Investigation of Structural Characteristics of III-V Semiconductor Nanowires Grown by Molecular Beam Epitaxy Zhi Zhang Master of Chemical Engineering

A thesis submitted for the degree of Doctor of Philosophy at The University of Queensland in 2015 School of Mechanical and Mining Engineering

Abstract

One-dimensional nanowires made of III-V semiconductors have attracted significant research interest in the recent decades due to their distinct physical and chemical properties that can potentially lead to a wide range of applications in nanoelectronics and optoelectronics. As a key class of III-V semiconductor nanowires, InAs nanowires have attracted special attention due to their narrow bandgap, relatively high electron mobility, and small electron effective mass, which made them a promising candidate for the applications in future optical and high- frequency electronic devices. One of the challenges in realizing these unique III-V nanowire properties in nanowire-based devices is to integrate nanowires on the nano-devices or chips with well-organized arrangement. To solve this issue, the epitaxial nanowire growth provides the uniqueness that well aligned nanowires can be grown on the chosen substrates, and by selecting substrates with particular orientations, specifically orientated nanowires can be grown. Au-assisted nanowire growth is one of the most common methods to grow epitaxial III-V nanowires via the vapor-liquid-solid (VLS) mechanism or vapor-solid-solid (VSS) mechanism. For III-V nanowires, one of the coherent problems is that the stacking faults and/or twin defects can easily be introduced in both wurtzite or zinc-blende structured nanowires due to the small energy differences between wurtzite and zinc-blende stacking sequences of their dense planes. For III-V nanowires to be practically useful, it is of significant importance to minimize the lattice defects in nanowires, and to control their crystal phase and structural quality. In this regard, in this thesis, we investigated the growth parameters on the growth behaviour, crystal phase and structural quality of III-V nanowires, particularly InAs and GaAs nanowires. By carefully tuning the growth parameters, we have successfully achieved the controlled growth of InAs and GaAs nanowires with defect-free wurtzite structure and/or zinc- blende structure. Meanwhile, although InAs nanowires have been grown by different growth techniques, such as metalorganic chemical vapor deposition (MOCVD) or chemical beam expitaxy (CBE), very few studies have been devoted to the growth of InAs nanowires in molecular beam epitaxy (MBE). In this thesis, we employed the MBE system to grow Au-catalysed InAs nanowires on the GaAs{111}B substrate. Our

I extensive experiments have determined the relatively narrow V/III ratio window for the growth of InAs nanowires in MBE. Furthermore, by designing a two-V/III-ratio procedure, the pure defect-free zinc-blende structured InAs nanowires were achieved.

Declaration by author

This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text. I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, and any other original research work used or reported in my thesis. The content of my thesis is the result of work I have carried out since the commencement of my research higher degree candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution. I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the policy and procedures of The University of Queensland, the thesis be made available for research and study in accordance with the Copyright Act 1968 unless a period of embargo has been approved by the Dean of the Graduate School.

I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis.

III

Publications during candidature

Peer-reviewed papers

1. Zhang, Z.; Lu, Z.; Chen, P.; Lu, W.; Zou, J. Defect-Free Zinc-Blende Structured InAs Nanowires Realized by in-situ Two-V/III Ratio Growth in Molecular Beam Epitaxy. Nanoscale 2015, 7, 12592-12597. 2. Zhang, Z.; Shi, S.; Chen, P.; Lu, W.; Zou, J. Influence of Substrate Orientation on the Structural Quality of GaAs Nanowires in Molecular Beam Epitaxy. Nanotechnology 2015, 26, 255601. 3. Zhang, Z.; Lu, Z.; Chen, P.; Lu, W.; Zou, J. Controlling the Crystal Phase and Structural Quality of Epitaxial InAs Nanowires by Tuning V/III Ratio in Molecular Beam Epitaxy. Acta Mater. 2015, 92，25-32. 4. Zhang, Z.; Zheng, K.; Lu, Z.; Chen, P.; Lu, W.; Zou, J. Catalyst Orientation Induced Growth of Defect-Free Zinc-Blende Structured <00 ̅> InAs Nanowires. Nano Lett. 2015, 15, 876-882. 5. Zhang, Z.; Lu, Z.; Xu, H.; Chen, P.; Lu, W.; Zou, J. Structure and Quality Controlled Growth of InAs Nanowires through Catalyst Engineering. Nano Res. 2014, 7, 1640-1649. 6. Zhang, Z.; Lu, Z.; Chen, P.; Xu, H.; Guo, Y.; Liao, Z.; Shi, S.; Lu, W.; Zou, J. Quality of Epitaxial InAs Nanowires Controlled by Catalyst Size in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 103, 073109. 7. Liao, Z; Chen, Z.; Lu, Z.; Xu, H.; Guo, Y.; Sun, W.; Zhang, Z.; Yang, L.; Chen,

P.; Lu, W.; Zou, J. Au Impact on GaAs Epitaxial Growth on GaAs (111)B Substrates in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 102, 063106. 8. Ye, D.; Wang, B.; Chen, Y.; Han, G.; Zhang, Z.; Hulicova-Jurcakova, D.; Zou, J.; Wang, L.; Understanding the Stepwise Capacity-increase of High Energy Low-Co Li-rich Cathode Materials for Lithium Ion Batteries. J. Mater. Chem. A 2014, 2, 18767-18774. 9. Lu, Z.; Zhang, Z.; Chen, P.; Shi, S.; Yao, L.; Zhou, C.; Zhou, X.; Zou, J.; Lu, W.; Bismuth-induced Phase Control of GaAs Nanowires Grown by Molecular Beam Epitaxy. Appl. Phys. Lett. 2014, 105, 162102. 10. Guo, N.; Hu, W.; Liao, L.; Yip, S.; Ho, J.; Miao, J.; Zhang, Z.; Zou, J.; Chen, X.; Lu, W.; Anomalous and Highly-efficient InAs Nanowire Phototransistors Based

on Majority Carrier Transport at Room Temperature. Adv. Mater. 2014, 26, 8232-8232. 11. Shi, S.; Lu, Z.; Zhang, Z.; Zhou C.; Chen, P.; Zou, J.; Lu W.; Morphology and Microstructure of InAs Nanowires on GaAs Substrate Grown by Molecular Beam Epitaxy. Chin. Phys. Lett. 2014, 31 (9): 098101. 12. Shi, S.; Zhang, Z.; Lu, Z.; Shu, H.; Chen, P.; Li, N.; Zou, J.; Lu, W. Evolution of Morphology and Microstructure of GaAs/GaSb Nanowire Heterostructures. Nanoscale Res. Lett. 2015, 10, 108. 13. Shao, Y.; Wu, S.; Zhang, Z.; Zou, J.; Jiang, Z. Evolution of Thin Protecting Si-

layer on Mn0.5Si0.5 Layer at Low Temperatures. Appl. Surf. Sci. 2015, 333, 54- 58. 14. Xu, B.; Wu, H.; Lin, C.; Wang, B.; Zhang, Z.; Zhao, X. Stabilization of Silicon Nanoparticles in Graphene Aerogel Framework for Lithium Ion Storage. RSC Adv. 2015, 5, 30624- 30630. 15. Zhang, X.; Shao, R.; Jin, L.; Wang, J.; Zheng, K.; Zhao, C.; Han, J.; Chen, B.; Sekiguchi, T.; Zhang, Z.; Zhang, Z.; Zou, J.; Song, B. Helical Growth of Aluminum Nitride: New Insights into its Growth Habit from Nanostructures to Single Crystals. Sci. Rep. 2015, 5, 10087. 16. Chen, H.; Wang, Q.; Lyu, M.; Zhang, Z.; Wang, L. Wavelength-Switchable

Photocurrent in a Hybrid TiO2/Ag Nanocluster Photoelectrode. Chem. Commun. 2015, 51, 12072-12075. 17. Alaskar, Y.; Arafin, S.; Lin, Q.; Wickramaratne, D.; McKay, J.; Norman, A. G.; Zhang, Z.; Yao, L.; Ding, F.; Zou, J.; Goorsky, M.; Lake, R.; Zurbuchen, M.; Wang, K. Theoretical and Experimental Study of Highly Textured GaAs on Silicon Using a Graphene Buffer Layer. J. Cryst. Growth 2015, 425, 268-273.

Conference abstracts

1. Zou, J.; Zhang, Z.; Xu, H.; Guo, Y.; Liao, Z.; Lu, Z.; Chen, P.; Lu, W.; Gao, Q.; Tan, H.; and Jagadish, C; Impact of Catalysts in the Epitaxial Growth of III-V Nanowires. ICONN2014-ACMM2. (Adelaide, 2014, Invited presentation)

2. Zhang, Z.; Lu, Z.; Chen, P.; Xu, H.; Guo, Y.; Liao, Z.; Shi, S.; Lu, W.; Zou, J.; Catalyst Size Dependent Structural Quality of InAs Nanowires Grown by Molecular Beam Epitaxy. ICAMP8. (Gold Coast, 2014.07, Oral presentation)

3. Zhang, Z.; Lu, Z.; Xu, H.; Chen, P.; Lu, W.; Zou, J.; Effect of V/III Ratio on the Structural Quality of InAs Nanowires. COMMAD. (Perth, 2014.12, Oral presentation)

4. Zhang, Z.; Lu, Z.; Xu, H.; Chen, P.; Lu, W.; Zou, J. Effect of V/III Ratio on the Structural Quality of InAs Nanowires. IEEE.2014 Conference on Optoelectronic and Microelectronic Materials & Devices 2014, 24-26. (DOI: 10.1109/commad.2014.7038642)

Publications included in this thesis

Zhang, Z.; Lu, Z.; Chen, P.; Xu, H.; Guo, Y.; Liao, Z.; Shi, S.; Lu, W.; Zou, J. Quality of Epitaxial InAs Nanowires Controlled by Catalyst Size in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 103, 073109. – incorporated as Chapter 4.

Contributor Statement of contribution Zhi Zhang (Candidate) Design the experiments (70%) Carried out characterization (80%) Carried out data analysis (70%) Prepared the paper (60%) Zhenyu Lu Carried out nanowire growth (70%) Pingping Chen Carried out nanowire growth (20%) Supervised the project (10%) Prepared the paper (5%) Hongyi Xu Carried out characterization (10%) Yanan Guo Carried out characterization (5%) Zhiming Liao Carried out characterization (5%) Suixing Shi Carried out nanowire growth (10%) Wei Lu Supervised the project (10%) Prepared the paper (5%) Jin Zou Design the experiments (30%) Carried out data analysis (30%) Supervised the project (80%) Prepared the paper (30%)

VII

Zhang, Z.; Lu, Z.; Xu, H.; Chen, P.; Lu, W.; Zou, J. Structure and Quality Controlled Growth of InAs Nanowires through Catalyst Engineering. Nano Res. 2014, 7, 1640- 1649. – incorporated as Chapter 5. Contributor Statement of contribution Zhi Zhang (Candidate) Design the experiments (70%) Carried out characterization (90%) Carried out data analysis (70%) Prepared the paper (60%) Zhenyu Lu Carried out nanowire growth (80%) Hongyi Xu Carried out characterization (10%) Pingping Chen Carried out nanowire growth (20%) Supervised the project (10%) Prepared the paper (5%) Wei Lu Supervised the project (10%) Prepared the paper (5%) Jin Zou Design the experiments (30%) Carried out data analysis (30%) Supervised the project (80%) Prepared the paper (30%)

VIII

Zhang, Z.; Zheng, K.; Lu, Z.; Chen, P.; Lu, W.; Zou, J. Catalyst Orientation Induced Growth of Defect-Free Zinc-Blende Structured <00 ̅> InAs Nanowires. Nano Lett. 2015, 15, 876-882. – incorporated as Chapter 6. Contributor Statement of contribution Zhi Zhang (Candidate) Design the experiments (70%) Carried out characterization (90%) Carried out data analysis (60%) Prepared the paper (60%) Kun Zheng Carried out characterization (10%) Carried out data analysis (10%) Zhenyu Lu Carried out nanowire growth (80%) Pingping Chen Carried out nanowire growth (20%) Supervised the project (10%) Prepared the paper (5%) Wei Lu Supervised the project (10%) Prepared the paper (5%) Jin Zou Design the experiments (30%) Carried out data analysis (30%) Supervised the project (80%) Prepared the paper (30%)

Zhang, Z.; Lu, Z.; Chen, P.; Lu, W.; Zou, J. Controlling the Crystal Phase and Structural Quality of Epitaxial InAs Nanowires by Tuning V/III Ratio in Molecular Beam Epitaxy. Acta Mater. 2015, 92，25-32. – incorporated as Chapter 7. Contributor Statement of contribution Zhi Zhang (Candidate) Design the experiments (70%) Carried out characterization (100%) Carried out data analysis (70%) Prepared the paper (60%) Zhenyu Lu Carried out nanowire growth (80%) Pingping Chen Carried out nanowire growth (20%) Supervised the project (10%) Prepared the paper (5%) Wei Lu Supervised the project (10%) Prepared the paper (5%) Jin Zou Design the experiments (30%) Carried out data analysis (30%) Supervised the project (80%) Prepared the paper (30%)

Zhang, Z.; Shi, S.; Chen, P.; Lu, W.; Zou, J. Influence of Substrate Orientation on the Structural Quality of GaAs Nanowires in Molecular Beam Epitaxy. Nanotechnology 2015, 26, 255601. – incorporated as Chapter 8. Contributor Statement of contribution Zhi Zhang (Candidate) Design the experiments (70%) Carried out characterization (100%) Carried out data analysis (70%) Prepared the paper (60%) Suixing Shi Carried out nanowire growth (80%) Pingping Chen Carried out nanowire growth (20%) Supervised the project (10%) Prepared the paper (5%) Wei Lu Supervised the project (10%) Prepared the paper (5%) Jin Zou Design the experiments (30%) Carried out data analysis (30%) Supervised the project (80%) Prepared the paper (30%)

Zhang, Z.; Lu, Z.; Chen, P.; Lu, W.; Zou, J. Defect-Free Zinc-Blende Structured InAs Nanowires Realized by in Situ Two V/III Ratio Growth in Molecular Beam Epitaxy. Nanoscale 2015, 7, 12592-12597. – incorporated as Chapter 9. Contributor Statement of contribution Zhi Zhang (Candidate) Design the experiments (70%) Carried out characterization (100%) Carried out data analysis (70%) Prepared the paper (60%) Zhenyu Lu Carried out nanowire growth (80%) Pingping Chen Carried out nanowire growth (20%) Supervised the project (10%) Prepared the paper (5%) Wei Lu Supervised the project (10%) Prepared the paper (5%) Jin Zou Design the experiments (30%) Carried out data analysis (30%) Supervised the project (80%) Prepared the paper (30%)

XII

Contributions by others to the thesis

“No contributions by others.”

Statement of parts of the thesis submitted to qualify for the award of another degree

“None”

XIII

Acknowledgements

First of all, I sincerely acknowledge my principal supervisor, Professor Jin Zou for his careful guidance to me in every aspect of my PhD research and his great patience with my learning. I am greatly impressed by his broad and deep knowledge, his passion and his seriousness in science. It is my great honour to join his family- like group. I will always appreciate those moments that he spent hours with me to teach me the skills on electron microscope and teach me how to write scientific papers. I benefit a lot from this process. I want to extend my thanks to my other two supervisors, Professor Wei Lu, Professor Pingping Chen and our co-workers Zhenyu Lu and Suixing Shi in SITP, CAS China for their generous help in sample growths and their precious knowledge in MBE growth. Also, I wish to thank Dr Zhigang Chen, Dr Yang Huang, Dr Yanan Guo, Dr Hongyi (Justin) Xu, Dr Wen Sun, Dr Guang Han, Dr Kun Zheng, Zhiming Liao, Lei Yang, Min Hong, Yichao Zou, Mun Soo, and Chen Zhou in our research group at The University of Queensland. Thank you very much for your great help and support in my study, research and life in Brisbane. In the meantime, I am very grateful for the staff of Centre for Microscopy and Microanalysis (CMM) at UQ, who helped me in so many ways. They are Richard Webb, Robyn Webb, Robert Gould, Wendy Amstrong, Bronwen Cribb, Kathryn Green, Eunice Grinan, Ying Yu, Kay Hodge, Andrew Stark, Kim Sewell, Ron Rasch, Gary Morgan and Jill Prescott. Special thanks go to Professor John Drennan for welcoming me into the CMM family in the first place. Because of him, I have been able to stay in such a nice office, shared with such a wonderful group of people during my candidature. Dr. Graeme Auchtermonie deserves another special thank. This is not only because he selflessly taught me everything he knows about practical TEM, but also we shared a lot of conversations about everything. The friendship and fellowship with him are the things I will always treasure. Last but most importantly, I want to give my great thanks to my family: my dad, my mum, and my little brother, my girlfriend Yaqin Wang, and all my best friends in China. Their endless love and encouragement are always the great power in my life.

XIV

Keywords III-V nanowires, semiconductor, molecular beam epitaxy, electron microscopy, crystal phase, structural quality

Australian and New Zealand Standard Research Classifications (ANZSRC)

ANZSRC code: 100712, Nanoscale Characterisation, 60%

ANZSRC code: 100706, Nanofabrication, Growth and Self Assembly, 30%

ANZSRC code: 091203, Compound Semiconductors, 10%

Fields of Research (FoR) Classification

FoR code: 0912, Materials Engineering, 50%

FoR code: 1007, Nanotechnology, 50%

Table of Contents

1 Introduction ...... 1

1.1 Background ...... 1

1.2 Objective and scopes ...... 3

1.3 Thesis outline ...... 3

2 Literature Review ...... 5

2.1 Semiconductor nanowire ...... 5

2.2 Nanowire growth techniques ...... 6

2.2.1 Epitaxial growth ...... 7

2.2.2 Growth techniques ...... 8

2.2.2.1 MOCVD ...... 8

2.2.2.2 MBE ...... 9

2.2.2.3 CBE ...... 10

2.3 Nanowire growth mechanism ...... 11

2.3.1 VLS ...... 11

2.3.2 VSS ...... 14

2.3.3 PIN ...... 16

2.3.4 VS ...... 17

2.4 Comparison of MBE and MOCVD ...... 18

2.5 Growth and characteristics of III-V nanowires ...... 19

2.5.1 Growth direction ...... 19

2.5.1.1 Determination of growth direction in TEM ...... 21

2.5.1.2 Determination of growth direction in SEM ...... 23

2.5.2 Nanowire growth rate ...... 25

2.5.2.1 Diameter effect ...... 26

2.5.2.2 Nanowire density effect ...... 27

2.5.2.3 V/III ratio effect ...... 28

XVI

2.5.2.4 Growth temperature ...... 30

2.5.3 Nanowire facet ...... 31

2.6 III-V nanowires crystal phase and structural quality ...... 33

2.7 Summary ...... 34

2.8 References ...... 35

3 Experimental Techniques ...... 51

3.1 III-V nanowires grown by MBE ...... 51

3.1.1 MBE ...... 51

3.1.2 Nanowire growth in MBE ...... 53

3.1.3 Nanowire growth procedures in MBE ...... 54

3.2 Electron microscopy ...... 55

3.2.1 A brief history of electron microscopy ...... 55

3.2.2 Interaction of incident electrons with a specimen ...... 56

3.2.3 Scanning electron microscopy (SEM) ...... 59

3.2.4 Transmission electron microscopy (TEM) ...... 63

3.3 Sample preparation for electron microscopy ...... 67

3.3.1 SEM sample preparation ...... 67

3.3.2 TEM sample preparation ...... 68

3.4 References ...... 69

4 Quality of Epitaxial InAs Nanowires Controlled by Catalyst Size in

Molecular Beam Epitaxy ...... 73

4.1 Introduction ...... 73

4.2 Journal Publication ...... 74

5 Structure and Quality Controlled Growth of InAs Nanowires through Catalyst Engineering ...... 91

5.1 Introduction ...... 91

5.2 Journal Publication ...... 92

XVII

6 Catalyst Orientation Induced Growth of Defect-Free Zinc-Blende

Structured <00 > InAs Nanowires ...... 115

6.1 Introduction ...... 115

6.2 Journal Publication ...... 116

7 Controlling the Crystal Phase and Structural Quality of Epitaxial

InAs Nanowires by Tuning V/III Ratio in Molecular Beam Epitaxy 147

7.1 Introduction ...... 147

7.2 Journal Publication ...... 148

8 Influence of Substrate Orientation on the Structural Quality of

GaAs Nanowires in Molecular Beam Epitaxy ...... 171

8.1 Introduction ...... 171

8.2 Journal Publication ...... 172

9 Defect-Free Zinc-Blende Structured InAs Nanowires Realized by in Situ Two V/III Ratio Growth in Molecular Beam Epitaxy ...... 191

9.1 Introduction ...... 191

9.2 Journal Publication ...... 192

10 Conclusions and Recommendations ...... 209

10.1 Conclusions ...... 209

10.2 Recommendations ...... 211

XVIII

List of Figures

Figure 2.1 Plot of bandgap vs. lattice constant.

Figure 2.2 Schematic illustration of the Aixtron 200/4 MOCVD reactor.

Figure 2.3 Schematic top view of MBE system.

Figure 2.4 Schematic overview of CBE equipment.

Figure 2.5 Schematic illustration of Si whisker growth from the reaction of SiCl4 and

H2 vapor phases.

Figure 2.6 Experimental and model droplet geometries of this new VLS mode.

Figure 2.7 Self-catalysed nanowire growths of (a) InAs and (b) GaAs.

Figure 2.8 Schematic illustration of VSS mechanism of Si nanowires catalysed by Al.

Figure 2.9 General terms in PIN and illustration of PIN.

Figure 2.10 InAs NWs image and schematic representation of the VS growth model.

Figure 2.11 Determination of nanowire growth direction in TEM. (a-g) shows an example using three set of SAED pattern to determine the growth direction of a <110> nanowire, and to determine the catalyst/nanowire interface.

Figure 2.12 Determine the growth direction of epitaxial nanowires in SEM. (a) 0 degree tilted images of the GaAs{111}B substrate. The inset in the SEM image of (a) is the sketch of all the <112> directions on the GaAs{111}B substrate. (b) 20° tilted images. (c) 35° tilted images when the axis of inclined nanowire is parallel to the electron beam. The unlabelled scale bar is 1µm. The inset in the SEM image of (c) is an enlarged SEM image showing the hexagonal-shape cross section.

Figure 2.13 Sketch showing the possible ways for Ga atoms contribute to the nanowire growth: (1) direct impingement on the droplet, (2) diffusion of atoms that impinge on the side facets, (3) diffusion of atoms that impinge on the substrate, (4) secondary adsorption.

Figure 2.14 Nanowire growth rate versus Au nanoparticle diameter.

Figure 2.15 Schematic nanowire growth model: (a) Competitive regime, (b) Synergetic regime, (c) Independent regime.

Figure 2.16 V/III ratio effect to the growth and morphology of nanowires.

XIX

Figure 2.17 Growth rate of GaAs nanowire as a function of As4 pressure.

Figure 2.18 InAs nanowire growth rate as a function of In and As flux.

Figure 2.19 Effect of Arsenic species on the growth rate.

Figure 2.20 Growth rate as a function of growth temperature.

Figure 2.21 (a-c) The growth parameter changes nanowire facets; (d) Top view SEM images revealing the sidewall facet transition between the tip and base.

Figure 2.22 The summary of reported cross-section facets of non-<111>B/<000 ̅> III-V nanowires.

Figure 3.1 A photograph of the solid-sourced Riber 32 MBE system located at Shanghai Institute of Technical Physics, Chinese Academy of Sciences.

Figure 3.2 Schematic illustrations of the interaction of an electron beam with a specimen.

Figure 3.3 A schematic illustration of the emission of characteristic X-rays.

Figure 3.4 A schematic illustration of a SEM (http://www.purdue.edu/rem/rs/sem.htm).

Figure 3.5 Illustration of interaction volume and emitted signals depth in a SEM.

Figure 3.6 Schematic illustrations of the effect of topography on SE yields (http://laser.phys.ualberta.ca/~egerton/SEM/sem.htm).

Figure 3.7 SEM images of Pt particles on alumina. Left SE image shows morphology only, while Pt particles appear with bright contrast in the right BSE image (http://www.microscopy.ethz.ch/bse.htm).

Figure 3.8 Schematic illustrations of TEM (http://www.ammrf.org.au/myscope/tem/introduction/).

List of Tables

Table 2.1 Comparison of MOCVD and MBE

Table 2.2 Summary of the non-<111>B/<0001>B growth directions of III-V nanowires

Table 3.1 A brief history of electron microscopy

Table 3.2 A comparison of SEM emission sources

Table 3.3 TEM sample preparations

XXI

List of Abbreviations used in the thesis

BF: bright-field EDS: energy dispersive X-ray spectroscopy EELS: electron energy loss spectroscopy HRTEM: high-resolution transmission electron microscopy MBE: molecular beam epitaxy MOCVD: metal-organic chemical vapor deposition CBE: chemical beam epitaxy SAED: selected area electron diffraction SEM: scanning electron microscope STEM: scanning transmission electron microscopy TEM: transmission electron microscopy VLS: vapor-liquid-solid VSS: vapor-solid-solid VS: vapor-solid WZ: wurtzite ZB: zinc-blende

XXII

XXIII

Nanowire circuit

1 Introduction

1.1 Background

With outstanding potential as the building blocks of future optical and optoelectronic devices and systems, semiconductor nanowires have attracted intense research interests during the past decade. Amongst semiconductor nanowires, III-V compound nanowires have always been one of the most promising materials with their direct bandgap structures and superior optical and optoelectronic properties.1-3 In order to investigate the growth mechanism and hence to better achieve the superior properties of these III-V nanowires, many growth techniques have been employed to epitaxially grow such semiconductor nanowires. One prominent growth technique to grow III-V nanowire is the molecular beam epitaxy (MBE),4-9 with the gold nanoparticles as the catalyst through the vapor-liquid-solid

(VLS) or vapor-solid-solid (VSS) growth mechnism, or without catalyst through the vapor-solid (VS) growth mechanism.

Generally speaking, the III-V nanowires are nucleated in the <111>B directions for zinc-blende structured nanowires or in the <000 ̅> direction for wurtzite structured nanowires due to the lowest surface energy in {111}/{0001} planes.10 In most cases, nanowires grown along the <111>B/<000 ̅ > directions with the {111}B/{000 ̅ } catalyst/nanowire interfaces tend to naturally contain uncontrolled mixture of polytypes and/or adopt planar defects,11-14 such as stacking faults or twins due to the small energetic differences for the stacking consequences in zinc-blende and 15 wurtzite structures along their <111>B/<000 ̅ > directions, although great efforts have been devoted to control the structure quality by tuning the growth parameters.16-19 For nanowires to be practically useful in nanowire-based devices, it is crucial to control their structural quality since the interruptions in the stacking sequence can act as scattering centres for electrons20 and can negatively affect the performance of nanowires.21-25 As a key III-V semiconductor, InAs has attracted special research interest due to its very high electron mobility,26 low-resistance ohmic contact,27 narrow band gap and small electron effective mass.28 The combination of these unique features and the NW distinct characteristics have made InAs NWs a promising candidate for applications in single-electron transistors,29 the Josephson junctions,30 resonant tunnelling diodes,31 and ballistic transistors.32 The InAs nanowires have been synthesized by different growth techniques such as metalorganic chemical vapor deposition (MOCVD),33-38 chemical beam expitaxy (CBE),39-41 or in a closed vapor growth system. Very few studies have been devoted to the catalyst-assisted MBE of InAs nanowires. The nanowire growth by elementary source MBE differs significantly from that by MOCVD or CBE due to the totally different growth environment. For example, in the MBE, the adsorption of the vapor phase species does not require a thermally activated decomposition of organometallic or hydride molecules. Although GaAs nanowires grown by MBE have been extensively researched, much less research work has been focused on InAs nanowires grown by MBE. Meanwhile, the investigation of the crystal phase and structural quality of InAs nanowires grown by MBE has been paid no attention. In addition, the growth directions of InAs nanowires has not been well investigated. This PhD thesis

addresses the crystal phase and structural quality controlled growth of InAs nanowires using electron microscopy.

1.2 Objective and scopes

The unique physical characteristics of InAs have made InAs nanowires highly promising, particularly for the use in the future high-frequency electron devices. Therefore, it is of significant importance to examine the structural characteristics and growth mechanism of InAs nanowires. The objective of this PhD thesis is to investigate the crystal phase and structural quality of Au-catalyzed InAs nanowires grown in MBE with advanced electron microscopy. The fulfilment of these objectives will provide insight into the growth mechanism of nanowires which are crucial for the design of nanowire based devices.

1.3 Thesis outline

In order to investigate the structural characteristics of III-V semiconductor nanowires, especially InAs and GaAs nanowires, nanowires are grown on GaAs

{111}B substrates in the MBE growth chamber, and are analysed with advanced electron microscopy.

Chapter 1 is the introduction of this thesis.

Chapter 2 presents a literature review of previous research work of nanowires. A few nanowire growth techniques and growth mechanisms are summarized. Then a simple comparison between metalorganic chemical vapour deposition (MOCVD) and MBE is stated. After that, this chapter discusses the growth parameters that influence the growth of nanowires. At the end of this chapter, a few unclear issues in the research of III-V nanowires are stated and these will form the objectives of this project.

Chapter 3 deals with the experimental techniques used in this thesis. The MBE technique is used to grow all the samples in this thesis. The theoretical and practical aspects of MBE are discussed in details. Then, some basic theories and experimental techniques of electron microscopy are introduced.

Chapter 4 demonstrate the effect of catalyst size on the structural quality of InAs nanowires grown on the GaAs {111}B substrates by MBE. This chapter is included as the Applied Physics Letter, 2013, 103, 073109.

Chapter 5 shows the structure and quality controlled growth of InAs nanowire through catalyst engineering. This chapter is included as the Nano Research, 2014, 7, 1640-1649.

Chapter 6 depicts the catalyst orientation induced growth of defect-free zinc- blende structured InAs nanowires. This chapter is included as the Nano Letter, 2015, 15, 876-882.

Chapter 7 demonstrate the effect of V/III ratio on the crystal phase and structural quality of InAs nanowires grown by MBE. This chapter is included as the Acta Materialia, 2015, 92，25-32.

Chapter 8 presents the results regarding the effect of substrate orientation on the structural quality of GaAs nanowires grown in MBE. This chapter is included as the Nanotechnology, 2015, 26, 255601.

Chapter 9 shows the controlled growth of pure defect-free zinc-blende structured InAs nanowires by in-situ two-V/III-ratio growth in MBE. This chapter is included as a Nanoscale, 2015, 7, 12592-12597.

Finally, Chapter 10 summarizes the conclusions of this PhD research work and the recommendations for future work.

2 Literature Review

2.1 Semiconductor nanowire

Semiconductors are of particular importance for electronics since their conductivity can be dynamically modified through the use of outside energy such as electronic field. Their electrical properties may be modified by the introduction of dopant, making them very versatile. The nature of semiconductors makes them the foundation of modern electronic and optoelectronic industry. When the physical size of materials decreases to the sub-100 nanometre scale, possessing the low-dimensional structures, quantum mechanical effects become

significant , and these low-dimensional materials such as quantum well, nanowire and quantum dot, will show significantly different properties compared to their bulk counterparts. According to the reported literatures, it can be noted that the research on nanowires has been rapidly expanded during the last decades. Semiconductor nanowires, with quantum confinement effect as well as high surface-to-volume ratio, have attracted tremendous research interest especially during the past decade due to their superior applications in optical and optoelectronic devices. Among various semiconductor nanowires, III-V nanowires show particular interesting characteristics. The superior electrical and optical properties, including their direct bandgap and high electron mobility, endow them a potential role in the advanced technologies like high performance field effect transistors, photo-detectors, chemical/biosensors and thermo electric devices.1-3 The practical applications of these nanowires greatly depend on the control over the morphology, positioning of nanowires, crystal phase and structural quality, which require extensive study on the growth parameters like growth temperature, V/III, absolute flow rate, choice of substrates, choice of catalyst type and size, and so on. In this regards, researcher have investigated the effects of these growth conditions on the controlled growth of nanowires in order to design and manufacture nanowires with desired properties.

2.2 Nanowire growth techniques

In general, there are two approaches to produce nanowires, namely, the top-down and bottom-up methods.42 Top-down method uses bulk material of desired composition, and achieves nanometre-scale dimensions by lithography and etching that essentially carve the desired structure out of the material. Top-down method has dominated the microelectronics industry over the last century, and it is still of significant importance for the production of electronic components. However, as the length scales of devices shrink, the top-down method becomes increasingly problematic according to Moore’s law. More importantly, the uniformity of bulk crystals on nanometre length scales is not very high, and the quality of structures becomes difficult to control. Furthermore, this process is intrinsically wasteful since most of the materials have

been discarded, and there are many challenges to overcome before this technology becomes a high-throughput and cost-effective means of nanowire fabrication. Bottom-up production method, on the other hand, mimics nature’s way of self- assembling atoms to form increasingly larger structures such as nanowires. Such technique involves controlled crystallization of materials from vapor or liquid sources, typically yielding uniform and highly ordered nanometre-scale structures. This bottom-up paradigm offers opportunities for fabrication of atomically precise, complex devices not possible with conventional top-down technology. In terms of nanowires, this method can achieve both axial and radial heterostructured nanowires with abrupt compositional change.

2.2.1 Epitaxial growth The bottom-up approach has been demonstrated by many growth techniques to be the more effective way of batch producing the semiconductor nanowires. This is especially true for the growth of III-V compound nanowires. Among various bottom- up approaches, Epitaxial growth is the most promising method to grow nanowires. Epitaxy is the general term for the oriented growth of a crystalline material on a single crystal structure.43 This means that the structure and orientation of the growing crystal will be influenced by that of the substrate. We use the terms “homoepitaxy” to refer to the growth of a crystal on the same substrate, and “heteroepitaxy” to refer to the growth of a crystal on a different substrate.

Figure 2.1 Plot of bandgap vs. lattice constant.44

Epitaxial growth of nanowires has become an exciting and rapidly expanding field in recent decades. Such nanowires have controlled crystal structure and usually length, and in many cases controlled diameter. The chemical composition is also controlled, and can even be varied over the length or radius of a nanowire, yielding axial or radial heterostructures. Epitaxial nanowire growth not only has the advantage that nanowire keeps the epitaxial relationship with the substrate as the case in epitaxial growth of film, they can also grow in the largely lattice-mismatched system (as shown in Figure 2.1) which is not possible for the growth of film. 2.2.2 Growth techniques

There are many growth techniques , mainly including Chemical vapour depiction (CVD),45 Metalorganic chemical vapour deposition (MOCVD),35 Molecular beam epitaxy (MBE),46 Chemical beam epitaxy (CBE),39 Laser-assisted catalytic growth (LCG),47 Oxide assisted growth (OAG),48 and so on. In this thesis, we employ the epitaxial growth process and conducted the growth of nanowires in MBE, and we will mainly introduce three mostly used nanowire growth techniques in recently years, namely, MOCVD, MBE and CBE.

2.2.2.1 MOCVD

Figure 2.2 Schematic illustration of the Aixtron 200/4 MOCVD reactor.49

MOCVD, also known as metalorganic vapour phase epitaxy (MOVPE) or organometallic vapour phase epitaxy (OMVPE), is an arranged chemical vapor deposition method capable of growing abrupt interfaces and high purity materials. The pioneering work of MOCVD was performed by Manasevit and colleagues50-51 in early 1970s in Japan. The schematic illustration of MOCVD reactor is shown in Figure 2.2. In MOCVD, ultra-pure gases are injected into a reactor and finely dosed to deposit onto a semiconductor wafer. Surface reaction of organic compounds or metalorganics and hydrides containing the required chemical elements creates conditions for crystalline growth - epitaxy of materials and compound semiconductors. In contrast to MBE, the growth of crystals in MOCVD is by chemical reaction and not physical deposition. This growth process takes place not in a vacuum, but from the gas phase at moderate pressures (2 to 100 kPa). As such, this technique is preferred for the formation of devices incorporating thermodynamically metastable alloys, and it has become a major process in the manufacture of optoelectronics. This advanced technique can produce almost all the III-V and II-V semiconductor compounds, and is readily scalable for industrial mass production.

2.2.2.2 MBE

Figure 2.3 Schematic top view of MBE system.52

MBE is a term used to denote the epitaxial growth process in which directed neutral thermal atomic and molecular beams impinge on a heated crystalline substrate under ultra-high vacuum (UHV) at about 10-9 torr. It was invented in the late 1960s at Bell Telephone Laboratory by J Arthur53 and A Cho.54 MBE system was first designed to grow epitaxial thin film, and it is of interest to note that the growth of nanowires in MBE was initially speculated to be impossible due to the fact that the material is supplied in the form of atom, where no catalytic activity of Au to decompose the precursors to atoms is required as in MOCVD. Figure 2.3 presents a schematic top view of MBE system. MBE system was not used to grow nanowires until 2000 and was proven to possess great advantages on the growth mechanism investigations of nanowires. Since this thesis focus on using MBE growth technique to grow III-V nanowires, the MBE technique will be discussed with more details in the experimental techniques part.

2.2.2.3 CBE

CBE was first demonstrated by W.T. Tsang in 1984.55 This technique was then described as a hybrid of MOCVD and MBE, which exploited the advantages of both techniques. While group III elements were derived from the pyrolysis of the precursors on the surface, the group V elements were obtained from the decomposition of the precursors by bringing them in contact with heated Tantalum (Ta) or Molybdenum (Mo) at 950-1200 °C and thermally cracked before entering the growth chamber as molecules.56 CBE forms an important class of deposition technique for semiconductor grown systems, especially III-V semiconductor systems. This form of epitaxial growth is performed in an ultrahigh vacuum system. A schematic overview of CBE equipment is illustrated in Figure 2.4. The term CBE is often used interchangeably with metalorganic molecular beam epitaxy (MOMBE). The nomenclature does differentiate between the two (slightly different) processes, however. When used in the strictest sense, CBE refers to the technique in which both components are obtained from gaseous sources, while MOMBE refers to the technique in which the group III component is obtained from a gaseous source and the group V component from a solid source.

Figure 2.4 Schematic overview of CBE equipment.57

2.3 Nanowire growth mechanism

With the rapid and vast development of research in both the experimental area and theoretical area, many growth mechanisms have been put forward to understand the growth behaviors of different kinds of nanowires, and hence to better control the growth and structure of nanowires.

2.3.1 VLS The vapor-liquid-solid (VLS) mechanism is most widely used to grow nanoscale one-dimensional semiconductor crystals (nanowhisker or nanowires). The VLS mechanism was proposed by Wagner and Ellis back in 196458 as an explanation for silicon whisker, with the diameter ranging from 100 nm to 0.2 mm, grown from the gas phase in the presence of liquid gold droplets placed upon a silicon substrate. Taking the Au-catalyzed Si nanowhisker growth for example, the proposed VLS mechanism is schematically illustrated in Figure 2.5. Firstly, the growth temperature is increased to higher than the eutectic temperature of the alloy which consists of gold and silicon. After that, the liquid alloy, which is considered as the catalyst, becomes the preferential site for the collection of Si. With the continuous supply of

precursors, the catalyst becomes supersaturated and excessive Si starts to precipitate at the interface between the liquid alloy and the substrate, leading to the anisotropic crystal growth of silicon whisker. This whisker, with the catalyst at the top of the whisker, can keep excellent epitaxial relationship with the silicon substrate. As shown in this process, the VLS growth mechanism includes three phases, i.e., the vapor phase that refers to the SiCl4 vapor, the liquid phase that refers to the Au-Si droplet at the top and the solid phase that refers to the grown crystal between the interface of droplet and substrate.

Figure 2.5 Schematic illustration of Si whisker growth from the reaction of SiCl4 and H2 vapor phases.

Although few research work had been devoted to the III-V crystal growth via the VLS growth mechanism by Holonyak59 and Takahash60 soon after the VLS growth of silicon whiskers, it is during the past two decades that the growth of III-V nanowires induced by the VLS have been swiftly expanded and investigated. Many literatures have reported the gold catalysed growth of III-V nanowires both in MOCVD61 and MBE.62 To further verify this VLS growth mechanism, Harmand63 and Tchernycheva64 have used RHEED patterns in MBE to show the supercooling of liquid particle with a hysteresis. They claimed that even though when the growth temperature was below the eutectic point of catalyst and nanowire component, the droplet at the top of nanowire was still liquid during the growth, which demonstrates the VLS growth

mechanism. This in-situ monitoring is the big strength of MBE compared to other growth techniques. At the same time, the VLS mechanism has also been adopted to induce the group III-assisted growth of III-V nanowires. Recently, self-catalyzed, also call group III- assisted, III-V nanowire growth has been achieved by both MOCVD65 and MBE.66 This type of growth mechanism has always been linked to the existence of a SiOx film or patterned SiOx surface. Normally, there always exists a gallium particle at the top of nanowire after growth confirming the gallium-assisted VLS growth mode. Dubrovskii67 recently proposed a new model of VLS mechanism to understand the growth of nanowire catalyzed by a lower surface energy metal. A model of a nonspherical elongated droplet shape in wetting of nanowire sidewall facets is developed, and this theoretical prediction is compared to the experimental data on the Ga-catalyzed growth of GaAs nanowires by MBE. The experimental and model droplet geometries of this new VLS mode are presented in the Figure 2.6.

Figure 2.6 Experimental and model droplet geometries of this new VLS mode.67

However, in contrast to the Ga-assisted GaAs nanowire, In-assisted InAs nanowires terminate with flat surface without particle at the top, which can be clearly seen in the compared two images in Figure 2.7. This phenomenon has intrigued researchers to employ two kinds of mechanism to explain this. One is the traditional VLS growth mechanism, and the other is the VS growth mechanism which we will discuss later. Mandl68 has investigated the growth mechanism of In-assisted growth of InAs nanowires in MOCVD. They assumed that the indium droplet formed at the

nanowire top during the growth and was consumed during the termination of growth. Through the experiment design of 1) growth interruption of arsenic source while keeping the indium source provided, InAs nanowire growth continued after the arsenic source was introduced again, and 2) growth interruption of indium while keeping arsenic source for a while then reintroduced the indiums source, the nanowire length is significantly short than those without interruption. Then they came to the conclusion that indium plays the central role in nanowire growth and forms the liquid droplet at the top of the nanowire during the growth.

Figure 2.7 Self-catalysed nanowire growths of (a) InAs and (b) GaAs.69-70

2.3.2 VSS The vapor-solid-solid mechanism71 was proposed recently to explain the growth temperature of Ti-catalysed Si nanowires under the eutectic point of the catalyst and the nanowire component alloy. This mechanism is similar to VLS mechanism except that the catalyst is in solid phase during the nanowire growth. The VSS mechanism of III-V nanowires was early proposed in 2004 in the growth of GaAs nanowires from the CBE growth chamber.72 The facts that the calculated Ga concentration in the catalyst during the growth is lower than level required for a eutectic melt under the growth temperature and the defocused diffraction pattern of the whisker shows the crystal phase made Persson et al. to put forward this new nanowire growth mechanism.72 Since then, other researches73-76 have also reported the nanowire growth under the eutectic temperature and claimed the VSS growth mechanism in this new growth phenomenon. Taking the process of Si nanowires growth catalyzed by the aluminium

catalyst as an example, the schematic illustration of VSS mechanism is shown in the Figure 2.8. Silicon nanowires were grown at temperatures ranging from 430 °C to 490 °C which are far away from the Si-Al eutectic temperature of 577 °C as shown from the phase diagram. Silicon atoms diffuse into Al nano-clusters and alloy with Al in the solid state. With the oversupply of silicon, the supersaturated alloy starts to deposit silicon underneath the solid nanocluster and initiate the Si nanowire growth.

Figure 2.8 Schematic illustration of VSS mechanism of Si nanowires catalysed by Al.73

The experimental phenomenon that InAs nanowire growth stopped when the growth temperature is higher than the eutectic point was reported by Dick and colleagues77 in 2005, and they asserted that the InAs nanowire growth is instead assisted by a solid particle and the temperature range of InAs nanowire growth is limited by the melting of the Au-In alloy. However, another work done by Dayeh and colleagues78 in 2007 reaches different conclusions when investigating InAs nanowires grown in the similar growth conditions. They claimed that the input V/III ratios strongly influence the growth temperature window of nanowire growth and cessation. Although this VSS has been widely cited as the growth mechanism to explain the growth temperature below the eutectic point of the catalyst and nanowire component, it should be noted that: 1) the nanowire growth rate via the VSS mechanism is not significantly lower than that of the VLS mechanism, and we should pay attention to the fact that liquid-state diffusion should be much more favorable to the solid-state diffusion.79-80 2) This mechanism is normally proposed by the MOCVD and CBE growth technique which in-situ monitoring is not possible and/or deduced from the phase diagram. On the other hand, Harmand63 and Tchernycheva64 used RHEED patterns in the MBE growth chamber to prove the supercooling the liquid particle with

a hysteresis. This RHEED information may provide insight to the catalyst phase state during the growth. 3) Although there has been research81 shown that the melting point of gold particle makes no difference to the bulk material until the diameter shrinks to 10nm, Sutter82 demonstrated that the phase diagram of the Au-Ge nanoscale drop deviates significantly from that of the bulk alloy. Taking these facts into consideration, this new developed VSS growth mechanism still remains open for further debate.

2.3.3 PIN As we discussed above, although VLS is the most widely accepted nanowire growth mechanism, VSS mechanism has also been proposed and in some aspects proved with experimental results. To this end, Wacaser83 reviewed and expanded the fundamental processes of the VLS growth mechanism, and summarized a general, system-independent mechanism, i.e., Preferential Interface Nucleation (PIN). They defined the previously called liquid droplet as the “collector” and assumed that the specific phase of the collector is not important to the mechanism itself, but the stability of the phase, the collection ability of this phase, and its interaction with the growing crystal at the interface are the important parameters. It means that any three phase system where the Three Phase Boundary and/or the collector-crystal interface acts as a preferred nucleation site can produce unidirectional growth of nanowires when nucleation is supressed at the supply- crystal interface. They have proposed the PIN growth model as shown in the Figure 2.9.

Figure 2.9 General terms in PIN and illustration of PIN.83

2.3.4 VS As we mentioned before, the self-catalyzed InAs nanowires always end up with flat top facet, not showing particle at the nanowire top as the in case of self-catalyzed GaAs nanowires. This interesting morphology intrigues research workers to come up with another growth mechanism to explain this phenomenon. Herttenberger84 challenged the VLS mechanism in the self-induced InAs nanowires on Si(111) and favored the VS mode. Their main argument was the fast nucleation of InAs on Si after initiating the growth process, implying that there is not sufficient time for indium droplets to form. Very soon after the VS model was proposed, growth of InAs nanowires on bare Si(111)70 was investigated systematically in MBE. According to their experimental results, the length of nanowire grown with the interruption of In flux under As flux is similar to nanowire with continuous growth. They also showed that this case is applicable to GaAs nanowires. A growth model was proposed to understand this growth mechanism as in Figure 2.10.

Figure 2.10 InAs NWs image and schematic representation of the VS growth model.70

There is also another work85 that has studied the strain-induced InGaAs nanowires grown on Si(111) substrate. The InGaAs nanowires also end with flat top surface without particles at the top. Although the author only indicated that the mechanism of strain-induced InGaAs nanowire growth is probably different from that of the metal- catalysed, including self-catalysed growth and VLS growth, their results tend to suggest the VS growth mechanism for these nanowires.

2.4 Comparison of MBE and MOCVD

Among all those growth techniques, MOCVD and MBE are two most popular candidates for epitaxial nanowire growth. Our group takes advantages of both techniques to grow III-V semiconductor nanowires, and all the III-V nanowires in this PhD thesis were grown in the MBE growth chamber. As has been discussed above, there are many differences between these two techniques, and some of the differences are shown in the Table 2.1.

Table 2.1 Comparison of MOCVD and MBE

MOCVD MBE

Growth Higher pressure High vacuum pressure

In-situ No Yes monitoring

Thermodynamically Strongly non equilibrium, not Growth state favorable thermodynamically favourable

Growth Higher Lower temperature

Slower: precise layer by layer Growth rate Faster growth and guarantee abrupt growth

Nanowire ZB and WZ，mainly ZB WZ except in other growth direction structure

Nanowire High Tg, easy tapering High Tg, less tapering tapering

Catalyst Lower group-III Higher group-III composition

Effect of group-V on Almost no effect Group-V has great influence growth rate

Huge amount of toxic Toxicity gases need to be Less toxic processed

2.5 Growth and characteristics of III-V nanowires

2.5.1 Growth direction

Generally speaking, the III-V nanowires are nucleated in the <111>B directions for zinc-blende structured nanowires or in the <000 ̅> direction for wurtzite structured nanowires due to the lowest surface energy in {111}/{0001} planes.10 To date, some reports have demonstrated the growth of both free-standing III-V nanowires and 15, 86- planar NWs grown on the substrate with non-<111>B/<000 ̅> growth directions, 90 as summarized in Table 2.2. It is of interest to note that most of these non-

<111>B/<000 ̅ > nanowires generally have high structural quality. In this regard, controlling the nanowire growth direction may provide a new possibility to obtain III-V nanowires with controlled crystal phase and high structural quality. From a fundamental scientific point of view, nanowire growth directions may influence the optical, electrical and mechanical properties of nanowires.86 In this regard, to clarify the property and application of nanowires, the first issue is to determine the nanowire growth directions. Basically, two main techniques have been adopted to determine the growth direction of III-V nanowires, i.e., selected area electron diffraction (SAED) pattern in transmission electron microscopy (TEM), and scanning electron microscopy (SEM). In the following section, we will discuss both techniques in detail.

Table 2.2 Summary of the non-<111>B/<0001>B growth directions of III-V nanowires.

Catalyst type-Growth III-V Growth directions References technique

GaAs <110> Au-MBE 87-88

<110> Ni-CVD 89-90

<110> Au-CVD 91

<001> Au-MBE 88, 92-96

<112> Au-MOCVD 88, 97-98

99 <111>A Au-MOCVD

<311> and high index Au-CVD 91

<10 ̅0> Ni-CVD 89

<110> planar Au-MOCVD 100-105

GaP <112> Au-MOCVD 106

InAs <001> Au-CBE 107

<001> Au-MOCVD 108-109

<001> Au-MBE 110-111

<110> In-CVD 112

<110> Pd-MOCVD 113

<110> Au-MBE 114

<110> planar Au-PVD 115

<112> Au-MOCVD 109

<112> planar Au-MOCVD 116

InP <001> Au-MOCVD 15, 108, 117-118

<001> Au-EBL-MOCVD 119-120

<110> planar Au-MOCVD 118

121-125 <111>A SA-MOCVD

<112> Au-MOCVD 117

GaN <1 ̅00> Au-MOCVD 126-129

<1 ̅00> Au-CVD 130-132

<1 ̅00> Ni-MOCVD 127-128

<1 ̅00> Co-CVD 133

<10 ̅0> Ga-CVD 134

135 <10 ̅0> Ga2O3-CVD

<11 ̅0> Ni-MOCVD 126, 136-140

<11 ̅0> Fe-MOCVD 127

<11 ̅0> Ga-CVD 134

High-index Au-MOCVD 127

High-index Ni-MOCVD 141

InN <11 ̅0> Au-CVD 142-146

<11 ̅0> Au-MOCVD 147

<10 ̅0> Au-CVD 148-150

<0 ̅11> Au-CVD 151

2.5.1.1 Determination of growth direction in TEM

The SAED pattern in TEM has been the mostly widely used technique in determining the nanowire growth directions. By correlating the bright-field TEM image of nanowire with the corresponding diffraction spot in the SAED pattern, the nanowire growth direction can be determined. In most studies,90-91 only one SAED pattern was collected from nanowire to determine its growth direction. However, one SAED pattern is not sufficient to exclusively determine the nanowire growth direction considering the possible fact that the nanowire could not be flat on the carbon/supporting film. In this situation, the viewing direction/zone axis is not perpendicular to the nanowire axis. Therefore, the determined direction based on the observed SAED pattern is the projection of nanowire growth direction, rather than the actual axis of nanowire. In order to explicitly determine the nanowire growth direction, nanowire needs to be rotated along its axial direction to different zone axes with the corresponding SAED patterns. To determine if nanowire was rotated along its axial direction, we need to check if both diffraction patterns indicate the same growth direction. A typical example using three set of diffraction patterns to determine nanowire growth direction are shown in Figure 2.11.

Figure 2.11a shows a bright-field TEM image of typical nanowire. To accurately determine the growth direction of this nanowire, the nanowire was tilted along its axis to three different zone axes. Figure 2.11b, d and f are three set of bright-field TEM images taken from the nanowire top, and Figure 2.11c, e and g are the corresponding diffraction patterns, taken along the [1 ̅0], [1 ̅ ̅] and [ ̅1 ̅] zone axes, respectively. Correlating the bright-field TEM images and with the corresponding SAED patterns, it can be noted that all the SAED patterns indicate that nanowire is grown along the [ ̅ ̅0] direction. It should be noted that orientations of the bright-field TEM images of nanowires and their corresponding SAED patterns have been carefully aligned. In this case, the nanowire growth direction can be unambiguously determined. Meanwhile, it can be noted that by tilting nanowire to a specific zone axis, the sharp interface between the catalyst and nanowire can be obtained, and in hence the crystallographic plane of catalyst/nanowire can be determined according

to the corresponding SAED pattern, as can be seen from Figure 2.11f, which will be useful to clarify the growth mechanism of the nanowires.113

2.5.1.2 Determination of growth direction in SEM

In addition to the SAED patterns in TEM, SEM has also been adopted to determine the nanowire growth directions. Since there is orientation relationship between the epitaxial nanowire and the substrate, it is possible and reliable to use the tilting process and tilting angle to exclusively determine the growth direction of the inclined nanowires grown on the substrate based on the crystallography.

Figure 2.12 demonstrates the tilting process using both the schematics and the corresponding SEM images to determine the growth direction of inclined nanowires.

Figure 2.12a shows a schematic and an SEM image of the GaAs {111}B substrate (cleavage planes being {1 ̅0} planes) without tilting, in which an inclined nanowire with the projection direction of <112> can be seen. Then, the substrate was tilted along the [1 ̅0] axis, and a typical image during the tilting process captured at 20° tilting angle is shown in the Figure 2.12b. It can be noted that the nanowire projection length becomes shorter. With further tilting, the axial direction of the inclined nanowire is parallel to the electron beam, with tilting angle of ~35° in this case, as shown in the Figure 2.12c. Based on the tilting angle and general crystallography geometry, the growth direction of the inclined nanowire can be exclusively determined as < ̅ ̅0> for this inclined nanowire. Meanwhile, when the nanowire axis has been tilted to be parallel to the electron beam, the side-wall facets of the inclined nanowire can be clearly observed. Using the substrate cleavage planes as reference, the side-wall planes can be determined. Also, the microtome technique to cut across the nanowire has also been employed to determine the cross-section of nanowire.

In addition, we can use the naturally cleavage side-view SEM investigations to further confirm the nanowire growth directions. From the side-view SEM image, the angle between the projection of nanowire and the substrate normal can be measured, and then the growth direction can be confirmed based on crystallography. However, it should be noted that since there are equivalent directions of one direction family, and side-view SEM images shows the projection of nanowires on the cleavage plane, there are both real angles and projected angles of one direction family. Therefore, crystallography attentions need to be paid when using the side-view SEM investigations to determine nanowire growth direction. It should be noted that the SEM technique using the tilting process to determine the nanowire growth direction is a universal technique which is suitable for the epitaxial grown nanowires since the determined growth direction of epitaxial nanowire is based on the orientation of the substrate.

2.5.2 Nanowire growth rate In the Au-assisted VLS growth of III-V nanowires, it has been commonly accepted that the Au particle only absorbs group-III atoms from the vapor and transport them to the growth interface. The group-V atoms do not alloy with Au particle to go through the catalyst, but incorporate into the nanowire through the triple phase line. This is due to the extremely different solubilities of group-III and group-V elements in Au within the temperature range of the III-V nanowire growth.152-153 Many theoretical154-155 and experimental research156 works have been tried to understand the effects that influence the nanowire growth rate. According to these works, apart from the direct impingement of growth species on the catalyst, they proposed that the atoms diffusion from sidewall and substrate also contribute to the nanowire growth. Very recently another work157 took the secondary adsorption (defined as the process that atoms that desorb from the substrate may re-adsorb to the nanowire sidewall facets) into account for the nanowire growth rate, and even assumed that this was a very important aspect. If we take the Au-assisted GaAs nanowire growth for example to understand this process, the schematic process demonstrating the possible ways for Ga atoms contribute to the nanowire growth can be illustrated in the Figure 2.13. Many growth parameters can affect the growth rate of nanowires, and it is even more complicated and interdependent in different growth systems. In the following discussions, we will mainly focus on the effects of growth parameters on the Au- assisted nanowire growth which is the method that nanowires are prepared in this thesis.

2.5.2.1 Diameter effect

Nanowire growth catalysed by Au can be grouped into two categories,158-159 the early kinetic model160 and the more recent diffusion model.161 The kinetic model predicts that smaller nanowires grow more slowly than larger ones due to the decrease of supersaturation in smaller nanoparticles as a result of the larger surface tension (Gibbs-Thomson effect), and nanowire growth cedes below a given critical nanowire dimension. Conversely, the diffusion model predicts that smaller nanowires grow faster than larger ones due to the larger sidewall collection area relative to the nanowire volume. The switchover between the two regimes is possible by tuning the V/III flux ratio.162-163 As shown in the Figure 2.14, the growth rate dependence on nanoparticle diameter shows opposite tendency under different V/III ratios, and this is due to the changing of growth mechanisms. Through theoretical calculation, Dubrovskii154 predicted the transition from decreasing to increasing length-diameter dependence at a certain critical diameter in the very wide-range sized Au-catalyzed nanowire growth in MBE. And they attributed this change to the transition from diffusion-induced growth to the traditional adsorption-induced VLS growth.

Figure 2.14 Nanowire growth rate versus Au nanoparticle diameter.163

The diameter influence on the growth rate has been systematically presented by K. Dick.164 Since the group-III element is incorporated into nanowire via the nanowire- catalyst interface, which scales with the square of the nanowire radius, R2 1) If the primary source of group-III that contributes to the nanowire growth comes from the alloy particle with volume scales with R3, this would lead to a R dependence (R3/ R2) of growth rate. 2) If the primary source of group-III is residual material on the nanowire side facets which scale with R, a R-1 dependence (R/ R2) of growth rate would be expected. 3) If the primary source of group-III is residual material on the alloy particle surface which scales with R2, no diameter dependence of growth rate (R2/ R2) would be expected.

2.5.2.2 Nanowire density effect

Normally, the growth rate can be influenced by the density of nanowires, namely higher density will result in lower growth rate33 due to the competition of growth materials between nanowires. However, when investigating the density effect on the nanowire growth rate, Borgstrom165 found the synergetic effect, that is, in certain nanowire spacing region, decreasing wire-to-wire distance could lead to increased growth rate. In this regard, he proposed the nanowire growth model in three regimes,

competitive regime, synergetic regime, and independent regime, as illustrated in the following Figure 2.15.

Figure 2.15 Schematic nanowire growth model: (a) Competitive regime, (b) Synergetic regime, (c) Independent regime.165

2.5.2.3 V/III ratio effect

Since group-III and group-V atoms incorporate into the growth front in different ways, it is straightforward to understand that the V/III ratio should play an important role in nanowire growth rate. However, according to the reported literatures, the V/III effect on the nanowire axial growth rate has proven to be different in different nanowire systems. In the MOCVD growth chamber, it is the normal case that the growth rate is mainly limited by the diffusion of less abundant species, namely group-III, to the growth interface.166 As can be noted in Figure 2.16, when V/III ratio is increased at fixed group III flow rate, there is a very marginal increase in axial growth rate.97 Beyond a V/III ratio of 46, the axial growth rate drops steeply. This was attributed to the non-

<111>B growth direction, the kinks of <111>B nanowires and the diminish of III diffusion length with increasing V/III ratio.167 When the absolute group III flux rate is increased, the nanowire growth rate increased rapidly. At low V/III ratio, an irregular layer covers the surface, and the gold nanoparticle failed to induce nanowire growth. As V/III ratio is increased, there are sufficient III to promote nanowire growth. When

the V/III ratio is too high, Au-assisted structures show island-like rather than nanowire-like morphology.

Figure 2.16 V/III ratio effect to the growth and morphology of nanowires.97

In the MBE growth chamber, there are very few investigations on the effect of V/III ratio on the nanowire growth. Few work168 reported the Ga-assisted nanowire growth in MBE. They found that axial nanowire growth is governed by the As pressure and has nothing to do with arriving Ga flux rate as can been seen from the Figure 2.17.

69 Figure 2.17 Growth rate of GaAs nanowire as a function of As4 pressure.

Recently, there is an interesting work169 in MBE growth as shown in Figure 2.18, which suggested that the increase of arsenic pressure can greatly promote nanowire growth rate while the increase of indium pressure could first increase the axial growth rate and then decrease growth rate.

Figure 2.18 InAs nanowire growth rate as a function of In and As flux.169

The type of arsenic species could also influence the growth rate of GaAs nanowire growth by MBE according to Sartel’s experimental results.170 Their results shown in

Figure 2.19 suggested that in a wide range of growth temperature, As4 leads to growth rate twice faster than As2, and they came to the conclusion that the diffusion length of Ga adtoms along the nanowire sidewalls is smaller under As2 flux as compared to that under As4 flux. This result made them propose that As2 flux is favorable to the formation of radial heterostructures on nanowire, whereas As4 flux is preferable to maintain pure axial nanowire growth.

Figure 2.19 Effect of Arsenic species on the growth rate.170

2.5.2.4 Growth temperature

As many research works171-173 on Au-assisted nanowire growth proved, the nanowire growth rate first increases with growth temperature, as a consequence to

the thermal activation of growth species and also the thermal decomposition of precursor in MOCVD. Then the growth rate will reach a maximum growth rate at certain growth temperature. After that, axial growth rate would decreases with increasing temperature due to the onset of significantly radial and substrate growth. The results are presented in Figure 2.20. Also, some experiment results demonstrate that the input V/III ratios strongly influence the temperature dependence of nanowire growth and cessation, i.e., the nanowire growth window.

Figure 2.20 Growth rate as a function of growth temperature.64, 78

2.5.3 Nanowire facet It has been well documented that, when the dimension of nanowires diminishes to the nanoscale, the increased surface to volume ratio can have significant effect on altering the nanowire surface to the stable state. In this regard, the grown nanowires always try to minimise their surface energy in order to reduce the total free energy of the system. Mostly, the III-V nanowires are nucleated in the <111>B directions for zinc-blende structured nanowires or in the <000 ̅> direction for wurtzite structured nanowires. As many experiments and theories stated, zinc-blende structured <111>B III-V nanowires tend to adopt either {110} or {112} facets, and even the mixed two 174-176 types of facets. While the wurtzite structured <000 ̅> III-V nanowires always take the {1 ̅00} or {11 ̅0} facets.177 In the gas-sourced MBE growth of Au-assisted GaAs nanowires, there has been literature178-180 reported that the transformation of sidewall facets occurring from six { ̅100} facets at the tip to six { ̅110} facets at the base of nanowire as presented in the Figure 2.21. They contributed this evolve toward the { ̅110} sidewall facets at the

nanowire base to the radial overgrowth, and assumed that this transformation is the result of the instability of sidewall facets. Another work94 suggested that growth parameters like Ga or As flux rate could change the facets of nanowires. As stated in their experiments, under lower As4 impingement flux, the sidewall facets of nanowires were { ̅ 100} family. While they increased As4 flux with or without increasing the Ga flux, the sidewall facets of nanowires change to { ̅110} family. Based on these observed experimental results, they suggested that this nanowire sidewall facets dependence on As4, although not fully understood at this stage, has 181 something to do with the surface energy variation due to the variation of As4 flux.

Figure 2.21 (a-c) The growth parameter changes nanowire facets; (d) Top view SEM images revealing the sidewall facet transition between the tip and base.94, 178

As we discussed earlier, for those inclined nanowires, since the electron beam is not parallel to the growth direction of the as-grown nanowire, the sidewall facets of these inclined nanowires are not straightforward shown. However, when the nanowire axis has been tilted to be parallel to the electron beam, the sidewall facets of the inclined nanowire can be clearly observed. Meanwhile, using the substrate cleavage planes as reference (for example, the cleavage planes of zinc-blende structured III-V substrate are {110} planes), the nanowire sidewall planes can be determined. In addition to the SEM tilting process to determine the nanowire facets, the microtome technique to cut across the nanowire has also been employed to

determine the cross-section of nanowire. The summary of reported cross-section facets of non-<111>B/<0001>B III-V nanowires is presented in Figure 2.22.

Figure 2.22 The summary of reported cross-section facets of non-<111>B/<000 ̅ > III-V nanowires.

2.6 III-V nanowires crystal phase and structural quality

Although the bulk III-V materials have the zinc-blende structure, III-V nanowire can adopt the wurtzite structure. In this regard, it is easy to introduce uncontrolled planar defects and the intermix of wurtzite structure and zinc-blende structure due to the small energetic differences of the stacking sequences in ZB and WZ structures along their <111>B/<000 ̅> directions. In order to realize and deliver the superior properties of one dimensional III-V nanowires in advanced nanoscale devices, minimizing the

lattice defects in nanowires has been extremely important, and tremendous efforts have been devoted to control the structural quality by tuning the nanowire growth parameters. For example, V/III ratio has been employed to successfully tune the structure of III-V nanowires between wurtzite and zinc-blende structures.16, 182 Also, it has been demonstrated that thin nanowires induced by small catalysts tend to have much better wurtzite structure quality.7 Meanwhile, a two-temperature process with a high nucleation temperature and a low nanowire growth temperature has been adapted to achieve defect-free zinc-blende structure.23 Furthermore, non-gold catalyst has been proved to be able to induce defect-free zinc-blende structured InAs nanowires through the formation of the unusual {113} catalyst/nanowire interface.113 In addition, it has been reported that the non-<111>/<0001> orientated nanowires tend to have less planar defects along the nanowire. Therefore, controlling the nanowire growth direction may provide a new insight into the improvement of nanowire structural quality. Furthermore, the substrate orientations may also be adopted to tune the structural quality of Au-catalyzed epitaxial GaAs nanowires in MBE due to different diffusion length of group-III atoms on different substrate.

2.7 Summary

This chapter has provided the background for the following chapters. A literature review on the III-V semiconductor nanowire growth techniques, growth mechanisms, growth direction, axial growth rate, sidewall facets information, crystal phase and structural quality of nanowires is presented. Based on these detailed review, it can be noted that although extensive research has been conducted to understand the growth of III-V nanowires grown by MOCVD and GaAs nanowires grown by MBE, little attention has been paid on the controlled growth of InAs nanowires.

2.8 References

1. Dayeh, S. A.; Yu, E. T.; Wang, D. Excess Indium and Substrate Effects on the Growth of Inas Nanowires. Small 2007, 3, 1683-1687. 2. Vurgaftman, I.; Meyer, J. R.; Ram-Mohan, L. R. Band Parameters for Iii-V Compound Semiconductors and Their Alloys. J. Appl. Phys. 2001, 89, 5815-5875. 3. Blakemore, J. S. Semiconducting and Other Major Properties of Gallium Arsenide. J. Appl. Phys. 1982, 53, R123-R181. 4. Wu, Z. H.; Mei, X. Y.; Kim, D.; Blumin, M.; Ruda, H. E. Growth of Au-Catalyzed Ordered Gaas Nanowire Arrays by Molecular-Beam Epitaxy. Appl. Phys. Lett. 2002, 81, 5177-5179. 5. Dubrovskii, V. G.; Cirlin, G. E.; Soshnikov, I. P.; Tonkikh, A. A.; Sibirev, N. V.; Samsonenko, Y. B.; Ustinov, V. M. Diffusion-Induced Growth of Gaas Nanowhiskers During Molecular Beam Epitaxy: Theory and Experiment. Phys. Rev. B 2005, 71, 205325. 6. Harmand, J. C.; Patriarche, G.; Pere-Laperne, N.; Merat-Combes, M. N.; Travers, L.; Glas, F. Analysis of Vapor-Liquid-Solid Mechanism in Au-Assisted Gaas Nanowire Growth. Appl. Phys. Lett. 2005, 87, 203101. 7. Zhang, Z.; Lu, Z. Y.; Chen, P. P.; Xu, H. Y.; Guo, Y. N.; Liao, Z. M.; Shi, S. X.; Lu, W.; Zou, J. Quality of Epitaxial Inas Nanowires Controlled by Catalyst Size in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 103, 073109. 8. Piccin, M.; Bais, G.; Grillo, V.; Jabeen, F.; De Franceschi, S.; Carlino, E.; Lazzarino, M.; Romanato, F.; Businaro, L.; Rubini, S., et al. Growth by Molecular Beam Epitaxy and Electrical Characterization of Gaas Nanowires. Physica E-Low- Dimensional Systems & Nanostructures 2007, 37, 134-137. 9. Plante, M. C.; LaPierre, R. R. Growth Mechanisms of Gaas Nanowires by Gas Source Molecular Beam Epitaxy. J. Cryst. Growth 2006, 286, 394-399. 10. Cahn, J. W.; Hanneman, R. E. (111) Surface Tensions of Iii–V Compounds and Their Relationship to Spontaneous Bending of Thin Crystals. Surf. Sci. 1964, 1, 387- 398. 11. Johansson, J.; Bolinsson, J.; Ek, M.; Caroff, P.; Dick, K. A. Combinatorial Approaches to Understanding Polytypism in Iii-V Nanowires. Acs Nano 2012, 6, 6142-6149.

12. Dick, K. A.; Thelander, C.; Samuelson, L.; Caroff, P. Crystal Phase Engineering in Single Inas Nanowires. Nano Lett. 2010, 10, 3494-3499. 13. Dheeraj, D. L.; Munshi, A. M.; Scheffler, M.; van Helvoort, A. T. J.; Weman, H.; Fimland, B. O. Controlling Crystal Phases in Gaas Nanowires Grown by Au-Assisted Molecular Beam Epitaxy. Nanotechnology 2013, 24, 015601. 14. Dubrovskii, V. G.; Sibirev, N. V.; Cirlin, G. E.; Bouravleuv, A. D.; Samsonenko, Y. B.; Dheeraj, D. L.; Zhou, H. L.; Sartel, C.; Harmand, J. C.; Patriarche, G., et al. Role of Nonlinear Effects in Nanowire Growth and Crystal Phase. Phys. Rev. B 2009, 80, 205305. 15. Krishnamachari, U.; Borgstrom, M.; Ohlsson, B. J.; Panev, N.; Samuelson, L.; Seifert, W.; Larsson, M. W.; Wallenberg, L. R. Defect-Free Inp Nanowires Grown in [001] Direction on Inp(001). Appl. Phys. Lett. 2004, 85, 2077-2079. 16. Joyce, H. J.; Wong-Leung, J.; Gao, Q.; Tan, H. H.; Jagadish, C. Phase Perfection in Zinc Blende and Wurtzite Iii-V Nanowires Using Basic Growth Parameters. Nano Lett. 2010, 10, 908-915. 17. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Jiang, N.; Tan, H. H.; Jagadish, C.; Zou, J. High-Density, Defect-Free, and Taper-Restrained Epitaxial Gaas Nanowires Induced from Annealed Au Thin Films. Cryst. Growth Des. 2012, 12, 2018-2022. 18. Shtrikman, H.; Popovitz-Biro, R.; Kretinin, A.; Houben, L.; Heiblum, M.; Bukala, M.; Galicka, M.; Buczko, R.; Kacman, P. Method for Suppression of Stacking Faults in Wurtzite Iii-V Nanowires. Nano Lett. 2009, 9, 1506-1510. 19. Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Kim, Y.; Fickenscher, M. A.; Perera, S.; Hoang, T. B.; Smith, L. M.; Jackson, H. E., et al. Unexpected Benefits of Rapid Growth Rate for Iii-V Nanowires. Nano Lett. 2009, 9, 695-701. 20. Ikonic, Z.; Srivastava, G. P.; Inkson, J. C. Electronic Properties of Twin Boundaries and Twinning Superlattices in Diamond-Type and Zinc-Blende-Type Semiconductors. Phys. Rev. B 1993, 48, 17181-17193. 21. Kretinin, A. V.; Popovitz-Biro, R.; Mahalu, D.; Shtrikman, H. Multimode Fabry- Perot Conductance Oscillations in Suspended Stacking-Faults-Free Inas Nanowires. Nano Lett. 2010, 10, 3439-3445. 22. Schroer, M. D.; Petta, J. R. Correlating the Nanostructure and Electronic Properties of Inas Nanowires. Nano Lett. 2010, 10, 1618-1622.

23. Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Kim, Y.; Zhang, X.; Guo, Y. N.; Zou, J. Twin-Free Uniform Epitaxial Gaas Nanowires Grown by a Two-Temperature Process. Nano Lett. 2007, 7, 921-926. 24. Woo, R. L.; Xiao, R.; Kobayashi, Y.; Gao, L.; Goel, N.; Hudait, M. K.; Mallouk, T. E.; Hicks, R. F. Effect of Twinning on the Photoluminescence and Photoelectrochemical Properties of Indium Phosphide Nanowires Grown on Silicon (111). Nano Lett. 2008, 8, 4664-4669. 25. Parkinson, P.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Zhang, X.; Zou, J.; Jagadish, C.; Herz, L. M.; Johnston, M. B. Carrier Lifetime and Mobility Enhancement in Nearly Defect-Free Core-Shell Nanowires Measured Using Time-Resolved Terahertz Spectroscopy. Nano Lett. 2009, 9, 3349-3353. 26. Dayeh, S. A.; Aplin, D. P. R.; Zhou, X.; Yu, P. K. L.; Yu, E. T.; Wang, D. High Electron Mobility Inas Nanowire Field-Effect Transistors. Small 2007, 3, 326-332. 27. Wright, S. L.; Marks, R. F.; Tiwari, S.; Jackson, T. N.; Baratte, H. In Situ Contacts to Gaas Based on Inas. Appl. Phys. Lett. 1986, 49, 1545-1547. 28. Milnes, A. G.; Polyakov, A. Y. Indium Arsenide: A Semiconductor for High Speed and Electro-Optical Devices. Mat. Sci. Eng. B 1993, 18, 237-259. 29. Thelander, C.; Martensson, T.; Bjork, M. T.; Ohlsson, B. J.; Larsson, M. W.; Wallenberg, L. R.; Samuelson, L. Single-Electron Transistors in Heterostructure Nanowires. Appl. Phys. Lett. 2003, 83, 2052-2054. 30. Doh, Y. J.; van Dam, J. A.; Roest, A. L.; Bakkers, E.; Kouwenhoven, L. P.; De Franceschi, S. Tunable Supercurrent through Semiconductor Nanowires. Science 2005, 309, 272-275. 31. Bjork, M. T.; Ohlsson, B. J.; Thelander, C.; Persson, A. I.; Deppert, K.; Wallenberg, L. R.; Samuelson, L. Nanowire Resonant Tunneling Diodes. Appl. Phys. Lett. 2002, 81, 4458-4460. 32. Zhou, X.; Dayeh, S. A.; Aplin, D.; Wang, D.; Yu, E. T. Direct Observation of Ballistic and Drift Carrier Transport Regimes in Inas Nanowires. Appl. Phys. Lett. 2006, 89, 053113. 33. Kim, Y.; Joyce, H. J.; Gao, O.; Tan, H. H.; Jagadish, C.; Paladugu, M.; Zou, J.; Suvorova, A. A. Influence of Nanowire Density on the Shape and Optical Properties of Ternary Ingaas Nanowires. Nano Lett. 2006, 6, 599-604.

34. Paladugu, M.; Zou, J.; Guo, Y. N.; Auchterlonie, G.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Kim, Y. Novel Growth Phenomena Observed in Axial Inas/Gaas Nanowire Heterostructures. Small 2007, 3, 1873-1877. 35. Zou, J.; Paladugu, M.; Wang, H.; Auchterlonie, G. J.; Guo, Y. N.; Kim, Y.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C. Growth Mechanism of Truncated Triangular Iii-V Nanowires. Small 2007, 3, 389-393. 36. Guo, Y. N.; Xu, H. Y.; Auchterlonie, G.; Burgess, T.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Shu, H. B.; Chen, X. S., et al. Phase Separation Induced by Au Catalysts in Ternary Ingaas Nanowires. Nano Lett. 2013, 13, 643-650. 37. Dick, K. A.; Deppert, K.; Larsson, M. W.; Martensson, T.; Seifert, W.; Wallenberg, L. R.; Samuelson, L. Synthesis of Branched 'Nanotrees' by Controlled Seeding of Multiple Branching Events. Nat. Mater. 2004, 3, 380-384. 38. Algra, R. E.; Verheijen, M. A.; Borgstrom, M. T.; Feiner, L.-F.; Immink, G.; van Enckevort, W. J. P.; Vlieg, E.; Bakkers, E. P. A. M. Twinning Superlattices in Indium Phosphide Nanowires. Nature 2008, 456, 369-372. 39. Jensen, L. E.; Bjork, M. T.; Jeppesen, S.; Persson, A. I.; Ohlsson, B. J.; Samuelson, L. Role of Surface Diffusion in Chemical Beam Epitaxy of Inas Nanowires. Nano Lett. 2004, 4, 1961-1964. 40. Persson, A. I.; Froberg, L. E.; Jeppesen, S.; Bjork, M. T.; Samuelson, L. Surface Diffusion Effects on Growth of Nanowires by Chemical Beam Epitaxy. J. Appl. Phys. 2007, 101, 034313. 41. Lugani, L.; Ercolani, D.; Sorba, L.; Sibirev, N. V.; Timofeeva, M. A.; Dubrovskii, V. G. Modeling of Inas-Insb Nanowires Grown by Au-Assisted Chemical Beam Epitaxy. Nanotechnology 2012, 23, 095602. 42. Samuelson, L. Self-Forming Nanoscale Devices. Materials Today 2003, 6, 22-31. 43. Dick, K. A. A Review of Nanowire Growth Promoted by Alloys and Non-Alloying Elements with Emphasis on Au-Assisted Iii-V Nanowires. Prog. Cryst. Growth Ch 2008, 54, 138-173. 44. Arthur, J. R. Molecular Beam Epitaxy. Surf. Sci. 2002, 500, 189-217. 45. Hwang, N. M.; Cheong, W. S.; Yoon, D. Y.; Kim, D. Y. Growth of Silicon Nanowires by Chemical Vapor Deposition: Approach by Charged Cluster Model. J. Cryst. Growth 2000, 218, 33-39.

46. Lu, Z. Y.; Zhang, Z.; Chen, P. P.; Shi, S. X.; Yao, L. C.; Zhou, C.; Zhou, X. H.; Zou, J.; Lu, W. Bismuth-Induced Phase Control of Gaas Nanowires Grown by Molecular Beam Epitaxy. Appl. Phys. Lett. 2014, 105, 162102. 47. Duan, X. F.; Lieber, C. M. Laser-Assisted Catalytic Growth of Single Crystal Gan Nanowires. J. Am. Chem. Soc. 2000, 122, 188-189. 48. Lee, S. T.; Zhang, Y. F.; Wang, N.; Tang, Y. H.; Bello, I.; Lee, C. S.; Chung, Y. W. Semiconductor Nanowires from Oxides. J. Mater. Res. 1999, 14, 4503-4507. 49. Joyce, H. J. Growth and Characterisation of Iii-V Semiconductor Nanowires for Optoelectronic Device Applications. Doctor of Philosophy thesis, The Australian National University 2009. 50. Manasevi, H. M. Single-Crystal Gallium Arsenide on Insulating Substrates. Appl. Phys. Lett. 1968, 12, 156-&. 51. Manasevi, H.; Simpon, W. The Use of Metal-Organics in the Preparation of Semiconductor Materials-I . Epitaxial Gallium‐ V Compounds. J. Electrochem. Soc. 1969, 116, 1725-1732. 52. Cho, A. Y.; Arthur, J. R. Molecular Beam Epitaxy. Prog. Solid State Ch. 1975, 10, Part 3, 157-191. 53. Arthur, J. R. Interaction of Ga and As2 Molecular Beams with Gaas Surfaces. J. Appl. Phys. 1968, 39, 4032-4034. 54. Cho, A. Y. Morphology of Epitaxial Growth of Gaas by a Molecular Beam Method: The Observation of Surface Structures J. Appl. Phys. 1970, 41, 2780-&. 55. Tsang, W. T. Chemical Beam Epitaxy of Inp and Gaas. Appl. Phys. Lett. 1984, 45, 1234-1236. 56. Tsang, W. T. From Chemical Vapor Epitaxy to Chemical Beam Epitaxy. J. Cryst. Growth 1989, 95, 121-131. 57. Rongen, R. T. H.; Leys, M. R.; Vanhall, P. J.; Vanes, C. M.; Vonk, H.; Wolter, J. H. Investigations on Indium Phosphide Grown by Chemical Beam Epitaxy. J. Electron. Mater. 1995, 24, 1391-1398. 58. Wagner, R. S.; Ellis, W. C. Vapor-Liquid-Solid Mechanism of Single Crystal Growth ( New Method Growth Catalyst from Impurity Whisker Epitaxial + Large Crystal Si E ). Appl. Phys. Lett. 1964, 4, 89-90. 59. Holonyak, N.; Wolfe, C. M.; Moore, J. S. Vapor-Liquid-Solid Growth of Gallium Phosphide (Whiskers 900 Degree C E). Appl. Phys. Lett. 1965, 6, 64-&.

60. Takahash, K.; Moriizum, T. Growth of Inas Whiskers in Wurtzite Structure. Jpn. J. Appl. Phys. 1966, 5, 657-662. 61. Kim, Y.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C.; Paladugu, M.; Zou, J., Iii-V Nanowires for Optoelectronies. In Optoeletronic Materials and Devices, Pts 1 and 2, Lee, Y. H. K. F. L. Y., Ed. 2006; Vol. 6352, pp U1115-U1127. 62. Lee, H. G.; Jeon, H. C.; Kang, T. W.; Kim, T. W. Gallium Arsenide Crystalline Nanorods Grown by Molecular-Beam Epitaxy. Appl. Phys. Lett. 2001, 78, 3319-3321. 63. Harmand, J. C.; Tchernycheva, M.; Patriarche, G.; Travers, L.; Glas, F.; Cirlin, G. Gaas Nanowires Formed by Au-Assisted Molecular Beam Epitaxy: Effect of Growth Temperature. J. Cryst. Growth 2007, 301–302, 853-856. 64. Tchernycheva, M.; Travers, L.; Patriarche, G.; Glas, F.; Harmand, J.-C.; Cirlin, G. E.; Dubrovskii, V. G. Au-Assisted Molecular Beam Epitaxy of Inas Nanowires: Growth and Theoretical Analysis. J. Appl. Phys. 2007, 102, 094313. 65. Mandl, B.; Stangl, J.; Martensson, T.; Mikkelsen, A.; Eriksson, J.; Karlsson, L. S.; Bauer, G.; Samuelson, L.; Seifert, W. Au-Free Epitaxial Growth of Inas Nanowires. Nano Lett. 2006, 6, 1817-1821. 66. Morral, A. F. i.; Colombo, C.; Abstreiter, G.; Arbiol, J.; Morante, J. R. Nucleation Mechanism of Gallium-Assisted Molecular Beam Epitaxy Growth of Gallium Arsenide Nanowires. Appl. Phys. Lett. 2008, 92, 063112. 67. Dubrovskii, V. G.; Cirlin, G. E.; Sibirev, N. V.; Jabeen, F.; Harmand, J. C.; Werner, P. New Mode of Vapor-Liquid-Solid Nanowire Growth. Nano Lett. 2011, 11, 1247-1253. 68. Mandl, B.; Stangl, J.; Hilner, E.; Zakharov, A. A.; Hillerich, K.; Dey, A. W.; Samuelson, L.; Bauer, G.; Deppert, K.; Mikkelsen, A. Growth Mechanism of Self- Catalyzed Group Iii-V Nanowires. Nano Lett. 2010, 10, 4443-4449. 69. Colombo, C.; Spirkoska, D.; Frimmer, M.; Abstreiter, G.; Fontcuberta i Morral, A. Ga-Assisted Catalyst-Free Growth Mechanism of Gaas Nanowires by Molecular Beam Epitaxy. Phys. Rev. B 2008, 77, 155326. 70. Dimakis, E.; Laehnemann, J.; Jahn, U.; Breuer, S.; Hilse, M.; Geelhaar, L.; Riechert, H. Self-Assisted Nucleation and Vapor-Solid Growth of in as Nanowires on Bare Si(111). Cryst. Growth Des. 2011, 11, 4001-4008. 71. Kamins, T. I.; Williams, R. S.; Basile, D. P.; Hesjedal, T.; Harris, J. S. Ti- Catalyzed Si Nanowires by Chemical Vapor Deposition: Microscopy and Growth Mechanisms. J. Appl. Phys. 2001, 89, 1008-1016.

72. Persson, A. I.; Larsson, M. W.; Stenstrom, S.; Ohlsson, B. J.; Samuelson, L.; Wallenberg, L. R. Solid-Phase Diffusion Mechanism for Gaas Nanowire Growth. Nat. Mater. 2004, 3, 677-681. 73. Wang, Y.; Schmidt, V.; Senz, S.; Goesele, U. Epitaxial Growth of Silicon Nanowires Using an Aluminium Catalyst. Nature Nanotechnology 2006, 1, 186-189. 74. Kodambaka, S.; Tersoff, J.; Reuter, M. C.; Ross, F. M. Germanium Nanowire Growth Below the Eutectic Temperature. Science 2007, 316, 729-732. 75. Lensch-Falk, J. L.; Hemesath, E. R.; Perea, D. E.; Lauhon, L. J. Alternative Catalysts for Vss Growth of Silicon and Germanium Nanowires. Journal of Materials Chemistry 2009, 19, 849-857. 76. Ross, F. M. Controlling Nanowire Structures through Real Time Growth Studies. Rep. Prog. Phys. 2010, 73, 114501. 77. Dick, K. A.; Deppert, K.; Martensson, T.; Mandl, B.; Samuelson, L.; Seifert, W. Failure of the Vapor-Liquid-Solid Mechanism in Au-Assisted Movpe Growth of Inas Nanowires. Nano Lett. 2005, 5, 761-764. 78. Dayeh, S. A.; Yu, E. T.; Wang, D. Iii-V Nanowire Growth Mechanism: V/Iii Ratio and Temperature Effects. Nano Lett. 2007, 7, 2486-2490. 79. Ghaisas, S. V.; Dassarma, S. Surface-Diffusion Length under Kinetic Growth- Condistions. Phys. Rev. B 1992, 46, 7308-7311. 80. Mohammad, S. N. General Theoretical Model for the Vapor-Phase Growth and Growth Rate of Semiconductor Nanowires. J. Vac. Sci. Technol. B 2010, 28, 329- 352. 81. Buffat, P.; Borel, J. P. Size Effect on the Melting Temperature of Gold Particles. Phys. Rev. A 1976, 13, 2287-2298. 82. Sutter, E.; Sutter, P. Phase Diagram of Nanoscale Alloy Particles Used for Vapor-Liquid-Solid Growth of Semiconductor Nanowires. Nano Lett. 2008, 8, 411- 414. 83. Wacaser, B. A.; Dick, K. A.; Johansson, J.; Borgstrom, M. T.; Deppert, K.; Samuelson, L. Preferential Interface Nucleation: An Expansion of the Vls Growth Mechanism for Nanowires. Adv. Mater. 2009, 21, 153-165. 84. Hertenberger, S.; Rudolph, D.; Bolte, S.; Doeblinger, M.; Bichler, M.; Spirkoska, D.; Finley, J. J.; Abstreiter, G.; Koblmueller, G. Absence of Vapor-Liquid-Solid Growth During Molecular Beam Epitaxy of Self-Induced Inas Nanowires on Si. Appl. Phys. Lett. 2011, 98, 123114.

85. Shin, J. C.; Choi, K. J.; Kim, D. Y.; Choi, W. J.; Li, X. Characteristics of Strain- Induced in(X)Ga(1-X)as Nanowires Grown on Si(111) Substrates. Cryst. Growth Des. 2012, 12, 2994-2998. 86. Fortuna, S. A.; Li, X. Metal-Catalyzed Semiconductor Nanowires: A Review on the Control of Growth Directions. Semicond. Sci. Tech. 2010, 25, 024005. 87. Wu, Z. H.; Mei, X.; Kim, D.; Blumin, M.; Ruda, H. E.; Liu, J. Q.; Kavanagh, K. L. Growth, Branching, and Kinking of Molecular-Beam Epitaxial < 110 > Gaas Nanowires. Appl. Phys. Lett. 2003, 83, 3368-3370. 88. Ihn, S.-G.; Song, J.-I.; Kim, T.-W.; Leem, D.-S.; Lee, T.; Lee, S.-G.; Koh, E. K.; Song, K. Morphology- and Orientation-Controlled Gallium Arsenide Nanowires on Silicon Substrates. Nano Lett. 2007, 7, 39-44. 89. Han, N.; Hui, A. T.; Wang, F.; Hou, J. J.; Xiu, F.; Hung, T.; Ho, J. C. Crystal Phase and Growth Orientation Dependence of Gaas Nanowires on Nixgay Seeds Via Vapor-Solid-Solid Mechanism. Appl. Phys. Lett. 2011, 99. 90. Han, N.; Wang, F.; Hui, A. T.; Hou, J. J.; Shan, G.; Xiu, F.; Hung, T.; Ho, J. C. Facile Synthesis and Growth Mechanism of Ni-Catalyzed Gaas Nanowires on Non- Crystalline Substrates. Nanotechnology 2011, 22, 285607. 91. Han, N.; Wang, F. Y.; Hou, J. J.; Yip, S.; Lin, H.; Fang, M.; Xiu, F.; Shi, X. L.; Hung, T.; Ho, J. C. Manipulated Growth of Gaas Nanowires: Controllable Crystal Quality and Growth Orientations Via a Supersaturation-Controlled Engineering Process. Cryst. Growth Des. 2012, 12, 6243-6249. 92. Ihn, S.-G.; Song, J.-I.; Kim, Y.-H.; Lee, J. Y.; Ahn, I.-H. Growth of Gaas Nanowires on Si Substrates Using a Molecular Beam Epitaxy. Ieee Transactions on Nanotechnology 2007, 6, 384-389. 93. Shtrikman, H.; Popovitz-Biro, R.; Kretinin, A.; Heiblumf, M. Stacking-Faults-Free Zinc Blende Gaas Nanowires. Nano Lett. 2009, 9, 215-219. 94. Li, X.; Guo, H.; Yin, Z.; Shi, T.; Wen, L.; Zhao, Z.; Liu, M.; Ma, W.; Wang, Y. Morphology and Crystal Structure Control of Gaas Nanowires Grown by Au-Assisted Mbe with Solid As4 Source. J. Cryst. Growth 2011, 324, 82-87. 95. Zhao, Z.-F.; Li, X.-H.; Wen, L.; Guo, H.-M.; Bu, S.-J.; Wang, Y.-Q. Orientation and Structure of Controllable Gaas Nanowires Grown on Gaas (311)B Substrates by Molecular Beam Epitaxiy. Chinese. Phys. Lett. 2012, 29.

96. Kavanagh, K. L.; Saveliev, I.; Blumin, M.; Swadener, G.; Ruda, H. E. Faster Radial Strain Relaxation in Inas-Gaas Core-Shell Heterowires. J. Appl. Phys. 2012, 111, 044301. 97. Joyce, H. J.; Gao, Q.; Wong-Leung, J.; Kim, Y.; Tan, H. H.; Jagadish, C. Tailoring Gaas, Inas, and Ingaas Nanowires for Optoelectronic Device Applications. IEEE. J. Sel. Top. Quant 2011, 17, 766-778. 98. Givargizov, E. I. Fundamental Aspects of Vls Growth. J. Cryst. Growth 1975, 31, 20-30. 99. Wacaser, B. A.; Deppert, K.; Karlsson, L. S.; Samuelson, L.; Seifert, W. Growth and Characterization of Defect Free Gaas Nanowires. J. Cryst. Growth 2006, 287, 504-508. 100. Mikkelsen, A.; Skold, N.; Ouattara, L.; Lundgren, E. Nanowire Growth and Dopants Studied by Cross-Sectional Scanning Tunnelling Microscopy. Nanotechnology 2006, 17, S362-S368. 101. Fortuna, S. A.; Wen, J.; Chun, I. S.; Li, X. Planar Gaas Nanowires on Gaas (100) Substrates: Self-Aligned, Nearly Twin-Defect Free, and Transfer-Printable. Nano Lett. 2008, 8, 4421-4427. 102. Ghosh, S. C.; Kruse, P.; LaPierre, R. R. The Effect of Gaas(100) Surface Preparation on the Growth of Nanowires. Nanotechnology 2009, 20, 115602. 103. Dowdy, R. S.; Walko, D. A.; Li, X. Relationship between Planar Gaas Nanowire Growth Direction and Substrate Orientation. Nanotechnology 2013, 24, 035304. 104. Zhang, G.; Tateno, K.; Gotoh, H.; Nakano, H. Parallel-Aligned Gaas Nanowires with < 110 > Orientation Laterally Grown on [311] B Substrates Via the Gold- Catalyzed Vapor-Liquid-Solid Mode. Nanotechnology 2010, 21, 095607. 105. Sun, W.; Guo, Y. N.; Xu, H. Y.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Polarity Driven Simultaneous Growth of Free-Standing and Lateral Gaasp Epitaxial Nanowires on Gaas (001) Substrate. Appl. Phys. Lett. 2013, 103, 223104. 106. Verheijen, M. A.; Algra, R. E.; Borgstrom, M. T.; Immink, G.; Sourty, E.; van Enckevort, W. J. P.; Vlieg, E.; Bakkers, E. P. A. M. Three-Dimensional Morphology of Gap-Gaas Nanowires Revealed by Transmission Electron Microscopy Tomography. Nano Lett. 2007, 7, 3051-3055. 107. Bjork, M. T.; Ohlsson, B. J.; Sass, T.; Persson, A. I.; Thelander, C.; Magnusson, M. H.; Deppert, K.; Wallenberg, L. R.; Samuelson, L. One-Dimensional

Heterostructures in Semiconductor Nanowhiskers. Appl. Phys. Lett. 2002, 80, 1058- 1060. 108. Seifert, W.; Borgstrom, M.; Deppert, K.; Dick, K. A.; Johansson, J.; Larsson, M. W.; Martensson, T.; Skold, N.; Svensson, C. P. T.; Wacaser, B. A., et al. Growth of One-Dimensional Nanostructures in Movpe. J. Cryst. Growth 2004, 272, 211-220. 109. Li, Z.-A.; Moeller, C.; Migunov, V.; Spasova, M.; Farle, M.; Lysov, A.; Gutsche, C.; Regolin, I.; Prost, W.; Tegude, F.-J., et al. Planar-Defect Characteristics and Cross-Sections of < 001 >, < 111 >, and < 112 > Inas Nanowires. J. Appl. Phys. 2011, 109, 114320. 110. Aagesen, M.; Johnson, E.; Sorensen, C. B.; Mariager, S. O.; FeidenhanS'L, R.; Spiecker, E.; Nygard, J.; Lindelof, P. E. Molecular Beam Epitaxy Growth of Freestanding Plane-Parallel Inas Nanoplates. Nature Nanotechnology 2007, 2, 761- 764. 111. Zhang Z.; Zheng K.; Lu Z. Y.; Chen P. P.; Lu W.; Zou J. Catalyst Orientation-Induced Growth of Defect-Free Zinc-Blende Structured <001 > Inas Nanowires. Nano Lett. 2015, 15, 876–882. 112. He, M. Q.; Fahmi, M. M. E.; Mohammad, S. N.; Jacobs, R. N.; Salamanca-Riba, L.; Felt, F.; Jah, M.; Sharma, A.; Lakins, D. Inas Nanowires and Whiskers Grown by Reaction of Indium with Gaas. Appl. Phys. Lett. 2003, 82, 3749-3751. 113. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Defect-Free <110> Zinc-Blende Structured Inas Nanowires Catalyzed by Palladium. Nano Lett. 2012, 12, 5744-5749. 114. Zhang, Z.; Lu, Z. Y.; Xu, H. Y.; Chen, P. P.; Lu, W.; Zou, J. Structure and Quality Controlled Growth of Inas Nanowires through Catalyst Engineering. Nano Res. 2014, 7, 1640-1649. 115. Zi, Y.; Jung, K.; Zakharov, D.; Yang, C. Understanding Self-Aligned Planar Growth of Inas Nanowires. Nano Lett. 2013, 13, 2786-2791. 116. Zhang, X.; Zou, J.; Paladugu, M.; Guo, Y. N.; Wang, Y.; Kim, Y.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C. Evolution of Epitaxial Inas Nanowires on Gaas (111)B. Small 2009, 5, 366-369. 117. Bao, J.; Bell, D. C.; Capasso, F.; Wagner, J. B.; Martensson, T.; Tragardh, J.; Samuelson, L. Optical Properties of Rotationally Twinned Inp Nanowire Heterostructures. Nano Lett. 2008, 8, 836-841.

118. Fonseka, H. A.; Caroff, P.; Wong-Leung, J.; Ameruddin, A. S.; Tan, H. H.; Jagadish, C. Nanowires Grown on Inp (100): Growth Directions, Facets, Crystal Structures, and Relative Yield Control. Acs Nano 2014, 8, 6945-6954. 119. Wang, J.; Plissard, S.; Hocevar, M.; Vu, T. T. T.; Zehender, T.; Immink, G. G. W.; Verheijen, M. A.; Haverkort, J.; Bakkers, E. P. A. M. Position-Controlled [100] Inp Nanowire Arrays. Appl. Phys. Lett. 2012, 100, 053107. 120. Wang, J.; Plissard, S. R.; Verheijen, M. A.; Feiner, L.-F.; Cavalli, A.; Bakkers, E. P. A. M. Reversible Switching of Inp Nanowire Growth Direction by Catalyst Engineering. Nano Lett. 2013, 13, 3802-3806. 121. Inari, M.; Takeda, J.; Motohisa, J.; Fukui, T. Selective Area Movpe Growth of Inp and Ingaas Pillar Structures for Inp-Based Two-Dimensional Photonic Crystals. Physica E-Low-Dimensional Systems & Nanostructures 2004, 21, 620-624. 122. Ikejiri, K.; Kitauchi, Y.; Tomioka, K.; Motohisa, J.; Fukui, T. Zinc Blende and Wurtzite Crystal Phase Mixing and Transition in Indium Phosphide Nanowires. Nano Lett. 2011, 11, 4314-4318. 123. Kitauchi, Y.; Kobayashi, Y.; Tomioka, K.; Hara, S.; Hiruma, K.; Fukui, T.; Motohisa, J. Structural Transition in Indium Phosphide Nanowires. Nano Lett. 2010, 10, 1699-1703. 124. Mohan, P.; Motohisa, J.; Fukui, T. Controlled Growth of Highly Uniform, Axial/Radial Direction-Defined, Individually Addressable Inp Nanowire Arrays. Nanotechnology 2005, 16, 2903-2907. 125. Ikejiri, K.; Ishizaka, F.; Tomioka, K.; Fukui, T. Bidirectional Growth of Indium Phosphide Nanowires. Nano Lett. 2012, 12, 4770-4. 126. Kuykendall, T. R.; Altoe, M. V. P.; Ogletree, D. F.; Aloni, S. Catalyst-Directed Crystallographic Orientation Control of Gan Nanowire Growth. Nano Lett. 2014, 14, 6767-6773. 127. Kuykendall, T.; Pauzauskie, P.; Lee, S. K.; Zhang, Y. F.; Goldberger, J.; Yang, P. D. Metalorganic Chemical Vapor Deposition Route to Gan Nanowires with Triangular Cross Sections. Nano Lett. 2003, 3, 1063-1066. 128. Zhou, X.; Chesin, J.; Crawford, S.; Gradecak, S. Using Seed Particle Composition to Control Structural and Optical Properties of Gan Nanowires. Nanotechnology 2012, 23, 285603.

129. Kuykendall, T.; Pauzauskie, P. J.; Zhang, Y. F.; Goldberger, J.; Sirbuly, D.; Denlinger, J.; Yang, P. D. Crystallographic Alignment of High-Density Gallium Nitride Nanowire Arrays. Nat. Mater. 2004, 3, 524-528. 130. Tang, Y. B.; Chen, Z. H.; Song, H. S.; Lee, C. S.; Cong, H. T.; Cheng, H. M.; Zhang, W. J.; Bello, I.; Lee, S. T. Vertically Aligned P-Type Single-Crystalline Gan Nanorod Arrays on N-Type Si for Heterojunction Photovoltaic Cells. Nano Lett. 2008, 8, 4191-4195. 131. Lim, S. K.; Crawford, S.; Gradecak, S. Growth Mechanism of Gan Nanowires: Preferred Nucleation Site and Effect of Hydrogen. Nanotechnology 2010, 21, 345604. 132. Chin, A. H.; Ahn, T. S.; Li, H.; Vaddiraju, S.; Bardeen, C. J.; Ning, C.-Z.; Sunkara, M. K. Photoluminescence of Gan Nanowires of Different Crystallographic Orientations. Nano Lett. 2007, 7, 626-631. 133. Chen, C. C.; Yeh, C. C.; Chen, C. H.; Yu, M. Y.; Liu, H. L.; Wu, J. J.; Chen, K. H.; Chen, L. C.; Peng, J. Y.; Chen, Y. F. Catalytic Growth and Characterization of Gallium Nitride Nanowires. J. Am. Chem. Soc. 2001, 123, 2791-2798. 134. Li, H. W.; Chin, A. H.; Sunkara, M. K. Direction-Dependent Homoepitaxial Growth of Gan Nanowires. Adv. Mater. 2006, 18, 216-+. 135. Nam, C. Y.; Tham, D.; Fischer, J. E. Effect of the Polar Surface on Gan Nanostructure Morphology and Growth Orientation. Appl. Phys. Lett. 2004, 85, 5676- 5678. 136. Wang, G. T.; Talin, A. A.; Werder, D. J.; Creighton, J. R.; Lai, E.; Anderson, R. J.; Arslan, I. Highly Aligned, Template-Free Growth and Characterization of Vertical Gan Nanowires on Sapphire by Metal-Organic Chemical Vapour Deposition. Nanotechnology 2006, 17, 5773-5780. 137. Qian, F.; Li, Y.; Gradecak, S.; Wang, D. L.; Barrelet, C. J.; Lieber, C. M. Gallium Nitride-Based Nanowire Radial Heterostructures for Nanophotonics. Nano Lett. 2004, 4, 1975-1979. 138. Qian, F.; Gradecak, S.; Li, Y.; Wen, C. Y.; Lieber, C. M. Core/Multishell Nanowire Heterostructures as Multicolor, High-Efficiency Light-Emitting Diodes. Nano Lett. 2005, 5, 2287-2291. 139. Ji, H.; Kuball, M.; Burke, R. A.; Redwing, J. M. Vibrational and Optical Properties of Gan Nanowires Synthesized by Ni-Assisted Catalytic Growth. Nanotechnology 2007, 18.

140. Li, Q.; Wang, G. T. Improvement in Aligned Gan Nanowire Growth Using Submonolayer Ni Catalyst Films. Appl. Phys. Lett. 2008, 93. 141. Burke, R. A.; Lamborn, D. R.; Weng, X.; Redwing, J. M. Growth and Process Modeling Studies of Nickel-Catalyzed Metalorganic Chemical Vapor Deposition of Gan Nanowires. J. Cryst. Growth 2009, 311, 3409-3416. 142. Ji, X.; Cheng, S.; Hu, H.; Li, H.; Wu, Z.; Yan, P. Effect of Catalyst Nanoparticle Size on Growth Direction and Morphology of Inn Nanowires. Aip Advances 2012, 2, 022150. 143. Liang, C. H.; Chen, L. C.; Hwang, J. S.; Chen, K. H.; Hung, Y. T.; Chen, Y. F. Selective-Area Growth of Indium Nitride Nanowires on Gold-Patterned Si(100) Substrates. Appl. Phys. Lett. 2002, 81, 22-24. 144. Tang, T.; Han, S.; Jin, W.; Liu, X. L.; Li, C.; Zhang, D. H.; Zhou, C. W.; Chen, B.; Han, J.; Meyyapan, M. Synthesis and Characterization of Single-Crystal Indium Nitride Nanowires. J. Mater. Res. 2004, 19, 423-426. 145. Cai, Z.; Garzon, S.; Chandrashekhar, M. V. S.; Webb, R. A.; Koley, G. Synthesis and Properties of High-Quality Inn Nanowires and Nanonetworks. J. Electron. Mater. 2008, 37, 585-592. 146. Chang, Y.-K.; Hong, F. C.-N. Synthesis and Characterization of Indium Nitride Nanowires by Plasma-Assisted Chemical Vapor Deposition. Mater. Lett. 2009, 63, 1855-1858. 147. Chen, R. S.; Yang, T. H.; Chen, H. Y.; Chen, L. C.; Chen, K. H.; Yang, Y. J.; Su, C. H.; Lin, C. R. Photoconduction Mechanism of Oxygen Sensitization in Inn Nanowires. Nanotechnology 2011, 22. 148. Liu, H.; Shi, L.; Geng, X.; Su, R.; Cheng, G.; Xie, S. Reactant-Governing Growth Direction of Indium Nitride Nanowires. Nanotechnology 2010, 21, 245601. 149. Huang, C.-T.; Song, J.; Tsai, C.-M.; Lee, W.-F.; Lien, D.-H.; Gao, Z.; Hao, Y.; Chen, L.-J.; Wang, Z. L. Single-Inn-Nanowire Nanogenerator with Upto 1 V Output Voltage. Adv. Mater. 2010, 22, 4008-4013. 150. Koley, G.; Cai, Z.; Bin Quddus, E.; Liu, J.; Qazi, M.; Webb, R. A. Growth Direction Modulation and Diameter-Dependent Mobility in Inn Nanowires. Nanotechnology 2011, 22. 151. Liu, H.; Cheng, G. Indium Predeposition-Enabled Vapor-Liquid-Solid Growth of Inn Nanowires. Applied Physics Express 2011, 4.

152. Tsai, C. T.; Williams, R. S. Solid Phase Equilibria in the Au-Ga-as, Au-Ga-Sb, Au-in-as, and Au-in-Sb Ternaries. J. Mater. Res. 1986, 1, 352-360. 153. Dick, K. A.; Deppert, K.; Karlsson, L. S.; Wallenberg, L. R.; Samuelson, L.; Seifert, W. A New Understanding of Au-Assisted Growth of Iii-V Semiconductor Nanowires. Adv. Funct. Mater. 2005, 15, 1603-1610. 154. Dubrovskii, V. G.; Sibirev, N. V.; Cirlin, G. E.; Harmand, J. C.; Ustinov, V. M. Theoretical Analysis of the Vapor-Liquid-Solid Mechanism of Nanowire Growth During Molecular Beam Epitaxy. Phys. Rev. E 2006, 73, 021603. 155. Glas, F. Chemical Potentials for Au-Assisted Vapor-Liquid-Solid Growth of Iii-V Nanowires. J. Appl. Phys. 2010, 108, 073506. 156. Ihn, S.-G.; Song, J.-I. Inas Nanowires on Si Substrates Grown by Solid Source Molecular Beam Epitaxy. Nanotechnology 2007, 18, 355603. 157. Rieger, T.; Heiderich, S.; Lenk, S.; Lepsa, M. I.; Gruetzmacher, D. Ga-Assisted Mbe Growth of Gaas Nanowires Using Thin Hsq Layer. J. Cryst. Growth 2012, 353, 39-46. 158. Cirlin, G. E.; Dubrovskii, V. G.; Sibirev, N. V.; Soshnikov, I. P.; Samsonenko, Y. B.; Tonkikh, A. A.; Ustinov, V. M. The Diffusion Mechanism in the Formation of Gaas and Algaas Nanowhiskers During the Process of Molecular-Beam Epitaxy. Semiconductors+ 2005, 39, 557-564. 159. Dubrovskii, V. G.; Sibirev, N. V. General Form of the Dependences of Nanowire Growth Rate on the Nanowire Radius. J. Cryst. Growth 2007, 304, 504- 513. 160. Dubrovskii, V. G.; Sibirev, N. V. Growth Rate of a Crystal Facet of Arbitrary Size and Growth Kinetics of Vertical Nanowires. Phys. Rev. E 2004, 70, 031604. 161. Johansson, J.; Svensson, C. P. T.; Martensson, T.; Samuelson, L.; Seifert, W. Mass Transport Model for Semiconductor Nanowire Growth. J. Phys. Chem. B 2005, 109, 13567-13571. 162. Bauer, J.; Gottschalch, V.; Paetzelt, H.; Wagner, G.; Fuhrmann, B.; Leipner, H. S. Movpe Growth and Real Structure of Vertical-Aligned Gaas Nanowires. J. Cryst. Growth 2007, 298, 625-630. 163. Soci, C.; Bao, X.-Y.; Aplin, D. P. R.; Wang, D. A Systematic Study on the Growth of Gaas Nanowires by Metal-Organic Chemical Vapor Deposition. Nano Lett. 2008, 8, 4275-4282.

164. Dick, K. A.; Bolinsson, J.; Borg, B. M.; Johansson, J. Controlling the Abruptness of Axial Heterojunctions in Iii-V Nanowires: Beyond the Reservoir Effect. Nano Lett. 2012, 12, 3200-3206. 165. Borgstrom, M. T.; Immink, G.; Ketelaars, B.; Algra, R.; Bakkers, E. P. A. M. Synergetic Nanowire Growth. Nature Nanotechnology 2007, 2, 541-544. 166. Reep, D. H.; Ghandhi, S. K. Deposition of Gaas Epitaxial Layers by Organometallic Cvd: Temperature and Orientation Dependence. J. Electrochem. Soc. 1983, 130, 675-680. 167. Zhang, Z. H.; Pickrell, G. W.; Chang, K. L.; Lin, H. C.; Hsieh, K. C.; Cheng, K. Y. Surface Morphology Control of Inas Nanostructures Grown on Ingaas/Inp. Appl. Phys. Lett. 2003, 82, 4555-4557. 168. Spirkoska, D.; Colombo, C.; Heiss, M.; Abstreiter, G.; Morral, A. F. i. The Use of Molecular Beam Epitaxy for the Synthesis of High Purity Iii-V Nanowires. J. Phys- condens. Mat. 2008, 20, 454225. 169. Babu, J. B.; Yoh, K. Growth Rate Enhancement of Inas Nanowire by Molecular Beam Epitaxy. J. Cryst. Growth 2011, 322, 10-14. 170. Sartel, C.; Dheeraj, D. L.; Jabeen, F.; Harmand, J. C. Effect of Arsenic Species on the Kinetics of Gaas Nanowires Growth by Molecular Beam Epitaxy. J. Cryst. Growth 2010, 312, 2073-2077. 171. Kohashi, Y.; Sato, T.; Ikejiri, K.; Tomioka, K.; Hara, S.; Motohisa, J. Influence of Growth Temperature on Growth of Ingaas Nanowires in Selective-Area Metal- Organic Vapor-Phase Epitaxy. J. Cryst. Growth 2012, 338, 47-51. 172. Koblmueller, G.; Hertenberger, S.; Vizbaras, K.; Bichler, M.; Bao, F.; Zhang, J. P.; Abstreiter, G. Self-Induced Growth of Vertical Free-Standing Inas Nanowires on Si(111) by Molecular Beam Epitaxy. Nanotechnology 2010, 21, 365602. 173. Babu, J. B.; Yoh, K. Effect of as/in-Flux on the Growth of Inas Nanowire by Molecular Beam Epitaxy. J. Cryst. Growth 2011, 323, 301-303. 174. Skold, N.; Wagner, J. B.; Karlsson, G.; Hernan, T.; Seifert, W.; Pistol, M.-E.; Samuelson, L. Phase Segregation in Allnp Shells on Gaas Nanowires. Nano Lett. 2006, 6, 2743-2747. 175. Johansson, J.; Wacaser, B. A.; Dick, K. A.; Seifert, W. Growth Related Aspects of Epitaxial Nanowires. Nanotechnology 2006, 17, S355-S361.

176. Yamashita, T.; Akiyama, T.; Nakamura, K.; Ito, T. Effects of Facet Orientation on Relative Stability between Zinc Blende and Wurtzite Structures in Group Iii-V Nanowires. Jpn. J. Appl. Phys. 2010, 49, 055003. 177. Hilner, E.; Hakanson, U.; Froeberg, L. E.; Karlsson, M.; Kratzer, P.; Lundgren, E.; Samuelson, L.; Mikkelsen, A. Direct Atomic Scale Imaging of Iii-V Nanowire Surfaces. Nano Lett. 2008, 8, 3978-3982. 178. Plante, M. C.; LaPierre, R. R. Au-Assisted Growth of Gaas Nanowires by Gas Source Molecular Beam Epitaxy: Tapering, Sidewall Faceting and Crystal Structure. J. Cryst. Growth 2008, 310, 356-363. 179. Plante, M. C.; LaPierre, R. R. Control of Gaas Nanowire Morphology and Crystal Structure. Nanotechnology 2008, 19, 495603. 180. Czaban, J. A.; Thompson, D. A.; LaPierre, R. R. Gaas Core-Shell Nanowires for Photovoltaic Applications. Nano Lett. 2009, 9, 148-154. 181. Leitsmann, R.; Bechstedt, F. Surface Influence on Stability and Structure of Hexagon-Shaped Iii-V Semiconductor Nanorods. J. Appl. Phys. 2007, 102, 063528. 182. Lehmann, S.; Wallentin, J.; Jacobsson, D.; Deppert, K.; Dick, K. A. A General Approach for Sharp Crystal Phase Switching in Inas, Gaas, Inp, and Gap Nanowires Using Only Group V Flow. Nano Lett. 2013, 13, 4099-4105.

3 Experimental Techniques

3.1 III-V nanowires grown by MBE

3.1.1 MBE MBE denote the epitaxial growth process in which directed neutral thermal atomic and molecular beams impinge on a heated crystalline substrate under ultra-high vacuum conditions. MBE was invented in the late 1960s at Bell Telephone Laboratory by John Arthur1 and Alfred Cho.2 The discovery of the MBE process

came about not with the intent of finding a new method of crystal growth, but rather as a study of surface-vapor interactions with a new, compact mass spectrometer when in 1968 John Arthur and John LePore were studying the reflection of pulsed molecular beams of Ga and As2 from GaAs surfaces in UHV in order to measure the energy of adsorption.1 Then Alfred Cho published landmark results reporting the first in-situ observations of MBE process using reflection high energy electron diffraction (RHEED).2-3 Since then, MBE has evolved into one of the most controllable and widely used non-equilibrium growth techniques for growing thin, epitaxial films of a wide variety of materials, both at research and industrial level. The principle underlying MBE growth is relatively simple. The epitaxial growth on the substrate is performed by evaporating the material from the effusion cells into a growth chamber maintained under ultrahigh vacuum condition. Because of ultrahigh vacuum and the extensive use of chilled walls surrounding the source ovens and the substrate, the beams make essentially a single pass through the chamber before condensing on the substrate. The ultrahigh vacuum chamber not only makes it possible to grow highly pure materials but also allows the in-situ characterization techniques such as RHEED to operate without damage from corrosive reaction with residual vapours. The RHEED system provides a diffraction pattern on a phosphor- coated window this is indicative of the ordering of the substrate surface. The flux of molecules or atoms towards the substrate can be abruptly released or blocked by controlling the shutters in front of the cell. This allows growing heterostructures with atomically sharp interface. A major difference between the MBE growth technique and other growth techniques like MOCVD is that it is far from thermodynamic equilibrium conditions, i.e., it is mainly governed by the kinetics of the surface processes. From the very beginning, MBE has benefited enormously from the inclusion of analytic tools, such as RHEED, that provides real time information on the topography of the surface and the growth process. Most MBE systems today either include an electron gun and phosphor screen for displaying RHEED patterns while the growth is going on, or include low energy electron diffraction (LEED) system for viewing the structure before and after growth. The great advantage of RHEED is that since the RHEED beam arrives at a glancing angle without any part positioned in front of sample, such geometry allows the RHEED to operate while the substrate is exposed to the molecular beams, and thus one can obtain real-time structural information.

The RHEED system provides a tool 1) for monitoring growth rate, 2) for a qualitative measure of surface topography including the surface roughness, and 3) for monitoring the surface structure that in certain instances can provide a measure of the surface composition. MBE has several advantages over other growth techniques:4 1) the ultra-high vacuum can increase mean free path of molecules and reduce contamination/oxidation of material surfaces. 2) The low growth temperature and low growth rate prevent inter-diffusion in the nanostructures, and can provide good interface, composition and doping control at the sub-monolayer level. 3) In situ monitoring of growth is possible. 4) Since all growth parameters can be adjusted precisely and separately, the intrinsic growth phenomena can be studied individually. 5) Strongly non-equilibrium growth conditions, mainly governed by the kinetics of the surface processes.

3.1.2 Nanowire growth in MBE The growth of nanowires with the assistance of catalysts in MBE was initially speculated to be impossible due to the fact that the materials are supplied in the form of atoms, where no catalytic activity of catalyst such as Au to decompose the precursors to atoms is required as in MOCVD.5 However, these speculations were proven to be incomplete until several groups6-8 demonstrated the Au-catalyzed growth of GaAs nanowires by MBE. The function of metal particle during nanowire growth is twofold: 1) absorption of atoms from vapor phases or substrate surfaces, and the driving force is to lower the chemical potential of the source atoms, and 2) precipitation or crystallization of the source materials at the particle-substrate interface. MBE technique provides an ideal and clean growth environment, and therefore the atomic structures, doping states and junctions or heterostructures can be well controlled. Combined with the VLS mechanism, MBE technique is able to produce high quality semiconductor nanowires. Different from other growth techniques, MBE works under ultra-high vacuum conditions. The mean free path of the source molecules under vacuum conditions of 10-5 Torr is about 0.2 m. The growth, surface structure and contamination can be monitored in situ by RHEED, Auger electron spectroscopy and other surface probing techniques.

Epitaxial III-V nanowire growth in the MBE system mainly includes: Self- catalysed,9-10 Selected area MBE (SA-MBE),11 Metal-assisted12-15 nanowire growth. The mostly used method is the metal-assisted, especially Au-assisted nanowire growth. Generally, the growth component, i.e. group III material, alloy with the gold, when the Au particle becomes supersaturated with the group III, the crystallization of materials at the interface between the catalyst particle and the substrate starts. During nanowire growth, nanowire lifts the catalyst particle as it grows with the nanowire diameter determined by the Au particle size.

3.1.3 Nanowire growth procedures in MBE There are mainly two MBE growth chambers: solid-sourced MBE and gas-sourced MBE to grow III-V nanowires. All the nanowires in this thesis were grown in the solid- sourced Riber 32 MBE system as shown in Figure 3.1. Au-assisted III-V nanowire growth represents one of the most investigated categories of nanowire growth method. In this work, III-V nanowires are grown on GaAs {111}B substrate in an solid source MBE system.

Figure 3.1 A photograph of the solid-sourced Riber 32 MBE system located at Shanghai Institute of Technical Physics, Chinese Academy of Sciences.

The nanowire growth procedures in MBE can be summarized as below. The first step is to degas the substrate for 10 minutes and then to desorb surface contaminants including surface dioxide at 580 °C under group V beam protection in the growth chamber. Then, a GaAs buffer layer is grown on the substrate in order to achieve an atomically flat substrate surface. After that, the sample is transferred to an Au evaporation chamber connected to the III-V growth chamber, and a thin Au film was deposited on substrate surface. Secondly, the Au coated substrate is transferred back into the growth chamber. Then the substrate temperature was increased to required temperature, normally higher than growth temperature, and the substrate is annealed to form equilibrium alloy droplets with Au. Annealing process is conducted under group V beam overpressure in order to prevent the decomposition of substrate. After annealing, the substrate temperature is lowed to the nanowire growth temperature and the growth species beams are introduced to the growth chamber to initiate nanowire growth. Then nanowire growth is performed under different designed growth parameters. When the growth time is reached, the growth species beams are closed and the sample is cooled down under group V beam protection, normally until the substrate temperature decreased to 300 °C. Eventually, the grown sample is removed from the growth chamber while the substrate temperature is close to room temperature.

3.2 Electron microscopy

3.2.1 A brief history of electron microscopy Historically before the evolution of electron microscope the term “microscope” refers to the visible light microscope, which was developed in sixteen century,16 working with the combination of glass lenses and visible light. In terms of the electron microscopy, we mainly focused on scanning electron microscope (SEM) and transmission electron microscope (TEM). It is of interest of note that it is the more complicated TEM that first being produced rather than the SEM as we can see from Table 3.1 which presents a brief history of electron microscopy. TEM has enabled researchers to investigate various materials in great depth due to its powerful characterization ability. High voltages in a modern TEM could accelerate the electrons to velocities that approach the speed of light and this

improvement has improved the resolution of TEM. Now, TEMs are widely available from many manufactures, such as JEOL, Hitachi, Carl Zeiss and FEI.

Table 3.1 A brief history of electron microscopy

Year Event

1897 J.J. Thomson17 discovered electrons

Louis de Broglie18 proposed that electrons possess wave-like 1924 characteristics with wave length of λ=h/mv

H. Busch19 demonstrated that “electron lens” can bend 1926 electron beams like what a normal lens can do to visible lights

G.P. Thomson and A. Reid,20 and C. Davisson and L. 1927 Germer21 demonstrated the wave nature of electrons by their experiments of electron diffraction through crystals

M. Knoll and E. Ruska22 first used the term TEM in their 1932 publication and demonstrated images taken by a microscope- like instrument

1936 First true TEM was built in the UK

M. von Ardenne23 constructed the scanning transmission 1938 electron microscope (STEM) by adding scanning coils to TEM

1939 First commercial TEM was started by Siemens in Germany

1941 V. Zworykin and E. Ramberg24 first developed the true SEM

R. Heidenreich25 thinned metal foils to electron transparency- 1949 a critical step for practical use of TEM in materials science

First commercial SEM “Stereoscan” was produced by the 1965 Cambridge Scientific Instrument Company

3.2.2 Interaction of incident electrons with a specimen An electron microscope uses a beam of electrons to illuminate a specimen and create a highly magnified image. Due to the wave-particle duality of fast moving electrons, the electrons can both transfer momentum and interfere constructively and destructively during their interaction with materials. When the incident high-energy electrons interact with the specimen in both elastic and inelastic scattering modes, a number of signals are generated, some of which are schematically shown in Figure

3.2. In elastic scattering mode the incident electrons don’t lose their kinetic energy but only change their directions. However, in inelastic scattering mode, the energy of the incident electrons are transferred to the atoms in the specimen.

Figure 3.2 Schematic illustrations of the interaction of an electron beam with a specimen.26

If the specimen is sufficiently thin (up to several hundred nm) and the energy of the electrons are sufficiently high (~100 keV or higher), the specimen will become transparent to the electron beam, and some of the electrons pass through the specimen as a transmitted beam. Those transmitted beam can include both elastically scattered electrons and inelastically scattered electrons, and these electrons are the basis for various TEM modes like imaging mode, diffraction mode and EELS mode.

If the specimen is not electron transparent, elastic backscattered electrons (BSE) and secondary electrons (SE) are generated from the specimen, and provide electron signals for SEM. When the incident electrons interact with the nucleus of the atom in the specimen and some electrons are scattered backwards the specimen surface without loss of energy due to high angle elastic scattering, those electrons are called BSE. SEs are those electrons emitted from the surface of specimen as a result of inelastic scattering of incident electrons. These incident electrons transfer kinetic energy to the outer-shell electrons of specimen atoms, and caused the emission of SEs with energy less than 50 eV. In both cases, X-rays are emitted from the specimen through the inelastic scattering of incident electrons, and can be detected by the attached equipment in both TEM and SEM for chemical analysis. The generation of X-rays is schematically shown in the Figure 3.3. When the incident high-energy electrons hit the atoms in a specimen, inner-shell electrons of the atoms could be ejected from their orbitals due to transfer of kinetic energy from incident electrons to inner-shell electrons, leaving the empty state inner-shell orbitals and possessing a high energy state. Therefore, in order to decay to the ground state, outer-shell electrons tend to fill in the inner-shell orbital, and accordingly a photon, in the form of X-ray, will be emitted concomitantly due to the energy conservation rule. Since different elements have different electronic structure, these characteristic energies of X-rays can be used to determine the element composition in the specimen.

Figure 3.3 A schematic illustration of the emission of characteristic X-rays.26

3.2.3 Scanning electron microscopy (SEM) In a SEM, an ultrafine electron beam is generated by electrons emission source and scans across the surface of a specimen. Therefore the electron beam quality has direct effect on the resolution of SEM. There are two types of electrons emission sources: thermionic emission and field emission (FE). Thermionic emission uses electrical current to heat up and to saturate a filament. Electrons will escape from the tip of the filament as long as the heat is high enough to overcome the work function of the filament material. However, the filament must not be turned up too quickly or else it will “blow” (burn out). Tungsten (W) and lanthanum hexaboride (LaB6) are two mostly used materials for filament.

Table 3.2 A comparison of SEM emission sources27

W LaB Thermal Emission source 6 Cold FE thermionic thermionic FE

Work function (eV) 4.5 2.7 4.5 2.9

Operating temperature 2700 1800 300 1800 (K)

Brightness (A/cm2.sr) ~105 ~106 ~108 ~107

Energy spread (eV) ~1.5 ~1.0 ~0.3 ~0.5

Current density (A/cm2) ~1 ~20 ~500 ~50,000

Source diameter (nm) ~2000 ~1000 ~10 ~20

Vacuum (Torr) 10-5 10-7 10-10 10-9

In order to overcome the disadvantages of thermionic emission such as: relatively low brightness, low resolution, evaporation of cathode material and thermal drift, field emission is introduced. Field emission electrons come from the surface of a sharp pointed metal, oxide or carbide emitter when a strong electric field is placed at their surface. FESEM employs a field emission gun (FEG) that uses an electric field to cause high energy electrons to be emitted from a chemically etched fine W crystal cathode with a tip of small tip radius (~100 nm). FEGs are divided into two types: 1) cold FEG that uses a strong electric field to extract electrons from cathode by the quantum tunnelling effect at room temperature, and 2) thermal (Schottky) FEG that

uses the electric field emission with the operating temperature of ~1800K, which occurs when a layer of zirconium oxide is supplied to the surface of the tungsten to lower the work function of the tungsten. Compared to cold FEG, thermal FEG can generate more stable and high probe current. A comparison of these SEM emission sources are presented in Table 3.2.

Figure 3.4 A schematic illustration of a SEM (http://www.purdue.edu/rem/rs/sem.htm).

Figure 3.4 shows a schematic illustration of SEM. Principally, the emitted beam is collimated by the electromagnetic condenser lens, focused by an objective lens and scanned across the surface of the specimen by electromagnetic scanning coils in a raster mode. Unlike the in light microscope, image in SEM is never a real image of the specimen, and it is formed by emitted signals such as SEs and BSEs which are collected by the detectors attached to the SEM. This digital image in SEM is reproduced from the intensity of the received electron signals correlated with the spatial positions of the scanning electron probe.

Figure 3.5 Illustration of interaction volume and emitted signals depth in a SEM.28

In a SEM the incident electron beam doesn’t transmit through the sample as in TEM but interacts with the surface of specimen. In addition, the accelerating voltage is much lower than that in TEM, around 1-50 kV. When the electron probe hit the specimen surface, the interaction volume (as shown in the Figure 3.5) between the incident electrons and specimen is directly proportional to the energy of incident electrons and inversely proportional to the average number of atoms in the specimen. The development of electron microscope (EM) is to take the higher resolution advantage of EM over optical microscope. The spatial resolution of SEM images is determined by many factors. Smaller spot size, smaller aperture size, smaller working distance and higher accelerating voltage will lead to higher resolution, vice versa. Other than the resolution, we also need to pay much attention to another important aspect regarding the quality SEM images, that is, the depth of field. Depth of field, referred to the amount of sample in acceptably sharp focus, is significantly important to those specimens with large protruding surface like nanowires. With higher depth of field, more of the specimen surface is in focus. The depth of field increases with working distance, and decreases with spot size and aperture size. According to this discussion, increase in work distance could increase the depth of field, but it would reduce the resolution of the image. While the decrease in aperture size could both increase resolution and depth of field, it would limit the signal strength. Therefore, the choice of aperture size and working distance can be compromised between signal strength, depth of filed and resolution.

Generally speaking, SEM mainly uses SEs and BSEs which are detected by suitable electron detectors as the signals for image formation. Secondary electrons (SEs) are those low energy (less than 50 eV) electrons emitted from the surface of specimen as a result of inelastic scattering of incident electrons. The number of SEs varies with the roughness of specimen surface, and this phenomenon is termed as the edge effect. Since the SEs escape from the same depth of the specimen, which is marked as d, and measured perpendicular to the surface, the volume of SE escape is larger for protruded surface than that of flat one (as shown in Figure 3.6). Hence, during the scanning of electron beam on specimen surface, when beam hits the protrusion area, more SE signals can be generated. This difference in the number of SEs leads to different contrast features in SEM image, i.e., higher SE signals show a higher contrast and thus make those features brighter. Therefore, these SE signals can be used to study the topography of the specimen.

Figure 3.6 Schematic illustrations of the effect of topography on SE yields (http://laser.phys.ualberta.ca/~egerton/SEM/sem.htm).

Backscattered electrons (BSEs) are the electrons that are scattered in the backward direction after interaction with the specimen with only small fraction loss or no loss of energy. Although high energy of BSE results in a large interaction volume and a low spatial resolution compared to the SE image, BSE image can provide chemical composition information of the specimen which can’t be attained by SE image. Since the heavy atoms with a high atomic number scatter incident electrons more strongly than light ones, the number of BSEs is propositional to the atomic number of constituent atoms. Therefore, this contrast caused by the different number

of BSEs can be used to reflect the composition difference across the sample. Figure 3.7 shows the comparison of SE image and BSE image.

Figure 3.7 SEM images of Pt particles on alumina. Left SE image shows morphology only, while Pt particles appear with bright contrast in the right BSE image (http://www.microscopy.ethz.ch/bse.htm).

The morphological characteristics, such as nanowire shape, nanowire length, sidewall facets and growth direction of III-V nanowires, and the surface morphology of substrate in this thesis were investigated by the FESEMs, including JEOL JSM7001F and the mainly used JEOL JSM7800F. This newly developed 7800F provides extreme resolution to our nanowire samples, i.e., 0.8 nm at 15 kV and 1.2 nm at 1 kV. The magnification of the images ranges from 25 to 1,000,000 times, with the accelerating voltage ranges from 0.01 to 30 kV.

3.2.4 Transmission electron microscopy (TEM) As described earlier, electron beam can be forward scattered during its interaction with a thin specimen in TEM, and these transmitted and/or scattered beams will be used to form the TEM image of the specimen. A schematic illustration of a typical TEM is present in Figure 3.8. In most cases, electrons in TEM are extracted from a filament or a field emission gun (FEG) with an accelerating voltage between 100-300 kV. FEG emitted electron beam shows a higher spatial coherency and a much higher brightness and behaves more monochromatic than that by a thermionic source. This emitted electron beam is paralleled by the condenser lens before it hits the specimen and transmits the thin specimen. After that, the transmitted beam is enlarged by the intermediate lens, transferred through the microscope column, further magnified by the projector lens, and finally forms an image of the specimen on screen.

Figure 3.8 Schematic illustrations of TEM (http://www.ammrf.org.au/myscope/tem/introduction/).

Same as the imaging principles in the optical system, the objective lens in TEM also forms the image of the specimen on its image plane using these transmitted electrons, while the electron diffraction (ED) pattern forms at the back focal plane of the objective lens. In TEM, we use intermediate lens to choose projecting the image or electron diffraction pattern onto viewing screen. There are basically two modes in the TEM, i.e. imaging mode and diffraction mode, which can be easily switched between these two modes in modern TEM.29 In the imaging mode: the transmission of the electron beam depends on the mass of the atoms in the specimen and thus can create the so-called mass-thickness contrast in the TEM image. After incident electron beam interacting with the specimen and transmitting the specimen, it will be focused by the objective lens. Then, the intermediate lens chooses to project the intermediate image of the specimen at the image plane onto the fluorescent viewing screen at the bottom of the column. These images then can be recorded by negative films or digital camera made of charge-coupled device (CCD). The objective aperture can also be inserted to increase the diffraction contrast in TEM image. In the diffraction mode: although there is mass-thickness contrast in TEM image, the principal strength of conventional TEM is the ability to generate diffraction contrast of the specimen using the coherent elastic scattered electrons. If the

specimen is crystalline in nature, the atomic planes of the crystal could diffract some incident beams due to the wave-like characteristic of electrons. These diffracted electrons will interfere with each other after scattering, leading to constructive scattering or destructive scattering defined by the Bragg’ law30 (2d sinө = nλ). Incident electrons that have been diffracted to the same angle are focused at the same position in the back focal plane of the objective lens, showing the electron diffraction (ED) pattern. To make this ED visible on the view screen, the intermediate lens is weakened, making the back focal plane of the objective lens coincide with the objective plane of the projector lens. Meanwhile, a selective area aperture can be inserted in the image plane to physically confine a specific specimen area that is illuminated by the electron beam, and this ED technique is termed as selective area electron diffraction (SAED). TEM is a versatile technique to do a variety of microscopic characterization. Other than traditional TEM imaging, other techniques like high resolution transmission electron microscopy (HRTEM), scanning transmission electron microscopy (STEM), X-ray energy dispersive spectroscopy (XEDS), electron energy loss spectroscopy (EELS) and energy filtered transmission electron microscopy (EFTEM) can also be performed on TEM. HRTEM: While conventional TEM imaging uses only the transmitted electrons or some of the forward scattered electrons to increase the diffraction contrast in image, HRTEM imaging uses the phase contrast, including both transmitted and diffracted electrons, to create an interfered image of specimen at the atomic level. In order to acquire HRTEM image, the resolution of the TEM needs to be smaller than the planar spacing of the crystal specimen. Otherwise, the small lattice fringes are beyond the limitation of TEM to resolve. In this regard, we need to tilt the crystalline specimen to the low-index zone axis in order to get the HRTEM image. HRTEM is mostly used to investigate the crystal phase and structural quality of III-V nanowires in this thesis. STEM: STEM works in a similar mode like SEM, i.e., a focused nanometre-sized electron probe scans across the specimen surface with the scanning coils, and then generates many transmitted electron signals which will be received by detectors rather than the viewing screen or CCD camera. If only the transmitted electrons are received by a small disk-like detector, a bright field (BF) STEM image is produced. If only the diffracted electrons are received by an annular detector which is a plate with

hole, a dark field (DF) STEM image is formed. With the development of DF STEM, the so-called high-angular annular dark field (HAADF) detector is invented to receive those high angular diffracted electrons. The contrast in a HAADF image can be used to determine the compositional information in the specimen, that is, material with higher atomic number behaves brighter than that with lower atomic number in a HAADF STEM image. Therefore, this contrast in a HAADF image is also called atomic number contrast (Z-contrast). XEDS: As we have discussed in the earlier section, XEDS detector can be equipped in both SEM and TEM to analyse the composition of the specimen. Due to the nature of TEM, XEDS in TEM possesses a higher resolution than in SEM. Specific peaks in the energy dispersive spectra of the X-rays induced from the interaction between incident electrons and nucleuses of atoms in the specimen can be treated as characteristic footprints of the elements in the specimen. EELS: This technique was developed by J Hiller and R Baker in 1944,31 but was not widely used over the next 50 years. It becomes more widespread in research in the 1990s due to the advances in microscope instrumentation and vacuum technology. EELS is an analytical technique that measures the change in kinetic energy of electrons after they undergo inelastic scattering through the specimen. The amount of energy loss can be measured by an electron spectrometer and be interpreted in terms of what causes this energy loss. EELS is often taken as a complementary technique to XEDS because the XEDS excels at identifying atomic composition in the specimen, is quite easy to use, and is very sensitive to heavier elements. Although EELS has historically been a more difficult technique, it is capable of measuring more aspects of specimen information than XEDS, such as, atomic composition, chemical bonding, valence and conduction band electronic properties and so on.32 EELS tends to work best at relatively low atomic numbers. The energy resolution in EELS is typically 1 eV but can approach 0.1 eV if an electron-beam monochromators is used. This great advantage of high energy resolution gives EELS the ability to “footprint” different forms of the same element, i.e. we can differentiate diamond graphite amorphous carbon and “mineral” carbon through EELS. Under the effect of a magnetic prism, inelastic scattering electrons with different energies are dispersed and are used to form the EELS spectrum. By choosing electrons having undergone a specific kinetic energy loss which

corresponds to a specific element, the distribution of the specific element in the specimen can be mapped. This technique is often named as EELS mapping. EFTEM: By using inelastic scattering electrons as in EELS, it is possible to place an adjustable slit to allow only electrons with a specific energy loss to pass through and reform an image, i.e., EFTEM image. As the inelastic scattering electron signal is often significantly smaller than the background signal, it is normally necessary to obtain more than one image at every energy point to remove background effect. EFTEM technique can be used to aid chemical analysis of the specimen in conjunction with other complementary techniques. In this thesis, we mainly used the Philips Tecnai F20 (operated at 200 kV) TEM and Philips Tecnai F30 (operated at 300 kV) to investigate the structures and composition information of III-V nanowires. The F20 is equipped with Schottky FEG, a twin lens system (70o α-tilt, 30o β-tilt), and an X-ray energy-dispersive spectroscopy (XEDS) detector from EDAX AMETEK Inc., with ultra-thin window. A Model 895 4k x 4k CCD camera from Gatan Inc. is attached to this TEM under the viewing screen. The wavelength of the electron beam is 0.25 pm. The point resolution and information limit are 0.27 nm and 0.18 nm respectively.

3.3 Sample preparation for electron microscopy

3.3.1 SEM sample preparation As the incident electrons with negative charges will continuously interact with the sample, it is very necessary to get rid of the negative charges accumulated on the sample. Thus, all the SEM samples need to be electrically conductive. Normally, a thin layer of conductive materials, such as carbon, platinum or gold, will be coated on the surface of the sample to make it conductive. Since all the SEM samples in this thesis are III-V nanowires which are semiconductors, all the nanowire samples are not pre-coated with any conductive layer. Therefore, the SEM samples were loaded into the SEM as the original form with as-grown nanowires on the substrate. In this thesis, we both stick the substrate bottom to the metallic sample stub to get the top views and birds-eye-views, and stick the facet of the substrate to the metallic sample stub to investigate the cross section of the sample.

3.3.2 TEM sample preparation Basically, there are three methods to prepare different nanowire samples suitable for TEM investigations: 1) individual nanowires preparation to investigate the structure and composition of single nanowire; 2) cross-section TEM sample preparation to study the nanowire base and the relationship between the nanowire base and substrate with the presence of the substrate; 3) nanowire cross-section sample preparation to study the nanowire cross-section, especially to study the core- shell structure of nanowire.

Table 3.3 TEM sample preparations

Sample type Preparation process

A small substrate with as-grown nanowires is cut and placed Individual in a small container with a tiny amount of pure ethanol. Then nanowires the container is put into a ultrasonicator for 10~40 min to break the nanowires from the substrate. A few drops of this nanowire-ethanol dispersion are dropped onto a holy carbon film coated copper grid.

First we glue together several semiconductor substrates with or without nanowires being shaken off. Then the glued Cross-section samples are manually polished in the cross-section direction TEM sample using tripod. After that, the ion milling system, i.e., Gatan Precision Ion Polishing System (PIPS) is used to further thin the specimen.

A small piece of substrate with as-grown nanowires is cut and then embedded in resin. Once the resin is set, we use Nanowire cross- razor blade to trim the sample to get away the substrate. section sample With the substrate removed, the nanowires are then cut in cross-section using ultra microtome.

3.4 References

1. Arthur, J. R. Interaction of Ga and As2 Molecular Beams with Gaas Surfaces. J. Appl. Phys. 1968, 39, 4032-4034. 2. Cho, A. Y. Morphology of Epitaxial Growth of Gaas by a Molecular Beam Method: The Observation of Surface Structures J. Appl. Phys. 1970, 41, 2780-&. 3. Cho, A. Y. Gaas Epitaxy by a Molecular Beam Method: Observations of Surface Structure on the (001) Face. J. Appl. Phys. 1971, 42, 2074-&. 4. Wang, N.; Cai, Y.; Zhang, R. Q. Growth of Nanowires. Mat. Sci. Eng. R. 2008, 60, 1-51. 5. Dheeraj, D. L.; Zhou, H. L.; Moses, A. F.; Hoang, T. B.; Helvoort, A. T. J. v.; Fimland, B. O.; Weman, H. Heterostructured Iii-V Nanowires with Mixed Crystal Phases Grown by Au-Assisted Molecular Beam Epitaxy. Nanowires 2010, 414. 6. Cirlin, G. E.; Dubrovskii, V. G.; Sibirev, N. V.; Soshnikov, I. P.; Samsonenko, Y. B.; Tonkikh, A. A.; Ustinov, V. M. The Diffusion Mechanism in the Formation of Gaas and Algaas Nanowhiskers During the Process of Molecular-Beam Epitaxy. Semiconductors+ 2005, 39, 557-564. 7. Harmand, J. C.; Patriarche, G.; Pere-Laperne, N.; Merat-Combes, M. N.; Travers, L.; Glas, F. Analysis of Vapor-Liquid-Solid Mechanism in Au-Assisted Gaas Nanowire Growth. Appl. Phys. Lett. 2005, 87, 203101. 8. Wu, Y. Y.; Yang, P. D. Direct Observation of Vapor-Liquid-Solid Nanowire Growth. J. Am. Chem. Soc. 2001, 123, 3165-3166. 9. Jabeen, F.; Grillo, V.; Rubini, S.; Martelli, F. Self-Catalyzed Growth of Gaas Nanowires on Cleaved Si by Molecular Beam Epitaxy. Nanotechnology 2008, 19, 275711. 10. Spirkoska, D.; Arbiol, J.; Gustafsson, A.; Conesa-Boj, S.; Glas, F.; Zardo, I.; Heigoldt, M.; Gass, M. H.; Bleloch, A. L.; Estrade, S., et al. Structural and Optical Properties of High Quality Zinc-Blende/Wurtzite Gaas Nanowire Heterostructures. Phys. Rev. B 2009, 80, 245325. 11. Wu, Z. H.; Mei, X. Y.; Kim, D.; Blumin, M.; Ruda, H. E. Growth of Au-Catalyzed Ordered Gaas Nanowire Arrays by Molecular-Beam Epitaxy. Appl. Phys. Lett. 2002, 81, 5177-5179.

12. Jabeen, F.; Piccin, M.; Felisari, L.; Grillo, V.; Bais, G.; Rubini, S.; Martelli, F.; d'Acapito, F.; Rovezzi, M.; Boscherini, F. Mn-Induced Growth of Inas Nanowires. J. Vac. Sci. Technol. B 2010, 28, 478-483. 13. Martelli, F.; Rubini, S.; Jabeen, F.; Felisari, L.; Grillo, V. On the Growth of Inas Nanowires by Molecular Beam Epitaxy. J. Cryst. Growth 2011, 323, 297-300. 14. Tivakornsasithorn, K.; Pimpinella, R. E.; Nguyen, V.; Liu, X.; Dobrowolska, M.; Furdyna, J. K. Magnetic Anisotropy of Gaas/Fe/Au Core-Shell Nanowires Grown by Mbe. J. Vac. Sci. Technol. B 2012, 30, 02B115. 15. Martelli, F.; Piccin, M.; Bais, G.; Jabeen, F.; Ambrosini, S.; Rubini, S.; Franciosi, A. Photoluminescence of Mn-Catalyzed Gaas Nanowires Grown by Molecular Beam Epitaxy. Nanotechnology 2007, 18, 125603. 16. Robinson, J. J. You Voted, We Counted: The 50 Greatest Moments in Materials, Part I : Nos. 50-11. Jom 2007, 59, 14-15. 17. Falconer, I. Corpuscles, Electrons and Cathode Rays: Jj. Thomson and the 'Discovery of the Electron'. The British Journal for the History of Science 1987, 20, 241-276. 18. De Broglie, L. A Tentative Theory of Light Quanta. Phil. Mag. Lett. 1924, 47, 446-458. 19. Busch, H. Calculation of the Channel of the Cathode Rays in Axial Symmetric Electromagnetic Fields. Ann. Phys-berlin. 1926, 81, 974-993. 20. Thomson, G. P.; Reid, A. Diffraction of Cathode Rays by a Thin Film. Nature 1927, 119, 890-890. 21. Davisson, C.; Germer, L. H. The Scattering of Electrons by a Single Crystal of Nickel. Nature 1927, 119, 558-560. 22. Knoll, M.; Ruska, E. The Electron Microscope. Zeitschrift Fur Physik 1932, 78, 318-339. 23. Von Ardenne, M. The Use of Electron Probes for Micro Manipulations. Naturwissenschaften 1938, 26, 562-562. 24. Zworykin, V. K.; Ramberg, E. G. Surface Studies with the Electron Microscope. J. Appl. Phys. 1941, 12, 692-695. 25. Heidenreich, R. D. Electron Microscope and Diffraction Study of Metal Crystal Textures by Means of Thin Sections J. Appl. Phys. 1949, 20, 993-1010. 26. Williams, D. B.; Carter, C. B., Transmission Electron Microscopy: A Textbook for Materials Science. Springer: 2009.

27. Goldstein, J.; Newbury, D. E.; Joy, D. C.; Lyman, C. E.; Echlin, P.; Lifshin, E.; Sawyer, L.; Michael, J. R., Scanning Electron Microscopy and X-Ray Microanalysis. Springer: 2003. 28. Unakar, N.; Tsui, J.; Harding, C. Scanning Electron Microscopy. Ophthalmic Res. 1981, 13, 20-35. 29. Fultz, B.; Howe, J., Transmission Electron Microscopy and Diffractometry of Materials. Springer: 2005. 30. Bragg, W. L.; Thomson, J. J. The Diffraction of Short Electromagnetic Waves by a Crystal. Proceedings of the Cambridge Philosophical Society 1914, 17, 43-57. 31. Hillier, J.; Baker, R. F. Microanalysis by Means of Electrons. J. Appl. Phys. 1944, 15, 663-675. 32. Egerton, R. F. Electron Energy-Loss Spectroscopy in the Tem. Rep. Prog. Phys. 2009, 72, 016502.

4 Quality of Epitaxial InAs Nanowires Controlled by Catalyst Size in Molecular Beam Epitaxy

4.1 Introduction

In this study, the structural quality of Au-catalyzed InAs nanowires grown by molecular beam epitaxy is investigated. Through detailed electron microscopy

characterizations and analysis of binary Au-In phase diagram, it was found that defect-free InAs nanowires can be induced by smaller catalysts with a high In concentration, while comparatively larger catalysts containing less In induce defected InAs nanowires. This study indicates that the structural quality of InAs nanowires can be controlled by the size of Au catalysts when other growth conditions remain as constants.

4.2 Journal Publication

The results in Chapter 4 are included as it appears in Applied Physics Letters 2013, 103, 073109. http://scitation.aip.org/content/aip/journal/apl/103/7/10.1063/1.4818682

Quality of epitaxial InAs nanowires controlled by catalyst size in molecular beam epitaxy

Zhi Zhang1, Zhen-Yu Lu2, Ping-Ping Chen2, Hong-Yi Xu1, Ya-Nan Guo1, Zhi-Ming Liao1, Sui-Xing Shi2, Wei Lu2, and Jin Zou1,3,a)

1 Materials Engineering, The University of Queensland, St. Lucia, Queensland 4072, Australia 2 National Laboratory for Infrared Physics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, 500 Yu-Tian Road, Shanghai 200083, People's Republic of China 3 Centre for Microscopy and Microanalysis, The University of Queensland, St. Lucia, Queensland 4072, Australia

In this study, the structural quality of Au-catalyzed InAs nanowires grown by molecular beam epitaxy is investigated. Through detailed electron microscopy characterizations and analysis of binary Au-In phase diagram, it was found that defect-free InAs nanowires can be induced by smaller catalysts with a high In concentration, while comparatively larger catalysts containing less In induce defected InAs nanowires. This study indicates that the structural quality of InAs nanowires can be controlled by the size of Au catalysts when other growth conditions remain as constants.

PACS: 81.16.-c, 82.65.+r, 68.35.bg, 61.46.Df, 81.05.Ea

Keywords: catalysts, electron microscopy, gold, III-V semiconductors, nanofabrication, nanoparticles, surface morphology

a) Email: [email protected]

III-V semiconductor nanowires have attracted extensive attentions in the recent decades due to their distinct physical properties and hence potential applications in nanoelectronics and optoelectronics.1,2 Currently, the most typical approach for nanowire growth is the vapor-liquid-solid (VLS)3 mechanism, in which reactant from ambient vapor is incorporated into metallic catalyst nanoparticles to form liquid droplets, and the supersaturation of reactant atoms leads to anisotropic nanowire growth. Metal-catalyzed nanowire growth under the eutectic temperature via the VLS process has also been reported and the existence of liquid droplets is attributed to the undercooling of nanodroplets and nanoscale size effect.4,5 As a key III-V semiconductor, InAs has attracted special research interest due to its very high electron mobility,6 low-resistance ohmic contact,7 narrow band gap and small electron effective mass.8 The combination of these unique features and the distinct characteristics of nanowires have made InAs nanowires a promising candidate for applications in single-electron transistors,9 the Josephon junctions,10 resonant tunnelling diodes,11 and ballistic transistors.12,13 For nanowires to be practically useful, minimizing the lattice defects in nanowires has been a challenging task. In this regard, several strategies have been implemented. For example, it has been demonstrated that structural quality of III-V nanowires can be well controlled by tuning growth parameters, such as V/III ratio and/or growth temperature.14 On the other hand, theoretical predictions has indicated that nanowire diameter15 and catalyst supersaturation16,17 can greatly influence the crystal structure of III-V nanowires and their qualities. The effect of nanowire diameters on the structural transition of wurtzite - zinc-blende structures in InAs nanowires grown by metal-organic chemical vapor deposition has been demonstrated.18 However, understanding the size-dependent growth of InAs nanowires and achieving high-quality InAs nanowires requires further investigations.19 In this letter, we investigated the structural characteristics of Au-catalyzed epitaxial InAs nanowires grown by molecular beam epitaxy (MBE). Through detailed structural and chemical characterizations using electron microscopy, it has been found that the structural quality of InAs nanowires can be well controlled by the size of Au catalysts. The physical reason behind these observed experimental results is explored. The InAs nanowires were grown in a Riber 32 MBE system using Au-assisted growth mode on a GaAs {111}B substrate. The substrate surface was firstly

degassed in the preparation chamber for 15 minutes at 250 °C. Then the substrate was transferred to the growth chamber to be thermally deoxidized at 610 °C, and a

~200 nm thick GaAs buffer layer was grown on the GaAs {111}B substrate at 580 °C to achieve atomically flat surface. After that, the substrate was transferred back to the preparation chamber, and a thin Au film was then deposited on the top of the GaAs buffer layer by the vacuum thermal evaporation. The Au-coated GaAs substrate was then transferred back to the MBE growth chamber and annealed at 500 °C for 5 mins under As ambient to agglomerate the Au thin film into nanoparticles via Oswald ripening. This thin film generated Au nanoparticles is a simple and cost efficient approach.20,21 After the annealing, the substrate temperature was lowered to 430 °C and the In source was introduced to initiate InAs nanowires growth for 100 mins with a V/III ratio of 20. The morphological characteristics of grown InAs nanowires were investigated by scanning electron microscopy (SEM, JEOL 7800F, operated at 10 kV), and their detailed structural and chemical characteristics were investigated by transmission electron microscopy [TEM, Philips Tecnai F20, operated at 200 kV, equipped with X- ray energy dispersive spectroscopy (EDS) for compositional analysis, and Philips Tecnai F30]. Individual nanowires for TEM analysis were prepared by ultrasonicating the nanowire samples in ethanol and then dispersing nanowires onto holey carbon films supported by Cu grids. Figure 1 is a typical tilted SEM image and shows that the majority nanowires are vertically grown on the substrate. As can be seen, the heights of these nanowires vary, similar to our early observation of InAs nanowires grown on the GaAs substrate.22 It should be noted that these nanowires have diverse diameters, as illustrated in the inset. Based on the nanowire morphology, no tapering was observed, although tapering, caused by lateral growth of nanowires,23 often exists in epitaxially grown nanowires. To understand the structural and chemical characteristics of grown nanowires, detailed TEM investigations were performed. Figure 2(a) is a bright-field TEM image of a typical thick InAs nanowire and shows uniform diameter (~45 nm) along axial direction of the nanowire. Figure 2(b) is a high-resolution TEM image of the nanowire section near the catalyst, in which the nanowire has the wurtzite structure with many planar defects along its axial direction. Interestingly, the post-growth catalyst on the top of the nanowire has a faceted morphology. Selected area electron diffraction

(SAED) was used to determine the crystal structure of the catalyst, and Fig. 2(c) is such a SAED pattern taken from the catalyst in Fig. 2(b) (note that the diffraction pattern of the nanowire section was also included in the SAED pattern). Using the diffraction spots of the nanowire section as a reference (note that the lattice parameters of wurtzite structured InAs are known: a = 0.427 nm and c = 0.703 nm),24 the lattice spacing(s) of diffraction spots of the catalyst can be determined. Our detailed SAED analysis indicates that the faceted catalyst is the hexagonal-close- pack structured Au-In ξ phase with lattice parameters of a = 0.290 nm and c = 0.478 nm.25 To confirm the composition of the catalyst and its underling nanowire, EDS analysis was performed and results are shown in Fig. 2(d), in which the catalyst contains Au and In. Our quantitative analysis of the EDS results indicates that the nanowire is indeed InAs and the composition of the catalyst is ~16 at% In and ~84 at% Au, the latter is in excellent agreement with the composition of the Au-In ξ phase (note that the solubility range of Au-In ξ phase is between ~13 at% In and ~21 at% In at a temperature of 430 °C). Figure 2(e) is a typical high-resolution TEM image taken from the middle of the nanowire shown in Fig. 2(a), from which many planer defects can be witnessed. To confirm these structural characteristics of the catalysts and their underlying nanowires, we have examined over a dozen of thick nanowires (all with their diameter being ~40 nm or thicker), in which faceted catalysts were confirmed in all cases. Similar TEM investigations were performed on thin nanowires. Figure 3(a) shows a TEM image of a typical thin nanowire with a constant diameter of ~30 nm along the nanowire. Figure 3(b) is a high-resolution TEM image of the nanowire near the catalyst. Different to the thick nanowires, the thin nanowire has hemisphere shaped catalyst on its top. SAED was also used to determine the crystal structure of the hemisphere shaped catalyst, as shown in Fig. 3(c), in which the superimposed SAED pattern of the catalyst and its underlying nanowire was taken from the area shown in Fig. 3(b). Our detailed SAED analysis indicates that the hemisphere shaped catalyst is the hexagonal structured Au-In ф phase with lattice parameters of a = 0.456 nm and c = 0.907 nm (PDF 26-0710). To confirm the composition of the catalyst and its underlying nanowire, EDS analysis was performed and results are shown in Fig. 3(d), in which the nanowire contains In and As and the catalyst contains Au and In. The quantitative analysis of the EDS profiles indicates that the thin nanowire is InAs and the catalyst has a chemical composition ~39 at% In and

~61 at% Au. This EDS result is consistent with the composition of the Au-In ф phase. Figure 3(e) shows a typical high-resolution TEM image taken from middle of the nanowire shown in Fig. 3(a), in which defect-free wurtzite structured nanowire can be seen. By examining the structural quality of the entire nanowire for a dozen of thin nanowires (all with the nanowire diameters of ~30 nm or thinner), we found that all examined nanowires have the defect-free wurtzite structure and their catalysts have the hemisphere shape. Based on our detailed characterization, we found that the smaller catalysts induce defect-free thin InAs nanowires, while larger catalysts induce defected thick InAs nanowires. To understand these findings, two questions needed to be addressed. (1) Why different sizes of catalysts result in the catalysts having different In compositions and (2) why smaller catalysts induce defect-free InAs nanowires and why larger catalysts induce defected InAs nanowires? To address the first question, we need to firstly make sure no nanowire growth after the formal nanowire growth procedure (often referred as necking,e.g.26 formed during the cooling stage), so that the determined chemical compositions of catalysts in the grown nanowires reflect the true chemical compositions during the nanowire growth. Since necking adopts the zinc-blende structure,20, 27 it can be clearly distinguished if the nanowires have the wurtzite dominated structure. From Figs. 2(b) and 3(b), no zinc-blende structured sections are shown, indicating no nanowire growth during the cooling stage.28 Part of the reason is due to the strong thermodynamic affinity between Au and In,29,30 which makes In difficult to be precipitated out from the catalysts during the cooling.31,32 As a consequence, the measured In concentration in the post-growth catalysts should reflect to the true In concentration during the nanowire growth. Therefore, it is necessary to answer why larger catalysts contain less In than the smaller catalysts. In this regard, we believe that the In diffusion behavior in Au catalysts should play a key role in developing the ultimate In concentrations in Au catalysts. Since the surface to volume ratio increases rapidly with decreasing the catalyst size, the foreign elements should be much easier to be diffused into smaller catalysts than that into larger catalysts33 in order to reach relatively stable concentration. Under this circumstance, for smaller catalysts, the fast In diffusion in smaller Au catalyst will lead to a high In concentration, reflecting that this is a kinetically dominated process.

To address the second question, we note two factors. (1) Theoretical calculations15 predicted that wurtzite structure is stable in thin nanowires while zinc- blende structure is stable in thick nanowires because the total system energy of thin nanowires with the wurtzite structure is lower than those with zinc-blende structure, while the total system energy of thick nanowires shows an opposite case. (2) The In solubility in Au depends upon the nanoparticle sizes, that is, with decreasing the Au nanoparticle size, the In solubility in Au is significantly increased.34,35 Therefore, it is anticipated that a Au catalyst with a small size tends to absorb more In atoms from the ambient vapor and leads to a higher In supersaturation, while a Au catalyst with a larger size tends to absorb less In atoms and forms a lower In supersaturation. It is generally accepted that wurtzite nucleation is favored at the high supersaturation, while with the lower supersaturation situation, zinc-blende structure is preferred.16,17 Taking these two factors into account, nanowires catalyzed by smaller Au catalysts with a high In supersaturation tend to have the wurtzite structure, while nanowires catalyzed by larger Au catalysts with a low In supersaturation tend to develop at least some of zinc-blende structured sections when compared with the nanowires induced by smaller Au catalysts, as proposed by Joyce et al..14 Therefore, the latter leads to the formation of defected wurtzite structures. Based on the experimental results and our analysis outlined above, a growth model of Au induced InAs nanowires in MBE can be proposed, as schematically illustrated in Fig. 4. When a thin layer of Au deposited on the GaAs {111}B substrate is annealed under the As-rich ambient, this thin Au film is spontaneously broken down and agglomerated into Au nanoparticles of different sizes due to the Ostwald ripening effect.36,37 With the introduction of In source, In atoms diffuse into these nanoparticles to form Au-In alloys. As a result of the In solubility difference in Au nanoparticles with different sizes, smaller Au nanoparticles may quickly absorb more In to form Au-In catalysts with a high In concentration. While larger Au nanoparticles tend to absorb In with a relatively slow rate, so that these Au-In catalysts contain a lower In concentration. As a result, InAs nanowires induced by small catalysts with a high In supersaturation have defect-free wurtzite structure, while defected InAs nanowires induced by larger catalysts with a low In supersaturation, as demonstrated by the high-resolution TEM images shown in Fig. 4. It should be noted that the size difference between small and large Au nanoparticles prior to the InAs nanowire growth is more significant than that after the nanowire growth, due to the

fact that high In residue is found in smaller post-growth catalysts than that in larger post-growth catalysts as shown by our extensive TEM analyses. In conclusion, the structural quality of epitaxial InAs nanowires catalyzed by a Au thin film in MBE has been found to be dependent upon the size and composition of Au catalysts in a single growth. It was found that the original Au breakdown sizes from the Au thin film play a key role in determining the composition of the catalysts during the nanowire growth, and in turn the structural quality of grown InAs nanowires. The fact of enhanced In diffusion in smaller Au catalysts leads to the formation of high In concentrated catalysts that induce defect-free InAs nanowires, while relatively slow In diffusion in larger Au catalysts results in the formation of low In concentrated catalysts that induce defected InAs nanowires. This finding provides an insight into the realization of defect-free wurtzite InAs nanowires in MBE.

Acknowledgements

This study is financially supported by the Australian Research Council, the National Basic Research Program of China (Grant No. 2011CB925604), and the National Science Foundation of China (Grant No. 91021015,10990103 and 91121009). Australian Microscopy & Microanalysis Research Facility is also gratefully acknowledged for providing microscopy facilities for this study.

Figures Captions

FIG. 1. SEM image showing Au-catalyzed one-dimensional InAs nanowires. The inset is a magnified image of two nanowires with different diameters.

FIG. 2. TEM images of a thick InAs nanowire with a faceted catalyst. (a) A bright-field TEM image. (b) Nanowire top section with the faceted catalyst. (c) SAED pattern taken from the top of the nanowire and catalyst. (d) EDS spectra taken from the faceted catalyst and its underlying nanowire. (e) A typical high-resolution TEM image taken from the middle of InAs nanowire from marked region in (a).

FIG. 3. TEM images of a thin InAs nanowire with a hemisphere shaped catalyst. (a) A bright- field TEM image. (b) Nanowire top section with the hemisphere shaped catalyst. (c) SAED pattern taken from the top of the nanowire and catalyst. (d) EDS spectra taken from the catalyst and its underlying nanowire. (e) A typical high-resolution TEM image taken at the middle of nanowire, marked in (a).

FIG. 4. Schematic illustrating the growth mechanism of InAs nanowires induced by Au catalysts with different sizes and compositions. The smaller hemisphere catalysts contain a high In concentration that induce defect-free InAs nanowires, and the larger faceted catalysts contain a low In concentration that induce defected InAs nanowires.

References 1. Kim, Y.; Joyce, H. J.; Gao, O.; Tan, H. H.; Jagadish, C.; Paladugu, M.; Zou, J.; Suvorova, A. A. Influence of Nanowire Density on the Shape and Optical Properties of Ternary Ingaas Nanowires. Nano Lett. 2006, 6, 599-604. 2. Xia, H.; Lu, Z. Y.; Li, T. X.; Parkinson, P.; Liao, Z. M.; Liu, F. H.; Lu, W.; Hu, W. D.; Chen, P. P.; Xu, H. Y., et al. Distinct Photocurrent Response of Individual Gaas Nanowires Induced by N-Type Doping. ACS Nano 2012, 6, 6005-6013. 3. Wagner, R. S.; Ellis, W. C. Vapor-Liquid-Solid Mechanism of Single Crystal Growth ( New Method Growth Catalyst from Impurity Whisker Epitaxial + Large Crystal Si E ). Appl. Phys. Lett. 1964, 4, 89-90. 4. Moutanabbir, O.; Isheim, D.; Blumtritt, H.; Senz, S.; Pippel, E.; Seidman, D. N. Colossal Injection of Catalyst Atoms into Silicon Nanowires. Nature 2013, 496, 78-82. 5. Park, H. D.; Gaillot, A.; Prokes, S. M.; Cammarata, R. C. Observation of Size Dependent Liquidus Depression in the Growth of Inas Nanowires. J. Cryst. Growth 2006, 296, 159-164. 6. Dayeh, S. A.; Aplin, D. P. R.; Zhou, X.; Yu, P. K. L.; Yu, E. T.; Wang, D. High Electron Mobility Inas Nanowire Field-Effect Transistors. Small 2007, 3, 326-332. 7. Wright, S. L.; Marks, R. F.; Tiwari, S.; Jackson, T. N.; Baratte, H. In Situ Contacts to Gaas Based on Inas. Appl. Phys. Lett. 1986, 49, 1545-1547. 8. Milnes, A. G.; Polyakov, A. Y. Indium Arsenide: A Semiconductor for High Speed and Electro-Optical Devices. Mat. Sci. Eng. B 1993, 18, 237-259. 9. Thelander, C.; Martensson, T.; Bjork, M. T.; Ohlsson, B. J.; Larsson, M. W.; Wallenberg, L. R.; Samuelson, L. Single-Electron Transistors in Heterostructure Nanowires. Appl. Phys. Lett. 2003, 83, 2052-2054. 10. Doh, Y. J.; van Dam, J. A.; Roest, A. L.; Bakkers, E.; Kouwenhoven, L. P.; De Franceschi, S. Tunable Supercurrent through Semiconductor Nanowires. Science 2005, 309, 272-275. 11. Bjork, M. T.; Ohlsson, B. J.; Thelander, C.; Persson, A. I.; Deppert, K.; Wallenberg, L. R.; Samuelson, L. Nanowire Resonant Tunneling Diodes. Appl. Phys. Lett. 2002, 81, 4458-4460. 12. Zhou, X.; Dayeh, S. A.; Aplin, D.; Wang, D.; Yu, E. T. Direct Observation of Ballistic and Drift Carrier Transport Regimes in Inas Nanowires. Appl. Phys. Lett. 2006, 89, 053113.

13. Chuang, S.; Gao, Q.; Kapadia, R.; Ford, A. C.; Guo, J.; Javey, A. Ballistic Inas Nanowire Transistors. Nano Lett. 2013, 13, 555-558. 14. Joyce, H. J.; Wong-Leung, J.; Gao, Q.; Tan, H. H.; Jagadish, C. Phase Perfection in Zinc Blende and Wurtzite Iii-V Nanowires Using Basic Growth Parameters. Nano Lett. 2010, 10, 908-915. 15. Akiyama, T.; Sano, K.; Nakamura, K.; Ito, T. An Empirical Potential Approach to Wurtzite-Zinc-Blende Polytypism in Group Iii-V Semiconductor Nanowires. Jpn. J. Appl. Phys. 2006, 45, L275-L278. 16. Glas, F.; Harmand, J.-C.; Patriarche, G. Why Does Wurtzite Form in Nanowires of Iii-V Zinc Blende Semiconductors? Phys. Rev. Lett. 2007, 99, 146101. 17. Dubrovskii, V. G.; Sibirev, N. V.; Harmand, J. C.; Glas, F. Growth Kinetics and Crystal Structure of Semiconductor Nanowires. Phys. Rev. B 2008, 78, 235301. 18. Johansson, J.; Dick, K. A.; Caroff, P.; Messing, M. E.; Bolinsson, J.; Deppert, K.; Samuelson, L. Diameter Dependence of the Wurtzite-Zinc Blende Transition in Inas Nanowires. J. Phys. Chem. C 2010, 114, 3837-3842. 19. Xu, H. Y.; Guo, Y. N.; Liao, Z. M.; Sun, W.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Catalyst Size Dependent Growth of Pd-Catalyzed One-Dimensional Inas Nanostructures. Appl. Phys. Lett. 2013, 102, 203108. 20. Harmand, J. C.; Patriarche, G.; Pere-Laperne, N.; Merat-Combes, M. N.; Travers, L.; Glas, F. Analysis of Vapor-Liquid-Solid Mechanism in Au-Assisted Gaas Nanowire Growth. Appl. Phys. Lett. 2005, 87, 203101. 21. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Jiang, N.; Tan, H. H.; Jagadish, C.; Zou, J. High-Density, Defect-Free, and Taper-Restrained Epitaxial Gaas Nanowires Induced from Annealed Au Thin Films. Cryst. Growth Des. 2012, 12, 2018-2022. 22. Zhang, X.; Zou, J.; Paladugu, M.; Guo, Y. N.; Wang, Y.; Kim, Y.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C. Evolution of Epitaxial Inas Nanowires on Gaas (111)B. Small 2009, 5, 366-369. 23. Zou, J.; Paladugu, M.; Wang, H.; Auchterlonie, G. J.; Guo, Y. N.; Kim, Y.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C. Growth Mechanism of Truncated Triangular Iii-V Nanowires. Small 2007, 3, 389-393. 24. Kriegner, D.; Panse, C.; Mandl, B.; Dick, K. A.; Keplinger, M.; Persson, J. M.; Caroff, P.; Ercolani, D.; Sorba, L.; Bechstedt, F., et al. Unit Cell Structure of Crystal Polytypes in Inas and Insb Nanowires. Nano Lett. 2011, 11, 1483-1489.

25. Hiscocks, S. E. R.; Hume-Rothery, W. The Equilibrium Diagram of the System Gold-Indium. Proc. R. Soc. (London) Ser. A 1964, 282, 318-330. 26. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Defect-Free <110> Zinc-Blende Structured Inas Nanowires Catalyzed by Palladium. Nano Lett. 2012, 12, 5744-5749. 27. Persson, A. I.; Larsson, M. W.; Stenstrom, S.; Ohlsson, B. J.; Samuelson, L.; Wallenberg, L. R. Solid-Phase Diffusion Mechanism for Gaas Nanowire Growth. Nat. Mater. 2004, 3, 677-681. 28. Park, H. D.; Prokes, S. M.; Cammarata, R. C. Growth of Epitaxial Inas Nanowires in a Simple Closed System. Appl. Phys. Lett. 2005, 87, 063110. 29. Borg, B. M.; Dick, K. A.; Ganjipour, B.; Pistol, M.; Wernersson, L.; Thelander, C. Inas/Gasb Heterostructure Nanowires for Tunnel Field-Effect Transistors. Nano Lett. 2010, 10, 4080-4085. 30. Guo, Y. N.; Xu, H. Y.; Auchterlonie, G.; Burgess, T.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Shu, H. B.; Chen, X. S., et al. Phase Separation Induced by Au Catalysts in Ternary Ingaas Nanowires. Nano Lett. 2013, 13, 643-650. 31. Paladugu, M.; Zou, J.; Guo, Y. N.; Auchterlonie, G.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Kim, Y. Novel Growth Phenomena Observed in Axial Inas/Gaas Nanowire Heterostructures. Small 2007, 3, 1873-1877. 32. Paladugu, M.; Zou, J.; Guo, Y. N.; Zhang, X.; Kim, Y.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C. Nature of Heterointerfaces in Gaas/Inas and Inas/Gaas Axial Nanowire Heterostructures. Appl. Phys. Lett. 2008, 93, 101911. 33. Sra, A. K.; Schaak, R. E. Synthesis of Atomically Ordered Aucu and Aucu3 Nanocrystals from Bimetallic Nanoparticle Precursors. J. Am. Chem. Soc. 2004, 126, 6667-6672. 34. Zhang, G. Q.; Tateno, K.; Sanada, H.; Tawara, T.; Gotoh, H.; Nakano, H. Synthesis of Gaas Nanowires with Very Small Diameters and Their Optical Properties with the Radial Quantum-Confinement Effect. Appl. Phys. Lett. 2009, 95, 123104. 35. Han, N.; Wang, F. Y.; Hou, J. J.; Yip, S.; Lin, H.; Fang, M.; Xiu, F.; Shi, X. L.; Hung, T.; Ho, J. C. Manipulated Growth of Gaas Nanowires: Controllable Crystal Quality and Growth Orientations Via a Supersaturation-Controlled Engineering Process. Cryst. Growth Des. 2012, 12, 6243-6249.

36. Xu, H. Y.; Guo, Y. N.; Sun, W.; Liao, Z. M.; Burgess, T.; Lu, H. F.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Quantitative Study of Gaas Nanowires Catalyzed by Au Film of Different Thicknesses. Nanoscale Res. Lett. 2012, 7, 589. 37. Voorhees, P. W. The Theory of Ostwald Ripening. J. Stat. Phys. 1985, 38, 231- 252.

5 Structure and Quality Controlled Growth of InAs Nanowires through Catalyst Engineering

5.1 Introduction

In this study, the structure and quality controlled growth of InAs nanowires using Au catalysts in a molecular beam epitaxy reactor is presented. By tuning the indium concentration in the catalyst, defect-free wurtzite structured and defect-free zinc-

blende structured InAs nanowires can be induced. It is found that these defect-free zinc-blende structured InAs nanowires grow along < ̅ ̅0> directions with four low- energy {111} and two {110} side-wall facets and adopt the ( ̅ ̅ ̅) catalyst/nanowire interface. Our structural and chemical characterizations and calculations identify the existence of catalyst supersaturation threshold for the InAs nanowire growth. When the In concentration in the catalyst is sufficiently high, defect-free zinc-blende structured InAs nanowires can be induced. This study provides an insight into the manipulation of crystal structure and structure quality of III-V semiconductor nanowires through catalyst engineering.

5.2 Journal Publication

The results in Chapter 5 are included as it appears in Nano Research 2014, 7(11), 1640-1649. http://link.springer.com/article/10.1007%2Fs12274-014-0524-x

Structure and Quality Controlled Growth of InAs Nanowires through Catalyst Engineering

Zhi Zhang1, Zhen-Yu Lu2, Hong-Yi Xu1, Ping-Ping Chen2, Wei Lu2, and Jin Zou1,3()

1 Materials Engineering The University of Queensland St. Lucia Queensland 4072 Australia 2 National Laboratory for Infrared Physics Shanghai Institute of Technical Physics Chinese Academy of Sciences 500 Yu-Tian Road Shanghai 200083 People's Republic of China 3 Centre for Microscopy and Microanalysis The University of Queensland St. Lucia Queensland 4072 Australia

Address correspondence to Jin Zou, [email protected]

ABSTRACT

In this study, the structure and quality controlled growth of InAs nanowires using Au catalysts in a molecular beam epitaxy reactor is present. By tuning the indium concentration in the catalyst, defect-free wurtzite structured and defect-free zinc- blende structured InAs nanowires can be induced. It is found that these defect-free zinc-blende structured InAs nanowires grow along < ̅ ̅0> directions with four low- energy {111} and two {110} side-wall facets and adopt the ( ̅ ̅ ̅) catalyst/nanowire interface. Our structural and chemical characterizations and calculations identify the existence of catalyst supersaturation threshold for the InAs nanowire growth. When the In concentration in the catalyst is sufficiently high, defect-free zinc-blende structured InAs nanowires can be induced. This study provides an insight into the manipulation of crystal structure and structure quality of III-V semiconductor nanowires through catalyst engineering.

1. Introduction III-V semiconductor nanowires have attracted extensive attentions in the recent decade due to their distinct physical and chemical properties that can potentially lead to a wide range of applications in optoelectronics and nanoelectronics.1-3 Among various III-V nanowires, InAs nanowires have attracted special research interest due to their relatively high electron mobility,4 narrow band gap and small electron effective mass,5 which have made them a promising candidate for applications in high-performance transistors, Josephon junctions, and gas detection systems.6-7 Au-assisted nanowire growth is one of the most common methods to grow III-V nanowires via the vapor-liquid-solid (VLS) mechanism.8 For III-V nanowires, one of the inherent problems is that stacking fault and/or twin defects can easily be introduced in both wurtzite or zinc-blende structured nanowires due to the small energy differences between wurtzite and zinc-blende stacking sequences of their dense planes.9 For nanowires to be practically useful, it is critical to control their structural quality. However, it is a challenging task to control the nanowire structure regardless of the growth methods. Despite the recent progress in the phase perfection of wurtzite and/or zinc-blende structured III-V nanowires achieved on a few occasions,10-12 the growth of defect-free zinc-blende structured InAs nanowires by molecular beam epitaxy (MBE) has not yet been realized. In this regard, to grow wurtzite structured and zinc-blende structured InAs nanowires with high structural quality by tuning the catalyst composition is desired. It has been well documented that III-V nanowires are generally grown along the

<111>B directions for zinc-blende structured nanowires or along the <000 ̅> for the wurtzite structured nanowires, while other low-index growth directions have also 9, 13-17 been occasionally reported. Most of these non-<111>B or <000 ̅> nanowires are generally defect-free and adopt zinc-blende crystal structure. Therefore, controlling the growth direction opens a new possibility to obtain defect-free zinc- blende structured III-V nanowires. In this study, we explored the growth mechanism of InAs nanowires grown in an MBE reactor. Our investigations show that there exist catalyst supersaturation threshold for InAs nanowires grown in MBE reactor. By manipulating the V/III ratio, in turn tuning the catalyst composition, we demonstrated the epitaxial growth of defect- free wurtzite structured and defect-free zinc-blende structured InAs nanowires on

GaAs{111}B substrate. The zinc-blended structured nanowires are grown along < ̅ ̅0> directions through the formation of four low-energy {111} and two {110} side-wall facets and adopt the ( ̅ ̅ ̅) catalyst/nanowire interface.

2. Experimental

The InAs nanowires were grown on the GaAs {111}B substrates in a Riber 32 MBE system using Au as catalysts. The substrates were first degassed in the MBE preparation chamber for 15 min at 250 °C, then were transferred to the growth chamber to be thermally deoxidized for 20 min at 610 °C. After that, for each case, a thin GaAs buffer layer was grown on the GaAs (111)B substrate at 580 °C to achieve atomically flat surface. The substrate was then transferred back to the preparation chamber, where a thin Au film was deposited on the top of the GaAs buffer layer by vacuum thermal evaporation. The substrate was then transferred to the growth chamber, and annealed at 500 °C for 5 minutes under As ambient (3.6 x 10-6 Torr) to agglomerate the Au thin film into nanoparticles via Oswald ripening.18 This thin film generated Au nanoparticles is a simple and cost efficient approach. After the annealing, the substrate temperature was lowered to 430 °C and the Indium source was introduced to initiate InAs nanowire growth, in which two V/III ratios were selected to grow InAs nanowires (denoting sample A for the sample grown under a V/III ratio of 54 and sample B for the sample grown under a V/III ratio of 15). Furthermore, two additional samples grown under V/III ratios of 70 and 7 were carried out to clarify the nanowire growth window. The morphological characteristics of grown InAs nanowires and their bases were investigated by SEM (JEOL 7800F, operated at 10 kV). Their detailed structural and chemical characteristics, together with their catalysts were investigated by TEM (Philips Tecnai F20, operated at 200 kV, equipped with EDS for compositional analysis, and FEI Tecnai F30, operated at 300 kV). Individual nanowires for TEM analysis were removed from the substrates by ultrasonicating in ethanol and then dispersed onto holey carbon films supported by Cu grids. The cross-sectional TEM specimens were prepared first by mechanical thinning, followed by ion-beam thinning using a Gatan precision ion polishing system.

3. Results and Discussion

Figure 1 shows scanning electron microscopy (SEM) images taken from both samples. Figure 1(a) is a tilted SEM image of sample A, where most of the nanowires with varied heights19 show vertical growth with respect to the substrate. It is of interest to note that some triangular pyramids with catalysts on their tops can be observed, as shown in Fig. 1(a). Figure 1(b) is a typical top-view SEM image of sample B and shows both vertical and inclined nanowires. According to the fact that the cleavage planes of zinc-blende structured GaAs are {110} planes, the crystallographic orientation of the substrate can be determined, as illustrated in the inset of Fig. 1(b). From which, the projection of the inclined nanowires with respect to the GaAs {111}B substrate surface can be determined to be along <112> directions, suggesting that general growth directions of the inclined nanowires may be along the <11n> directions where n is an integer if the nanowires have the zinc-blende structure. In addition, for each inclined nanowire, a triangular pyramid base, similar to those shown in sample A (refer to Fig. 1(a)), can be observed in sample B. Therefore, the bottom edges of the pyramids can be determined to be parallel to <110> directions on the substrate surface. Based on this crystallographic information, the facets of the triangular pyramids in both samples can be determined as {11m} planes with m being an integer. Figure 1(c) is a ~35° tilted (with respect to the top view) SEM image, in which a typical inclined nanowire is viewed edge-on, and shows pronounced tapering morphology.20-21 Through careful crystallographic analysis, the growth direction of this nanowire can be determined as a < ̅ ̅0> direction. In addition, the hexagonal-shape faceted side walls of this < ̅ ̅ 0> nanowire can be clearly observed. Taking the cleavage planes of the GaAs substrate as a reference, the side facets of < ̅ ̅0> nanowires can also be determined as four {111} planes and two {110} planes, as illustrated in Fig. 1(d).

Figure 1 SEM investigations of InAs nanowires from sample A and B. (a) Tilted SEM image of sample A. (b) Top-view SEM image of sample B. The inset is the sketch of the orientation relationship between <112> and <110> directions on the (111)B substrate, solid red arrow and dashed blue arrow representing <112> and <110> directions, respectively. (c) 35° tilted SEM image showing a typical inclined nanowire, edged-on, with facet details (the tilt axis is <1 ̅0>). (d) Schematic illustration of facets of the inclined nanowire.

As can be seen from Fig. 1, vertical nanowires are observed in both samples. To understand their structural and chemical characteristics, TEM investigations were performed on over a dozen of nanowires from each sample. Figures 2(a) and 2(d) show bright-field transmission electron microscopy (TEM) images of a typical vertical nanowire from each of the two samples, respectively, where no tapering was observed in both cases. Figures 2(b) and 2(e) are their corresponding high- resolution TEM images of the nanowire sections near their catalysts, in which both nanowires have the wurtzite structure and diameters of ~25 nm. It is of interest to note that planar defects can be observed in sample A, while no planar defects can be seen from the nanowires taken from sample B, by investigating entire lengths of a

dozen of nanowires. To understand this structure difference, X-ray energy dispersive spectroscopy (EDS) analysis was performed on these nanowires and their catalysts, and the results are shown respectively in Figs. 2(c) and 2(f). Our quantitative EDS analysis on several nanowires from each sample indicates that the nanowires are indeed InAs as anticipated for both cases, but the indium concentrations in the catalysts are 20-25 at.% for sample A and 35-40 at.% for sample B, respectively. It has been well documented that the indium concentration in the catalysts is determined by the indium partial pressure in the surrounding vapor through direct deposition of In atoms to the catalysts22 and by the surface diffusion of In atoms to the catalysts.23 Since, during the MBE growth, only the growth species were introduced in the MBE growth chamber, the practical nominal V/III ratio can be used to establish the indium partial pressure in the vapor, and in turn the indium concentration in the catalyst. Therefore, during our nanowire growth, the indium concentration in the catalysts should be lower in sample A (the high V/III ratio with a relatively low III concentration in the vapor) than that in sample B (the low V/III ratio with a relatively high III concentration in the vapor). On the other hand, since the III diffusion length of group III elements diminishes with increasing the V/III ratio,24 a high V/III ratio leads to less In atoms diffusing to the nanowire growth front, resulting in a low In concentration in the catalysts. Accordingly, both direct deposition and surface diffusion of In atoms result in a low In concentration in the catalysts under a high V/III ratio. Since the Au-In catalyst supersaturation increases with increasing the indium concentration in the catalyst,25-26 the catalyst supersaturation in sample B with a higher indium concentration must be higher than that in sample A with a lower indium concentration. It is well accepted that the wurtzite nucleation is favored at high supersaturation, while with the lower supersaturation, zinc-blende structure is preferred.25, 27 In addition, according to theoretical prediction and experimental results,28 InAs nanowires, especially with smaller diameters, prefer to adopt the wurtzite structure. Based on these analyses, nanowires in sample B grown at a low V/III ratio tend to have a high indium concentration in the catalysts and in turn prefers to grow pure wurtzite structure, while nanowires in sample A grown at a high V/III ratio tend to have a low indium concentration in the catalysts and thus lead to the formation of defected wurtzite structure.

Figure 2 TEM characterizations of vertical nanowires from samples A and B. (a,d) Bright- field TEM images of typical nanowires from samples A and B, respectively. (b,e) Corresponding high-resolution TEM images taken from the top of both nanowires, and the insets are the diffraction patterns taken along <11 ̅ > direction. (c,f) Corresponding EDS spectra taken from the catalysts and nanowires of both cases.

As outlined earlier, inclined nanowires were grown simultaneously on GaAs {111}B substrate in sample B. To fundamentally understand these inclined nanowires, detailed TEM investigation was also performed to determine their structural and chemical characteristics. Figure 3(a) is a bright-field TEM image of a typical < ̅ ̅0> nanowire. Clearly, the nanowire has a tapered morphology with a thin top and thick bottom (caused by the lateral growth of nanowires), which is in excellent agreement with our SEM observations. Since the catalyst/nanowire interface is crucial in determining the nanowire growth, it is essential to investigate the catalyst/nanowire interface. In this regard, a set of TEM images and their corresponding selective area electron diffraction (SAED) patterns were obtained by tilting the nanowire along its growth axis, defined as [ ̅ ̅0] in this case. Figures 3(b) and 3(c) show bright-field TEM images near the top of the nanowire with catalyst diameter of ~ 25 nm, and Figs. 3(e) and 3(f) show their corresponding SAED patterns, taken respectively along

the [ ̅ 1 ̅ ] and [1 ̅ 0] zone axes. According to these SAED patterns, the crystal structure of the inclined nanowire can be determined to be zinc-blende structure, which is distinctly different to the wurtzite structure of those vertical nanowires in both samples. Through careful investigation of these TEM images, a sharp catalyst/nanowire interface can be witnessed, as shown in Fig. 3(c), and the high- resolution TEM image of the interface is shown in Fig. S1 in the Electronic Supplementary Material (ESM). By carefully correlating Figs. 3(c) and 3(f), this interface can be determined as the ( ̅ ̅ ̅) plane since the interface normal is parallel to the ̅ ̅ ̅*, the diffraction spot shown in Fig. 3(f) (note that orientations of bright-field TEM images and their corresponding SAED patterns are accurately aligned). Our extensive investigations on a dozen of the inclined < ̅ ̅0> nanowires confirmed that this ( ̅ ̅ ̅ ) catalyst/nanowire interface commonly exists. To understand the composition of the catalyst and its underlying nanowire, EDS analysis was employed and the results are shown in Fig. 3(d), in which the catalyst contains Au and In. Our quantitative EDS analysis of a number of inclined nanowires shows the nanowires are indeed InAs and the indium concentration in their catalysts is in the range of 35- 40 at.%, similar to the case of vertical nanowires in sample B. To examine the structural quality of < ̅ ̅ 0> nanowires, high-resolution TEM investigation was performed. Figure 3(g) is a high-resolution TEM image taken from the middle of the nanowire, and confirms that the nanowire has the zinc-blende structure and shows no lattice defects. Our extensive high-resolution TEM studies on over a dozen of < ̅ ̅0> nanowires confirmed this defect-free zinc-blende structure nature. In general, Au-catalyzed III-V nanowires often show randomly distributed planar defects along its growth direction, such as stacking faults, rotational twin planes or polytypic structures,29 as a result of the small energetic differences for hexagonal or cubic stacking sequences along the <0001>/<111> directions.9 Since the lattice defects can be deleterious to the performance of nanowire-based devices, our defect-free < ̅ ̅0> InAs nanowires would lead to superior improvement of device performance.

100

Figure 3 Detailed TEM investigations of a typical < ̅ ̅0> InAs nanowire. (a) TEM image showing its overview. (b, c) TEM images showing the top region, viewed along the [ ̅1 ̅] and [1 ̅0] zone axis, respectively. (e, f) Corresponding SAED patterns. (d) EDS spectra taken from the catalyst and its underlying nanowire. (g) High-resolution TEM image taken at the middle of the nanowire.

As shown in Fig. 1(a), some triangular pyramids with catalysts on their tops can be observed. To understand why such catalysts cannot induce the nanowire growth, TEM investigation was performed on the cross-sectional TEM specimen of sample A. Figure 4(a) is a typical low-magnification cross-sectional <110> TEM image, and shows a catalyst surrounded by non-flatted surface, agreed well with the pyramid- like base observed above by SEM (refer to Fig. 1(a)). Figure 4(b) is a <110> high- resolution TEM image taken from the catalyst region, in which a hemisphere-shaped catalyst can be observed, under which zinc-blende structured base can be witnessed. It can be noted that the projected angles of the pyramid are asymmetrical when viewing in a <110> zone axis. From Fig. 4(b), the projected angles α ~ 16° and β ~ 30° can be measured. By carefully crystallographic analysis and incorporated with

101

SEM investigations, facets of the triangular pyramid surfaces can be determined as { ̅ ̅ ̅} planes. The chemical composition of the catalyst as well as the underlying pyramid was investigated by EDS, as shown in Fig. 4(c), in which the pyramid contains In and As and the catalyst contains In and Au. The quantitative analysis of the EDS profiles indicates that the pyramid is InAs and the indium concentration in the catalyst is ~10 at.% only, much lower than those in the catalysts of the vertically grown nanowires in sample A. This experimental result might indicate that, for these catalysts with underlying triangular pyramid bases, their indium concentration have not reached the supersaturation threshold required for the nanowire growth. As a consequence, they cannot induce the nanowire growth. However, we cannot rule out the possibility that, during the growth and/or cooling, some catalysts with high In concentration might have lost the catalytic activity and may not be able to induce nanowire growth. Interestingly, through carefully studying the interfaces of the catalysts that did not induce the nanowire growth and their underlying InAs, we found that the interfaces are not sharp as for the catalyst/nanowire interfaces (refer to Fig. 4(b)). We anticipate that this could be due to the interdiffusion, and this non-sharp interfaces could be the other reason to the failure of inducing InAs nanowire growth since the formation of catalyst/nanowire interface is critical to the nanowire growth.30 Therefore, when the catalysts are reactive and form the sharp interfaces between catalysts and their underlying materials, the nanowire growth can be triggered. Otherwise, the catalysts would remain stationary.

102

Figure 4 TEM characterizations of the pyramid on the substrate. (a) Cross-sectional <110> TEM image of sample A showing a catalyst with non-flatted surface. (b) Corresponding high- resolution TEM image showing zinc-blend structured base. (c) EDS spectra of the catalyst and its underlying base.

Base on the experimental results and discussion outlined above, the growth model of Au-induced InAs nanowires under different V/III ratios in MBE can be schematically illustrated in Fig. 5. With the introduction of indium source, the Au nanoparticles developed during the annealing process will capture indium from the vapor to form Au-In alloy droplets (as shown in Fig. 5(a)). With a high V/III ratio, only limited indium in the vapor can be incorporated into the catalysts with different catalysts experienced different indium incorporation. For those catalysts that are not reactive and do not form the sharp interfaces between catalysts and their underlying materials, they remain stationary on the substrate surface, but attracted a large amount of indium through surface diffusion and in turn form pyramid bases, as illustrated in Fig. 5(b). When the catalysts are reactive and sharp catalyst/nanowire interfaces are formed, vertical nanowires with the wurtzite structure containing lattice defects can be grown, as illustrated in Fig. 5(c). On the other hand, with a low V/III ratio, a large amount of indium in the vapor can be incorporated into the catalysts, leading to the indium concentration in the catalysts well over their supersaturation

103

threshold, so that nanowires can be grown. In our case, both vertically grown defect- free wurtzite structured nanowires and inclined defect-free zinc-blende structured < ̅ ̅0> nanowires are grown, as illustrate respectively in Figs. 5(d) and 5(e). It should be noted that the growth window of InAs nanowires in MBE is very narrow as no nanowires can be grown under the V/III ratios of 70 and 7 (SEM results of these are shown in the Fig. S2 in the ESM).

Figure 5 Schematic diagram of InAs nanowires growth under different V/III ratios. (a) Annealed substrate with droplets on the surface. (b,c) Growth under high V/III ratio. (d,e) Growth under low V/III ratio. The red lines indicate stacking faults found in the nanowires.

To clarify these extraordinary observations, two issues need to be addressed. (1) Why is the formation of < ̅ ̅0> InAs nanowires possible, and (2) why were only <000 ̅> nanowires induced under high V/III ratio, and why can both <000 ̅> and < ̅ ̅0> nanowires be triggered under low V/III ratio? To address the first issue, we note two factors. (i) The fact that the catalyst/nanowire interface of the < ̅ ̅0> InAs nanowire is ( ̅ ̅ ̅) plane (the lowest interfacial energy in the zinc-blende structure) suggests that the nanowire nucleation should be dominated by the energy minimization of the catalyst/nanowire interfacial energy. (ii) The sidewall facets of our < ̅ ̅0> InAs nanowires contains four {111} planes and two {110} planes, all with low surface energies, indicating that the nanowire growth is governed by the minimizing surface energy of side facets.31 It is of interest to note that this faceted

104

configuration is different to those reported hexagonal facet of < ̅ ̅0> nanowires with four {111} planes and two {001} planes.17, 32 The fact that the surface energy of {110} planes is lower than that of {001} planes33 indicates that the faceted configuration of our < ̅ ̅0> nanowires must be more energetically favorable. With such an interesting facet configuration, the surface energy density 훶 of our < ̅ ̅0> nanowires can be determined. Under the assumption that all the facet planes have approximately the same dimension, we can define a geometrical parameter,34 q, to represent the percentage of {111} facets. Therefore, 훶 can be expressed as:

훶 훶 훶 , (1) where q is 4/6 in our case, 훶 is the surface energy of the {hkl} planes, in which 2 2 33 훶 = 0.84 J/m and 훶 = 1 J/m can be obtained from Ref. . Accordingly, 2 2 훶 ̅ ̅ = ~ 0.89 J/m can be determined, which is comparable to 훶 ̅ is 0.91 J/m for the vertically grown <000 ̅ > nanowires with six { ̅ 00} side facets. This calculation indicates the energetic possibility for the growth of our zinc-blende structured < ̅ ̅0> nanowires. To address the second issue, we consider the nucleation kinetics of VLS nanowire growth,27 and this nucleation kinetics has successfully explained several nanowire growth models.35-37 According to the nucleation kinetics model schematically shown in Fig. 6, the nucleation energy barrier ΔG* for the creation of a nucleus can be 37 calculated as:

ΔG* , (2) where A is the base area of the nucleus, h is nucleus height, P is nucleus perimeter, Δµ is chemical potential difference between the supersaturated catalyst and solid nucleus, and 훤 is effective lateral surface energy, which can be expressed as:37

훤 훶 , (3)

where x is the fraction of the nucleus perimeter in contact with the vapor, 훶LS is the liquid-solid interface energy, 훶SV is the solid-vapor surface energy of the side facet of the nucleus in contact with the vapor, 훶LV is the liquid-vapor surface energy, θ is the tilt angle of side facet, δ = β – θ – π/2 is the contact angle between the droplet and side facet, and β is the contact angle between the droplet and the interface between the catalyst and the nanowire. Since cosδ = cos(β – θ – π/2) = sin(β – θ) = sinβ cosθ – cosβ sinθ, the effective surface energy can be rewritten as

105

훤 훶 훶 훶 ), (4)

By examining Eq. (4), is appeared in the terms of 훶 , so that they need to be assessed when comparing 훤 for nuclei with different facet geometry under the same growth condition (maintaining other terms the same). In this regard, it can be determined that the effective surface energy 훤 of vertical nucleus with =1 is lower than that of inclined nucleus with an inclined angle θ (leading to < 1).

Meanwhile, the term for <000 ̅> nanowire can be estimated as ~3.46 (using p = 6

L and A = 3√ /2 with L being the facet side length), which is lower than that for < ̅ ̅0> nanowires (~4.31). Therefore, it is expected that the nucleation energy barrier ΔG* for the vertical <000 ̅> nanowires is lower than that of inclined < ̅ ̅0> nanowire, leading to only the growth of <000 ̅> nanowires under a high V/III ratio, as observed in our experiments.

Figure 6 Schematic showing the nucleation model at the triple phase line during VLS growth. (a) Nucleation with vertical side facet. (b) Nucleation with inclined side facet.

Under a lower V/III ratio, following facts can be noted. (i) The In concentration in the catalysts of sample B is much higher than that of sample A. The high In concentration in the catalyst denotes high In chemical potential in the catalysts and hence a high chemical potential difference Δµ between the catalyst and the solid nucleus is anticipated. (ii) The liquid-vapor surface energy 훶LV increases with increasing the In partial pressure, corresponding to an increasing In concentration in the catalysts. This fact can be understood as, when In atoms are absorbed by Au droplets, the number of surface dangling bonds per unit area at the liquid-vapor 38 interface increases, and in turn the liquid-vapor surface energy 훶LV increases,

106

leading to the increase of the effective lateral surface energy 훤 when other growth parameters remain the same according to Eq. (3). Therefore, by decreasing the V/III ratio (increasing the Δµ and decreasing the 훤), decreasing ΔG*<110> can be achieved according to Eq. (4), which promotes the growth of < ̅ ̅0> nanowires under the nanowire growth with a low V/III ratio. For the zinc-blende structured < ̅ ̅ 0> nanowires, the introduction of wurtzite structure or stacking fault requires the formation of new facets or/and the change of nanowire growth direction, which both requires extra energy. Therefore, it is anticipated that defect-free zinc-blende structure can be realized in thin nanowires.

4. Conclusions

In conclusion, we demonstrate the structure and quality controlled growth of InAs nanowires through tuning catalyst compositions, from which we, for the first time, realize the growth of defect-free zinc-blende InAs nanowires in MBE. Our direct evidences show that there exists threshold for inducing InAs nanowire growth. When catalysts are reactive and sharp catalyst/nanowire interfaces are formed, vertically grown <000 ̅> wurtzite structured InAs nanowires with planar defects can be grown on the GaAs {111}B substrate. With increasing the In concentration in the Au catalysts, a high supersaturation threshold is reached, inducing the growth of defect- free wurtzite structured and defect-free zinc-blende structured < ̅ ̅0> InAs nanowires. The fact that these < ̅ ̅0> nanowires adopt ( ̅ ̅ ̅) catalyst/nanowire interface and have four low-energy {111} and two {110} side-wall facets suggests that the growth of these unusual III-V semiconductor nanowires are governed by the thermodynamics. This study provides a strategy to grow defect-free wurtzite structured and defect-free zinc-blende structured III-V semiconductor nanowires in MBE.

Acknowledgements

107

Australian Microscopy & Microanalysis Research Facility is also gratefully acknowledged for providing microscopy facilities for this study.

Electronic Supplementary Material: supplementary material (high-resolution TEM image of the catalyst/nanowire interface from <110> nanowire, SEM images of samples grown at V/III ratios of 70 and 7) is available in the online version of this article at http://dx.doi.org/10.1007/s12274-***-****-* (automatically inserted by the publisher).

108

Electronic Supplementary Material

Structure and Quality Controlled Growth of InAs Nanowires through Catalyst Engineering

Zhi Zhang1, Zhen-Yu Lu2, Hong-Yi Xu1, Ping-Ping Chen2, Wei Lu2, and Jin Zou1,3()

Address correspondence to Jin Zou, [email protected]

TEM Investigations of the defect-free zinc-blende <110> nanowire Figure S1 shows the TEM investigations of the <110> nanowire. It has been found that the <110> nanowire have defect-free zinc-blende and possess the {111} catalyst/nanowire interface. Figure S1(c) is a high resolution TEM image taken along the <110> zone axis from the top of the nanowire, from which a sharp interface between the catalyst and nanowire can be clearly observed.

SEM images of InAs growth under V/III ratios of 70 and 7 Figure S2 shows the tilted SEM images of InAs growth under V/III ratios of 70 and 7, respectively, from which we can observe that no nanowires are induced by the Au catalyst. These results indicate that the growth window of InAs nanowires in MBE is very narrow.

109

Figure S1 (a) The TEM overview of nanowire. (b) Bright-field TEM image taken from the nanowire top. (c) High-resolution TEM image taken along <110> zone axis showing the catalyst/nanowire interface.

110

Figure S2 (a) Tilted SEM image of InAs growth under V/III ratio of 70. (b) Tilted SEM image of InAs growth under V/III ratio of 7.

111

References 1. Duan, X. F.; Huang, Y.; Cui, Y.; Wang, J. F.; Lieber, C. M. Indium Phosphide Nanowires as Building Blocks for Nanoscale Electronic and Optoelectronic Devices. Nature 2001, 409, 66-69. 2. Xia, H.; Lu, Z. Y.; Li, T. X.; Parkinson, P.; Liao, Z. M.; Liu, F. H.; Lu, W.; Hu, W. D.; Chen, P. P.; Xu, H. Y., et al. Distinct Photocurrent Response of Individual Gaas Nanowires Induced by N-Type Doping. ACS Nano 2012, 6, 6005-6013. 3. Thelander, C.; Caroff, P.; Plissard, S. b.; Dey, A. W.; Dick, K. A. Effects of Crystal Phase Mixing on the Electrical Properties of Inas Nanowires. Nano Lett. 2011, 11, 2424-2429. 4. Dayeh, S. A.; Yu, E. T.; Wang, D. Excess Indium and Substrate Effects on the Growth of Inas Nanowires. Small 2007, 3, 1683-1687. 5. Milnes, A. G.; Polyakov, A. Y. Indium Arsenide: A Semiconductor for High Speed and Electro-Optical Devices. Mat. Sci. Eng. B 1993, 18, 237-259. 6. Chuang, S.; Gao, Q.; Kapadia, R.; Ford, A. C.; Guo, J.; Javey, A. Ballistic Inas Nanowire Transistors. Nano Lett. 2013, 13, 555-558. 7. Offermans, P.; Crego-Calama, M.; Brongersma, S. H. Gas Detection with Vertical Inas Nanowire Arrays. Nano Lett. 2010, 10, 2412-2415. 8. Wagner, R. S.; Ellis, W. C. Vapor-Liquid-Solid Mechanism of Single Crystal Growth ( New Method Growth Catalyst from Impurity Whisker Epitaxial + Large Crystal Si E ). Appl. Phys. Lett. 1964, 4, 89-90. 9. Krishnamachari, U.; Borgstrom, M.; Ohlsson, B. J.; Panev, N.; Samuelson, L.; Seifert, W.; Larsson, M. W.; Wallenberg, L. R. Defect-Free Inp Nanowires Grown in [001] Direction on Inp(001). Appl. Phys. Lett. 2004, 85, 2077-2079. 10. Joyce, H. J.; Wong-Leung, J.; Gao, Q.; Tan, H. H.; Jagadish, C. Phase Perfection in Zinc Blende and Wurtzite Iii-V Nanowires Using Basic Growth Parameters. Nano Lett. 2010, 10, 908-915. 11. Shtrikman, H.; Popovitz-Biro, R.; Kretinin, A.; Heiblumf, M. Stacking-Faults-Free Zinc Blende Gaas Nanowires. Nano Lett. 2009, 9, 215-219. 12. Dick, K. A.; Caroff, P.; Bolinsson, J.; Messing, M. E.; Johansson, J.; Deppert, K.; Wallenberg, L. R.; Samuelson, L. Control of Iii-V Nanowire Crystal Structure by Growth Parameter Tuning. Semicond. Sci. Tech. 2010, 25, 024009. 13. Ikejiri, K.; Ishizaka, F.; Tomioka, K.; Fukui, T. Bidirectional Growth of Indium Phosphide Nanowires. Nano Lett. 2012, 12, 4770-4.

112

14. Bjork, M. T.; Ohlsson, B. J.; Sass, T.; Persson, A. I.; Thelander, C.; Magnusson, M. H.; Deppert, K.; Wallenberg, L. R.; Samuelson, L. One-Dimensional Heterostructures in Semiconductor Nanowhiskers. Appl. Phys. Lett. 2002, 80, 1058- 1060. 15. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Defect-Free <110> Zinc-Blende Structured Inas Nanowires Catalyzed by Palladium. Nano Lett. 2012, 12, 5744-5749. 16. Ghosh, S. C.; Kruse, P.; LaPierre, R. R. The Effect of Gaas(100) Surface Preparation on the Growth of Nanowires. Nanotechnology 2009, 20, 115602. 17. Wu, Z. H.; Mei, X.; Kim, D.; Blumin, M.; Ruda, H. E.; Liu, J. Q.; Kavanagh, K. L. Growth, Branching, and Kinking of Molecular-Beam Epitaxial < 110 > Gaas Nanowires. Appl. Phys. Lett. 2003, 83, 3368-3370. 18. Voorhees, P. W. The Theory of Ostwald Ripening. J. Stat. Phys. 1985, 38, 231- 252. 19. Zhang, Z.; Lu, Z. Y.; Chen, P. P.; Xu, H. Y.; Guo, Y. N.; Liao, Z. M.; Shi, S. X.; Lu, W.; Zou, J. Quality of Epitaxial Inas Nanowires Controlled by Catalyst Size in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 103, 073109. 20. Zou, J.; Paladugu, M.; Wang, H.; Auchterlonie, G. J.; Guo, Y. N.; Kim, Y.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C. Growth Mechanism of Truncated Triangular Iii-V Nanowires. Small 2007, 3, 389-393. 21. Guo, Y. N.; Burgess, T.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Polarity- Driven Nonuniform Composition in Ingaas Nanowires. Nano Lett. 2013, 13, 5085- 5089. 22. Guo, Y. N.; Xu, H. Y.; Auchterlonie, G.; Burgess, T.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Shu, H. B.; Chen, X. S., et al. Phase Separation Induced by Au Catalysts in Ternary Ingaas Nanowires. Nano Lett. 2013, 13, 643-650. 23. Jensen, L. E.; Bjork, M. T.; Jeppesen, S.; Persson, A. I.; Ohlsson, B. J.; Samuelson, L. Role of Surface Diffusion in Chemical Beam Epitaxy of Inas Nanowires. Nano Lett. 2004, 4, 1961-1964. 24. Dayeh, S. A.; Yu, E. T.; Wang, D. Iii-V Nanowire Growth Mechanism: V/Iii Ratio and Temperature Effects. Nano Lett. 2007, 7, 2486-2490. 25. Johansson, J.; Dick, K. A.; Caroff, P.; Messing, M. E.; Bolinsson, J.; Deppert, K.; Samuelson, L. Diameter Dependence of the Wurtzite-Zinc Blende Transition in Inas Nanowires. J. Phys. Chem. C 2010, 114, 3837-3842.

113

26. Schwarz, K. W.; Tersoff, J. Elementary Processes in Nanowire Growth. Nano Lett. 2011, 11, 316-320. 27. Glas, F.; Harmand, J.-C.; Patriarche, G. Why Does Wurtzite Form in Nanowires of Iii-V Zinc Blende Semiconductors? Phys. Rev. Lett. 2007, 99, 146101. 28. Akiyama, T.; Sano, K.; Nakamura, K.; Ito, T. An Empirical Potential Approach to Wurtzite-Zinc-Blende Polytypism in Group Iii-V Semiconductor Nanowires. Jpn. J. Appl. Phys. 2006, 45, L275-L278. 29. Caroff, P.; Dick, K. A.; Johansson, J.; Messing, M. E.; Deppert, K.; Samuelson, L. Controlled Polytypic and Twin-Plane Superlattices in Iii-V Nanowires. Nature Nanotechnology 2009, 4, 50-55. 30. Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Kim, Y.; Zhang, X.; Guo, Y. N.; Zou, J. Twin-Free Uniform Epitaxial Gaas Nanowires Grown by a Two-Temperature Process. Nano Lett. 2007, 7, 921-926. 31. Schmidt, V.; Senz, S.; Gosele, U. Diameter-Dependent Growth Direction of Epitaxial Silicon Nanowires. Nano Lett. 2005, 5, 931-935. 32. Ma, D. D. D.; Lee, C. S.; Au, F. C. K.; Tong, S. Y.; Lee, S. T. Small-Diameter Silicon Nanowire Surfaces. Science 2003, 299, 1874-1877. 33. Sibirev, N. V.; Timofeeva, M. A.; Bol'shakov, A. D.; Nazarenko, M. V.; Dubrovskii, V. G. Surface Energy and Crystal Structure of Nanowhiskers of Iii-V Semiconductor Compounds. Phys. Solid State 2010, 52, 1531-1538. 34. Cai, Y.; Chan, S. K.; Sou, I. K.; Chan, Y. T.; Su, D. S.; Wang, N. The Size- Dependent Growth Direction of Znse Nanowires. Adv. Mater. 2006, 18, 109-114. 35. Dubrovskii, V. G.; Sibirev, N. V.; Harmand, J. C.; Glas, F. Growth Kinetics and Crystal Structure of Semiconductor Nanowires. Phys. Rev. B 2008, 78, 235301. 36. Algra R. E.; Verheijen M. A.; Feiner L.-F.; Immink G. G. W.; Theissmann R.; van Enckevort W. J. P.; Vlieg E.; Bakkers E. P. A. M. Paired Twins and {112 Morphology in Gap Nanowires. Nano Lett. 2010, 10, 2349-2356. 37. Wang, J.; Plissard, S. R.; Verheijen, M. A.; Feiner, L.-F.; Cavalli, A.; Bakkers, E. P. A. M. Reversible Switching of Inp Nanowire Growth Direction by Catalyst Engineering. Nano Lett. 2013, 13, 3802-3806. 38. Algra, R. E.; Verheijen, M. A.; Feiner, L.-F.; Immink, G. G. W.; van Enckevort, W. J. P.; Vlieg, E.; Bakkers, E. P. A. M. The Role of Surface Energies and Chemical Potential During Nanowire Growth. Nano Lett. 2011, 11, 1259-1264.

114

6 Catalyst Orientation Induced Growth of Defect-Free Zinc-Blende Structured <00 ̅> InAs Nanowires

6.1 Introduction

In this study, we demonstrate the epitaxial growth of <00 ̅> defect-free zinc-blende structured InAs nanowires on GaAs {111}B substrate using Au catalysts in molecular

115

beam epitaxy. It has been found that the catalysts and their underlying <00 ̅ > nanowires have the orientation relationship of { ̅ }C//{ ̅}InAs and [ ̅ ]C//[1 ̅ ]InAs due to their small in-plane lattice mismatches between their corresponding lattice spacings perpendicular to the { ̅ atomic planes of the nanowires, leading to the formation of the { ̅ catalyst/nanowire interfaces, and consequently the formation of <00 ̅> nanowires. This study provides a practical approach to manipulate the crystal structure and structural quality of III-V nanowires through carefully control the crystal phase of the catalysts.

6.2 Journal Publication

The results in Chapter 6 are included as it appears in Nano Letters 2015, 15, 876- 882. http://pubs.acs.org/doi/abs/10.1021/nl503556a

116

Catalyst Orientation Induced Growth of Defect-Free Zinc-Blende Structured <00 ̅> InAs Nanowires

Zhi Zhang,† Kun Zheng,‡,§ Zhen-Yu Lu,# Ping-Ping Chen,# Wei Lu,# and Jin Zou*,†,‡

† Materials Engineering, The University of Queensland, St. Lucia, QLD 4072, Australia ‡ Centre for Microscopy and Microanalysis, The University of Queensland, St. Lucia, QLD 4072, Australia § Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, St. Lucia, QLD 4072, Australia # National Laboratory for Infrared Physics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, 500 Yu-Tian Road, Shanghai 200083, China

ABSTRACT

In this study, we demonstrate the epitaxial growth of <00̅ > defect-free zinc- blende structured InAs nanowires on GaAs {111}B substrate using Au catalysts in molecular beam epitaxy. It has been found that the catalysts and their underlying <00 ̅ > nanowires have the orientation relationship of

{ ̅ }C//{ ̅ }InAs and [̅ ]C//[1̅ ]InAs due to their small in-plane lattice mismatches between their corresponding lattice spacings perpendicular to the { ̅ atomic planes of the nanowires, leading to the formation of the { ̅ catalyst/nanowire interfaces, and consequently the formation of <00 ̅ > nanowires. This study provides a practical approach to manipulate the crystal structure and structural quality of III-V nanowires through carefully control the crystal phase of the catalysts.

KEYWORDS: Epitaxial growth, defect-free, InAs nanowires, MBE

*Address correspondence to: [email protected]

117

One-dimensional nanowires made of III-V compound semiconductors have attracted significant attention in the recent decade due to their distinct physical and chemical properties that can potentially lead to a wide range of applications in the nanoelectronics and optoelectronics.1-4 As a key class of III- V semiconductor nanowires, InAs nanowires have attracted special research interest due to their narrow band gap, relatively high electron mobility and small electron effective mass,5,6 which have made them a promising candidate for applications in Josephson junctions, gas detection systems, and high- performance transistors.7-9 It has been well documented that III-V nanowires are generally grown along the <111>B directions for zinc-blende structured nanowires or along the <000̅ > for the wurtzite structured nanowires via the vapor-liquid-solid (VLS) mechanism,10,11 or via the vapor-solid-solid (VSS) mechanism.12,13 The drawback of growing nanowires in the <111>B/<000̅ > directions is that the nanowires tend to have planar defects in comparison to other growth directions.14 To date, few reports have demonstrated the growth of III-V 15-20 nanowires other than <111>B/<000 ̅ > directions. These non-

<111>B/<000̅ > nanowires are generally defect-free and adopt zinc-blende crystal structure. Therefore, controlling the growth direction opens a new possibility to obtain defect-free III-V nanowires with controlled crystal structures. However, the growth mechanism of these non-<111>B/<000̅ > nanowires are still under debate. For example, Bjork et al.15 proposed that the compressive strain induced by lattice mismatch between the GaAs (111)B substrate and the grown InAs/InP nanowires can change the growth direction 16 from <111>B to <001>. Krishnamachari et al. reported the homoepitaxial growth of <001> oriented InP nanowires on a InP (001) substrate. They attributed the change of growth directions to the effect of removing the annealing step prior to the nanowire growth. However, Li and co-workers17 recently discovered the growth of <00 > oriented InAs nanowires on the InAs (001) substrate when they performed an annealing step on the Au nanoparticle-loaded substrate prior to the nanowire growth. Recently, Wang and co-workers18,21 reported the position-controlled <001> InP nanowire arrays on InP (001) substrate by electron beam lithography. However, the

118

driving mechanism for the preferential <001> growth direction is still not fully understood. Furthermore, the growth of defect-free zinc-blende <00 > InAs nanowires on the GaAs {111}B substrate by molecular beam epitaxy (MBE) has not been yet realized. In this study, we demonstrate the epitaxial growth of <00̅ > InAs nanowires on the GaAs {111}B substrate by carefully controlling the catalyst phase in a MBE reactor. Through detailed structural and chemical characterizations using electron microscopy, it has been found that the <00̅ > InAs nanowires have defect-free zinc-blende structure and adopt the {00 ̅ } catalyst/nanowire interface. Specifically, the growth of our <00̅ > InAs nanowire is governed by the small lattice mismatch of specific orientation between the catalyst and nanowire through the VSS growth mechanism. The epitaxial growth of InAs nanowires was carried out in a Riber 32 MBE system using Au as catalyst on the GaAs {111}B substrate. The substrate surface was first degassed in the preparation chamber for 15 min at 250 °C, then was transferred to the growth chamber to be thermally deoxidized at 610

°C, and a thin GaAs buffer layer was grown on the GaAs {111}B substrate at 580 °C to achieve atomically flat surface. After that, the substrate was transferred back to the preparation chamber, and a thin Au film was deposited on top of the GaAs buffer layer by the vacuum thermal evaporation. The Au- coated GaAs substrate was then transferred to the MBE growth chamber and annealed at 550 °C for 5 min under As ambient to agglomerate the Au thin film into nanoparticles. This thin film generated Au nanoparticles is a simple and cost-efficient approach.22,23 After the annealing, the substrate temperature was lowered to 410 °C and the indium source was introduced to initiate InAs nanowires growth with a V/III ratio of ~20. The morphological characteristics and sidewall facets of the grown InAs nanowires were investigated by scanning electron microscopy (SEM, JEOL 7800F, operated at 5 kV), and their detailed structural and chemical characteristics were investigated by transmission electron microscopy (TEM, Philips Tecnai F20, operated at 200 kV, equipped with X-ray energy dispersive spectroscopy (EDS) for compositional analysis, and Philips Tecnai F30, operated at 300 kV). Individual nanowires for TEM analysis were prepared by

119

ultrasonicating the InAs nanowire sample in ethanol and then dispersing nanowires onto holey carbon films supported by Cu grids. Figure 1 show SEM investigations of Au-catalyzed InAs nanowires grown on the GaAs {111}B substrate. Figure 1a is a typical side-view SEM image, and shows that both vertical and inclined nanowires were grown simultaneously on the substrate. Since our substrate surfaces are {111}B, the vertical epitaxial nanowires should have the [000 ̅]/<111>B growth direction, and inclined nanowires would have other oriented growth directions. As can be seen, the heights of nanowires vary, similar to our early observation of InAs nanowires grown on the GaAs substrates.24,25 Statistically speaking, the majority of inclined nanowires have two inclined angles with respect to the substrate normal, i.e., ~35° and ~55°. In order to determine the growth direction(s) of these inclined nanowires, the top-view SEM specimen was tilted along the [1̅ 0] axis until the electron beam is parallel to the axial directions of some inclined nanowires. The detailed tilting geometry is schematically illustrated in Figure S1. In such a manner, the growth direction and sidewall facets of these inclined nanowires can be determined explicitly. When the substrate was titled ~55° along the [1̅ 0] axis, the edge-on nanowires can be seen, and an example is shown in Figure 1b and 1c where the electron beam is focused at the bottom and top of nanowire, respectively. From which, the nanowire growth direction can be determined as [ ̅ ] according to crystallography. As can be clearly seen, the nanowire catalyst shows a faceted morphology and the nanowire have a rectangular cross-section. Since the cleavage planes of GaAs are {110} (refer to the Figure S2), the nanowire sidewall facets can then be determined as the four {110} surfaces, namely ( ̅ ), ( ), (̅ ) and (̅ ̅ ) surfaces. Since the [110] growth rate is higher than that of [1̅ 0] due to the higher density of dangling bonds at the steps of {110} surfaces under As-rich environment,26,27 this leads to the anisotropic lateral growth on different sidewalls, as illustrated in Figure 1d. For a GaAs

{111}B substrate, there are two other equivalent <00̅ > directions to the [00̅ ] direction, i.e., [̅ 00] and [0̅ 0] directions, respectively; and all these three equivalent directions should be the possible growth directions. Accordingly, when viewed along the [1̅ 0] direction, the projection of the [00̅ ] nanowires on

120

the ( ̅ 0) plane is [00̅ ], while the projections of [̅ 00] and [0̅ 0] nanowires on the ( ̅ 0) plane are both along the [̅ ̅ 0] direction. Since the substrate normal is [̅ ̅ ̅ ], the [00̅ ] nanowires result in an inclined angle of 54.7° (between [00̅ ] and [̅ ̅ ̅ ]) when viewed along [1̅ 0] direction, while the projection of [̅ 00] and [0̅ 0] nanowires on the ( ̅ 0) viewing plane leads to an inclined angle of 35.3° (between [̅ ̅ 0] and [̅ ̅ ̅ ]). From this analysis, we can attribute those observed inclined angles (i.e. ~35° and ~55°) between the axial direction of the nanowires and the substrate normal, as shown in Figure 1a, to the projection of nanowires with all possible <00̅ > growth directions viewed along the [1̅ 0] direction. Therefore, the inclined nanowires with <00̅ > growth directions are confirmed in the side-view SEM image. To understand the structural and chemical characteristics of the vertical

InAs nanowires on the GaAs {111}B substrate, TEM investigations were performed. Figure 2a is a bright-field TEM image of a typical vertical InAs nanowire and shows uniform diameter (~50 nm) along its axial direction, indicating no nanowire lateral growth in the vertical nanowires.28,29 Figure 2b is a high-resolution TEM image of the nanowire near the catalyst, viewed along <11̅ 0> zone axis, in which the nanowire has the wurtzite structure with planar defects along its axial direction, and grows along the <000̅ > direction. It is of interest to note that no “necking”22 was observed below the catalyst, indicating no nanowire growth during the cooling down stage. Therefore, the determined chemical composition of catalyst in the grown nanowire reflects the true chemical composition during the nanowire growth.30 Besides, it can be noted that the catalyst shape is faceted, suggesting that the growth of InAs nanowires should be induced by the VSS mechanism.12 To investigate the chemical composition of the catalyst and its underlying nanowire, energy dispersive spectroscopy (EDS) analysis was performed and the results are shown in Figure 2c, in which the catalyst contains Au and In and the nanowire contains In and As. It should be noted that the Cu peaks in the EDS spectra are due to the Cu TEM grids used to support the nanowires. The quantitative analysis of the EDS profiles indicates that the nanowire is InAs and the catalyst has a chemical composition of ~16 at% In and ~84 at% Au. These

121

structural and chemical characteristics of the vertical InAs nanowires are confirmed by examining over a dozen of vertical nanowires. According to our SEM investigation, other than vertical nanowires, inclined nanowires were also grown simultaneously on the GaAs {111}B substrate. To understand their structural characteristics, detailed TEM investigations were performed. Figure 3a shows a bright-field TEM image of a typical inclined nanowire with a catalyst diameter of ~50 nm and with tapered morphology caused by the nanowire lateral growth. It should be noted that the ripple-like contrast in the TEM image is the bend-contour in the TEM diffraction contrast, caused by the bent nanostructure due to its surface tension.31 By tilting along its growth axis, a set of TEM images and corresponding selected area electron diffraction (SAED) patterns were obtained. Figure 3b and 3c show a bright- field TEM image taken near the tip of the nanowire and its corresponding SAED pattern taken along the [1̅ ] zone axis, respectively. According to the SAED pattern, the crystal structure of the inclined nanowire can be determined as the zinc-blende structure, which is different to the wurtzite structure observed for those vertical nanowires. By carefully examining the enlarged bright-field TEM image (refer to the inset of Figure 3b), a sharp interface between the catalyst and nanowire can be seen, and this interface is assigned as the crystallographic interface between the catalyst and nanowire. By combining SEM results and both TEM image shown in Figure 3b and the SAED pattern shown in Figure 3c, the catalyst/nanowire interface can be determined as the (00̅ ) plane as the interface normal is parallel to the 00̅ * diffraction spot in Figure 3c (note that the orientations of the TEM image and its corresponding SAED pattern are accurately aligned). Therefore, the growth direction of this inclined nanowire can be determined as [00̅ ]. It is of interest to note that, similar to the catalysts in the vertical nanowires, the catalyst has faceted morphology (confirmed by our extensive TEM investigations), indicating that the catalyst must be in the solid form during the nanowire growth, so that the < ̅ > nanowires must be induced by the VSS growth mechanism. According to the crystallographic geometry, if a <00̅ > nanowire has the {00 ̅ } catalyst/nanowire interface, this sharp interface should be observed when this nanowire is rotated along its axial direction. Figure 3d and

122

3e show a pair of bright-field TEM image and the corresponding SAED pattern, both taken along the [0̅ ] zone axis. The sharp interface between the catalyst and nanowire can still be witnessed in Figure 3d, which is also assigned as the (00̅ ) plane based on the corresponding diffraction spots. Detailed TEM investigations on a dozen of nanowires showed this {00̅ } catalyst/nanowire interface in all cases, indicating that the {00̅ } interfaces are common in the <00̅ > InAs nanowires. High-resolution TEM investigations were performed to investigate the structural quality of these < ̅ > nanowires. Figures 3f and 3g are high-resolution TEM images taken along the <110> direction near the catalyst region and at the middle of the nanowire, respectively, in which the nanowire shows defect-free zinc-blende structure. By examining the structural characteristics of over a dozen of < ̅ > nanowires, the defect-free zinc-blende structure along the whole nanowire length are confirmed. As in most cases, nanowires grown along

<111>B/<000̅ > directions with {111}B/{000̅ } catalyst/nanowire interfaces can easily adopt planar defects, such as twins or stacking faults, due to the small energetic differences for the stacking sequences in zinc-blende and wurtzite 32 structures along their <111>B/<000̅ > directions. However, our < ̅ > InAs nanowires possess defect-free zinc-blende structure and have the {00 ̅ } catalyst/nanowire interface. The lack of planar defects is due to the fact that the creation of stacking faults in the <00̅ > nanowires needs to overcome a relatively high energy barrier.33 In general, as lattice defects may negatively affect the performance of nanowire-based devices, we expect that our defect- free InAs nanowires would have superior properties. To understand the growth mechanism of <00 ̅ > InAs nanowires, we investigate their structural and chemical characteristics with a focus on the catalysts and catalyst/nanowire interfaces. Figure 4a is a bright-field TEM image of a typical <00̅ > nanowire near the tip, in which a facet-shaped catalyst and a sharp catalyst/nanowire interface can be clearly observed. Figure 4b shows a <110> high-resolution TEM image of the catalyst/nanowire interface, from which the defect-free zinc-blende structure of nanowire and an atomic-sharp {00 ̅ } catalyst/nanowire interface can be confirmed. To investigate the chemical compositions of the catalysts and its underlying

123

nanowires, EDS analysis was performed, and an example is shown in Figure 4c, in which the catalyst contains Au and In and the nanowire contains In and As. The quantitative analysis of the EDS profiles indicates that the catalyst has a chemical composition of ~30 at% In and ~70 at% Au and the nanowire is InAs as expected. By examining over a dozen of <00 ̅ > nanowires, we confirmed that this compositional characteristic was true for <00̅ > nanowires. Figure 4d shows the SAED pattern taken from the region containing the catalyst and its underlying nanowire, where superimposed electron diffraction patterns with a certain orientation relationship between the catalyst and its underlying nanowire can be observed (note that the camera length has been carefully calibrated as demonstrated in Figure S3 in the supporting information using standard single crystal Si). Based on the nanowire diffraction spots (some circled in Figure 4d), the zone axis of the nanowire can be determined as [1̅ ]. Using the diffraction spots of the nanowire section as a reference (note that the lattice parameter of zinc-blende structured InAs is a = 0.606 nm34), the lattice spacing(s) of diffraction spots of the catalyst can be determined. Our detailed SAED analysis indicates that the faceted catalyst is the hexagonal structured Au-In γ' phase35 with lattice parameters of a = 1.22 nm and c = 0.85 nm (PDF 29-0648), which is consistent with the EDS analysis. In addition, based on the Au-In binary phase diagram, the Au-In γ' phase has a melting point of ~479 °C, which is much higher than our growth temperature (410 °C ), suggesting that the catalyst should be in the solid state during the nanowire growth, which further confirms the VSS growth mechanism.13 According to their superimposed electron diffraction patterns and their indexing, the catalyst and the nanowire have the orientation relationship of { ̅ }C//{ ̅ }InAs and [̅ ]C//[1̅ ]InAs. To understand this crystallographic orientation relationship, it is of interest to note from the superimposed diffraction spots that the distance between three {220}InAs 36 atomic planes matches the distance between two {22̅ 0}C atomic planes. Our theoretical analysis based on equilibrium lattice parameters of the nanowire and catalyst (refer to Figure 5) indicates that the lattice mismatch between 3d{220}InAs and 2d{22 ̅ 0}C is ~4.9%. The fact of a superimposed electron diffraction spot indicates the potential distortion(s) of the nanowire

124

and catalyst lattices. Indeed, the diffraction pattern of the nanowire is found slightly distorted with laterally extended and vertically compressed when compared with free-standing nanowires, and the diffraction pattern of the catalyst is also slightly distorted with laterally compressed and vertically extended. Furthermore, the other (perpendicular) in-plane lattice mismatch 37 between 4d{̅ 02}C and 5d{2̅ 0}InAs is calculated as ~1%. Therefore, a good lattice coherent interface between the nanowire and catalyst is expected in our case, as schematically illustrated in Figure 5. We therefore anticipate that the coherent orientation relationship between the catalyst and its underlying nanowire, and the small in-plane lattice mismatch between the solid catalyst and the nanowire is the driving force for the formation of { ̅ } catalyst/nanowire interfaces, and hence results in the growth of < ̅ > nanowires through the VSS growth mechanism. Furthermore, kinked nanowires with the same inclined angle as those <00̅ > nanowires can also be observed in the SEM image as well (refer to the mark in Figure 1a). In this regard, the TEM investigations on the structural and chemical characteristics of these kinked nanowires were performed. Figure 6a shows a bright-field TEM image of a typical kinked nanowire. Figure 6b is a <1̅ 0> high-resolution TEM image taken from the top of nanowire, in which a sharp interface between the catalyst and nanowire and a faceted catalyst at the tip of nanowire can be clearly observed (note that the dark line between the nanowire and the outside amorphous layer is the Fresnel contrast caused by the defocus of the TEM image). Similar to the inclined nanowires, the growth direction of the kinked section was determined to be < ̅ >, and the catalyst/nanowire interface was determined as {00̅ } plane. EDS analysis was performed to determine the chemical composition of the catalysts in kinked nanowires. By investigating over half a dozen of kinked nanowires, it has been found that the catalyst of kinked nanowires has the same chemical composition as the catalyst in the < ̅ > nanowires, i.e. ~30 at% In and ~70 at% Au. Figure 6c is a high-resolution TEM image taken from the area marked in Figure 6b, in which defect-free zinc-blende structure can be witnessed. Our detailed investigations along the entire kinked section from over half a dozen of kinked nanowires confirmed this defect-free zinc-blende structural

125

characteristic. Figure 6d shows a high-resolution TEM image taken from the kinked junction of the nanowire as marked in Figure 6a, from which the bottom section with wurtzite structure and upper section with zinc-blende structure can be clarified. It is of interest to note that the interface between the wurtzite section and the zinc-blende section is atomically sharp, and this provides a new approach to manipulate III-V nanowire heterostructures with different crystal structures. Based on these analyses, it can be concluded that the kinked nanowires have the same catalyst composition as the < ̅ > nanowires, and they both adopt the {00̅ } catalyst/nanowire interface, suggesting that they are induced by the same growth mechanism. According to these detailed investigations, four facts can be noted. (1) The indium concentration in the catalyst of < ̅> nanowires with the zinc-blende structure is higher than that of vertical nanowires with the wurtzite structure. (2) The catalyst size of both <000 ̅> and <00 ̅> nanowires are distrusted in a narrow range of 45 ± 5 nm (refer to Figure S4 in the supporting information), suggesting that the catalyst size should not determine the growth directions of nanowires in our case. (3) The catalyst of kinked nanowires have the same indium concentration as the catalyst of < ̅> nanowires, and both kinked nanowires and < ̅ > nanowires adopt the {00 ̅ } catalyst/nanowire interfaces. (4) The in-plane lattice mismatch between the catalyst and its underlying nanowire in the < ̅ > nanowires is small. Based on the experimental results and analyses outlined above, a growth model of Au induced wurtzite structured < ̅> InAs nanowires and defect-free zinc-blende structured < ̅> InAs nanowires in MBE can be proposed. With the introduction of indium source, indium atoms diffuse into the Au nanoparticles developed during annealing process to form Au-In alloy nanoparticles. Au-In nanoparticles with lower indium concentration tend to initiate vertical < ̅> nanowire growth on the substrate, while Au-In nanoparticles with higher indium concentration form the Au-In γ' phase inducing the epitaxial growth of < ̅> nanowires, governed by the small in-plane lattice mismatch on the catalyst/nanowire interface via the VSS growth mechanism. Meanwhile, the increase of indium concentration in the catalyst caused by the environmental fluctuation during the nanowire growth will result in the formation of Au-In γ' phase for the catalysts. Consequently the coherence between the solid

126

catalysts and nanowires alters the nanowire growth direction from < ̅> to < ̅>, forming defect-free zinc-blende structure. To understand the effect of the V/III ratio on the growth of inclined < ̅ > InAs nanowires, we repeated the nanowire growths with a higher V/III ratio of 45 and with a lower V/III ratio of 10 by keeping the same indium pressure. According our SEM investigation (as presented in Figure S5 in the supporting information), nanowires were predominantly grown vertically under the higher V/III ratio; and inclined nanowires can be often observed when a lower V/III ratio of 20 is used to grow InAs nanowires. In fact, by reducing the V/III ratio, the indium vapor partial pressure increases, causing more indium atoms diffused into catalysts. When the Au-In γ' phase is formed in some catalysts inclined < ̅ > InAs nanowires can then be induced by these high indium concentrated catalysts via the VSS mechanism, driven by the small in-plane lattice mismatch between the Au-In γ' phase and InAs. With further decreasing the V/III ratio to 10, the SEM investigation (refer to Figure S5c) shows an increased percentage of inclined nanowires. However, these inclined nanowires were grown randomly on the substrate with irregular morphologies and poor structural quality. The statistics of the effect of V/III ratio on the percentage of inclined nanowire nanowires are shown in Table S1 in the supporting information. These detailed experimental results indicate that the growth of defect-free <00 ̅ > InAs nanowires are sensitive to growth parameters and they can only be induced in an optimized V/III ratio. In conclusion, we demonstrated the epitaxial growth of zinc-blende and defect-free <00̅ > InAs nanowires with {00̅ } catalyst/nanowire interface on the GaAs {111}B substrate in MBE. Our detailed investigations indicate that the growth of <00̅ > nanowire is governed by the small in-plane lattice mismatch between catalysts and nanowires via the VSS growth mechanism. This study provides a new approach to control the growth of defect-free zinc-blende InAs nanowires by controlling the Au-In catalyst phase.

AUTHOR INFORMATION

Corresponding Author

127

*E-mail: [email protected]

Notes

The authors declare no competing financial interest.

ASSOCIATED CONTENT

Supporting Information. Description of titling experiment and results, calibration of SAED patterns, statistic analysis of catalysts size distribution, SEM investigation of nanowires grown under different V/III ratios. This material is available free of charge via the Internet at http://pubs.acs.org.

ACKNOWLEDGMENT

This study is financially supported by the Australian Research Council, the National Basic Research Program of China (Grant No. 2011CB925604), the National Science Foundation of China (Grant No. 61376015, 91321311, 11334008 and 91121009) and Shanghai Science and Technology Foundation (Grant No. 13JC1405901). Australian Microscopy & Microanalysis Research Facility is also gratefully acknowledged for providing microscopy facilities for this study.

128

Figure 1. SEM images of Au-catalyzed one-dimensional InAs nanowires. (a) Side- view (viewed along the [1̅ 0] direction) SEM image of InAs nanowires. (b, c) 55° tilted SEM images showing facet details and catalyst shape of a typical [00̅ nanowire, respectively. The tilt axis is [1̅ 0]. (d) Schematic illustration of the facets of [00̅ ] nanowire.

129

Figure 2. TEM investigations of <0001> oriented InAs nanowire. (a) The overview of a typical <000̅ > oriented InAs nanowire. (b) A <11̅ 0> high-resolution TEM image of the top of nanowire. (c) EDS spectra taken from the catalyst and its underling nanowire.

130

Figure 3. Detailed TEM investigations of <00̅ > InAs nanowire. (a) The overview of a typical <00̅ > InAs nanowire. (b, d) Bright-field TEM images of the top region of the same nanowire viewed along the [1̅ ] and [0̅ ] zone axes, respectively. (c, e) Corresponding SAED patterns taken along the [1 ̅ ] and [0 ̅ ] zone axes, respectively. (f) A high-resolution TEM image taken from the top of nanowire marked in inset of (b). (g) A high-resolution TEM image taken from the middle of nanowire marked in (a).

131

Figure 4. (a) A bright-field TEM image of a typical nanowire at the region near the catalyst. (b) High-resolution TEM image from the area marked in (a). (c) EDS spectra taken from the catalyst and its underling nanowire. (d) SAED pattern taken from the top of nanowire and catalyst. Circled diffraction spots belong to the nanowire, and arrowed diffraction spots belong to the catalyst.

132

Figure 5. Schematic diagram showing near coherence of the atomic planes at the catalyst/nanowire interface.

133

Figure 6. TEM investigations of kinked nanowire. (a) The overview of a typical kinked nanowire. (b) A bright-field TEM image from the top of nanowire viewed along the <1̅ 0> zone axis. The inset is the corresponding SAED pattern. (c) High-resolution TEM image taken from the top section as marked in (b). (d) A high-resolution TEM image taken from the kinked area as marked in (a).

134

Supporting information for

“Catalyst Orientation Induced Growth of Defect-Free Zinc-Blende Structured <00 ̅> InAs Nanowires”

Zhi Zhang,† Kun Zheng,‡,§ Zhen-Yu Lu,# Ping-Ping Chen,# Wei Lu,# and Jin Zou*,†,‡

*Address correspondence to: [email protected]

135

Schematics showing the process of tilting inclined nanowire to be parallel to the electron beam and the edge-on SEM images showing the nanowire side facets with respect to the substrate

Figure S1 schematically illustrates the tilting process of an inclined <00 ̅> nanowire until its axial direction is parallel to the electron beam. Figure S1a shows an image of the GaAs {111}B substrate (cleavage planes being {1 ̅0} planes) without tilting, in which an inclined nanowire with the projection direction of <112> can be seen. Then, the substrate was tilted along the [1 ̅0] axis, and a typical image during the tilting process captured at a certain tilting angle is shown in the Figure S1b. It can be noted that the nanowire projection length becomes shorter. With further tilting, the axial direction of the inclined nanowire is parallel to the electron beam, as shown in the Figure S1c. Based on the tilting angle and general crystallography, the growth direction of the inclined nanowire can be determined. In addition, the side-wall facets of the inclined nanowire can be clearly observed.

Figure S1. Schematic illustration of the tilting process. (a) 0 degree tilted image of the

GaAs{111}B substrate with the cleavage planes of {1 ̅ 0}. An inclined nanowire with the projection direction on the substrate being one of the <112> directions is grown on the substrate. The inset is the sketch of all the <112> directions on the GaAs{111}B substrate. (b) Image of the substrate with certain tilting angle during the tilting process. (c) Image of the substrate when the axial direction of inclined nanowire is parallel to the electron beam. The tilting axis is [1 ̅0].

136

Figure S2 is a low-magnification ~55° tilted SEM image showing the side facets of <00 ̅> nanowires with respect to the GaAs substrate, in which the cleavage plane is (1 ̅0). Based on the tilting process shown in Figure S1, two sidewalls of an edge-on nanowire are parallel to the cleavage plane while the other two sidewalls are perpendicular to the cleavage plane. Therefore, these two sidewalls parallel to the cleavage plane can be determined as (1 ̅0) and ( ̅10) planes. According to the crystallography, in the (00 ̅) plane, the only crystal planes perpendicular to the ±(1 ̅0) planes are (110) and ( ̅ ̅0) planes. Therefore, the sidewalls of the nanowire can be determined to be four {110} planes, namely, (1 ̅ 0), (110), ( ̅ 10), and ( ̅ ̅ 0), as schematically illustrated in Figure S1c.

Figure S2. 55° tilted low-magnification SEM image indicating the substrate cleavage plane and the nanowire facet orientation with respect to the GaAs substrate. The inset is a magnified SEM image of an edge-on <00 ̅> nanowire as indicated by the red arrow.

Selected area electron diffraction (SAED) pattern calibration

Figure S3 shows the SAED pattern of the single crystal silicon taken along the [1 ̅0] zone axis. By carefully measuring the d{002} spacing on the SAED pattern based on the scale bar, an error bar of 0.3% is found when compared with the theoretical d{002}, suggesting that the camera length and scale bar in the TEM have been well calibrated.

137

Figure S3. The SAED pattern of the single crystal silicon taken along the [1 ̅0] zone axis.

Statistics on the catalyst size distribution

Since the size of catalysts can be directly reflected by the lateral dimension of their underlying nanowires, the size distribution of catalysts can be statistically measured by measuring the distribution of nanowire diameters. Figure S4 shows the statistic measurements of nanowire diameters for the <000 ̅> and <00 ̅> nanowires. As can be seen, catalyst size of both types of nanowires are distrusted in a narrow range of 45 ± 5 nm, suggesting that the catalyst size should not be the driving force for the change of the growth directions in our case.

138

Figure S4. The distributions of nanowire diameters of <000 ̅> and <00 ̅> nanowires.

139

SEM investigations of InAs nanowires grown under different V/III ratios

Figure S5 show side-view SEM images of the InAs nanowires grown under V/III ratios of 45,

20 and 10, respectively. As can be noted, under a high V/III ratio of 45, nearly all nanowires were vertically grown on the substrate, as shown in the Figure S5a. When the V/III ratio was decreased to 20, in addition to the vertical nanowires, more inclined <00 ̅> nanowires can be observed, refer to the Figure S5b. With further decreasing the V/III ratio to ~10, the SEM investigation (refer to Figure S5c) shows an increased percentage of inclined nanowires.

However, these nanowires were grown randomly on the substrate with irregular morphologies. The effect of V/III on the percentage of inclined nanowires are shown in Table

S1.

140

Figure S5. Side-view SEM images (viewed along the [1 ̅0] direction) of the InAs nanowires grown under different V/III ratios on the GaAs{111}B substrate. (a) V/III ratio of ~45; (b) V/III ratio of ~20; (c) V/III ratio of ~10.

Table S1. The statistics of the effect of V/III ratio on the percentage of inclined nanowires.

V/III ratio 45 20 10

Inclined ~ 0 ~ 40 ~80 nanowires %

141

REFERENCES

1. Lieber, C. M. Semiconductor Nanowires: A Platform for Nanoscience and Nanotechnology. Mrs. Bull. 2011, 36, 1052-1063. 2. Xia, H.; Lu, Z. Y.; Li, T. X.; Parkinson, P.; Liao, Z. M.; Liu, F. H.; Lu, W.; Hu, W. D.; Chen, P. P.; Xu, H. Y., et al. Distinct Photocurrent Response of Individual Gaas Nanowires Induced by N-Type Doping. ACS Nano 2012, 6, 6005-6013. 3. Duan, X. F.; Huang, Y.; Cui, Y.; Wang, J. F.; Lieber, C. M. Indium Phosphide Nanowires as Building Blocks for Nanoscale Electronic and Optoelectronic Devices. Nature 2001, 409, 66-69. 4. Thelander, C.; Caroff, P.; Plissard, S. b.; Dey, A. W.; Dick, K. A. Effects of Crystal Phase Mixing on the Electrical Properties of Inas Nanowires. Nano Lett. 2011, 11, 2424-2429. 5. Dayeh, S. A.; Yu, E. T.; Wang, D. Excess Indium and Substrate Effects on the Growth of Inas Nanowires. Small 2007, 3, 1683-1687. 6. Milnes, A. G.; Polyakov, A. Y. Indium Arsenide: A Semiconductor for High Speed and Electro-Optical Devices. Mat. Sci. Eng. B 1993, 18, 237-259. 7. Doh, Y. J.; van Dam, J. A.; Roest, A. L.; Bakkers, E.; Kouwenhoven, L. P.; De Franceschi, S. Tunable Supercurrent through Semiconductor Nanowires. Science 2005, 309, 272-275. 8. Offermans, P.; Crego-Calama, M.; Brongersma, S. H. Gas Detection with Vertical Inas Nanowire Arrays. Nano Lett. 2010, 10, 2412-2415. 9. Chuang, S.; Gao, Q.; Kapadia, R.; Ford, A. C.; Guo, J.; Javey, A. Ballistic Inas Nanowire Transistors. Nano Lett. 2013, 13, 555-558. 10. Wagner, R. S.; Ellis, W. C. Vapor-Liquid-Solid Mechanism of Single Crystal Growth ( New Method Growth Catalyst from Impurity Whisker Epitaxial + Large Crystal Si E ). Appl. Phys. Lett. 1964, 4, 89-90. 11. Hillerich, K.; Dick, K. A.; Messing, M. E.; Deppert, K.; Johansson, J. Simultaneous Growth Mechanisms for Cu-Seeded Inp Nanowires. Nano Res. 2012, 5, 297-306. 12. Persson, A. I.; Larsson, M. W.; Stenstrom, S.; Ohlsson, B. J.; Samuelson, L.; Wallenberg, L. R. Solid-Phase Diffusion Mechanism for Gaas Nanowire Growth. Nat. Mater. 2004, 3, 677-681.

142

13. Dick, K. A.; Deppert, K.; Martensson, T.; Mandl, B.; Samuelson, L.; Seifert, W. Failure of the Vapor-Liquid-Solid Mechanism in Au-Assisted Movpe Growth of Inas Nanowires. Nano Lett. 2005, 5, 761-764. 14. Wilhelm, C.; Larrue, A.; Dai, X.; Migas, D.; Soci, C. Anisotropic Photonic Properties of Iii-V Nanowires in the Zinc-Blende and Wurtzite Phase. Nanoscale 2012, 4, 1446-1454. 15. Bjork, M. T.; Ohlsson, B. J.; Sass, T.; Persson, A. I.; Thelander, C.; Magnusson, M. H.; Deppert, K.; Wallenberg, L. R.; Samuelson, L. One-Dimensional Heterostructures in Semiconductor Nanowhiskers. Appl. Phys. Lett. 2002, 80, 1058- 1060. 16. Krishnamachari, U.; Borgstrom, M.; Ohlsson, B. J.; Panev, N.; Samuelson, L.; Seifert, W.; Larsson, M. W.; Wallenberg, L. R. Defect-Free Inp Nanowires Grown in [001] Direction on Inp(001). Appl. Phys. Lett. 2004, 85, 2077-2079. 17. Li, Z.-A.; Moeller, C.; Migunov, V.; Spasova, M.; Farle, M.; Lysov, A.; Gutsche, C.; Regolin, I.; Prost, W.; Tegude, F.-J., et al. Planar-Defect Characteristics and Cross-Sections of < 001 >, < 111 >, and < 112 > Inas Nanowires. J. Appl. Phys. 2011, 109, 114320. 18. Wang, J.; Plissard, S.; Hocevar, M.; Vu, T. T. T.; Zehender, T.; Immink, G. G. W.; Verheijen, M. A.; Haverkort, J.; Bakkers, E. P. A. M. Position-Controlled [100] Inp Nanowire Arrays. Appl. Phys. Lett. 2012, 100, 053107. 19. Shtrikman, H.; Popovitz-Biro, R.; Kretinin, A.; Heiblumf, M. Stacking-Faults-Free Zinc Blende Gaas Nanowires. Nano Lett. 2009, 9, 215-219. 20. Zhang, Z.; Lu, Z. Y.; Xu, H. Y.; Chen, P. P.; Lu, W.; Zou, J. Structure and Quality Controlled Growth of Inas Nanowires through Catalyst Engineering. Nano Res. 2014, 7, 1640-1649. 21. Wang, J.; Plissard, S. R.; Verheijen, M. A.; Feiner, L.-F.; Cavalli, A.; Bakkers, E. P. A. M. Reversible Switching of Inp Nanowire Growth Direction by Catalyst Engineering. Nano Lett. 2013, 13, 3802-3806. 22. Harmand, J. C.; Patriarche, G.; Pere-Laperne, N.; Merat-Combes, M. N.; Travers, L.; Glas, F. Analysis of Vapor-Liquid-Solid Mechanism in Au-Assisted Gaas Nanowire Growth. Appl. Phys. Lett. 2005, 87, 203101. 23. Liao, Z. M.; Chen, Z. G.; Lu, Z. Y.; Xu, H. Y.; Guo, Y. N.; Sun, W.; Zhang, Z.; Yang, L.; Chen, P. P.; Lu, W., et al. Au Impact on Gaas Epitaxial Growth on Gaas

(111)B Substrates in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 102, 063106.

143

24. Zhang, X.; Zou, J.; Paladugu, M.; Guo, Y. N.; Wang, Y.; Kim, Y.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C. Evolution of Epitaxial Inas Nanowires on Gaas (111)B. Small 2009, 5, 366-369. 25. Zhang, Z.; Lu, Z. Y.; Chen, P. P.; Xu, H. Y.; Guo, Y. N.; Liao, Z. M.; Shi, S. X.; Lu, W.; Zou, J. Quality of Epitaxial Inas Nanowires Controlled by Catalyst Size in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 103, 073109. 26. Asai, H. Anisotropic Lateral Growth in Gaas Mocvd Layers on (001) Substrates. J. Cryst. Growth 1987, 80, 425-433. 27. Seifert, W.; Borgstrom, M.; Deppert, K.; Dick, K. A.; Johansson, J.; Larsson, M. W.; Martensson, T.; Skold, N.; Svensson, C. P. T.; Wacaser, B. A., et al. Growth of One-Dimensional Nanostructures in Movpe. J. Cryst. Growth 2004, 272, 211-220. 28. Zou, J.; Paladugu, M.; Wang, H.; Auchterlonie, G. J.; Guo, Y. N.; Kim, Y.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C. Growth Mechanism of Truncated Triangular Iii-V Nanowires. Small 2007, 3, 389-393. 29. Guo, Y. N.; Xu, H. Y.; Auchterlonie, G.; Burgess, T.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Shu, H. B.; Chen, X. S., et al. Phase Separation Induced by Au Catalysts in Ternary Ingaas Nanowires. Nano Lett. 2013, 13, 643-650. 30. Park, H. D.; Prokes, S. M.; Cammarata, R. C. Growth of Epitaxial Inas Nanowires in a Simple Closed System. Appl. Phys. Lett. 2005, 87, 063110. 31. Pan, Z. W.; Dai, Z. R.; Wang, Z. L. Nanobelts of Semiconducting Oxides. Science 2001, 291, 1947-1949. 32. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Defect-Free <110> Zinc-Blende Structured Inas Nanowires Catalyzed by Palladium. Nano Lett. 2012, 12, 5744-5749. 33. Hornstra, J. Dislocations in the Diamond Lattice. J. Phys. Chem. Solids 1958, 5, 129-141. 34. Takahash, K.; Moriizum, T. Growth of Inas Whiskers in Wurtzite Structure. Jpn. J. Appl. Phys. 1966, 5, 657-662. 35. Dick, K. A.; Deppert, K.; Karlsson, L. S.; Wallenberg, L. R.; Samuelson, L.; Seifert, W. A New Understanding of Au-Assisted Growth of Iii-V Semiconductor Nanowires. Adv. Funct. Mater. 2005, 15, 1603-1610. 36. Yoshiie, T.; Bauer, C. L.; Milnes, A. G. Interfacial Reactions between Gold Thin Films and Gaas Substrates Thin Solid Films 1984, 111, 149-166.

144

37. Zhang, W. Z.; Weatherly, G. C. On the Crystallography of Precipitation. Prog. Mater. Sci. 2005, 50, 181-292.

145

146

7 Controlling the Crystal Phase and Structural Quality of Epitaxial InAs Nanowires by Tuning V/III Ratio in Molecular Beam Epitaxy

7.1 Introduction

In this study, we demonstrated the control of crystal phase and structural quality of

Au-catalyzed InAs nanowires grown on the GaAs {111}B substrates by tuning the V/III ratio in molecular beam epitaxy. It has been found that InAs nanowires can only

147

be grown in a relatively narrow window of the V/III ratio. It is also demonstrated that the V/III ratio can be used to control the structural quality of wurtzite structured and zinc-blende structured InAs nanowires under low V/III ratios, and defect-free wurtzite structured and zinc-blende structured InAs nanowires were successfully achieved. This study provides an insight into the controlled growth of high-quality wurtzite structured and zinc-blende structured InAs nanowires through the V/III ratio engineering.

7.2 Journal Publication

The results in Chapter 7 are included as it appears in Acta Materialia 2015, 92, 25- 32. http://www.sciencedirect.com/science/article/pii/S1359645415002220

148

Controlling the Crystal Phase and Structural Quality of Epitaxial InAs Nanowires by Tuning V/III Ratio in Molecular Beam Epitaxy

Zhi Zhang, a Zhen-Yu Lu, b Ping-Ping Chen, b Wei Lu, b and Jin Zou a,c,*

a Materials Engineering, The University of Queensland, St. Lucia, QLD 4072, Australia b National Laboratory for Infrared Physics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, 500 Yu-Tian Road, Shanghai 200083, China c Centre for Microscopy and Microanalysis, The University of Queensland, St. Lucia, QLD 4072, Australia

Abstract

In this study, we demonstrated the control of crystal phase and structural quality of

Au-catalyzed InAs nanowires grown on the GaAs {111}B substrates by tuning the V/III ratio in molecular beam epitaxy. It has been found that InAs nanowires can only be grown in a relatively narrow window of the V/III ratio. It is also demonstrated that the V/III ratio can be used to control the structural quality of wurtzite structured and zinc-blende structured InAs nanowires under low V/III ratios, and defect-free wurtzite structured and zinc-blende structured InAs nanowires were successfully achieved. This study provides an insight into the controlled growth of high-quality wurtzite structured and zinc-blende structured InAs nanowires through the V/III ratio engineering.

Keywords: InAs nanowires; Crystal phase; Structural quality; MBE

*Corresponding author. Tel: +61-7-33463195; E-mail address: [email protected] (J. Zou)

149

1. Introduction III-V compound semiconductor nanowires (NWs) have attracted extensive attentions in the recent decades due to their distinct physical properties and hence potential applications in nanoelectronics and optoelectronics.1-3 Currently, the most typical techniques to grow epitaxial III-V NWs are metal-organic chemical vapour deposition (MOCVD), molecular beam epitaxy (MBE) and chemical beam epitaxy through the vapor-liquid-solid (VLS)4 and/or vapor-solid-solid (VSS)5 mechanisms. As a key III-V semiconductor, InAs has attracted special research interest due to its very high electron mobility,6 low-resistance ohmic contact,7 narrow band gap and small electron effective mass.8 The combination of these unique features and the NW distinct characteristics have made InAs NWs a promising candidate for applications in single-electron transistors,9 the Josephson junctions,10 resonant tunnelling diodes,11 and ballistic transistors.12 For NWs to be practically useful in many nanodevices, it is of critical importance to control their crystal phase and structural quality. In general, most epitaxial III-V NWs are grown along the <111>B directions for zinc-blende (ZB) structured NWs or along the <000 ̅> for the wurtzite (WZ) structured NWs,13 while other low-index growth directions have also been occasionally reported.14-16 However, the majority of epitaxial III-V NWs naturally contain uncontrolled planar defects due to the small energetic differences of the stacking sequences in ZB and WZ structures along their

<111>B/<000 ̅> directions. In this regard, several approaches have been employed to control the structural quality of Au-catalyzed III-V NWs. Controlling crystal structures of III-V NWs has been reported by varying the nominal V/III ratio, mostly for NWs grown by MOCVD, but seemingly different tendencies were found. Some studies17-18 show that III-V NWs grow preferably in the WZ structure than ZB structure with increasing the V/III ratio. However, others19-21 demonstrated that III-V NWs prefer to nucleate in the ZB phase with increasing the V/III ratio. Since MBE has a distinct growth environment when compared to MOCVD,22 the impact of the V/III ratio on the crystal phase and structural quality of III-V NWs can vary from MBE to MOCVD. Furthermore, it has been well documented that Au-catalyzed III-V NWs grown by MBE prefer to adopt the WZ structure with planar defects.23-25 Very few studies have been reported the ZB structured InAs NWs grown by MBE. In this regard, it is scientifically important and practically necessary to control the structural

150

quality of WZ structured InAs NWs and to achieve the growth of ZB structured InAs NWs in MBE. In this study, we reported a systematic investigation of the crystal phase and structural quality controlled growth of epitaxial InAs NWs on the GaAs {111}B substrate by merely tuning the V/III ratio in MBE using Au as catalysts. Our results indicate that the V/III ratio window to grow InAs NWs is comparatively narrow. It was found that the V/III ratio can be used to tune the structural quality of WZ structured InAs NWs. By manipulating the V/III ratio, we also demonstrated the successful growth of defect-free ZB structured InAs NWs.

2. Experimental

The InAs NWs were grown on the GaAs {111}B substrates in a Riber 32 MBE system using Au as catalysts. The substrate surface was first degassed in the MBE preparation chamber for 15 min at 250 °C, then was transferred to the growth chamber to be thermally deoxidized for 20 min at 610 °C. After that, for each growth, a thin GaAs buffer layer was grown on the GaAs {111}B substrate at 580 °C to achieve atomically flat surface. The substrate was then transferred back to the preparation chamber, and a thin Au film was deposited on the top of the GaAs buffer layer by vacuum thermal evaporation. The Au-coated GaAs substrate was finally transferred to the growth chamber, and annealed at 500 °C for 5 min under As ambient to agglomerate the Au thin film into nanoparticles. This thin film generated Au nanoparticles is a simple and cost-efficient approach to prepare Au catalysts.26 After the annealing, the substrate temperature was lowered to 430 °C and the indium source was introduced to initiate the InAs NW growth. Five InAs samples were grown on the GaAs {111}B substrates under different V/III ratios (with fixed indium beam pressure of 1.0 x 10-7 Torr) from a V/III ratio ~70 to a V/III ratio ~5 (sample A: V/III ~70, sample B: V/III ~45, sample C: V/III ~20, sample D: V/III ~10, sample E: V/III ~5) . The morphological characteristics of the as-grown InAs NWs and their bases were investigated by scanning electron microscopy (SEM, JEOL 7800F, operated at 10 kV), and their detailed structural and chemical characteristics, together with their catalysts, were investigated by transmission electron microscopy (TEM, Philips Tecnai F20, operated at 200 kV, equipped with X-ray energy dispersive spectroscopy (EDS) for compositional analysis, and FEI Tecnai F30, operated at 300

151

kV). Individual NWs for TEM analysis were prepared by ultrasonicating the as-grown samples in ethanol for 20 min and then dispersing individual NWs onto holey carbon films supported by Cu grids.

3. Results and discussion Fig. 1 shows the SEM images of InAs NWs grown under different V/III ratios. Fig. 1a is a top view SEM image of sample A (grown at V/III ratio of ~70), from which island-like bases with catalysts on the top can be seen, as shown in the magnified SEM image in the inset of Fig. 1a. To clarify the growth status of these nanostructures, the SEM specimen was tilted and viewed, as shown in Fig. 1b. It is of interest to note that no NW can be observed on the substrate, indicating that catalysts did not induce NW growth under such a high V/III ratio, in which the material growth is in two-dimensional that lead to the formation of island-like nanostructures only. When the InAs NW growth was conducted under a lower V/III ratio of ~45 (sample B), InAs NWs were grown vertically on the substrate, as shown Fig. 1c and 1d (viewed respectively from the top and side of the SEM specimens). As can be seen, the NW lengths is varied, which is a typical characteristic of InAs NWs grown on the GaAs substrates.27 Fig. 1e and 1f show respectively the top-view and [1 ̅0] side-view SEM images of sample C (grown at V/III ratio of ~20), in which, apart from conventional vertical NWs, inclined NWs projected in certain crystallographic orientations can be observed. Using the fact that the cleavage planes of ZB structured GaAs substrate are {110}, the projections of all the inclined NWs can be determined as <112> directions (see discussion later). With further reducing the V/III ratio to ~10, SEM investigations (refer to Fig. 1g and 1h) show an increased percentage of inclined NWs (many grown randomly on the substrate with irregular morphologies). When the V/III ratio was lowered to ~5, the Au nanoparticles failed to induce NW growth, as evidenced in Fig. 1i and 1j, in which an irregular layer and many particles are found on the substrate. Based on these experimental results, it can be noted that, compared to GaAs NW growth.23 InAs NWs can only be induced in a relatively narrow window for the V/III ratio, and the window for only growing vertical InAs NWs is even narrow. The failure to induce NW growth under the high V/III ratio of ~70 can be understood by the following three facts. Firstly, it is well documented that the As- trimer surface reconstruction takes place on the GaAs {111}B substrate under a high

152

As partial pressure (corresponding to high V/III ratio), and this surface reconstruction is believed to make <111>B/<000 ̅> growth direction less energetically favorable and 28 therefore decrease the growth rate in the <111>B/<000 ̅> directions. Secondly, the decreased diffusion length of group III atoms under a high V/III ratio also hinders axial NW growth while favors the two-dimensional planar growth on the substrate.29 Thirdly, the indium concentration in the catalyst may have not reached the catalyst supersaturation threshold required for InAs NW growth, which is determined by the low indium partial pressure in the vapor due to the high V/III ratio.30 These three facts may play a synergetic role to restraint the one-dimensional NW growth. In this regard, such a high V/III ratio sets the upper limit of the V/III ratio for the InAs NW growth. On the other end, the failure to trigger NW growth under a low V/III ratio of ~5 is due to the fact that InAs is very reactive with Au and it requires a high arsenic overpressure to prevent InAs decomposition from Au.31 Therefore, under such a low V/III ratio, there is no sufficient As species to promote InAs NW growth. In this regard, this sets the lower limit of the V/III ratio for the growth of InAs NWs. According to the SEM investigation shown in Fig. 1, vertical InAs NWs can be observed in samples B, C and D. To understand their structural and chemical characteristics, TEM investigations were performed on over a dozen of NWs from each sample. Fig. 2a, 2c and 2e are bright-field TEM images showing the overview of typical vertical NWs taken from samples B, C and D, respectively, from which no tapering was observed in all cases. Fig. 2b, 2d and 2f are their corresponding high- resolution TEM images taken along the <11 ̅ > zone axis from the middle of NWs, in which all NWs show the WZ structure. It is of interest to note that planar defects can be observed in the NW of sample B, while no planar defects can be found in the NWs of samples C and D by investigating its entire length. To confirm this conclusion, we have investigated a dozen of vertical NWs from each sample. To understand the difference of their structural qualities, we consider the nucleation kinetics in Au- catalyzed NW growth. The ratio between nucleation barriers (훥퐺 and 훥퐺 ) for the formation of WZ and ZB structures can be expressed as32-33

휉 (1)

where 훥휇 is the liquid-solid supersaturation that is the chemical potential difference between the catalyst and the solid phase which increases with increasing the group-

III concentration in the catalysts 휓 is the extra cohesive energy associated with

153

WZ formation and η is the ratio between the change in effective surface energies due to the formation of a WZ and a ZB nucleus (훤 and 훤 ) as expressed below

𝜂 (2)

where 훶 , 훶 , 훶 are the interfacial energies of the vapor-liquid-solid system x is the share of nucleus perimeter in contact with the vapor is the contact angle and τ

= 훶 /훶 is the ratio between the lateral solid-vapor surface energies of WZ and ZB nuclei which is expected to be slightly less than 1.

According to the Eq. (1) both 훥휇 and η can affect the NW crystal phase namely higher 훥휇 and lower η favors the formation of WZ structure. Otherwise the formation of ZB structure will be preferred. Let us consider the effect of η first. Based on Eq. (1) and (2) the newly formed monolayers will adopt the ZB structure for higher values of η. Accordingly we anticipate that a lower 훶 or more precisely an increase of the 훶 /훶 ratio will increase η and in turn increase the probability of ZB formation. It has been well documented that arsenic can act as surface active agent when absorbed on the Au nanoparticle surface and hence it can lower the 34 surface energy 훶 . Arsenic may also lower the Au/NW interfacial energy 훶 but this effect is expected to be much less significant due to the rapid consumption of any arsenic present in the growth interface into the NW growth.35 On this basis in sample B NWs grown under a high V/III ratio (relatively more arsenic in the vapor) tends to decrease 훶 more significantly when compared with NWs in samples C and D (grown under relatively low V/III ratios) and in turn NWs in sample B prefers to form the ZB nucleation while NWs in samples C and D tend to form the WZ nucleation. Furthermore, our extensive EDS characterizations (over a dozen of NWs from each sample) show that the indium concentration in the catalysts is lower (~25 at.%) in sample B (the high V/III ratio with a relatively low III partial pressure) than that (~40 at.%) in sample C (the low V/III ratio with a relatively high III partial pressure). Surprisingly, the indium concentration in the catalysts from sample D (the even lower V/III ratio) is ~25 at.%, which has the same indium concentration level as in sample B, but is lower than that in sample C. Since the Au-In catalyst supersaturation increases with increasing the indium concentration in the catalyst,36-37 the catalyst supersaturation in sample C with a higher indium concentration should be higher

154

than that in sample B with a lower indium concentration, while the catalyst supersaturation in samples D is close to that in sample B. In addition, according to the theoretical prediction and experimental investigation,38 InAs NWs, especially with small diameters, prefer to adopt the WZ structure. Based on these detailed analyses considering the effects of both 훥휇 and η, it is well expected that InAs NWs in samples C and D grown at low V/III ratios tend to grow pure WZ structure, while NWs in sample B grown at a high V/III ratio prefer to adopt defected WZ structure. Therefore, the structural quality of the WZ structured InAs NWs can be controlled by tuning the V/III ratio. To understand the catalyst composition differences in NWs grown under different V/III ratios, we consider the indium absorption process on the catalyst surface. It has been well documented that the indium concentration in the catalyst is determined by the indium partial pressure in the vapor according to thermodynamical requirements 30. In this regard, the indium concentration in the catalysts in sample B (the high V/III ratio with a relatively low III partial pressure) should be lower than that in sample C (the low V/III ratios with a relatively high III partial pressure). However, when the V/III ratio is extremely low as in sample D, the catalyst surface energy becomes comparatively high due to the lack of arsenic in the vapor, resulting in the increased indium vapor pressure in the catalyst droplet. In this situation, the indium absorption by the Au catalyst may be restricted due to the decreased driving force of transporting indium from the vapor to the catalyst caused by the high indium vapor pressure in the catalyst droplet. Consequently, the indium concentration in the catalysts in sample D grown under extremely low V/III ratio is lower than that in sample C. These results suggest that there might exist an arsenic partial pressure threshold, under which the indium absorption kinetics plays a key role. As shown in Fig. 1e and 1f, both vertical and inclined NWs were grown simultaneously on the substrate in sample C. To understand the morphology, growth direction(s), and facet information of the inclined NWs, detailed SEM investigations were performed. Fig. 3a is a typical top-view SEM image, from which many inclined NWs with respect to the substrate can be confirmed, and the projections of the inclined NWs are along <112> directions, suggesting that general growth directions of the inclined NWs may be along the <100> or <11n> directions where n is an integer if the NWs have the ZB structure. Fig. 3b and 3c are high-magnification SEM

155

images of two typical inclined NWs projected in the same direction, and show different morphologies, indicating that there exist two types of inclined NWs. In order to determine the growth direction(s) of these inclined epitaxial NWs, the top-view SEM specimen was tilted along the [1 ̅0] axis until the electron beam is parallel to these inclined NWs. In this way, the growth direction and sidewall facets of individual NWs can be determined explicitly. Fig. 3d and 3e are a pair of ~35° tilted SEM images of a NW that are viewed edge-on, focused at the bottom and top of the NW, respectively. Through careful crystallographic analysis, the growth direction of this NW can be determined as the [ ̅ ̅ 0] direction, as marked in Fig. 3e. The comparison of the top and bottom of the NW indicates the pronounced tapering morphology with a large base and a small top, caused by the lateral growth of the NW during its axial growth.39 The side facets of this NW show an elongated hexagon cross section, consistent with the our previous reported side facets of < ̅ ̅0> NWs.40 Taking the cleavage planes of the GaAs substrate being {110} as a reference, the side facets can be determined as four {111} planes and two {110} planes (all have low surface energies), as illustrated in Fig. 3f. When the substrate was titled ~55° along the [1 ̅0] axis, another type of edge-on NW can be seen, and an example is shown in Fig. 3g and 3h. In this case, the growth direction of the NW can be determined as [00 ̅]. It can be observed that the NW sidewall has a rectangular cross-section with four {110} facets (typical side facets feature of <00 ̅> NWs) 16, namely the ( ̅10), (110), ( ̅0), and ( ̅ ̅0) planes, as illustrated in Fig. 3i. Based on our detailed analyses, we found two kinds of inclined NWs with their growth directions being < ̅ ̅0> or <00 ̅ >. Although the projections of < ̅ ̅0> and

<00 ̅> NWs on the GaAs {111}B substrate are the same, being the <112> directions they are clearly distinguishable in SEM images since they have distinct morphology (as shown in Fig. 3b and 3c), and sidewall facet shape (as shown in Fig. 3d and 3g). To understand the structural characteristics of these inclined NWs in sample C, TEM investigations were performed. Fig. 4a is a bright-field TEM image of a typical < ̅ ̅0> NW, in which the NW has a tapered morphology due to the lateral growth of NW which is consistent with our SEM observation. Fig. 4b shows a selected area electron diffraction (SAED) pattern of nanowire taken along the [ ̅10] zone axis. According to the SAED pattern, the growth direction of this inclined NW can be determined as [ ̅ ̅0], and the crystal phase of NW can be determined to be the ZB

156

structure which is different to the WZ structure of the vertical NWs. Fig. 4c is a high- resolution TEM image taken from the middle of the NW, which confirms that the NW has the ZB structure and shows no planar defects along the whole NW length. Our extensive high-resolution TEM studies on over a dozen of < ̅ ̅0> NWs confirmed the defect-free ZB structure nature along the entire NW length. Similar TEM investigations were performed on the <00 ̅> NWs. Fig. 4d shows the overview of a typical <00 ̅> NW, and Fig. 4e is the corresponding SAED pattern, from which the NW growth direction can be determined to be [ ̅], and the NW crystal phase can be identified as the ZB structure as well. High-resolution TEM image (Fig. 4f) taken from the middle of the NW indicates that the nanowire has defect-free ZB structure. To confirm this structural characteristic, we have examined over a dozen of < ̅> NWs, in which the defect-free ZB structure nature along the entire NW length was confirmed in all cases. Based on our detailed electron microscopy investigations, we found that all the inclined NWs grown under this V/III ratio adopt the defect-free ZB structure along their entire NW lengths. When compared with sample C, inclined NWs grown in sample D tend to be randomly. Fig. 5a is a TEM image taken from a typical inclined NW, in which irregular NW morphology can be observed. Fig. 5b is a high-resolution TEM image taken from the middle of the NW viewed along the <110> zone axis, in which the NW adopts ZB structure and has a high density of stacking faults on both {111} planes. Fig. 5c shows a high-resolution TEM image of the NW top, in which the defected ZB structural characteristic can be confirmed. Through careful study of the catalyst and its underlying nanowire, we note that the catalyst/NW interface is not flat. The fact that the {111} planes in both <111> directions adopt stacking faults suggests that the catalyst and catalyst/nanowire interface were not stable during the NW growth under such a low V/III ratio, leading to the formation of non-flat catalyst/nanowire interface, and in turn leading to the formation defected structure in NW.15 Our extensive high- resolution TEM results show that nearly all the inclined nanowires have defected ZB structure along the whole NW length. Based on these structural investigations, it can be concluded that, under this low V/III ratio, although ZB structured NWs can be induced, they tend to grow in an irregular manner and adopt many planar defects in the NWs.

157

In this regard, the V/III ratio engineering provides a new strategy to control the structural quality of ZB structured InAs NWs and can be used to grow defect-free ZB structured InAs NWs in MBE. Since the crystal phase and structure quality is critical to the performance of NW-based devices, we expect that our defect-free ZB InAs NWs would lead to superior device performance. Based on the detailed experimental results and discussion outlined above, the effect of V/III ratio on the growth of InAs NWs on the GaAs {111}B substrates can be summarized, as schematically illustrated in Fig. 6. Under a high V/III ratio (~70), InAs NWs cannot be induced and only pyramidal bases with Au catalysts on their tops are formed on the substrate surface. This is due to the competition between layer growth and one-dimensional NW growth, substrate surface reconstruction, and/or the low indium composition in the catalyst which did not reach the catalyst supersaturation threshold required to induce NW growth. According to the well-proved nanowire nucleation model,15, 40 the nucleation energy barrier ΔG* for the growth of inclined

NWs is higher than that of vertical <111>B/<0001> NWs. In this regard, the formation of inclined NWs with ZB structure is preferred at lower V/III ratio and/or higher catalyst supersaturation. When the V/III ratio is decreased to ~45, the nucleation energy barrier ΔG* for the growth of inclined NWs has not reached under such condition. Therefore, only vertical NWs with defected WZ structure can be triggered. With further decreasing the V/III ratio to ~20, both vertical and inclined NWs can be induced. It was found that the structure quality of the vertical NWs was significantly improved and defect-free WZ structured InAs NWs were achieved. Meanwhile, it has been found that these inclined NWs have defect-free ZB structure. With even lower V/III ratio (~10), more inclined NWs were induced on the substrate. However, these inclined NWs grow in a random manner and tend to adopt a high density of planar defects. When the V/III ratio goes to extremely low (~5), NWs cannot be induced again due to the insufficient arsenic protection of the InAs decomposition by the Au, and many particles were formed on the surface. It can be noted that the V/III ratio window for InAs NW growth is relatively narrow, suggesting that special attentions need to be paid when the growth of InAs NWs is performed.

4. Conclusion In conclusion, we have demonstrated the control of the crystal phase and structural quality of Au-catalyzed InAs NWs on the GaAs {111}B substrates by tuning

158

the V/III ratio in MBE, and have determined the narrow V/III ratio window suitable for InAs NW growth. Our experimental results indicate that the structural quality of both WZ structured and ZB structured InAs NWs can be controlled by governing the V/III ratio. Furthermore, we have achieved the growth of defect-free ZB structured and defect-free WZ structured InAs NWs in MBE. This comprehensive study provides a significant insight into the crystal phase and structural quality controlled growth of InAs NWs through V/III ratio engineering.

Acknowledgement

159

Fig. 1. SEM images of grown InAs on GaAs substrates under different V/III ratios. (a, b) Top- view and tilted-view SEM images of sample A, respectively. The inset in (a) is a magnified SEM showing the catalyst on top of the pyramid. (c, d) Top-view and side-view SEM images of sample B, respectively. (e, f) Top-view and side-view SEM images of sample C, respectively. The inset in (e) shows the <112> directions on the {111}B substrate. (g, h) Top- view and side-view SEM images of sample D, respectively. (i, j) Top-view and tilted-view SEM images of sample E, respectively.

160

Fig. 2. TEM investigations of vertical InAs NWs. (a, c, e) The overview of typical NWs taken from samples B, C and D, respectively. (b) High-resolution TEM image showing defected WZ structure from sample B. (d, f) High-resolution TEM image showing defect-free WZ structure from sample C and D, respectively.

161

Fig. 3. (a) Top-view SEM image with an inset of sketching the orientation relationship between <110> and <112> directions on the (111)B substrate, solid red arrow and dashed blue arrow representing <110> and <112> directions, respectively. (b, c) Magnified SEM images of inclined NWs with different morphologies. (d, e) 35° tilted SEM images showing facet details of a typical < ̅ ̅0> NW, focused at the bottom and top of NW, respectively. (f) Schematic illustration of facets of the < ̅ ̅0> NW. (g, h) 55° tilted SEM images showing facet details of a typical <00 ̅ > NW, focused at the bottom and top of NW, respectively. (i) Schematic illustration of facets of the <00 ̅> NW.

162

Fig. 4. (a) TEM image of a typical < ̅ ̅0> NW. (b) The SAED pattern taken from the NW. (d) Schematic illustration of < ̅ ̅0> NW. (c) High-resolution TEM image of the NW showing defect-free ZB structure. (d) TEM image of a typical <00 ̅> NW. (e) The SAED pattern of the NW. (f) High-resolution TEM image showing defect-free ZB structure.

163

Fig. 5. (a) TEM image of a typical inclined NW taken from sample D. (b) High-resolution TEM image taken from the marked area in (a) showing defected ZB structure. (c) High-resolution TEM image of the NW top region.

164

Fig. 6. Schematic illustration of the effect of V/III ratio on the InAs NW growth.

165

References

1. Duan, X. F.; Huang, Y.; Cui, Y.; Wang, J. F.; Lieber, C. M. Indium Phosphide Nanowires as Building Blocks for Nanoscale Electronic and Optoelectronic Devices. Nature 2001, 409, 66-69. 2. Xia, H.; Lu, Z. Y.; Li, T. X.; Parkinson, P.; Liao, Z. M.; Liu, F. H.; Lu, W.; Hu, W. D.; Chen, P. P.; Xu, H. Y., et al. Distinct Photocurrent Response of Individual Gaas Nanowires Induced by N-Type Doping. ACS Nano 2012, 6, 6005-6013. 3. Guo, N.; Hu, W. D.; Liao, L.; Yip, S.; Ho, J. C.; Miao, J. D.; Zhang, Z.; Zou, J.; Jiang, T.; Wu, S. J., et al. Anomalous and Highly Efficient Inas Nanowire Phototransistors Based on Majority Carrier Transport at Room Temperature. Adv. Mater. 2014, 26, 8232-8232. 4. Wagner, R. S.; Ellis, W. C. Vapor-Liquid-Solid Mechanism of Single Crystal Growth ( New Method Growth Catalyst from Impurity Whisker Epitaxial + Large Crystal Si E ). Appl. Phys. Lett. 1964, 4, 89-90. 5. Persson, A. I.; Larsson, M. W.; Stenstrom, S.; Ohlsson, B. J.; Samuelson, L.; Wallenberg, L. R. Solid-Phase Diffusion Mechanism for Gaas Nanowire Growth. Nat. Mater. 2004, 3, 677-681. 6. Dayeh, S. A.; Aplin, D. P. R.; Zhou, X.; Yu, P. K. L.; Yu, E. T.; Wang, D. High Electron Mobility Inas Nanowire Field-Effect Transistors. Small 2007, 3, 326-332. 7. Wright, S. L.; Marks, R. F.; Tiwari, S.; Jackson, T. N.; Baratte, H. In Situ Contacts to Gaas Based on Inas. Appl. Phys. Lett. 1986, 49, 1545-1547. 8. Milnes, A. G.; Polyakov, A. Y. Indium Arsenide: A Semiconductor for High Speed and Electro-Optical Devices. Mat. Sci. Eng. B 1993, 18, 237-259. 9. Thelander, C.; Martensson, T.; Bjork, M. T.; Ohlsson, B. J.; Larsson, M. W.; Wallenberg, L. R.; Samuelson, L. Single-Electron Transistors in Heterostructure Nanowires. Appl. Phys. Lett. 2003, 83, 2052-2054. 10. Doh, Y. J.; van Dam, J. A.; Roest, A. L.; Bakkers, E.; Kouwenhoven, L. P.; De Franceschi, S. Tunable Supercurrent through Semiconductor Nanowires. Science 2005, 309, 272-275. 11. Bjork, M. T.; Ohlsson, B. J.; Thelander, C.; Persson, A. I.; Deppert, K.; Wallenberg, L. R.; Samuelson, L. Nanowire Resonant Tunneling Diodes. Appl. Phys. Lett. 2002, 81, 4458-4460.

166

12. Zhou, X.; Dayeh, S. A.; Aplin, D.; Wang, D.; Yu, E. T. Direct Observation of Ballistic and Drift Carrier Transport Regimes in Inas Nanowires. Appl. Phys. Lett. 2006, 89, 053113. 13. Caroff, P.; Dick, K. A.; Johansson, J.; Messing, M. E.; Deppert, K.; Samuelson, L. Controlled Polytypic and Twin-Plane Superlattices in Iii-V Nanowires. Nature Nanotechnology 2009, 4, 50-55. 14. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Defect-Free <110> Zinc-Blende Structured Inas Nanowires Catalyzed by Palladium. Nano Lett. 2012, 12, 5744-5749. 15. Wang, J.; Plissard, S. R.; Verheijen, M. A.; Feiner, L.-F.; Cavalli, A.; Bakkers, E. P. A. M. Reversible Switching of Inp Nanowire Growth Direction by Catalyst Engineering. Nano Lett. 2013, 13, 3802-3806. 16. Krishnamachari, U.; Borgstrom, M.; Ohlsson, B. J.; Panev, N.; Samuelson, L.; Seifert, W.; Larsson, M. W.; Wallenberg, L. R. Defect-Free Inp Nanowires Grown in [001] Direction on Inp(001). Appl. Phys. Lett. 2004, 85, 2077-2079. 17. Paiman, S.; Gao, Q.; Tan, H. H.; Jagadish, C.; Pemasiri, K.; Montazeri, M.; Jackson, H. E.; Smith, L. M.; Yarrison-Rice, J. M.; Zhang, X., et al. The Effect of V/Iii Ratio and Catalyst Particle Size on the Crystal Structure and Optical Properties of Inp Nanowires. Nanotechnology 2009, 20, 225606. 18. Dick, K. A.; Bolinsson, J.; Messing, M. E.; Lehmann, S.; Johansson, J.; Caroff, P. Parameter Space Mapping of Inas Nanowire Crystal Structure. J. Vac. Sci. Technol. B 2011, 29, 04D103. 19. Joyce, H. J.; Wong-Leung, J.; Gao, Q.; Tan, H. H.; Jagadish, C. Phase Perfection in Zinc Blende and Wurtzite Iii-V Nanowires Using Basic Growth Parameters. Nano Lett. 2010, 10, 908-915. 20. Algra, R. E.; Verheijen, M. A.; Feiner, L.-F.; Immink, G. G. W.; van Enckevort, W. J. P.; Vlieg, E.; Bakkers, E. P. A. M. The Role of Surface Energies and Chemical Potential During Nanowire Growth. Nano Lett. 2011, 11, 1259-1264. 21. Lehmann, S.; Wallentin, J.; Jacobsson, D.; Deppert, K.; Dick, K. A. A General Approach for Sharp Crystal Phase Switching in Inas, Gaas, Inp, and Gap Nanowires Using Only Group V Flow. Nano Lett. 2013, 13, 4099-4105. 22. Johansson, J.; Karlsson, L. S.; Svensson, C. P. T.; Martensson, T.; Wacaser, B. A.; Deppert, K.; Samuelson, L.; Seifert, W. Structural Properties of <111>B-Oriented Iii-V Nanowires. Nat. Mater. 2006, 5, 574-580.

167

23. Shtrikman, H.; Popovitz-Biro, R.; Kretinin, A.; Houben, L.; Heiblum, M.; Bukala, M.; Galicka, M.; Buczko, R.; Kacman, P. Method for Suppression of Stacking Faults in Wurtzite Iii-V Nanowires. Nano Lett. 2009, 9, 1506-1510. 24. Lu, Z. Y.; Chen, P. P.; Liao, Z. M.; Shi, S. X.; Sun, Y.; Li, T. X.; Zhang, Y. H.; Zou, J.; Lu, W. Impact of Growth Parameters on the Morphology and Microstructure of Epitaxial Gaas Nanowires Grown by Molecular Beam Epitaxy. J. Alloy. Compd. 2013, 580, 82-87. 25. Liao, Z. M.; Chen, Z. G.; Lu, Z. Y.; Xu, H. Y.; Guo, Y. N.; Sun, W.; Zhang, Z.; Yang, L.; Chen, P. P.; Lu, W., et al. Au Impact on Gaas Epitaxial Growth on Gaas

(111)B Substrates in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 102, 063106. 26. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Jiang, N.; Tan, H. H.; Jagadish, C.; Zou, J. High-Density, Defect-Free, and Taper-Restrained Epitaxial Gaas Nanowires Induced from Annealed Au Thin Films. Cryst. Growth Des. 2012, 12, 2018-2022. 27. Zhang, Z.; Lu, Z. Y.; Chen, P. P.; Xu, H. Y.; Guo, Y. N.; Liao, Z. M.; Shi, S. X.; Lu, W.; Zou, J. Quality of Epitaxial Inas Nanowires Controlled by Catalyst Size in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 103, 073109. 28. Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Kim, Y.; Fickenscher, M. A.; Perera, S.; Hoang, T. B.; Smith, L. M.; Jackson, H. E., et al. High Purity Gaas Nanowires Free of Planar Defects: Growth and Characterization. Adv. Funct. Mater. 2008, 18, 3794-3800. 29. Reep, D. H.; Ghandhi, S. K. Electrical Properties of Organometallic Chemical Vapor Deposited Gaas Epitaxial Layers - Temperature Dependence. J. Electrochem. Soc. 1984, 131, 2697-2702. 30. Guo, Y. N.; Xu, H. Y.; Auchterlonie, G.; Burgess, T.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Shu, H. B.; Chen, X. S., et al. Phase Separation Induced by Au Catalysts in Ternary Ingaas Nanowires. Nano Lett. 2013, 13, 643-650. 31. Pugh, J. H.; Williams, R. S. Entropy-Driven Loss of Gas Phase Group V Species from Gold/Iii-V Compound Semiconductor Systems. J. Mater. Res. 1986, 1, 343-351. 32. Glas, F.; Harmand, J.-C.; Patriarche, G. Why Does Wurtzite Form in Nanowires of Iii-V Zinc Blende Semiconductors? Phys. Rev. Lett. 2007, 99, 146101. 33. Wallentin, J.; Ek, M.; Wallenberg, L. R.; Samuelson, L.; Deppert, K.; Borgstrom, M. T. Changes in Contact Angle of Seed Particle Correlated with Increased Zincblende Formation in Doped Inp Nanowires. Nano Lett. 2010, 10, 4807-4812.

168

34. Copel, M.; Reuter, M. C.; Kaxiras, E.; Tromp, R. M. Surfactants in Epitaxial Growth. Phys. Rev. Lett. 1989, 63, 632-635. 35. Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Kim, Y.; Fickenscher, M. A.; Perera, S.; Hoang, T. B.; Smith, L. M.; Jackson, H. E., et al. Unexpected Benefits of Rapid Growth Rate for Iii-V Nanowires. Nano Lett. 2009, 9, 695-701. 36. Johansson, J.; Dick, K. A.; Caroff, P.; Messing, M. E.; Bolinsson, J.; Deppert, K.; Samuelson, L. Diameter Dependence of the Wurtzite-Zinc Blende Transition in Inas Nanowires. J. Phys. Chem. C 2010, 114, 3837-3842. 37. Schwarz, K. W.; Tersoff, J. Elementary Processes in Nanowire Growth. Nano Lett. 2011, 11, 316-320. 38. Akiyama, T.; Sano, K.; Nakamura, K.; Ito, T. An Empirical Potential Approach to Wurtzite-Zinc-Blende Polytypism in Group Iii-V Semiconductor Nanowires. Jpn. J. Appl. Phys. 2006, 45, L275-L278. 39. Zou, J.; Paladugu, M.; Wang, H.; Auchterlonie, G. J.; Guo, Y. N.; Kim, Y.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C. Growth Mechanism of Truncated Triangular Iii-V Nanowires. Small 2007, 3, 389-393. 40. Zhang, Z.; Lu, Z. Y.; Xu, H. Y.; Chen, P. P.; Lu, W.; Zou, J. Structure and Quality Controlled Growth of Inas Nanowires through Catalyst Engineering. Nano Res. 2014, 7, 1640-1649.

169

170

8 Influence of Substrate Orientation on the Structural Quality of GaAs Nanowires in Molecular Beam Epitaxy

8.1 Introduction

171

therefore to alter the structural quality of GaAs nanowires grown on different substrates. Defect-free wurtzite structured GaAs nanowires grown on the GaAs (110) substrate have been achieved under our growth conditions.

8.2 Journal Publication

The results in Chapter 8 are included as it appears in Nanotechnology, 2015, 26, 255601. http://iopscience.iop.org/article/10.1088/0957-4484/26/25/255601/meta

172

Influence of substrate orientation on the structural quality of GaAs nanowires in molecular beam epitaxy

Zhi Zhang,1 Sui-Xing Shi,2 Ping-Ping Chen,2 Wei Lu,2 and Jin Zou1,3,*

1 Materials Engineering, The University of Queensland, St. Lucia, QLD 4072, Australia 2 National Laboratory for Infrared Physics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, 500 Yu-Tian Road, Shanghai 200083, China 3 Centre for Microscopy and Microanalysis, The University of Queensland, St. Lucia, QLD 4072, Australia

* Email: [email protected]

Abstract

In this study, the effect of substrate orientation on the structural quality of Au- catalyzed epitaxial GaAs nanowires grown by molecular beam epitaxy reactor has been investigated. It was found that the substrate orientations can be used to manipulate the nanowire catalyst composition and the catalyst surface energy, and therefore to alter the structural quality of GaAs nanowires grown on different substrates. Defect-free wurtzite structured GaAs nanowires grown on the GaAs (110) substrate have been achieved under our growth conditions.

173

III-V epitaxial semiconductor nanowires have been considered as building blocks for the advanced functional devices due to their potential applications in nanoelctronics and optoelectronics1-4 based on their superior physical and chemical properties, such as direct bandgap, high carrier mobility, and high photoluminescence efficiency.5-6 In general, III-V nanowires can be homoepitaxially or heterepitaxially grown on the III-V substrate via the vapor-liquid-solid mechanism 7 or vapor-solid-solid mechanism8 by dedicated growth techniques, such as, metalorganic chemical vapor deposition,9-11 chemical beam epitaxy,12 and molecular beam epitaxy (MBE).13-15 In order to realize and deliver the superior properties of one dimensional III-V nanowires in advanced nanoscale devices, minimizing the lattice defects in nanowires has been extremely important, and tremendous efforts have been devoted to control the structural quality by tuning the nanowire growth parameters16- 17. For example, the V/III ratio has been employed to successfully tune the structure of III-V nanowires between wurtzite and zinc-blende structures.18 Also, it has been demonstrated that thin nanowires induced by small catalysts tend to have much better wurtzite structure quality.19-20 Meanwhile, a two-temperature process with a high nucleation temperature and a low nanowire growth temperature has been adapted to achieve defect-free zinc-blende structure.21 Furthermore, non-gold catalyst has been proved to be able to induce defect-free zinc-blende structured III-V nanowires.22 In addition, it has been reported that the non-<111>/<0001> orientated nanowires tend to have less planar defects along the nanowires,23-24 so that controlling the nanowire growth direction provides a new approach to improve the nanowire structural quality. However, the effect of substrate orientation on the structural quality of Au-catalyzed epitaxial III-V nanowires grown by MBE has not been well studied. In this study, we demonstrate the effect of the substrate orientation on the structural quality of Au-catalyzed epitaxial GaAs nanowires grown by MBE. It was found that the orientations of different substrates can be used to control the Ga composition in the catalysts and the catalyst surface energy, and hence to control the structural quality of wurtzite structured GaAs nanowires. Defect-free wurtzite structured GaAs nanowires have been achieved for nanowires grown on the GaAs (110) substrate in our growth conditions.

174

The GaAs nanowires were simultaneously grown on the GaAs (111)B, GaAs

(211)A, and GaAs (110) substrates in a Riber 32 MBE system using Au as catalysts. The substrate surfaces were first degassed in the MBE preparation chamber at 250 °C, then were transferred to the growth chamber to be thermally deoxidized at

610 °C. After that, a thin GaAs buffer layer was grown on each substrate at 580 °C to achieve atomically flat surface, and the substrates were transferred back to the preparation chamber, and a thin Au film was deposited on the top of each GaAs buffer layer by the vacuum thermal evaporation. The Au-coated GaAs substrates were then transferred back to the growth chamber, and annealed at 560 °C for 5 minutes under As ambient to agglomerate the Au thin film into nanoparticles. After the annealing, the substrate temperature was lowered to 500 °C and the Ga source was introduced to initiate GaAs nanowire growth on different substrates simultaneously under the V/III ratio of ~20. The morphology and growth direction(s) of as-grown GaAs nanowires were analyzed by scanning electron microscopy (SEM, JEOL 7800F, operated at 5 kV), and their detailed structural characteristics and their catalyst compositions were investigated by transmission electron microscopy [TEM, Philips Tecnai F20, operated at 200 kV, equipped with X-ray energy dispersive spectroscopy (EDS) for compositional analysis]. Individual nanowires for TEM analysis were prepared by ultrasonicating the samples in ethanol for 20 min and then dispersing nanowires onto holey carbon films supported by Cu grids. Figure 1 shows SEM images of GaAs nanowires grown on GaAs substrates with different orientations. Figure 1(a) is a top-view SEM image of nanowires grown on the (111)B substrate, from which many island-like bases with catalysts on the top can be observed. To clarify their growth status, the <1 ̅0> side-view SEM specimen was investigated. The result is shown in Figure 1(b), in which the majority of nanowires were grown perpendicular to the substrate, suggesting that these epitaxial nanowires were grown along the <111>B direction of the GaAs substrate. Figure 1(c) schematically illustrates the nanowire growth on the (111)B substrate. Figure 1(d) shows the top-view SEM image of nanowires grown on the (211)A substrate, in which nanowires are inclined with respect to the substrate normal, and are projected in two directions on the (211)A surface. To determine the growth direction(s) of these inclined nanowires, the substrate was tilted until the axial direction of nanowires projected in one direction become parallel to the electron beam. Based on the tilting

175

geometry, the axial direction of these nanowires can be determined as <111>B as well. Similarly, nanowires projected in the other direction can also be determined as

<111>B direction. These results are consistent with the crystallography that there are two equivalent <111>B directions ([1 ̅ 1] and [11 ̅ ]) with respect to the (211)A substrate. Table I summarizes the possibility of <111>B directions on zinc-blende structured substrates with different orientations. To further confirm the <111>B growth directions of these inclined nanowires, <0 ̅1> side-view SEM investigations were performed, as shown in figure 1(e). It is of interest to note that all the nanowires are projected in the same direction, leading to the same inclined angle of ~35° respect to the substrate normal. Based on crystallography, both the projections of [1 ̅1] and [11 ̅] directions on the (0 ̅1) plane are the [100] direction. In this regard, the projections of [1 ̅1] and [11 ̅] nanowires on the (0 ̅1) plane result in the same inclined angle of ~35° with respect to the substrate normal (between [100] and [211]), as schematically illustrated in figure 1(f). Figure 1(g) and 1(h) show respectively top- view and <1 ̅0> side-view SEM images of nanowires grown on the (110) substrate, in which almost all the nanowires were projected in the same direction. According to the theoretical analyses (Table I), there is only one possible <111>B direction with respect to the (110) substrate, namely [11 ̅] with an elevation angle (defined as the angle between nanowire and substrate surface) of ~55°. By tilting nanowires until their axial direction parallel to the electron beam, the growth direction of inclined nanowires on the (110) substrate can be unambiguously determined as [11 ̅], which is schematically illustrated in figure 1(i). To understand the structural quality of nanowires grown on different substrates and the chemical composition of the nanowire catalysts, detailed TEM investigations were performed. Figure 2(a) is an overview TEM image of a typical nanowire grown on the (111)B substrate. Figures 2(b) and 2(c) are a bright-field (BF) TEM image and a high-resolution TEM (HRTEM) image, both taken from the middle of the nanowire, in which a large amount of planar defects are found to be perpendicular to the nanowire axial direction. Figure 2(d) is a corresponding selected-area electron diffraction (SAED) pattern, showing the wurtzite structured [11 ̅0] zone axis. Our extensive TEM investigations over a dozen of nanowires verified the defected wurtzite structure nature along the entire nanowires grown on the (111)B substrate.

Similar TEM investigations were performed on nanowires grown on the (211)A and

176

(110) substrates, and detailed results are shown in figures 2(f) – 2(i) and figures 2(k) – 2(n), respectively. It is of interest to note that the structural quality of nanowires grown on the (211)A and (110) substrates were improved. In particular, defect-free wurtzite structured nanowires were grown on the (110) substrate. By examining over a dozen of nanowires from each sample, the structural characteristics of nanowires were confirmed in all cases. These detailed experimental results indicate that the nanowire structural quality can be tuned by different substrates. Since structural quality of III-V nanowires can be controlled by the nature of catalysts,19 it is necessary to clarify the nature of catalysts in nanowires grown on different substrates. Accordingly, EDS analysis was performed on the catalysts of nanowires grown on the (111)B, (211)A, and (110) substrates, and results are shown respectively in figures 2(e), 2(j) and 2(o), in which the catalyst contains Au with or without Ga. It should be noted that the catalyst size of nanowires grown on three different substrates are distributed in the similar range of 25 ± 5 nm, suggesting that the catalyst diameter is not the dominant factor in controlling the nanowire structural quality. Our quantitative analysis of the EDS results indicate that the Ga concentration in the catalysts for the three different cases are 0 at.%, ~4 at.% and ~10 at.%, respectively. It has been well documented that, for III-V nanowire growth, the zinc-blende nucleation is favored at low supersaturation, while at higher supersaturation, the wurtzite structure is preferred.19, 25 On this basis, GaAs nanowires induced by Au catalysts with a high Ga supersaturation tend to form defect-free wurtzite structure when they are grown on the (110) substrate, while GaAs nanowires catalyzed by a very low Ga supersaturation prefer to form wurtzite structure with intermixing of zinc-blende structure when they are grown on the (111)B substrate, which match perfectly with our extensive TEM observations. As a consequence, our experiments provide direct evidences on how the substrate orientations affect the structural quality of GaAs nanowires, and suggest that under our growth conditions, the nanowire structural quality can be tuned by growing nanowires on substrates with different crystallographic orientations. To understand why catalyst composition changes for nanowires grown on GaAs substrates with different orientation, we consider the contribution of Ga sources to the nanowire growth front during the nanowire growth. Generally speaking, there are three main pathways for Ga atoms to be absorbed by the catalyst and in turn contributing to the Ga concentration in the catalyst:26-28 (1) direct impingement of

177

source Ga atoms on the catalyst surface,29 in which the Ga source impinges on the catalyst surface and then the Ga atoms can be absorbed by the catalyst; (2) diffusion Ga adatoms from nanowire side-walls to the catalyst;27 and (3) diffusion Ga adatoms from the substrate surface to the catalyst.30 Since the nanowire growth on different substrates was conducted in one batch under the same growth environment, the impingement rate and amount on the catalyst surface should be the same, i.e. the contribution of pathway (1) should be the same for all catalysts. In terms of the Ga adatom diffusion from nanowire side-walls, we note that, in our MBE growth chamber, the Ga injection beams are nearly perpendicularly introduced to the substrate surface. For the (111)B substrate, nanowires are grown perpendicular on the substrate and hence their side-walls are parallel to the incoming Ga beam. Therefore, the incoming Ga atoms should have less chance to impinge directly on the nanowire side-walls. However, since there is an elevation angle of nanowires grown on the

(211)A and (110) substrates (as stated in Table I), those inclined nanowires grown on these two substrates may have higher chances to collect Ga atoms onto their sidewalls that may diffuse to the catalysts and consequently contribute to the Ga concentration in the catalysts. In this regard, Ga concentration in catalysts of nanowires grown on the (211)A and (110) substrates tend to be higher than that of nanowires grown on the (111)B substrate. As for the third pathway, it is well documented that in the MBE system, the Ga adatoms diffused from the substrate surface make the main contribution to the catalyst composition.29-30 Based on this, the observed variations of catalyst compositions found in nanowires grown on different substrates may be controlled by the third pathway in the initial stage of nanowire growth. When the length of nanowire becomes sufficiently long, the second pathway may become dominant due to the fact that the inclined nanowires may collect more adatoms on their sidewalls. In order to clarity the effect of the substrate orientation on the Ga surface diffusion and hence on the catalyst composition, we note that, in MBE, Ga surface diffusion length increases in the order of the (111)B-related, (111)A-related, and (110) surfaces,31 with a very large Ga diffusion length on the (110) surface due to the surface reconstruction of the (110) surface.32-33 To understand the Ga diffusion length on the substrates with different orientations in our specific cases, we investigated the surface atom configurations on different surfaces. Figures 3(a) – 3(c) show respectively the side-view atomic configurations of the (111)B, (211)A and (110)

178

surfaces, all viewed along the <1 ̅0> direction. As can be noted from figure 3(b), the

(211)A surface is not atomically flat but can be dissociated into (111)A and (200) planes.34-35 Based on the facts that (1) Ga diffusion length on the Ga-terminated

(111)A surface is larger than that on the As-terminated (111)B surface, and (2) the

(n11)A surface with n ≤ 6 behaves like the (111)A surface while the (n11)A surface with n ≥ 7 behaves like the (200) surface,31 it is anticipated that the Ga diffusion length on the (211)A substrate is larger than that on the (111)B substrate. As shown in figures 3(c) and 3(d), in order to lower the surface potential energy, the GaAs (110) surface tends to be reconstructed. This is because each Ga surface atom rehybridizes into the sp2 configuration and bonds with its three nearest neighbors, while each As atom forms an s2p3 rehybridization with three p electrons forming bonds with its three nearest neighbors and two s electrons forming s2 orbits (so- called “lone pairs” and known to be very stable).36 In this case, any incoming Ga atom can easily diffuse on the reconstructed (110) surface and have large diffusion length on the substrate since they hardly react with the satisfied As surface atoms. Therefore, the Ga diffusion length on the (110) surface is way much larger than that on other surfaces, i.e., (111)B, (211)A surfaces. Based on these analyses, the Ga diffusion length λGa on different crystallographic surfaces can have the following sequence: λGa(111)B < λGa(211)A << λGa(110), as stated in figure 3(e). Taking all these three pathways of Ga atoms incorporated in the catalysts into account, we anticipate that the nanowire catalysts for the (110) substrate tend to absorb more Ga atoms, leading to a higher Ga concentration and in turn a higher Ga supersaturation in the catalysts, and hence lead to a better nanowire structure quality. However, we cannot rule out other possibilities that are responsible for the different structural qualities of GaAs nanowires grown on the different substrates, such as that inclined nanowires may collect more adatoms on the nanowire sidewalls. Another alternative explanation to the structural quality difference could be related to effect of contact angle between the catalysts and their underlying nanowires, which is closely correlated to the catalyst surface energy.37 According to the well- proved nucleation kinetics model in Au-catalyzed nanowires,25, 38 the nucleation barrier ratio between the formation of wurtzite and zinc-blende structures (훥퐺 and 37 훥퐺 ) can be expressed as:

휉 (1)

179

where 훥휇 is the liquid-solid chemical potential difference 휓 is the extra cohesive energy associated with wurtzite formation and η is the ratio between the changes in effective surface energies due to the formation of a wurtzite and a zinc-blende nucleus (훤 and 훤 ) which is expressed as:

𝜂 (2)

where 훶 , 훶 , 훶 are the interfacial energies of the vapor-liquid-solid system x is the share of nucleus perimeter in contact with the vapor is the contact angle and τ

= 훶 /훶 is the ratio between the lateral solid-vapor surface energies of wurtzite and ZB nuclei which is slightly less than 1. According to the Eq. (1) lower η favors the formation of wurtzite structure. Otherwise the formation of zinc-blende structure will be preferred. Based on Eq. (2) we anticipate that a higher 훶 or more precisely an decrease of the 훶 /훶 ratio will decease η and in turn increase the probability of wurtzite formation. It has been well documented that the surface energy 훶 increases with increasing the Ga concentration in the catalysts.39 In this regard we expect that nanowires grown on the (110) substrate with higher Ga concentration in the catalysts tend to increase

훶 more significantly and in turn promote the formation of wurtzite nucleation and hence have better wurtzite structural quality. In conclusion, we demonstrated the effect of substrate orientations on the structural quality of Au-catalyzed epitaxial GaAs nanowires in a single growth by MBE. It was found that the substrates with different crystallographic orientations can be used to control the nanowire catalyst composition and catalyst surface energy, and hence to control the structural quality of nanowires. Our directed evidences suggest that, due to the larger Ga diffusion length and tilted morphology of nanowires on the (110) substrate, defect-free GaAs nanowires can be achieved when they are grown on the (110) substrate under our growth conditions.

Acknowledgements This study is financially supported by the Australian Research Council, the National Basic Research Program of China (Grant No. 2011CB925604), the National Science Foundation of China (Grant No. 61376015, 91321311, 11334008 and 91121009) and Shanghai Science and Technology Foundation

180

(Grant No. 13JC1405901). Australian Microscopy & Microanalysis Research Facility is also gratefully acknowledged for providing microscopy facilities for this study.

181

TABLE I. List of possible <111>B nanowires on different zinc-blende substrate orientations.

Number of Substrate Elevation angle Specific <111> equivalent <111> B orientation B (degree) directions directions (111)B 1 90 [ ̅ ̅ ̅] (211)A 2 28.1 [1 ̅1], [11 ̅] (110) 1 54.7 [11 ̅]

182

Figure 1. SEM investigations of GaAs nanowires grown on GaAs substrates with different orientations. (a-c) Top-view, side-view and schematic illustration of nanowires grown on the

(111)B substrate, respectively. (d-f) Top-view, side-view and schematic illustration of nanowires grown on the (211)A substrate, respectively. (g-i) Top-view, side-view and schematic illustration of nanowires grown on the (110) substrate, respectively.

183

Figure 2. (a,f,k) TEM images of typical nanowires grown on (111)B, (211)A and (110) substrates, respectively. (b,g,l) The corresponding BF TEM images taken along the [11 ̅0] zone axis. (c,h,m) The corresponding high-resolution TEM images. (d,i,n) The corresponding SAED patterns. (e,j,o) The Corresponding EDS spectra of the nanowire catalysts.

184

Figure 3. (a-d) Schematic illustration showing the atom configuration of different surfaces projected in the <1 ̅0> direction, and (e) the Ga diffusion length order on different surfaces.

185

References

1. Lieber, C. M. Semiconductor Nanowires: A Platform for Nanoscience and Nanotechnology. Mrs. Bull. 2011, 36, 1052-1063. 2. Bjork, M. T.; Ohlsson, B. J.; Thelander, C.; Persson, A. I.; Deppert, K.; Wallenberg, L. R.; Samuelson, L. Nanowire Resonant Tunneling Diodes. Appl. Phys. Lett. 2002, 81, 4458-4460. 3. Xia, H.; Lu, Z. Y.; Li, T. X.; Parkinson, P.; Liao, Z. M.; Liu, F. H.; Lu, W.; Hu, W. D.; Chen, P. P.; Xu, H. Y., et al. Distinct Photocurrent Response of Individual Gaas Nanowires Induced by N-Type Doping. ACS Nano 2012, 6, 6005-6013. 4. Guo, N.; Hu, W. D.; Liao, L.; Yip, S.; Ho, J. C.; Miao, J. D.; Zhang, Z.; Zou, J.; Jiang, T.; Wu, S. J., et al. Anomalous and Highly Efficient Inas Nanowire Phototransistors Based on Majority Carrier Transport at Room Temperature. Adv. Mater. 2014, 26, 8232-8232. 5. Blakemore, J. S. Semiconducting and Other Major Properties of Gallium Arsenide. J. Appl. Phys. 1982, 53, R123-R181. 6. Vurgaftman, I.; Meyer, J. R.; Ram-Mohan, L. R. Band Parameters for Iii-V Compound Semiconductors and Their Alloys. J. Appl. Phys. 2001, 89, 5815-5875. 7. Wagner, R. S.; Ellis, W. C. Vapor-Liquid-Solid Mechanism of Single Crystal Growth ( New Method Growth Catalyst from Impurity Whisker Epitaxial + Large Crystal Si E ). Appl. Phys. Lett. 1964, 4, 89-90. 8. Persson, A. I.; Larsson, M. W.; Stenstrom, S.; Ohlsson, B. J.; Samuelson, L.; Wallenberg, L. R. Solid-Phase Diffusion Mechanism for Gaas Nanowire Growth. Nat. Mater. 2004, 3, 677-681. 9. Guo, Y. N.; Xu, H. Y.; Auchterlonie, G.; Burgess, T.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Shu, H. B.; Chen, X. S., et al. Phase Separation Induced by Au Catalysts in Ternary Ingaas Nanowires. Nano Lett. 2013, 13, 643-650. 10. Xu, H. Y.; Guo, Y. N.; Liao, Z. M.; Sun, W.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Catalyst Size Dependent Growth of Pd-Catalyzed One-Dimensional Inas Nanostructures. Appl. Phys. Lett. 2013, 102, 203108. 11. Dowdy, R. S.; Walko, D. A.; Li, X. Relationship between Planar Gaas Nanowire Growth Direction and Substrate Orientation. Nanotechnology 2013, 24, 035304.

186

12. Lugani, L.; Ercolani, D.; Sorba, L.; Sibirev, N. V.; Timofeeva, M. A.; Dubrovskii, V. G. Modeling of Inas-Insb Nanowires Grown by Au-Assisted Chemical Beam Epitaxy. Nanotechnology 2012, 23, 095602. 13. Zhang, Z.; Lu, Z. Y.; Xu, H. Y.; Chen, P. P.; Lu, W.; Zou, J. Structure and Quality Controlled Growth of Inas Nanowires through Catalyst Engineering. Nano Res. 2014, 7, 1640-1649. 14. Liao, Z. M.; Chen, Z. G.; Lu, Z. Y.; Xu, H. Y.; Guo, Y. N.; Sun, W.; Zhang, Z.; Yang, L.; Chen, P. P.; Lu, W., et al. Au Impact on Gaas Epitaxial Growth on Gaas

(111)B Substrates in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 102, 063106. 15. Tchernycheva, M.; Harmand, J. C.; Patriarche, G.; Travers, L.; Cirlin, G. E. Temperature Conditions for Gaas Nanowire Formation by Au-Assisted Molecular Beam Epitaxy. Nanotechnology 2006, 17, 4025-4030. 16. Plante, M. C.; LaPierre, R. R. Control of Gaas Nanowire Morphology and Crystal Structure. Nanotechnology 2008, 19, 495603. 17. Joyce, H. J.; Wong-Leung, J.; Gao, Q.; Tan, H. H.; Jagadish, C. Phase Perfection in Zinc Blende and Wurtzite Iii-V Nanowires Using Basic Growth Parameters. Nano Lett. 2010, 10, 908-915. 18. Lehmann, S.; Wallentin, J.; Jacobsson, D.; Deppert, K.; Dick, K. A. A General Approach for Sharp Crystal Phase Switching in Inas, Gaas, Inp, and Gap Nanowires Using Only Group V Flow. Nano Lett. 2013, 13, 4099-4105. 19. Zhang, Z.; Lu, Z. Y.; Chen, P. P.; Xu, H. Y.; Guo, Y. N.; Liao, Z. M.; Shi, S. X.; Lu, W.; Zou, J. Quality of Epitaxial Inas Nanowires Controlled by Catalyst Size in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 103, 073109. 20. Shtrikman, H.; Popovitz-Biro, R.; Kretinin, A.; Houben, L.; Heiblum, M.; Bukala, M.; Galicka, M.; Buczko, R.; Kacman, P. Method for Suppression of Stacking Faults in Wurtzite Iii-V Nanowires. Nano Lett. 2009, 9, 1506-1510. 21. Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Kim, Y.; Zhang, X.; Guo, Y. N.; Zou, J. Twin-Free Uniform Epitaxial Gaas Nanowires Grown by a Two-Temperature Process. Nano Lett. 2007, 7, 921-926. 22. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Defect-Free <110> Zinc-Blende Structured Inas Nanowires Catalyzed by Palladium. Nano Lett. 2012, 12, 5744-5749.

187

23. Wang, J.; Plissard, S.; Hocevar, M.; Vu, T. T. T.; Zehender, T.; Immink, G. G. W.; Verheijen, M. A.; Haverkort, J.; Bakkers, E. P. A. M. Position-Controlled [100] Inp Nanowire Arrays. Appl. Phys. Lett. 2012, 100, 053107. 24. Zhang Z.; Zheng K.; Lu Z. Y.; Chen P. P.; Lu W.; Zou J. Catalyst Orientation- Induced Growth of Defect-Free Zinc-Blende Structured <001 > Inas Nanowires. Nano Lett. 2015, 15, 876–882. 25. Glas, F.; Harmand, J.-C.; Patriarche, G. Why Does Wurtzite Form in Nanowires of Iii-V Zinc Blende Semiconductors? Phys. Rev. Lett. 2007, 99, 146101. 26. Dubrovskii, V. G.; Cirlin, G. E.; Soshnikov, I. P.; Tonkikh, A. A.; Sibirev, N. V.; Samsonenko, Y. B.; Ustinov, V. M. Diffusion-Induced Growth of Gaas Nanowhiskers During Molecular Beam Epitaxy: Theory and Experiment. Phys. Rev. B 2005, 71, 205325. 27. Dalacu, D.; Kam, A.; Austing, D. G.; Wu, X. H.; Lapointe, J.; Aers, G. C.; Poole, P. J. Selective-Area Vapour-Liquid-Solid Growth of Inp Nanowires. Nanotechnology 2009, 20, 395602. 28. Johansson, J.; Svensson, C. P. T.; Martensson, T.; Samuelson, L.; Seifert, W. Mass Transport Model for Semiconductor Nanowire Growth. J. Phys. Chem. B 2005, 109, 13567-13571. 29. Dubrovskii, V. G.; Sibirev, N. V.; Cirlin, G. E.; Soshnikov, I. P.; Chen, W. H.; Larde, R.; Cadel, E.; Pareige, P.; Xu, T.; Grandidier, B., et al. Gibbs-Thomson and Diffusion-Induced Contributions to the Growth Rate of Si, Inp, and Gaas Nanowires. Phys. Rev. B 2009, 79, 205316. 30. Lu, Z. Y.; Zhang, Z.; Chen, P. P.; Shi, S. X.; Yao, L. C.; Zhou, C.; Zhou, X. H.; Zou, J.; Lu, W. Bismuth-Induced Phase Control of Gaas Nanowires Grown by Molecular Beam Epitaxy. Appl. Phys. Lett. 2014, 105, 162102. 31. Takebe, T.; Fujii, M.; Yamamoto, T.; Fujita, K.; Watanabe, T. Orientation- Dependent Ga Surface Diffusion in Molecular Beam Epitaxy of Gaas on Gaas Patterned Substrates. J. Appl. Phys. 1997, 81, 7273-7281. 32. Lopez, M.; Nomura, Y. Surface Diffusion Length of Ga Adatoms in Molecular- Beam Epitaxy on Gaas(100)–(110) Facet Structures. J. Cryst. Growth 1995, 150, 68- 72. 33. Tong, S. Y.; Xu, G.; Hu, W. Y.; Puga, M. W. Vacancy Buckling Model for the (111) Surface of Iii–V Compound Semiconductors J. Vac. Sci. Technol. B 1985, 3, 1076-1078.

188

34. Zou, J.; Paladugu, M.; Wang, H.; Auchterlonie, G. J.; Guo, Y. N.; Kim, Y.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C. Growth Mechanism of Truncated Triangular Iii-V Nanowires. Small 2007, 3, 389-393. 35. Guo, Y. N.; Burgess, T.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Polarity- Driven Nonuniform Composition in Ingaas Nanowires. Nano Lett. 2013, 13, 5085- 5089. 36. Bringans, R. D.; Olmstead, M. A.; Uhrberg, R. I. G.; Bachrach, R. Z. Formation of the Interface between Gaas and Si: Implications for Gaas‐on‐Si Heteroepitaxy. Appl. Phys. Lett. 1987, 51, 523-525. 37. Wallentin, J.; Ek, M.; Wallenberg, L. R.; Samuelson, L.; Deppert, K.; Borgstrom, M. T. Changes in Contact Angle of Seed Particle Correlated with Increased Zincblende Formation in Doped Inp Nanowires. Nano Lett. 2010, 10, 4807-4812. 38. Zhang, Z.; Lu, Z.; Chen, P.; Lu, W.; Zou, J. Controlling the Crystal Phase and Structural Quality of Epitaxial Inas Nanowires by Tuning V/Iii Ratio in Molecular Beam Epitaxy. Acta Mater. 2015, 92, 25-32. 39. Algra, R. E.; Verheijen, M. A.; Feiner, L.-F.; Immink, G. G. W.; van Enckevort, W. J. P.; Vlieg, E.; Bakkers, E. P. A. M. The Role of Surface Energies and Chemical Potential During Nanowire Growth. Nano Lett. 2011, 11, 1259-1264.

189

190

9 Defect-Free Zinc-Blende Structured InAs Nanowires Realized by in Situ Two V/III Ratio Growth in Molecular Beam Epitaxy

9.1 Introduction

In this study, we devised a two-V/III-ratio procedure to control the Au-assisted growth of defect-free InAs nanowires in molecular beam epitaxy. The demonstrated two-V/III-ratio procedure consists of a first high-V/III-ratio growth step to prepare the nanowire foundation on the substrate surface, followed by a low-V/III-ratio step to induce the nanowire growth. By manipulating the V/III ratios in different steps, we

191

have achieved the controlled growth of pure defect-free zinc-blende structured InAs nanowires on the GaAs { ̅ ̅ ̅} substrates. This study provides an approach to control not only the crystal structure of semiconductor nanowires, but also their structural qualities.

9.2 Journal Publication

The results in Chapter 9 are included as it appeared in Nanoscale, 2015, 7, 12592- 12597. http://pubs.rsc.org/en/Content/ArticleLanding/2015/NR/C5NR03503A#!divAbstract

192

Defect-Free Zinc-Blende Structured InAs Nanowires Realized by in situ Two V/III Ratio Growth in Molecular Beam Epitaxy

Zhi Zhang,a Zhen-Yu Lu,b Ping-Ping Chen,b Wei Lu,b and Jin Zou*a ,c

In this study, we devised a two-V/III-ratio procedure to control the Au-assisted growth of defect-free InAs nanowires in molecular beam epitaxy. The demonstrated two V/III ratio procedure consists of a first high-V/III-ratio growth step to prepare the nanowire foundation on the substrate surface, followed by a low V/III ratio step to induce the nanowire growth. By manipulating the V/III ratios in different steps, we have achieved the controlled growth of pure defect-free zinc-blende structured InAs nanowires on the GaAs { ̅ ̅ ̅} substrates. This study provides an approach to control not only the crystal structure of semiconductor nanowires, but also their structural qualities.

193

Introduction One-dimensional III-V compound semiconductor nanowires have attracted extensive research interest in recent years due to their distinct physical properties and hence potential applications in nanoelectronics and optoelectronics.1-4 Currently, the most adopted techniques to grow epitaxial III-V nanowires are metal-organic chemical vapor deposition (MOCVD),5-7 molecular beam epitaxy (MBE)8-10 and chemical beam epitaxy11 through the vapor-liquid-solid mechanism12 or vapor-solid-solid mechanism.13 In order to achieve the practical applications of nanowires in nanodevices, it is of critical importance to control their crystal structure and structural quality. It has been found that most epitaxial III-V nanowires are generally grown along the <000 ̅> for the wurtzite structured nanowires or along the < ̅ ̅ ̅ > directions for zinc-blende structured nanowires.14-18 However, the majority of epitaxial <000 ̅ >/< ̅ ̅ ̅ > III-V nanowires naturally contain planar defects due to the small energetic differences of the stacking sequences in wurtzite and zinc-blende structures along their <000 ̅>/< ̅ ̅ ̅> directions. In this regard, several approaches have been employed to control the structural quality of III-V nanowires, such as, varying the nominal V/III ratio in MOCVD,19-22 adopting a two-temperature process,23-24 introducing a small amount of dopant to nanowires,25 and using different kinds of catalyst (such as Pd).26 Recently, the growth of III-V nanowires in the non-<000 ̅>/< ̅ ̅ ̅> growth directions has been reported, and these nanowires are generally defect-free and adopt zinc- blende structure.27-31 It has been well documented that Au-catalyzed III-V nanowires grown by MBE prefer to adopt the wurtzite structure with planar defects,32-34 and only very few studies have reported the growth of zinc-blende structured InAs nanowires in MBE.30-31 Based on our previous study on the investigations of InAs nanowires grown under a given V/III ratio at various V/III ratios on the GaAs { ̅ ̅ ̅} substrates in MBE,35 we found that the nanowire growth windows of varying the V/III ratio is narrow. With a relatively high V/III ratio, conventional vertical nanowires with defected wurtzite structure can be obtained. With reducing the V/III ratio, inclined nanowires with defect-free zinc-blende structure can be obtained with a tendency of the lower the V/III ratio, the more the defect-free inclined nanowires as long as the V/III ratio is within the growth window. Since both conventional vertical nanowires and inclined nanowires are always co-existed, alternative strategies should be

194

developed to secure the growth of defect-free zinc-blende structured nanowires exclusively on the GaAs { ̅ ̅ ̅} substrates. In this study, we demonstrate a two-step growth to induce pure defect-free zinc- blende structured InAs nanowires on the GaAs { ̅ ̅ ̅} substrates, in which two V/III ratios were used during the InAs nanowire growth. In the first step, a high V/III ratio is used to prepare the nanowire foundation on the substrate surface; which is followed by the second step where the nanowire growth is taken place at a low V/III ratio to induce high-quality zinc-blende structured nanowires.

Experimental

The InAs nanowires were grown on the GaAs { ̅ ̅ ̅} substrates in a Riber 32 MBE system using coated Au as catalysts and with In as the indium source and As4 as the arsenic source. The substrate surface was first degassed in the MBE preparation chamber at 250 °C, then was transferred to the growth chamber to be thermally deoxidized at 610 °C. After that, for each growth, a thin GaAs buffer layer was grown on the GaAs { ̅ ̅ ̅ } substrate at 580 °C to achieve atomically flat surface. The substrate was then transferred back to the preparation chamber, and a thin Au film was deposited on the top of the GaAs buffer layer by vacuum thermal evaporation. The Au-coated GaAs substrate was finally transferred to the growth chamber, and annealed at 500 °C for 5 min under arsenic ambient to agglomerate the Au thin film into nanoparticles. After the annealing, the substrate temperature was lowered to 400 °C and the indium source was introduced to initiate the InAs nanowire growth at different V/III ratios. To achieve the growth of exclusively defect-free zinc-blende structured nanowires on the GaAs { ̅ ̅ ̅ } substrates, a two-step growth was employed with the first step using a high-V/III ratio followed by a low-V/III-ratio in the second step. The morphological characteristics of the as-grown InAs nanowires and their bases were investigated by scanning electron microscopy (SEM, JEOL 7800F, operated at 10 kV), and their detailed structural characteristics were investigated by transmission electron microscopy (TEM, Philips Tecnai F20, operated at 200 kV). Individual nanowires for TEM analysis were prepared by ultrasonicating the as-grown nanowire samples in ethanol and then dispersing individual nanowires onto holey carbon films, supported by the Cu grids.

195

Results and discussion Fig. 1a and 1b show a top-view and a side-view SEM images taken from a nanowire sample grown under a V/III ratio of ~45 first for 30 min followed by a V/III ratio of ~15 for 15 min (the switch of V/III ratio was conducted in-situ to avoid the interruption of the growth). From which two kinds of nanowires can be seen: (1) long nanowires with vertical bottom section and kinked top section and (2) short inclined nanowires with irregular morphology. To clarify the growth behaviors of these nanowires, we repeated the InAs nanowire growth with the growth terminated after the first step (i.e. InAs nanowires were grown under a single high V/III ratio of ~45). Fig. 1c is a titled SEM image taken from this sample, in which two features associated with catalysts can be seen: in addition to those Au catalysts induced vertical nanowires, many Au catalysts failed to induce the nanowire growth, and they stationarily stayed on the substrate surface. Interestingly, such Au catalysts are always located on the tops of triangular pyramids (arrowed in Fig. 1c), which is formed by the attraction of indium through surface diffusion during the growth.32 This result suggests that in our two-V/III-ratio grown nanowires, for the long nanowires, the vertical sections correspond to the first step, while the kinked sections are related to the second step. On the other hand, the short nanowires were grown only in the second step and originated from those triangular pyramids on the substrate surface.

196

Fig. 1(a) Top-view SEM image and (b) side-view SEM image of InAs nanowires grown by a two-V/III-ratio procedure. (c) Tilt SEM image of InAs nanowires grown by a single-V/III-ratio at ~45.

To understand the structural characteristics of these long and short nanowires, TEM investigations were performed on over a dozen of nanowires from each type. Fig. 2a shows a bright-field TEM image of a typical long nanowire, from which the kinked morphology can be clearly observed. The inset is a <11 ̅0> zone-axis high- resolution TEM (HRTEM) image showing that the vertical section adopts the wurtzite structure with many planar defects. Our extensive HRTEM studies confirmed the nature of defected wurtzite structure in the vertical section of the long nanowires. On the other hand, Fig. 2b shows the overview of a typical short inclined nanowire, in which the nanowire shows irregular morphology, which is consistent with our SEM study (as can be seen from Fig. 1b, the morphology of inclined nanowires is not well controlled and the inclined features tend to be random). Fig. 2c is a bright-field TEM

197

image taken from the middle of the nanowire, from which many planar defects can be clearly observed. Fig. 2d is the corresponding <110> zone-axis HRTEM image of the nanowire, and shows that the short nanowire has the zinc-blende structure with planar defects on their 36 atomic planes, which is different to the long nanowires with the wurtzite structure. By investigating over a dozen of short nanowires, this defected zinc-blende structure characteristic can be confirmed. It is of interest to note that, the imperfection of the side-walls of these short inclined nanowires are caused by locations of planar defects, leading to irregular side-walls (arrowed in Fig. 2c).

Fig. 2 TEM investigations of both kinds of InAs nanowires. (a) The overview of a typical long nanowire with inset of HRTEM image showing defected wurtzite structure. (b) The overview of a typical short inclined nanowire. (c) Bright-field TEM image taken from a section of the nanowire shown in (b). (d) HRTEM image taken from the marked area in (c).

Based on the experimental results outlined above, we anticipate that using the two-V/III-ratio procedure may induce the growth of pure defect-free zinc-blende structured InAs nanowires. Based on Fig. 1c, the growth with a high V/III ratio of ~45 leads to the formation of both vertical nanowires and triangular pyramids with Au catalysts on their tops. Since these vertical nanowires have the defected wurtzite structure (refer to Fig. 2a), they need to be restrained. Therefore, during the first step of the two-V/III-ratio procedure, the V/III ratio should be further increased to secure

198

the formation of triangular pyramids only on the substrate surface. To clarify this assumption, we performed a nanowire growth with a single-V/III-ratio ratio ~55, and Fig. 3a shows the result where a tilt SEM image shows only triangular pyramids are formed. On the other hand, the quality of short inclined nanowires in our two-V/III-ratio grown nanowires is poor (refer to Fig. 2b), which may be due to the V/III ratio in the second step being set too low.35 In this regard, in order to improve the structural quality of these short nanowires, a relatively higher V/III ratio should be set in the second step. Accordingly, we redesign a two-V/III-ratio procedure to exclusively grow defect-free zinc-blende structured InAs nanowires on the GaAs { ̅ ̅ ̅} substrates. In this case, we set the growth to initiate at a high V/III ratio of ~55 for 35 min, then switch to a low V/III ratio of ~20 for 7 min. Fig. 3b is the top-view SEM image taken from this newly designed sample, and shows many inclined nanowires with certain orientations grown on the substrate. Based on the fact the cleavage planes of zinc- blende structured GaAs are {110} planes, the crystallographic orientation of the substrate can be determined, as marked in Fig. 3b. Therefore, the projections of the inclined nanowires with respect to the GaAs { ̅ ̅ ̅ } substrate surface can be determined as the <112> directions, suggesting that the growth directions of these inclined nanowires are where m and n are integers if the nanowires adopt the zinc-blende structure. Fig. 3c is a side-view SEM image, in which only inclined nanowires were found to be grown directly from the substrate as expected, and these inclined nanowires tend to have regular morphology, suggesting the success growth of purely inclined nanowires on the GaAs { ̅ ̅ ̅} substrate.

199

Fig. 3 (a) Tilt SEM image of InAs growth at a single-V/III-ratio of ~55. (b, c) SEM images of InAs nanowires grown by the two-V/III-ratio procedure (first at a V/III ratio of ~55, and switch to a V/III ratio of ~20). (b) Top-view SEM image. The inset showing a sketch of the orientation relationship between <112> (purple) and <110> (red) directions. (c) Side-view SEM image.

TEM investigations were performed on these inclined nanowires to determine their structural characteristics. Fig. 4a shows the overview of a typical inclined nanowire. In strong contrast to the short inclined nanowires found in the previous sample (refer to Fig. 1 and 2), the inclined nanowires in this sample have perfect morphology. Fig. 4b is low-magnified HRTEM image showing no lattice defects in a larger area and Fig. 4c is a <110> zone-axis HRTEM image, in which the nanowire is shown to have the zinc-blende structure as expected. The inset in Fig. 4c is the corresponding fast Fourier transform (FFT). From which, the <110> growth direction of the nanowire can be clarified. Our extensive HRTEM studies on over a dozen of inclined nanowires confirmed this defect-free zinc-blende structure nature along entire nanowires and

200

they are indeed grown along the <110> directions. In this regard, the growth of pure defect-free zinc-blende structured nanowires has been successfully achieved by our two-V/III-ratio procedure in MBE. It should be noted that the growth directions of our inclined nanowires is different to those inclined nanowires induced by the multiple- order twinning.37-38 In their cases, multiple-order twinning induced non-vertical nanowires still belong to one of the <111> directions, however, our inclined nanowires belong to the non-<111> directions. It has been documented that the growth directions of nanowires can affect their properties,39-40 so that our defect-free nanowires grown in the non-<111> directions would have different properties when compared with those conventional <111> nanowires.

Fig. 4 TEM investigations of InAs nanowires from newly designed growth. (a) The overview of a typical nanowire. (b) Low-magnified HRTEM image taken from the top of nanowire as marked in (a). (c) <110> zone-axis HRTEM image taken from the marked area in (b), and the inset is the corresponding FFT.

Based on these experimental results and discussion outlined above, different growth models of Au-catalyzed InAs nanowires on the GaAs { ̅ ̅ ̅} substrates in MBE can be summarized, as schematically illustrated in Fig. 5. After the annealing

201

process, catalyst droplets were formed on the GaAs substrate, as shown in Fig. 5a. When the growth was conducted by a single V/III ratio procedure with a high V/III ratio (~45), only the vertical nanowires and triangular pyramids can be induced, as illustrated in Fig. 5b, and experimentally confirmed in Fig. 5c. With a nanowire growth under a lower single V/III ratio (~20), both vertical and inclined nanowires can be induced simultaneously, as illustrated in Fig. 5d and experimentally demonstrated in Fig. 5e.35 The formation of our inclined nanowires under a low V/III ratio can be explained by the well-proved nanowire nucleation model28, 31, which suggested that the nucleation energy barrier for the growth of inclined nanowires is higher than that of vertical nanowires. Therefore, the inclined nanowires can be induced at a relatively low V/III ratio and/or at a higher catalyst supersaturation which would decrease the nucleation energy barrier. Importantly, if the nanowire growth was carried out under a two-V/III-ratio procedure with a high-V/III ratio of ~55 first followed by a low-V/III ratio of ~20, defect-free and zinc-blende structured nanowires can be induced. As illustrated in Fig. 5f , with the growth of a high V/III ratio (~55), only the triangular pyramids were formed during InAs growth, and when the V/III ratio is consequently lowered (to ~20), defect-free and zinc-blende structured nanowires can be grown from these pyramids, as shown schematically in Fig. 5f and experimentally in Fig. 5h. Therefore, as can be witnessed in SEM images shown in Fig. 5c, 5e and 5h, by manipulating the V/III ratio and performing a two-V/III-ratio procedure, controlled growths of solely vertical wurtzite structured InAs nanowires, the mixture of vertical and inclined nanowires, and solely inclined defect-free zinc- blende structured nanowires can be realized.

Fig. 5 Schematic illustrating the controlled growth of InAs nanowires under different V/III ratios and different growth procedures.

202

Conclusions In conclusion, we demonstrated the controlled growth of high-quality zinc-blende structured InAs nanowires in MBE by a two-V/III-ratio procedure which consists of a high-V/III ratio growth step to prepare the nanowire foundation on the substrate surface, followed by a low-V/III ratio step to induce nanowire growth. By tuning the V/III ratios in the two-V/III-ratio procedure, we have successfully achieved the growth of pure defect-free zinc-blende structured InAs nanowires in MBE. This novel growth approach provides a new strategy to control the crystal structure and quality of semiconductor nanowires.

Acknowledgements This study is financially supported by the Australian Research Council, the National Basic Research Program of China, and the National Science Foundation of China. Australian Microscopy & Microanalysis Research Facility is also gratefully acknowledged for providing microscopy facilities for this study.

Notes and references a Materials Engineering, The University of Queensland, St. Lucia, QLD 4072, Australia b National Laboratory for Infrared Physics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, 500 Yu-Tian Road, Shanghai 200083, China c Centre for Microscopy and Microanalysis, The University of Queensland, St. Lucia, QLD 4072, Australia

203

References 1. Duan, X. F.; Huang, Y.; Cui, Y.; Wang, J. F.; Lieber, C. M. Indium Phosphide Nanowires as Building Blocks for Nanoscale Electronic and Optoelectronic Devices. Nature 2001, 409, 66-69. 2. Xia, H.; Lu, Z. Y.; Li, T. X.; Parkinson, P.; Liao, Z. M.; Liu, F. H.; Lu, W.; Hu, W. D.; Chen, P. P.; Xu, H. Y., et al. Distinct Photocurrent Response of Individual Gaas Nanowires Induced by N-Type Doping. ACS Nano 2012, 6, 6005-6013. 3. Guo, N.; Hu, W. D.; Liao, L.; Yip, S.; Ho, J. C.; Miao, J. D.; Zhang, Z.; Zou, J.; Jiang, T.; Wu, S. J., et al. Anomalous and Highly Efficient Inas Nanowire Phototransistors Based on Majority Carrier Transport at Room Temperature. Adv. Mater. 2014, 26, 8232-8232. 4. Krogstrup, P.; Jorgensen, H. I.; Heiss, M.; Demichel, O.; Holm, J. V.; Aagesen, M.; Nygard, J.; Fontcuberta i Morral, A. Single-Nanowire Solar Cells Beyond the Shockley-Queisser Limit. Nature Photonics 2013, 7, 306-310. 5. Guo, Y. N.; Xu, H. Y.; Auchterlonie, G.; Burgess, T.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Shu, H. B.; Chen, X. S., et al. Phase Separation Induced by Au Catalysts in Ternary Ingaas Nanowires. Nano Lett. 2013, 13, 643-650. 6. Zou, J.; Paladugu, M.; Wang, H.; Auchterlonie, G. J.; Guo, Y. N.; Kim, Y.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C. Growth Mechanism of Truncated Triangular Iii- V Nanowires. Small 2007, 3, 389-393. 7. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Jiang, N.; Tan, H. H.; Jagadish, C.; Zou, J. High-Density, Defect-Free, and Taper-Restrained Epitaxial Gaas Nanowires Induced from Annealed Au Thin Films. Cryst. Growth Des. 2012, 12, 2018-2022. 8. Zhang, Z.; Lu, Z. Y.; Chen, P. P.; Xu, H. Y.; Guo, Y. N.; Liao, Z. M.; Shi, S. X.; Lu, W.; Zou, J. Quality of Epitaxial Inas Nanowires Controlled by Catalyst Size in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 103, 073109. 9. Dubrovskii, V. G. Influence of the Group V Element on the Chemical Potential and Crystal Structure of Au-Catalyzed Iii-V Nanowires. Appl. Phys. Lett. 2014, 104, 053110. 10. Rieger, T.; Luysberg, M.; Schaepers, T.; Gruetzmacher, D.; Lepsa, M. I. Molecular Beam Epitaxy Growth of Gaas/Inas Core-Shell Nanowires and Fabrication of Inas Nanotubes. Nano Lett. 2012, 12, 5559-5564.

204

11. Jensen, L. E.; Bjork, M. T.; Jeppesen, S.; Persson, A. I.; Ohlsson, B. J.; Samuelson, L. Role of Surface Diffusion in Chemical Beam Epitaxy of Inas Nanowires. Nano Lett. 2004, 4, 1961-1964. 12. Wagner, R. S.; Ellis, W. C. Vapor-Liquid-Solid Mechanism of Single Crystal Growth ( New Method Growth Catalyst from Impurity Whisker Epitaxial + Large Crystal Si E ). Appl. Phys. Lett. 1964, 4, 89-90. 13. Persson, A. I.; Larsson, M. W.; Stenstrom, S.; Ohlsson, B. J.; Samuelson, L.; Wallenberg, L. R. Solid-Phase Diffusion Mechanism for Gaas Nanowire Growth. Nat. Mater. 2004, 3, 677-681. 14. Caroff, P.; Dick, K. A.; Johansson, J.; Messing, M. E.; Deppert, K.; Samuelson, L. Controlled Polytypic and Twin-Plane Superlattices in Iii-V Nanowires. Nature Nanotechnology 2009, 4, 50-55. 15. Glas, F.; Harmand, J.-C.; Patriarche, G. Why Does Wurtzite Form in Nanowires of Iii-V Zinc Blende Semiconductors? Phys. Rev. Lett. 2007, 99, 146101. 16. Shi, S.; Zhang, Z.; Lu, Z.; Shu, H.; Chen, P.; Li, N.; Zou, J.; Lu, W. Evolution of Morphology and Microstructure of Gaas/Gasb Nanowire Heterostructures. Nanoscale Res. Lett. 2015, 10, 108. 17. Fonseka, H. A.; Caroff, P.; Wong-Leung, J.; Ameruddin, A. S.; Tan, H. H.; Jagadish, C. Nanowires Grown on Inp (100): Growth Directions, Facets, Crystal Structures, and Relative Yield Control. Acs Nano 2014, 8, 6945-6954. 18. Johansson, J.; Bolinsson, J.; Ek, M.; Caroff, P.; Dick, K. A. Combinatorial Approaches to Understanding Polytypism in Iii-V Nanowires. Acs Nano 2012, 6, 6142-6149. 19. Algra, R. E.; Verheijen, M. A.; Feiner, L.-F.; Immink, G. G. W.; van Enckevort, W. J. P.; Vlieg, E.; Bakkers, E. P. A. M. The Role of Surface Energies and Chemical Potential During Nanowire Growth. Nano Lett. 2011, 11, 1259-1264. 20. Lehmann, S.; Wallentin, J.; Jacobsson, D.; Deppert, K.; Dick, K. A. A General Approach for Sharp Crystal Phase Switching in Inas, Gaas, Inp, and Gap Nanowires Using Only Group V Flow. Nano Lett. 2013, 13, 4099-4105. 21. Ikejiri, K.; Kitauchi, Y.; Tomioka, K.; Motohisa, J.; Fukui, T. Zinc Blende and Wurtzite Crystal Phase Mixing and Transition in Indium Phosphide Nanowires. Nano Lett. 2011, 11, 4314-4318. 22. Ameruddin, A. S.; Fonseka, H. A.; Caroff, P.; Wong-Leung, J.; Veld, R. L. O. h.;

Boland, J. L.; Johnston, M. B.; Tan, H. H.; Jagadish, C. Inxga1−Xas Nanowires with

205

Uniform Composition, Pure Wurtzite Crystal Phase and Taper-Free Morphology. Nanotechnology 2015, 26, 205604. 23. Joyce, H. J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Kim, Y.; Zhang, X.; Guo, Y. N.; Zou, J. Twin-Free Uniform Epitaxial Gaas Nanowires Grown by a Two-Temperature Process. Nano Lett. 2007, 7, 921-926. 24. Hou, J. J.; Han, N.; Wang, F.; Xiu, F.; Yip, S.; Hui, A. T.; Hung, T.; Ho, J. C. Synthesis and Characterizations of Ternary Ingaas Nanowires by a Two-Step Growth Method for High-Performance Electronic Devices. Acs Nano 2012, 6, 3624- 3630. 25. Zhuang, Q. D.; Anyebe, E. A.; Chen, R.; Liu, H.; Sanchez, A. M.; Rajpalke, M. K.; Veal, T. D.; Wang, Z. M.; Huang, Y. Z.; Sun, H. D. Sb-Induced Phase Control of Inassb Nanowires Grown by Molecular Beam Epitaxy. Nano Lett. 2015, 15, 1109- 1116. 26. Xu, H. Y.; Wang, Y.; Guo, Y. N.; Liao, Z. M.; Gao, Q.; Tan, H. H.; Jagadish, C.; Zou, J. Defect-Free <110> Zinc-Blende Structured Inas Nanowires Catalyzed by Palladium. Nano Lett. 2012, 12, 5744-5749. 27. Krishnamachari, U.; Borgstrom, M.; Ohlsson, B. J.; Panev, N.; Samuelson, L.; Seifert, W.; Larsson, M. W.; Wallenberg, L. R. Defect-Free Inp Nanowires Grown in [001] Direction on Inp(001). Appl. Phys. Lett. 2004, 85, 2077-2079. 28. Wang, J.; Plissard, S. R.; Verheijen, M. A.; Feiner, L.-F.; Cavalli, A.; Bakkers, E. P. A. M. Reversible Switching of Inp Nanowire Growth Direction by Catalyst Engineering. Nano Lett. 2013, 13, 3802-3806. 29. Shtrikman, H.; Popovitz-Biro, R.; Kretinin, A.; Heiblumf, M. Stacking-Faults-Free Zinc Blende Gaas Nanowires. Nano Lett. 2009, 9, 215-219. 30. Zhang, Z.; Zheng, K.; Lu, Z. Y.; Chen, P. P.; Lu, W.; Zou, J. Catalyst Orientation- Induced Growth of Defect-Free Zinc-Blende Structured <001 > Inas Nanowires. Nano Lett. 2015, 15, 876–882. 31. Zhang, Z.; Lu, Z. Y.; Xu, H. Y.; Chen, P. P.; Lu, W.; Zou, J. Structure and Quality Controlled Growth of Inas Nanowires through Catalyst Engineering. Nano Res. 2014, 7, 1640-1649. 32. Liao, Z. M.; Chen, Z. G.; Lu, Z. Y.; Xu, H. Y.; Guo, Y. N.; Sun, W.; Zhang, Z.; Yang, L.; Chen, P. P.; Lu, W., et al. Au Impact on Gaas Epitaxial Growth on Gaas

(111)B Substrates in Molecular Beam Epitaxy. Appl. Phys. Lett. 2013, 102, 063106.

206

33. Dheeraj, D. L.; Patriarche, G.; Zhou, H.; Hoang, T. B.; Moses, A. F.; Gronsberg, S.; van Helvoort, A. T. J.; Fimland, B.-O.; Weman, H. Growth and Characterization of Wurtzite Gaas Nanowires with Defect-Free Zinc Blende Gaassb Inserts. Nano Lett. 2008, 8, 4459-4463. 34. Martelli, F.; Rubini, S.; Jabeen, F.; Felisari, L.; Grillo, V. On the Growth of Inas Nanowires by Molecular Beam Epitaxy. J. Cryst. Growth 2011, 323, 297-300. 35. Zhang, Z.; Lu, Z.; Chen, P.; Lu, W.; Zou, J. Controlling the Crystal Phase and Structural Quality of Epitaxial Inas Nanowires by Tuning V/Iii Ratio in Molecular Beam Epitaxy. Acta Mater. 2015, 92, 25-32. 36. Akabori, M.; Takeda, J.; Motohisa, J.; Fukui, T. Ingaas Nano-Pillar Array Formation on Partially Masked Inp(Iii)B by Selective Area Metal-Organic Vapour Phase Epitaxial Growth for Two-Dimensional Photonic Crystal Application. Nanotechnology 2003, 14, 1071-1074. 37. Uccelli, E.; Arbiol, J.; Magen, C.; Krogstrup, P.; Russo-Averchi, E.; Heiss, M.; Mugny, G.; Morier-Genoud, F.; Nygard, J.; Ramon Morante, J., et al. Three- Dimensional Multiple-Order Twinning of Self-Catalyzed Gaas Nanowires on Si Substrates. Nano Lett. 2011, 11, 3827-3832. 38. Russo-Averchi, E.; Heiss, M.; Michelet, L.; Krogstrup, P.; Nygard, J.; Magen, C.; Morante, J. R.; Uccelli, E.; Arbiol, J.; Fontcuberta i Morral, A. Suppression of Three Dimensional Twinning for a 100% Yield of Vertical Gaas Nanowires on Silicon. Nanoscale 2012, 4, 1486-1490. 39. Chin, A. H.; Ahn, T. S.; Li, H.; Vaddiraju, S.; Bardeen, C. J.; Ning, C.-Z.; Sunkara, M. K. Photoluminescence of Gan Nanowires of Different Crystallographic Orientations. Nano Lett. 2007, 7, 626-631. 40. Bao, J.; Bell, D. C.; Capasso, F.; Wagner, J. B.; Martensson, T.; Tragardh, J.; Samuelson, L. Optical Properties of Rotationally Twinned Inp Nanowire Heterostructures. Nano Lett. 2008, 8, 836-841.

207

208

10 Conclusions and Recommendations

10.1 Conclusions

This thesis investigated the structural characteristics of Au-catalyzed III-V nanowires grown by MBE. Through detailed electron microscopy investigations, the crystal phase and structural quality controlled growth of III-V nanowires, specifically, GaAs and InAs nanowires, has been successfully achieved, and the growth mechanisms were proposed to understand their growth behaviours. Detailed experimental

209

investigations have demonstrated that nanowire size, catalyst composition, catalyst orientation, substrate orientation, and V/III ratio can be used to tune the crystal phase and structural quality of III-V nanowires. The conclusions of this PhD thesis are summarized as follows.

 Size of the catalyst has been proved to affect the structural quality of InAs nanowires. Detailed electron microscopy investigations indicate that defect- free wurtzite structured InAs nanowires can be induced by smaller catalysts with a high In concentration, while comparatively larger catalysts containing less In induce defected wurtzite structured InAs nanowires. This study indicates that the structural quality of InAs nanowires can be controlled by the size of Au catalysts  Through catalyst engineering, the manipulation of crystal structure and structure quality of III-V semiconductor nanowires can be achieved. By tuning the indium concentration in the catalyst, defect-free wurtzite structured and defect-free zinc-blende structured InAs nanowires can be induced. It is found that these defect-free zinc-blende structured InAs nanowires grow along < ̅ ̅0> directions with four low-energy {111} and two {110} side-wall facets and adopt the ( ̅ ̅ ̅ ) catalyst/nanowire interface. Our structural and chemical characterizations and calculations identify the existence of catalyst supersaturation threshold for the InAs nanowire growth. When the In concentration in the catalyst is sufficiently high, defect-free zinc-blende structured InAs nanowires can be induced.  Crystal phase of the catalysts has been proved to manipulate the crystal structure and structural quality of III-V nanowires. It has been found that the catalysts and their underlying <00 ̅ > nanowires have the orientation

relationship of { ̅ }C//{ ̅}InAs and [ ̅ ]C//[1 ̅ ]InAs due to their small in- plane lattice mismatches between their corresponding lattice spacings perpendicular to the { ̅ atomic planes of the nanowires, leading to the formation of the { ̅ catalyst/nanowire interfaces, and consequently the formation of <00 ̅> nanowires.  Through the V/III ratio engineering, the controlled growth of high-quality wurtzite structured and zinc-blende structured InAs nanowires has been achieved. It has been found that InAs nanowires can only be grown in a

210

relatively narrow window of the V/III ratio. It is also demonstrated that the V/III ratio can be used to control the structural quality of wurtzite structured and zinc-blende structured InAs nanowires under low V/III ratios, and defect-free wurtzite structured and zinc-blende structured InAs nanowires were successfully achieved.  The effect of substrate orientation on the structural quality of Au-catalyzed epitaxial GaAs nanowires grown by molecular beam epitaxy reactor has been investigated. It was found that the substrate orientations can be used to manipulate the nanowire catalyst composition and the catalyst surface energy, and therefore to alter the structural quality of GaAs nanowires grown on different substrates. Defect-free wurtzite structured GaAs nanowires grown on the GaAs (110) substrate have been achieved under our growth conditions.  A two-V/III-ratio procedure has been proposed to control the Au-assisted growth of defect-free InAs nanowires in molecular beam epitaxy. The demonstrated two-V/III-ratio procedure consists of a first high-V/III-ratio growth step to prepare the nanowire foundation on the substrate surface, followed by a low-V/III-ratio step to induce the nanowire growth. By manipulating the V/III ratios in different steps, we have achieved the controlled growth of pure defect-free zinc-blende structured InAs nanowires on the GaAs { ̅ ̅ ̅} substrates.

10.2 Recommendations

The observations and conclusions that were made in the PhD thesis can be extended to explore the properties and applications of III-V nanowires. The suggestions are stated as follows.

 InAs nanowires have attracted special attention due to their narrow bandgap, relatively high electron mobility, and small electron effective mass, which made them a promising candidate for the applications in future optical and high-frequency electronic devices. With the obtained InAs nanwoires with different structural quality, it is necessary and interesting to investigate the effect of structural quality on their electrical properties.

211

 In addition, it has been demonstrated in this PhD thesis that InAs nanowires with different growth directions and crystal structures have been successfully grown by tuning the nanowire growth parameters. Due to the one-dimensional characteristic of nanowires, the crystal orientation of nanowires may also affect their performances and may lead to different applications in nanowire- based devices.

212