, 22, 254. (70) Fratzky, D.; Götze, T.; Worzala, H.; Meisel, M. Materials Research Bulletin 1998, 33, 635. (71) Giorgetti, M.; Passerini, S.; Smyrl, W. H.; Berrettoni, M. Inorganic Chemistry 2000, 39, 1514. (72) Petkov, V.; Trikalitis, P. N.; Bozin, E. S.; Billinge, S. J. L.; Vogt, T.; Kanatzidis, M. G. Journal of the American Chemical Society 2002, 124, 10157. (73) Fontenot, C. J.; Wiench, J. W.; Schrader, G. L.; Pruski, M. Journal of the American Chemical Society 2002, 124, 8435. (74) Evans, H. T. Inorganic Chemistry 1966, 5, 967. (75) Heath, E.; Howarth, O. W. Journal of the Chemical Society, Dalton Transactions 1981, 0, 1105. (76) Wéry, A. S. J.; Gutiérrez-Zorrilla, J. M.; Luque, A.; Ugalde, M.; Román, P. Chemistry of Materials 1996, 8, 408. (77) Chirayil, T.; Zavalij, P. Y.; Whittingham, M. S. Chemistry of Materials 1998, 10, 2629. (78) Riou, D.; Roubeau, O.; Férey, G. Zeitschrift für anorganische und allgemeine Chemie 1998, 624, 1021. (79) Livage, J. Coordination Chemistry Reviews 1998, 178–180, Part 2, 999. (80) Zhou, F.; Zhao, X.; Yuan, C.; Li, L. Crystal Growth & Design 2007, 8, 723. (81) Muster, J.; Kim, G. T.; Krstić, V.; Park, J. G.; Park, Y. W.; Roth, S.; Burghard, M. Advanced Materials 2000, 12, 420. (82) Gu, G.; Schmid, M.; Chiu, P.-W.; Minett, A.; Fraysse, J.; Kim, G.-T.; Roth, S.; Kozlov, M.; Munoz, E.; Baughman, R. H. Nat Mater 2003, 2, 316. (83) Chan, C. K.; Peng, H.; Twesten, R. D.; Jarausch, K.; Zhang, X. F.; Cui, Y. Nano Letters 2007, 7, 490. (84) Liu, J.; Wang, X.; Peng, Q.; Li, Y. Advanced Materials 2005, 17, 764. (85) Krumeich, F.; Muhr, H. J.; Niederberger, M.; Bieri, F.; Schnyder, B.; Nesper, R. Journal of the American Chemical Society 1999, 121, 8324. (86) Pan, D.; Shuyuan, Z.; Chen, Y.; Hou, J. G. Journal of Materials Research 2002, 17, 1981. (87) Pavasupree, S.; Suzuki, Y.; Kitiyanan, A.; Pivsa-Art, S.; Yoshikawa, S. Journal of Solid State Chemistry 2005, 178, 2152. (88) Nesper, R.; Muhr, H.-J. CHIMIA International Journal for Chemistry 1998, 52, 571. (89) Sun, D.; Kwon, C. W.; Baure, G.; Richman, E.; MacLean, J.; Dunn, B.; Tolbert, S. H. Advanced Functional Materials 2004, 14, 1197. (90) Nordlinder, S.; Nyholm, L.; Gustafsson, T.; Edström, K. Chemistry of Materials 2005, 18, 495. (91) Patzke, G. R.; Krumeich, F.; Nesper, R. Angewandte Chemie International Edition 2002, 41, 2446.

163

(92) Petkov, V.; Zavalij, P. Y.; Lutta, S.; Whittingham, M. S.; Parvanov, V.; Shastri, S. Physical Review B 2004, 69, 085410. (93) Davidson, P. Comptes Rendus Chimie 2010, 13, 142. (94) Grigorieva, A. V.; Goodilin, E. A.; Anikina, A. V.; Kolesnik, I. V.; Tretyakov, Y. D. Mendeleev Communications 2008, 18, 71. (95) Corr, S. A.; Grossman, M.; Furman, J. D.; Melot, B. C.; Cheetham, A. K.; Heier, K. R.; Seshadri, R. Chemistry of Materials 2008, 20, 6396. (96) Vera-Robles, L. I.; Campero, A. The Journal of Physical Chemistry C 2008, 112, 19930. (97) Pang, S.; Li, G.; Zhang, Z. Macromolecular Rapid Communications 2005, 26, 1262. (98) Cao, A.-M.; Hu, J.-S.; Liang, H.-P.; Wan, L.-J. Angewandte Chemie International Edition 2005, 44, 4391. (99) Chen, Y.; Liu, H.; Ye, W.-L. Scripta Materialia 2008, 59, 372. (100) Lim, H. S.; Kwak, D.; Lee, D. Y.; Lee, S. G.; Cho, K. Journal of the American Chemical Society 2007, 129, 4128. (101) Liu, X.; Huang, C.; Yi, S.; Xie, G.; Li, H.; Luo, Y. Solid State Communications 2007, 144, 259. (102) Wu, C. Z.; Xie, Y.; Lei, L. Y.; Hu, S. Q.; OuYang, C. Z. Advanced Materials 2006, 18, 1727. (103) O'Dwyer, C.; Navas, D.; Lavayen, V.; Benavente, E.; Santa Ana, M. A.; González, G.; Newcomb, S. B.; Sotomayor Torres, C. M. Chemistry of Materials 2006, 18, 3016. (104) Liu, W.; Wang, X. F.; Cao, L. X.; Su, G.; Zhang, L.; Wang, Y. G. Science China-Chemistry 2011, 54, 724. (105) Zhang, J.; Sun, L.-D.; Jiang, X.-C.; Liao, C.-S.; Yan, C.-H. Crystal Growth & Design 2003, 4, 309. (106) Chiu, W. S.; Khiew, P. S.; Isa, D.; Cloke, M.; Radiman, S.; Abd-Shukor, R.; Abdullah, M. H.; Huang, N. M. Chemical Engineering Journal 2008, 142, 337. (107) Geng, F.; Zhao, Z.; Cong, H.; Geng, J.; Cheng, H.-M. Materials Research Bulletin 2006, 41, 2238. (108) Yin, Z.; Sakamoto, Y.; Yu, J.; Sun, S.; Terasaki, O.; Xu, R. Journal of the American Chemical Society 2004, 126, 8882. (109) Desai, S. D.; Cussler, E. L. Langmuir 1997, 13, 1496. (110) Desai, S. D.; Cussler, E. L. Langmuir 1998, 14, 277. (111) Lee, J.-K.; Kim, G.-P.; Song, I. K.; Baeck, S.-H. Electrochemistry Communications 2009, 11, 1571. (112) da Silva, D. L.; Delatorre, R. G.; Pattanaik, G.; Zangari, G.; Figueiredo, W.; Blum, R.-P.; Niehus, H.; Pasa, A. A. Journal of The Electrochemical Society 2008, 155, E14. (113) Hu, C.-C.; Chang, K.-H.; Huang, C.-M.; Li, J.-M. Journal of The Electrochemical Society 2009, 156, D485. (114) Takahashi, K.; Limmer, S. J.; Wang, Y.; Cao, G. The Journal of Physical Chemistry B 2004, 108, 9795. (115) Wang, Y.; Takahashi, K.; Shang, H.; Cao, G. The Journal of Physical Chemistry B 2005, 109, 3085. (116) Bennett, J. A.; Iii, J. E. P.; Neiswonger, M. A. Journal of Electroanalytical Chemistry 2011, 654, 1. (117) Wang, Y.; Takahashi, K.; Shang, H. M.; Cao, G. Z. J Phys Chem B 2005, 109, 3085. (118) Pourbaix, M. Atlas D'equilibres Electrochimiques; Pergamon Press, 1966. (119) Potiron, E.; Le Gal La Salle, A.; Verbaere, A.; Piffard, Y.; Guyomard, D. Electrochimica Acta 1999, 45, 197. 164

(120) Lakshmi, B. B.; Patrissi, C. J.; Martin, C. R. Chemistry of Materials 1997, 9, 2544. (121) Al Zoubi, M.; Farag, H. K.; Endres, F. J Mater Sci 2009, 44, 1363. (122) Mooney, J. B.; Radding, S. B. Annu. Rev. Mater. Sci. 1982, 12, 81. (123) Feng, C. Q.; Wang, S. Y.; Zeng, R.; Guo, Z. P.; Konstantinov, K.; Liu, H. K. J. Power Sources 2008, 184, 485. (124) Ng, S. H.; Patey, T. J.; Buechel, R.; Krumeich, F.; Wang, J. Z.; Liu, H. K.; Pratsinis, S. E.; Novak, P. Phys Chem Chem Phys 2009, 11, 3748. (125) Xuchuan, J.; Qinghua, Z.; Aibing, Y. Nanotechnology 2006, 17, 4929. (126) Jiang; Zeng; Yu Langmuir 2007, 23, 2218. (127) Qinghua, Z.; Xuchuan, J.; Aibing, Y.; Gaoqing, L. Nanotechnology 2007, 18, 035708. (128) Jiang, X.; Yu, A. Langmuir 2008, 24, 4300. (129) Jiang, X.; Yu, A. Journal of Nanoscience and Nanotechnology 2010, 10, 7829. (130) Xuchuan, J. Q., Zeng; Aibing, Yu In New Nanotechnology Developments; Barrañón, A., Ed.; Nova Science Publishers: NY, 2009, p 145. (131) Chen, S.; Carroll, D. L. Nano Lett. 2002, 2, 1003. (132) Zhang, J. H.; Liu, H. Y.; Zhan, P.; Wang, Z. L.; Ming, N. B. Adv. Funct. Mater. 2007, 17, 1558. (133) Jin, R.; Cao, Y.; Mirkin, C. A.; Kelly, K. L.; Schatz, G. C.; Zheng, J. G. Science 2001, 294, 1901. (134) Jin, R.; Charles Cao, Y.; Hao, E.; Metraux, G. S.; Schatz, G. C.; Mirkin, C. A. Nature 2003, 425, 487. (135) Faraday, M. Philos. Trans. R. Soc. London 1857, 147, 145. (136) Mie, G. Annalen der Physik 1908, 330, 377. (137) Turkevich, J.; Stevenson, P. C.; Hillier, J. Discussions of the Faraday Society 1951, 11, 55. (138) Daniel, M.-C.; Astruc, D. Chem. Rev. 2004, 104, 293. (139) Giersig, M.; Mulvaney, P. Langmuir 1993, 9, 3408. (140) Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Soc., Chem. Commun. 1994, 801. (141) Murphy, C. J.; Gole, A. M.; Hunyadi, S. E.; Stone, J. W.; Sisco, P. N.; Alkilany, A.; Kinard, B. E.; Hankins, P. Chem. Commun. 2008, 544. (142) Lal, S.; Clare, S. E.; Halas, N. J. Acc. Chem. Res. 2008, 41, 1842. (143) Niidome, T.; Yamagata, M.; Okamoto, Y.; Akiyama, Y.; Takahashi, H.; Kawano, T.; Katayama, Y.; Niidome, Y. J. Controlled Release 2006, 114, 343. (144) Sun, Y.; Xia, Y. J. Am. Chem. Soc. 2004, 126, 3892. (145) Wang, Z. J. Phys. Chem. B 2000, 104, 1153. (146) Barbosa, G. N.; Mac Leod, T. C. O.; Guedes, D. F. C.; Assis, M. D.; Oliveira, H. P. J Sol- Gel Sci Techn 2008, 46, 99. (147) Bayati, M. R.; Golestani-Fard, F.; Moshfegh, A. Z. Catalysis Letters 2010, 134, 162. (148) Zhu, D.; Liu, H.; Lv, L.; Yao, Y. D.; Yang, W. Z. Scripta Materialia 2008, 59, 642. (149) Zampronio, E. C.; Greggio, D. N.; Oliveira, H. P. J Non-Cryst Solids 2003, 332, 249. (150) Li, G. C.; Zhang, C. Q.; Peng, H. R.; Chen, K. Z. Macromolecular Rapid Communications 2009, 30, 1841. (151) Malta, M.; Silva, L. H.; Galembeck, A.; Korn, M. Macromolecular Rapid Communications 2008, 29, 1221. (152) Huguenin, F.; Girotto, E. M.; Torresi, R. M.; Buttry, D. A. Journal of Electroanalytical Chemistry 2002, 536, 37. (153) Jin, A. P.; Chen, W.; Zhu, Q. Y.; Yang, Y.; Volkov, V. L.; Zakharova, G. S. Solid State Ionics 2008, 179, 1256. 165

(154) Boyano, I.; Bengoechea, M.; de Meatza, I.; Miguel, O.; Cantero, I.; Ochoteco, E.; Rodriguez, J.; Lira-Cantu, M.; Gomez-Romero, P. J Power Sources 2007, 166, 471. (155) De, S.; Dey, A.; De, S. K. Solid State Communications 2006, 137, 662. (156) Demets, G. J. F.; Toma, H. E. Electrochem Commun 2003, 5, 73. (157) Huguenin, F.; Girotto, E. M.; Torresi, R. M.; Buttry, D. A. J Electroanal Chem 2002, 536, 37. (158) Demets, G. J. F.; Anaissi, F. J.; Toma, H. E. Electrochim Acta 2000, 46, 547. (159) Harreld, J.; Wong, H. P.; Dave, B. C.; Dunn, B.; Nazar, L. F. Journal of Non-Crystalline Solids 1998, 225, 319. (160) Jin, A. P.; Zhu, Q. Y.; Chen, W.; Volkov, V. L.; Zakharova, G. S.; Liu, H. X.; Zhou, J.; Xu, Q. Sol St Phen 2007, 124-126, 363. (161) Guerra, E. M.; Ciuffi, K. J.; Oliveira, H. P. Journal of Solid State Chemistry 2006, 179, 3814. (162) Boehme, J. L.; Mudigonda, D. S. K.; Ferraris, J. P. Chem Mater 2001, 13, 4469. (163) Liu, Y. J.; Schindler, J. L.; DeGroot, D. C.; Kannewurf, C. R.; Hirpo, W.; Kanatzidis, M. G. Chem Mater 1996, 8, 525. (164) Liu, Y. J.; Degroot, D. C.; Schindler, J. L.; Kannewurf, C. R.; Kanatzidis, M. G. Chem Mater 1991, 3, 992. (165) Dejene, F. B.; Ocaya, R. O. Current Applied Physics 2010, 10, 508. (166) Ilda, Y.; Kanno, Y. Jpn J Appl Phys 2008, 47, 667. (167) Schlecht, U.; Guse, B.; Raible, I.; Vossmeyer, T.; Burghard, M. Chem Commun 2004, 2184. (168) Kim, H. K.; Ok, Y. W.; Seong, T. Y.; Jeon, E. J.; Cho, W.; Yoon, Y. S. Journal of Vacuum Science & Technology a-Vacuum Surfaces and Films 2001, 19, 2549. (169) Bochenkov, V. E.; Sergeev, G. B. In Metal Oxide Nanostructures and Their Application; Ahmad, U., Hanh, Y. B., Eds.; Americal Scientific Publishers: 2010, p 31. (170) Liu, J. H.; Yang, R.; Li, S. M. J Rare Earth 2007, 25, 173. (171) Ivanova, T.; Harizanova, A. Mater Res Bull 2005, 40, 411. (172) Matralis, H. K.; Papadopoulou, C.; Kordulis, C.; Elguezabal, A. A.; Corberan, V. C. Appl Catal a-Gen 1995, 126, 365. (173) Siqintana, B.; Xu, A.; Zhang, Y.; Zhaorigetu, B. Chinese J Catal 2009, 30, 991. (174) Liu, H. R.; Chu, Y. Q.; Fu, Z. W.; Qin, Q. Z. J Power Sources 2003, 124, 163. (175) Chary, K. V. R.; Kumar, C. P.; Reddy, K. R.; Bhaskar, T.; Rajiah, T. Catal Commun 2002, 3, 7. (176) Najbar, M.; Broclawik, E.; Gora, A.; Camra, J.; Bialas, A.; Weselucha-Birczynska, A. Chem Phys Lett 2000, 325, 330. (177) Najbar, M.; Mizukami, F.; Bialas, A.; Camra, J.; Weselucha-Birczynska, A.; Izutsu, H.; Gora, A. Top Catal 2000, 11, 131. (178) Ozer, N.; Lampert, C. M. Thin Solid Films 1999, 349, 205. (179) BeginColin, S.; LeCaer, G.; Mocellin, A.; Jurenka, C.; Zandona, M. Journal of Solid State Chemistry 1996, 127, 98. (180) Cavani, F.; Trifiro, F.; Bartolini, A.; Ghisletti, D.; Nalli, M.; Santucci, A. J Chem Soc Faraday T 1996, 92, 4321. (181) Satsuma, A.; Okada, F.; Hattori, A.; Miyamoto, A.; Hattori, T.; Murakami, Y. Appl Catal 1991, 72, 295. (182) Pal, M.; Hirota, K.; Sakata, H. Physica Status Solidi a-Applied Research 2003, 196, 396. (183) Sidkey, M. A.; ElMallawany, R.; Nakhla, R. I.; AbdElMoneim, A. Physica Status Solidi a- Applied Research 1997, 159, 397. (184) Chopra, N.; Mansingh, A.; Chadha, G. K. Journal of Non-Crystalline Solids 1990, 126, 194. 166

(185) Liu, J. H.; Yang, R.; Li, S. M. Rare Metals 2006, 25, 636. (186) Moggi, P.; Morselli, S.; Lucarelli, C.; Sarzi-Amade, M.; Devillers, M. Stud Surf Sci Catal 2005, 155, 427. (187) Barbosa, G. N.; Oliveira, H. P. Journal of Non-Crystalline Solids 2006, 352, 3009. (188) Figueiredo, M. A.; de Faria, A. L.; Assis, M. D.; Oliveira, H. P. Journal of Non-Crystalline Solids 2005, 351, 3624. (189) Carrazan, S. R. G.; Benitez, P. J.; Rives, V. Vib Spectrosc 1993, 5, 295. (190) Cazzanelli, E.; Capoleoni, S.; Papalino, L.; Ceccato, R.; Carturan, G. Gnsr 2001: State of Art and Future Development in Raman Spectroscopy and Related Techniques 2002, 118. (191) Sharma, B. I.; Robi, P. S.; Srinivasan, A. Mater Lett 2003, 57, 3504. (192) Lin, Y. C.; Hohn, K. L. Catal Lett 2006, 107, 215. (193) Zampronio, E. C.; Barbosa, G. N.; Graeff, C. F. O.; Lassali, T. A. F.; Oliveira, H. P. Mater Res Soc Symp P 2005, 835, 27. (194) Barbosa, G. N.; Brunello, C. A.; Graeff, C. F. O.; Oliveira, H. P. Journal of Solid State Chemistry 2004, 177, 960. (195) Wang, M. G.; Zhong, S. H. Chinese J Catal 2007, 28, 124. (196) Shin, M. Y.; Nam, C. M.; Park, D. W.; Chung, J. S. Appl Catal a-Gen 2001, 211, 213. (197) Quaranta, N. E.; Soria, J.; Corberan, V. C.; Fierro, J. L. G. J Catal 1997, 171, 1. (198) Dias, C. R.; Portela, M. F.; GalanFereres, M.; Banares, M. A.; Granados, M. L.; Pena, M. A.; Fierro, J. L. G. Catal Lett 1997, 43, 117. (199) Li, G.; Jiang, L.; Peng, H. Macromolecules 2007, 40, 7890. (200) Pang, S. P.; Li, G. C.; Zhang, Z. K. Macromol Rapid Comm 2005, 26, 1262. (201) Qazilbash, M. M.; Brehm, M.; Chae, B.-G.; Ho, P.-C.; Andreev, G. O.; Kim, B.-J.; Yun, S. J.; Balatsky, A. V.; Maple, M. B.; Keilmann, F.; Kim, H.-T.; Basov, D. N. Science 2007, 318, 1750. (202) Rini, M.; Hao, Z.; Schoenlein, R. W.; Giannetti, C.; Parmigiani, F.; Fourmaux, S.; Kieffer, J. C.; Fujimori, A.; Onoda, M.; Wall, S.; Cavalleri, A. APPLIED PHYSICS LETTERS 2008, 92, 181904. (203) Huang, Z.; Chen, S.; Lv, C.; Huang, Y.; Lai, J. APPLIED PHYSICS LETTERS 2012, 101, 191905. (204) McWhan, D. B.; Remeika, J. P.; Rice, T. M.; Brinkman, W. F.; Maita, J. P.; Menth, A. Phys Rev Lett 1971, 27, 941. (205) Jayaraman, A.; McWhan, D. B.; Remeika, J. P.; Dernier, P. D. Phys Rev B 1970, 2, 3751. (206) Zhou, J.; Gao, Y.; Liu, X.; Chen, Z.; Dai, L.; Cao, C.; Luo, H.; Kanahira, M.; Sun, C.; Yan, L. Phys Chem Chem Phys 2013, 15, 7505. (207) Marezio, M.; McWhan, D. B.; Remeika, J. P.; Dernier, P. D. Phys Rev B 1972, 5, 2541. (208) Brückner, W.; Gerlach, U.; Brückner, H. P.; Moldenhauer, W.; Oppermann, H. physica status solidi (a) 1977, 42, 295. (209) Tan, X.; Yao, T.; Long, R.; Sun, Z.; Feng, Y.; Cheng, H.; Yuan, X.; Zhang, W.; Liu, Q.; Wu, C.; Xie, Y.; Wei, S. Scientific reports 2012, 2, 466. (210) Kang, M.; Kim, S. W.; Ryu, J. W.; Noh, T. AIP Advances 2012, 2. (211) Kang, L.; Gao, Y.; Luo, H. 2011; Vol. 687, p 791. (212) Kang, L.; Gao, Y.; Zhang, Z.; Du, J.; Cao, C.; Chen, Z.; Luo, H. The Journal of Physical Chemistry C 2010, 114, 1901. (213) Whittingham, M. S. Journal of The Electrochemical Society 1975, 122, 713. (214) Pan, A.; Zhang, J.-G.; Nie, Z.; Cao, G.; Arey, B. W.; Li, G.; Liang, S.-q.; Liu, J. J Mater Chem 2010, 20, 9193. (215) Zhai, T.; Liu, H.; Li, H.; Fang, X.; Liao, M.; Li, L.; Zhou, H.; Koide, Y.; Bando, Y.; Golberg, D. advanced materials 2010, 22, 2547. 167

(216) Harreld, J. H.; Dunn, B.; Cheng, W.; Leroux, F.; Nazar, L. F. Organic/Inorganic Hybrid Materials 1998, 519, 191. (217) Kuwabata, S.; Masui, S.; Tomiyori, H.; Yoneyama, H. Electrochim Acta 2000, 46, 91. (218) Cepak, V. M.; Hulteen, J. C.; Che, G.; Jirage, K. B.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R.; Yoneyama, H. Chem Mater 1997, 9, 1065. (219) Min, M. R.; Kim, J. H.; Kim, E. K.; Kim, Y. K. A.; Ha, J. S.; Kim, K. T. J Korean Phys Soc 2006, 49, 1097. (220) Takahashi, K.; Limmer, S. J.; Wang, Y.; Cao, G. Z. Jpn J Appl Phys 1 2005, 44, 662. (221) Mandal, T. K.; Fleming, M. S.; Walt, D. R. Chem Mater 2000, 12, 3481. (222) Wang, Y.-h.; Liu, H.; Zhu, D.; Guo, Z.-p.; Liu, H.-k.; Dou, S.-x. Transactions of Nonferrous Metals Society of China 2011, 21, 1303. (223) Cao, A. M.; Hu, J. S.; Liang, H. P.; Wan, L. J. Angew Chem Int Edit 2005, 44, 4391. (224) Xu, Y.; Zheng, L.; Wu, C.; Qi, F.; Xie, Y. Chemistry – A European Journal 2011, 17, 384. (225) Krusin-Elbaum, L.; Newns, D. M.; Zeng, H.; Derycke, V.; Sun, J. Z.; Sandstrom, R. Nature 2004, 431, 672. (226) Vavilova, E.; Hellmann, I.; Kataev, V.; Täschner, C.; Büchner, B.; Klingeler, R. Physical Review B 2006, 73, 144417. (227) Kweon, H.; Lee, K. W.; Lee, E. M.; Park, J.; Kim, I. M.; Lee, C. E.; Jung, G.; Gedanken, A.; Koltypin, Y. Phys Rev B 2007, 76, 045434. (228) Commeinhes, X.; Davidson, P.; Bourgaux, C.; Livage, J. advanced materials 1997, 9, 900. (229) Chin, S. F.; Iyer, K. S.; Raston, C. L. Cryst Growth Des 2009, 9, 2685. (230) Demishev, S. V.; Chernobrovkin, A. L.; Glushkov, V. V.; Grigorieva, A. V.; Goodilin, E. A.; Sluchanko, N. E.; Samarin, N. A.; Semeno, A. V. Jetp Lett. 2010, 91, 11. (231) Demishev, S. V.; Chernobrovkin, A. L.; Glushkov, V. V.; Goodilin, E. A.; Grigorieva, A. V.; Ishchenko, T. V.; Kuznetsov, A. V.; Sluchanko, N. E.; Tretyakov, Y. D.; Semeno, A. V. Fullerenes, Nanotubes and Carbon Nanostructures 2010, 19, 27. (232) Gründler, P. Chemical Sensors: An Introduction for Scientists and Engineers; Springer: Berlin, 2007. (233) Law, M.; Kind, H.; Messer, B.; Kim, F.; Yang, P. Angewandte Chemie International Edition 2002, 41, 2405. (234) Dan, Y.; Evoy, S.; Johnson, A. arXiv preprint arXiv:0804.4828 2008. (235) Maciak, E.; Opilski, Z.; Urbanczyk, M. J Phys Iv 2005, 129, 137. (236) Gou, X.; Wang, G.; Kong, X.; Wexler, D.; Horvat, J.; Yang, J.; Park, J. Chemistry – A European Journal 2008, 14, 5996. (237) Schilling, O.; Colbow, K. Sensor Actuat B-Chem 1994, 21, 151. (238) Ramirez, R.; Casal, B.; Utrera, L.; Ruiz-Hitzky, E. The Journal of Physical Chemistry 1990, 94, 8960. (239) Andersson, A. Journal of Solid State Chemistry 1982, 42, 263. (240) Fiermans, L.; Vennik, J. Surf Sci 1968, 9, 187. (241) Lambrecht, W.; Djafari-Rouhani, B.; Vennik, J. Surf Sci 1983, 126, 558. (242) Narayana, K. V.; Masthan, S. K.; Venugopal, A.; Rao, V. V.; Rao, K. S. R.; Rao, P. K. Recent Advances in Basic and Applied Aspects of Industrial Catalysis 1998, 113, 947. (243) Narayana, K. V.; Venugopal, A.; Rao, K. S. R.; Masthan, S. K.; Rao, V. V.; Rao, P. K. Appl Catal a-Gen 1998, 167, 11. (244) Mathews, S.; Garg, A.; Prasad, R. Indian J Technol 1991, 29, 195. (245) Sanati, M.; Andersson, A. Ind Eng Chem Res 1991, 30, 312. (246) Satsuma, A.; Takenaka, S.; Tanaka, T.; Nojima, S.; Kera, Y.; Miyata, H. Chem Lett 1996, 1115. 168

(247) Najbar, M.; Banas, J.; Korchowiec, J.; Bialas, A. Catal Today 2002, 73, 249. (248) Bielański, A.; Najbar, M. Applied Catalysis A: General 1997, 157, 223. (249) Brédas, J. L.; Chance, R. R. Conjugated Polymeric Materials: Opportunities in Electronics, Optoelectronics, and Molecular Electronics; Springer, 1990. (250) GANG GU, M. S., PO-WEN CHIU,ANDREW MINETT, JERÔME FRAYSSE,GYU-TAE KIM, SIEGMAR ROTH, MIKHAIL KOZLOV, EDGAR MUÑOZ AND RAY H. BAUGHMAN nature materials 2003, 2, 316. (251) Chain, E. E. Appl Optics 1991, 30, 2782. (252) Eden, D. D. Opt Eng 1981, 20, 203377. (253) Batista, C.; Ribeiro, R.; Carneiro, J.; Teixeira, V. J Nanosci Nanotechno 2009, 9, 4220. (254) Reddy, C. V. S.; Mho, S. I.; Kalluru, R. R.; Williams, Q. L. J Power Sources 2008, 179, 854. (255) Spahr, M. E.; Bitterli, P.; Nesper, R.; Müller, M.; Krumeich, F.; Nissen, H. U. Angew. Chem. Int. Ed. 1998, 37, 1263. (256) Nesper, R.; Muhr, H. J. Chimia 1998, 52, 571. (257) Wang, Y.; Cao, G. Adv. Mater. 2008, 20, 2251. (258) Chernova, N. A.; Roppolo, M.; Dillon, A. C.; Whittingham, M. S. Journal of Materials Chemistry 2009, 19, 2526. (259) Wu, C.; Xie, Y. Energy & Environmental Science 2010, 3, 1191. (260) Schlecht, U.; Besnard, I.; Yasuda, A.; Vossmeyer, T.; Burghard, M. Molecular Nanostructures 2003, 685, 491. (261) O'Dwyer, C.; Lavayen, V.; Newcomb, S. B.; Ana, M. A. S.; Benavente, E.; Gonzalez, G.; Torres, C. M. S. Journal of The Electrochemical Society 2007, 154, K29. (262) O'Dwyer, C.; Lavayen, V.; Tanner, D. A.; Newcomb, S. B.; Benavente, E.; Gonzalez, G.; Torres, C. M. S. Adv Funct Mater 2009, 19, 1736. (263) Su, Q.; Huang, C. K.; Wang, Y.; Fan, Y. C.; Lu, B. A.; Lan, W.; Wang, Y. Y.; Liu, X. Q. J. Alloys Compd. 2009, 475, 518. (264) Hu, C.-C.; Huang, C.-M.; Chang, K.-H. J. Power Sources 2008, 185, 1594. (265) Pinna, N.; Willinger, M.; Weiss, K.; Urban, J.; Schlogl, R. NANO LETTERS 2003, 3, 1131. (266) Liu, J.; Li, Q.; Wang, T.; Yu, D.; Li, Y. Angew. Chem. Int. Ed. 2004, 43, 5048. (267) Sun, Y.; Xia, Y. Science 2002, 298, 2176. (268) Tang, X.; Tsuji, M.; Jiang, P.; Nishio, M.; Jang, S.-M.; Yoon, S.-H. Colloids Surf., A 2009, 338, 33. (269) Weeks, C.; Song, Y. N.; Suzuki, M.; Chernova, N. A.; Zavalij, P. Y.; Whittingham, M. S. J Mater Chem 2003, 13, 1420. (270) Jiang, X.; Wang, Y.; Herricks, T.; Xia, Y. Journal of Materials Chemistry 2004, 14, 695. (271) C. O’Dwyer, D. N., V. Lavayen,*, E. Benavente,M. A. Santa Ana, G. Gonza´lez, S. B. Newcomb,| and C. M. Sotomayor Torres Chem. Mater. 2006, 2006, 3016. (272) O'Dwyer, C.; Lavayen, V.; Fuenzalida, D.; Newcomb, S. B.; Santa Ana, M. A.; Benavente, E.; González, G.; Sotomayor Torres, C. M. physica status solidi (b) 2007, 244, 4157. (273) Fei, H. L.; Liu, M.; Zhou, H. J.; Sun, P. C.; Ding, D. T.; Chen, T. H. Solid State Sci 2009, 11, 102. (274) Kang, L.; Gao, Y.; Luo, H. ACS Applied Materials & Interfaces 2009, 1, 2211. (275) Ragupathy, P.; Shivakumara, S.; Vasan, H. N.; Munichandraiah, N. J. Phys. Chem. C 2008, 112, 16700. (276) Shi, S.; Cao, M.; He, X.; Xie, H. Crystal Growth & Design 2007, 7, 1893. (277) Luca, V.; Hook, J. M. Chemistry of Materials 1997, 9, 2731.

169

(278) Luca, V.; MacLachlan, D. J.; Hook, J. M.; Withers, R. Chemistry of Materials 1995, 7, 2220. (279) Li, B.; Rong, G.; Xie, Y.; Huang, L.; Feng, C. Inorganic Chemistry 2006, 45, 6404. (280) Fu, X.; Feng, J.; Wang, H.; Ng, K. M. Nanotechnology 2009, 20, 375601. (281) Yang, H. G.; Zeng, H. C. J. Phys. Chem. B 2004, 108, 3492. (282) Sohn, J. R.; Doh, I. J.; Pae, Y. I. Langmuir 2002, 18, 6280. (283) Guerra, E. M.; Silva, G. R.; Mulato, M. Solid State Sci 2009, 11, 456. (284) Eid, C.; Arnaud, B.; Vincent, S.; Jean-Claude, P.; Roy, A.; Yves, M.; Randa, K.; Antonio, K.; Philippe, M. Nanotechnology 2010, 21, 125701. (285) Xu, P.; Han, X. J.; Liu, X. R.; Zhang, B.; Wang, C.; Wang, X. H. Materials Chemistry and Physics 2009, 114, 556. (286) Guifen, Y.; Xiaobin, Z.; Fu, L.; Jipeng, C.; Yuhong, M.; Huai, Z. In Nano/Micro Engineered and Molecular Systems, 2006. NEMS '06. 1st IEEE International Conference on 2006, p 649. (287) Zhao, Z.; Liu, Y.; Cao, H.; Gao, S.; Tu, M. Powder Technology 2008, 181, 31. (288) Lide, D. R. Handbook of Chemistry and Physics 87 ed.; CRC Press, 1998. (289) Sun, B.; Horvat, J.; Kim, H. S.; Kim, W.-S.; Ahn, J.; Wang, G. J. Phys. Chem. C 2010, 114, 18753. (290) Rella, R.; Siciliano, P.; Cricenti, A.; Generosi, R.; Girasole, M.; Vanzetti, L.; Anderle, M.; Coluzza, C. Thin Solid Films 1999, 349, 254. (291) A. Dhayal Raj , T. P., P. Suresh Kumar , D. Mangalaraj , D. Nataraj , N. Ponpandian Current Applied Physics 2009, 2009, 1. (292) Sahay, P. P. Journal of Materials Science 2005, 40, 4383. (293) Rezlescu, N.; Iftimie, N.; Rezlescu, E.; Doroftei, C.; Popa, P. D. Sensors and Actuators B: Chemical 2006, 114, 427. (294) Qin, L.; Xu, J.; Dong, X.; Pan, Q.; Cheng, Z.; Xiang, Q.; Li, F. Nanotechnology 2008, 19, 185705. (295) Aguilar-Leyva, J.; Maldonado, A.; de la L. Olvera, M. Materials Characterization, 58, 740. (296) Epifani, M.; Arbiol, J.; Pellicer, E.; Comini, E.; Siciliano, P.; Faglia, G.; Morante, J. R. Crystal Growth & Design 2008, 8, 1774. (297) Déramond, E.; Savariault, J. M.; Galy, J. Acta Crystallographica Section C 1994, 50, 164. (298) Leising, R. A.; Thiebolt, W. C.; Takeuchi, E. S. Inorganic Chemistry 1994, 33, 5733. (299) Van Den Berg, J.; Broersma, A.; Van Dillen, A. J.; Geus, J. W. Thermochimica Acta 1983, 63, 123. (300) Rozier, P.; Savariault, J.-M.; Galy, J. Journal of Solid State Chemistry 1996, 122, 303. (301) Takeuchi, E. S.; Thiebolt, W. C. Journal of The Electrochemical Society 1988, 135, 2691. (302) Mao, C.; Wu, X.; Pan, H.; Zhu, J.; Chen, H. Nanotechnology 2005, 16, 2892. (303) Liang, C. C.; Bolster, M. E.; Murphy, R. M. In United States Patent Wilson Greatbatch Ltd.: U.S., 1982. (304) Shi, H.; Li, Z.; Kou, J.; Ye, J.; Zou, Z. The Journal of Physical Chemistry C 2010, 115, 145. (305) Liang, Y.; Zhu, L. F.; Liu, P.; Li, H. B.; Xiao, J.; Ji, X. W.; Yang, G. W. CrystEngComm 2013. (306) Brooks, S.; Wright, R. NIVA Report 2008, 5698. (307) Hammer, G.; Lübcke, T.; Kettner, R.; Pillarella, M. R.; Recknagel, H.; Commichau, A.; Neumann, H.-J.; Paczynska-Lahme, B. In Ullmann's Encyclopedia of Industrial Chemistry; Wiley- VCH Verlag GmbH & Co. KGaA: 2000. (308) Murray, B. J.; Walter, E. C.; Penner, R. M. Nano Lett. 2004, 4, 665. (309) Scholtens, B. B. Materials Research Bulletin 1976, 11, 1533.

170

(310) Wenda, E. Journal of Thermal Analysis 1985, 30, 879. (311) Ivanova, Y. Y.; Dimitriev, Y. B. Materials Chemistry and Physics 1985, 12, 397. (312) Bertoni, M. I.; Kidner, N. J.; Mason, T. O.; Albrecht, T. A.; Sorensen, E. M.; Poeppelmeier, K. R. J Electroceram 2007, 18, 189. (313) Jiang, X. C.; Yu, A. B. Journal of Materials Processing Technology 2009, 209, 4558. (314) Zheng, Y.; Chen, C.; Zhan, Y.; Lin, X.; Zheng, Q.; Wei, K.; Zhu, J. The Journal of Physical Chemistry C 2008, 112, 10773. (315) Liu, S.; Yue, J.; Gedanken, A. Advanced Materials 2001, 13, 656. (316) Yu, D.; Yam, V. W.-W. Journal of the American Chemical Society 2004, 126, 13200. (317) Liu, J.; Xue, D. Nanoscale Res Lett 2010, 5, 1619. (318) Lu, C.; Shen, Q.; Zhao, X.; Zhu, J.; Guo, X.; Hou, W. Sensors and Actuators B: Chemical 2010, 150, 200. (319) Sauvage, F.; Bodenez, V.; Tarascon, J.-M.; Poeppelmeier, K. R. J Am Chem Soc 2010, 132, 6778. (320) Chemical Abstract, 1970; Vol. 72:22811d. (321) Bouhaouss, A.; Aldebert, P. Mater Res Bull 1983, 18, 1247. (322) Gomez-Romero, P. Advanced Materials 2001, 13, 163. (323) Frattini, A.; Pellegri, N.; Nicastro, D.; Sanctis, O. d. Materials Chemistry and Physics 2005, 94, 148. (324) Ohnishi, H.; Sasaki, H.; Matsumoto, T.; Ippommatsu, M. Sensors and Actuators: B. Chemical 1993, 14, 677. (325) Bochenkov, V. E.; Sergeev, G. B. Advances in Colloid and Interface Science 2005, 116, 245. (326) Xu, C.; Tamaki, J.; Miura, N.; Yamazoe, N. Sensors and Actuators B: Chemical 1991, 3, 147. (327) Hu, J.; Odom, T. W.; Lieber, C. M. Accounts of Chemical Research 1999, 32, 435. (328) Cui, Y.; Wei, Q.; Park, H.; Lieber, C. M. Science 2001, 293, 1289. (329) Zach, M. P.; Ng, K. H.; Penner, R. M. Science 2000, 290, 2120. (330) Fedutik, Y.; Temnov, V.; Woggon, U.; Ustinovich, E.; Artemyev, M. J. Am. Chem. Soc. 2007, 129, 14939. (331) Shanmukh, S.; Jones, L.; Driskell, J.; Zhao, Y.; Dluhy, R.; Tripp, R. A. Nano Lett. 2006, 6, 2630. (332) McFarland, A. D.; Young, M. A.; Dieringer, J. A.; Van Duyne, R. P. J. Phys. Chem. B 2005, 109, 11279. (333) Roy, S.; Chen, X.; Li, M.-H.; Peng, Y.; Anariba, F.; Gao, Z. J. Am. Chem. Soc. 2009, 131, 12211. (334) Silvis-Cividjian, N.; Hagen, C. W.; Kruit, P.; Stam, M. A. J. v. d.; Groen, H. B. Applied Physics Letters 2003, 82, 3514. (335) Kramer, N.; Birk, H.; Jorritsma, J.; Schonenberger, C. Applied Physics Letters 1995, 66, 1325. (336) Chaney, S. B.; Shanmukh, S.; Dluhy, R. A.; Zhao, Y.-P. Applied Physics Letters 2005, 87, 031908. (337) Chen, C. L.; Furusho, H.; Mori, H. Nanotechnology 2009, 20, 405605. (338) Zhao, W.-B.; Zhu, J.-J.; Chen, H.-Y. Journal of Crystal Growth 2003, 258, 176. (339) Eisele, D. r. M.; Berlepsch, H. v.; B ttcher, C.; Stevenson, K. J.; Vanden Bout, D. A.; Kirstein, S.; Rabe, J. r. P. J. Am. Chem. Soc. 2010, 132, 2104. (340) Kaper, H.; Willinger, M. G.; Djerdj, I.; Gross, S.; Antonietti, M.; Smarsly, B. M. J Mater Chem 2008, 18, 5761. 171

(341) Peppler, K.; Janek, J. Electrochimica Acta 2007, 53, 319. (342) Fang, J.; Hahn, H.; Krupke, R.; Schramm, F.; Scherer, T.; Ding, B.; Song, X. Chem. Commun. 2009, 1130. (343) Murphy, C. J.; Jana, N. R. Adv. Mater. 2002, 14, 80. (344) Sun, Y.; Tao, Z.; Chen, J.; Herricks, T.; Xia, Y. J. Am. Chem. Soc. 2004, 126, 5940. (345) Jiang, X.; Chen, W.; Chen, C.; Xiong, S.; Yu, A. Nanoscale Res Lett 2011, 6, 32. (346) Yuan, G.; Chang, X.; Zhu, G. Particuology 2011, 9, 644. (347) Zhang, D.; Qi, L.; Yang, J.; Ma, J.; Cheng, H.; Huang, L. Chem. Mater. 2004, 16, 872. (348) Jana, N. R.; Gearheart, L.; Murphy, C. J. Chem. Commun. 2001, 617. (349) Caswell, K. K.; Bender, C. M.; Murphy, C. J. Nano Lett. 2003, 3, 667. (350) Halder, A.; Ravishankar, N. Adv. Mater. 2007, 19, 1854. (351) Xu, X.; Zhuang, J.; Wang, X. J. Am. Chem. Soc. 2008, 130, 12527. (352) Zhou, B.; He, D. Y. J Raman Spectrosc 2008, 39, 1475. (353) Portehault, D.; Cassaignon, S.; Baudrin, E.; Jolivet, J.-P. Chem. Mater. 2007, 19, 5410. (354) Sun, S.; Deng, D.; Kong, C.; Zhang, Y.; Song, X.; Ding, B.; Yang, Z. CrystEngComm 2011, 13. (355) Wen, X.; Xie, Y.-T.; Mak, W. C.; Cheung, K. Y.; Li, X.-Y.; Renneberg, R.; Yang, S. Langmuir 2006, 22, 4836. (356) Shen, X. S.; Wang, G. Z.; Hong, X.; Xie, X.; Zhu, W.; Li, D. P. J. Am. Chem. Soc. 2009, 131, 10812. (357) Jiang, X. C.; Xiong, S. X.; Tian, Z. A.; Chen, C. Y.; Chen, W. M.; Yu, A. B. The Journal of Physical Chemistry C 2011, 115, 1800. (358) Jiang, X.; Xiong, S.; Chen, C.; Chen, W.; Yu, A. J. Nanopart. Res. 2011, 13, 5087. (359) Long, C.; Ustin, S. A.; Ho, W. J. Appl. Phys. 1999, 86, 2509. (360) Silversmit, G.; Depla, D.; Poelman, H.; Marin, G. B.; De Gryse, R. Journal of Electron Spectroscopy and Related Phenomena 2004, 135, 167. (361) Lide, D. R. CRC Hangdbook of Chemistry and Physics; 84 ed., 2003. (362) de Dios, M.; Barroso, F.; Tojo, C.; Blanco, M. C.; López-Quintela, M. A. Colloids Surf., A 2005, 270–271, 83. (363) Wei, G.; Wang, L.; Zhou, H.; Liu, Z.; Song, Y.; Li, Z. Applied Surface Science 2005, 252, 1189. (364) Zhu, W.; Zhu, S.; Xiang, L. CrystEngComm 2009, 11, 1910. (365) Jiang, X. C.; Zeng, Q. H.; Chen, C. Y.; Yu, A. B. Journal of Materials Chemistry 2011, 21, 16797. (366) Washio, I.; Xiong, Y.; Yin, Y.; Xia, Y. Adv. Mater. 2006, 18, 1745. (367) Pileni, M.-P. Nat Mater 2003, 2, 145. (368) Braun, E.; Eichen, Y.; Sivan, U.; Ben-Yoseph, G. Nature 1998, 391, 775. (369) Yu, H.; Peng, J.; Zhai, M.; Li, J.; Wei, G. Physica E: Low-dimensional Systems and Nanostructures 2008, 40, 2694. (370) Zong, R.-L.; Zhou, J.; Li, Q.; Du, B.; Li, B.; Fu, M.; Qi, X.-W.; Li, L.-T.; Buddhudu, S. J. Phys. Chem. B 2004, 108, 16713. (371) Dupuis, A. C.; Abu Haija, M.; Richter, B.; Kuhlenbeck, H.; Freund, H. J. Surface Science 2003, 539, 99. (372) Michaelis, M.; Henglein, A.; Mulvaney, P. J. Phys. Chem. 1994, 98, 6212. (373) Kosuge, K.; Takada, T.; Kachi, S. Journal of the Physical Society of Japan 1963, 18, 318. (374) C. J. Murphy, T. K. S., A. Gole, and C. J. Orendorff MRS Bullet 2005, 30, 349. (375) Sun, Y.; Gates, B.; Mayers, B.; Xia, Y. Nano Lett. 2002, 2, 165. (376) Sun, Y.; Yin, Y.; Mayers, B. T.; Herricks, T.; Xia, Y. Chem. Mater. 2002, 14, 4736. 172

(377) Xue, M.; Chen, H.; Zhang, H.; Auroux, A.; Shen, J. Applied Catalysis A: General 2010, 379, 7. (378) Teo, B. K.; Keating, K.; Kao, Y. H. J. Am. Chem. Soc. 1987, 109, 3494. (379) Toshima, N.; Harada, M.; Yamazaki, Y.; Asakura, K. J. Phys. Chem. 1992, 96, 9927. (380) Itakura, T.; Torigoe, K.; Esumi, K. Langmuir 1995, 11, 4129. (381) Liz-Marzan, L. M.; Philipse, A. P. J. Phys. Chem. 1995, 99, 15120. (382) Torigoe, K.; Esumi, K. Langmuir 1993, 9, 1664. (383) Schmid, G.; West, H.; Mehles, H.; Lehnert, A. Inorg. Chem. 1997, 36, 891. (384) Schmid, G.; Lehnert, A.; Malm, J.-O.; Bovin, J.-O. Angew. Chem., Int. Ed. 1991, 30, 874. (385) Toshima, N.; Hirakawa, K. Polym. J. 1999, 31, 1127. (386) Lu, P.; Dong, J.; Toshima, N. Langmuir 1999, 15, 7980. (387) Freeman, R. G.; Hommer, M. B.; Grabar, K. C.; Jackson, M. A.; Natan, M. J. J. Phys. Chem. 1996, 100, 718. (388) Lee, A. F.; Baddeley, C. J.; Hardacre, C.; Ormerod, R. M.; Lambert, R. M.; Schmid, G.; West, H. ChemInform 1995, 26, no. (389) Tsuji, M.; Miyamae, N.; Lim, S.; Kimura, K.; Zhang, X.; Hikino, S.; Nishio, M. Cryst. Growth Des. 2006, 6, 1801. (390) Tsuji, M.; Matsuo, R.; Jiang, P.; Miyamae, N.; Ueyama, D.; Nishio, M.; Hikino, S.; Kumagae, H.; Kamarudin, K. S. N.; Tang, X.-L. Cryst. Growth Des. 2008, 8, 2528. (391) Link, S.; Wang, Z. L.; El-Sayed, M. A. J. Phys. Chem. B 1999, 103, 3529. (392) Srnová-Šloufová, I.; Lednický, F.; Gemperle, A.; Gemperlová, J. Langmuir 2000, 16, 9928. (393) Ayyappan, S.; Gopalan, R. S.; Subbanna, G. N.; Rao, C. N. R. J. Mater. Res. 1997, 12, 398. (394) Han, J.; Zhou, Z.; Yin, Y.; Luo, X.; Li, J.; Zhang, H.; Yang, B. CrystEngComm 2012. (395) Pal, A.; Shah, S.; Devi, S. Colloids Surf., A 2007, 302, 483. (396) Brown, K. R.; Natan, M. J. Langmuir 1998, 14, 726. (397) Shahjamali, M. M.; Bosman, M.; Cao, S.; Huang, X.; Saadat, S.; Martinsson, E.; Aili, D.; Tay, Y. Y.; Liedberg, B.; Loo, S. C. J.; Zhang, H.; Boey, F.; Xue, C. Adv. Funct. Mater. 2012, 22, 849. (398) Ma, Y.; Li, W.; Cho, E. C.; Li, Z.; Yu, T.; Zeng, J.; Xie, Z.; Xia, Y. ACS Nano 2010, 4, 6725. (399) Métraux, G. S.; Cao, Y. C.; Jin, R.; Mirkin, C. A. Nano Lett. 2003, 3, 519. (400) Aherne, D.; Gara, M.; Kelly, J. M.; Gun'ko, Y. K. Adv. Funct. Mater. 2010, 20, 1329. (401) Lin, L.; Shang, K.; Xu, X.; Chu, C.; Ma, H.; Lee, Y.-I.; Hao, J.; Liu, H.-G. J. Phys. Chem. B 2011, 115, 11113. (402) Haruta, M. The Chemical Record 2003, 3, 75. (403) E. Bailie, J.; J. Hutchings, G. Chem. Commun. 1999, 2151. (404) Han, J.; Liu, Y.; Guo, R. J. Am. Chem. Soc. 2009, 131, 2060. (405) Hughes, M. D.; Xu, Y.-J.; Jenkins, P.; McMorn, P.; Landon, P.; Enache, D. I.; Carley, A. F.; Attard, G. A.; Hutchings, G. J.; King, F.; Stitt, E. H.; Johnston, P.; Griffin, K.; Kiely, C. J. Nature 2005, 437, 1132. (406) Haruta, M. Nature 2005, 437, 1098. (407) Zeng, J.; Zhang, Q.; Chen, J.; Xia, Y. Nano Lett. 2009, 10, 30. (408) Peceros, K. E.; Xu, X.; Bulcock, S. R.; Cortie, M. B. J. Phys. Chem. B 2005, 109, 21516. (409) Koga, H.; Tokunaga, E.; Hidaka, M.; Umemura, Y.; Saito, T.; Isogai, A.; Kitaoka, T. Chem. Commun. 2010, 46, 8567. (410) Tang, S.; Vongehr, S.; Meng, X. J. Mater. Chem. 2010, 20, 5436. (411) Kuroda, K.; Ishida, T.; Haruta, M. J. Mol. Catal. A: Chem. 2009, 298, 7. (412) Wu, F.; Yang, Q. Nano Res. 2011, 4, 861. 173

(413) Sharma, J.; Vivek, J. P.; Vijayamohanan, K. P. J. Nanosci. Nanotechnol. 2006, 6, 3464. (414) Qu, Y.; Xue, T.; Zhong, X.; Lin, Y.-C.; Liao, L.; Choi, J.; Duan, X. Adv. Funct. Mater. 2010, 20, 3005. (415) Fu, H.; Yang, X.; Jiang, X.; Yu, A. Particuology 2012. (416) Yue, J.; Jiang, X.; Yu, A. J. Phys. Chem. C 2012, 116, 8145. (417) Jiang, X. C.; Yu, A. B. Langmuir 2008, 24, 4300. (418) Mukherjee, P.; Nandi, A. K. Journal of Colloid and Interface Science 2010, 344, 30. (419) Li, X.; Cao, M.; Zhang, H.; Zhou, L.; Cheng, S.; Yao, J.-L.; Fan, L.-J. Journal of Colloid and Interface Science 2012, 382, 28. (420) Zhang, Z.; Shao, C.; Zou, P.; Zhang, P.; Zhang, M.; Mu, J.; Guo, Z.; Li, X.; Wang, C.; Liu, Y. Chem. Commun. 2011, 47, 3906. (421) Sun, Y. J. Phys. Chem. C 2010, 114, 2127. (422) Som, T.; Karmakar, B. Nano Res. 2009, 2, 607. (423) Sánchez-Ramírez, J. F.; Pal, U.; Nolasco-Hernández, L.; Mendoza-Álvarez, J.; Pescador- Rojas, J. A. J. Nanomater. 2008, 2008. (424) Du, Y.; Chen, H.; Chen, R.; Xu, N. Appl. Catal., A 2004, 277, 259. (425) Corbett, J. F. Dyes Pigm. 1999, 41, 127. (426) Rode, C. V.; Vaidya, M. J.; Chaudhari, R. V. Org. Process Res. Dev. 1999, 3, 465. (427) Mei, Y.; Lu, Y.; Polzer, F.; Ballauff, M.; Drechsler, M. Chem. Mater. 2007, 19, 1062. (428) Panigrahi, S.; Basu, S.; Praharaj, S.; Pande, S.; Jana, S.; Pal, A.; Ghosh, S. K.; Pal, T. J. Phys. Chem. C 2007, 111, 4596. (429) Oh, S.-D.; Kim, M.-R.; Choi, S.-H.; Chun, J.-H.; Lee, K.-P.; Gopalan, A.; Hwang, C.-G.; Sang-Ho, K.; Hoon, O. J. Journal of Industrial and Engineering Chemistry 2008, 14, 687.

174

Figure A1-1 JCPDS file of V2O5.

175