Copyright Undertaking · 2020. 7. 9. · enhancement of graphene field-effect transistor based on vertical tunnelling heterostructures, Advanced Materials. 28, 10048 (2016). 5. Weng-Fu

Copyright Undertaking

This thesis is protected by copyright, with all rights reserved.

By reading and using the thesis, the reader understands and agrees to the following terms:

1. The reader will abide by the rules and legal ordinances governing copyright regarding the use of the thesis.

2. The reader will use the thesis for the purpose of research or private study only and not for distribution or further reproduction or any other purpose.

3. The reader agrees to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

IMPORTANT If you have reasons to believe that any materials in this thesis are deemed not suitable to be distributed in this form, or a copyright owner having difficulty with the material being included in our database, please contact [email protected] providing details. The Library will look into your claim and consider taking remedial action upon receipt of the written requests.

Pao Yue-kong Library, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong

http://www.lib.polyu.edu.hk

2D FERROELECTRICITY AND

PIEZOELECTRICITY FOR ELECTRONIC

DEVICES

YUAN Shuoguo

PhD

The Hong Kong Polytechnic University

2020

The Hong Kong Polytechnic University

Department of Applied Physics

2D ferroelectricity and piezoelectricity for electronic devices

YUAN Shuoguo

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

of Doctor of Philosophy

August 2019

CERTIFICATE OF ORIGINALITY

I hereby declare that this thesis is my own work and that, to the best of my knowledge

and belief, it reproduces no material previously published or written, nor material that

has been accepted for the award of any other degree or diploma, except where due

acknowledgement has been made in the text.

(Signed)

YUAN Shuoguo (Name of student)

fffI YUAN Shuoguo

THE HONG KONG POLYTECHNIC UNIVERSITY

Abstract

Abstract

Two-dimensional (2D) layered materials have attracted enormous interests in

fundamental research and industrial applications due to their excellent physical

properties, such as electronic, optoelectric, thermal, optical, mechanical and magnetic

properties. Recently, ferroelectricity and piezoelectricity in 2D layered material

family were addressed, which leads to great potential applications for atomic-scale

smart electronic devices. The emerging 2D ferroelectricity and piezoelectricity can

yield fascinating behaviors with various functional devices.

In this thesis, firstly, the MoTe2 samples were prepared using mechanical

exfoliation and laser process. By combining first-principle calculations and

experimental studies, the robust room-temperature out-of-plane ferroelectricity is

realized in the monolayer MoTe2 with unexploited distorted 1T (d1T) phase. The

ferroelectricity origin in d1T-MoTe2 results from the spontaneous symmetry breaking

due to the relative atomic displacements of Mo atoms and Te atoms. A large ON/OFF

resistance ratio is achieved in ferroelectric devices composed of MoTe2 based van der

Waals (vdW) heterostructure. Secondly, piezoelectricity in 2D material family has

largely been predicted by theoretical calculations, but still few experimental

observations from them. The out-of-plane piezoelectric response can be achieved in

2D layered material In2Se3. In addition, a high-precision actuator device based on the

fffII YUAN Shuoguo


Abstract

piezoelectricity in 2D atomic scale is demonstrated. Moreover, the vertical

ferroelectric polarization characteristic of In2Se3 is investigated by piezoresponse

force microscopy, which can retain the ferroelectric behaviour down to 6 nm. Thirdly,

the vertical piezoelectric response of several 2D layered materials is theoretically

absent via electric field or strain applied perpendicular to their surface due to the

inversion symmetry. Recently, vertical piezoelectric properties of In2Se3 were

reported, which exhibits low piezoelectric response. A remarkably enhanced

out-of-plane piezoelectric performance in 2D layered vdW heterostructure is reported.

In particular, the vdW heterostructure based piezoelectricity is not restricted by

layer-dependence and edge-effect. The enhanced piezoelectric properties result from

the band offset. In addition, the vdW heterostructure can be extended to other 2D

materials like WS2. The heterostructure is fabricated onto silicon substrate which is

compatible with state-of-the-art microfabrication technology.

The 2D ferroelectricity and piezoelectricity have been studied. The ferroelectric,

piezoelectric properties and related device applications of 2D materials have been

investigated. These studies provide a platform for designing new functionalities and

achieving unexpected applications in atomic-scale smart electronic devices and

electromechanical systems.

fffIII YUAN Shuoguo


List of Publications


Journal Papers

1. Shuoguo Yuan, Xin Luo, Hung Lit Chan, Chengcheng Xiao, Yawei Dai, Maohai

Xie, and Jianhua Hao*. Room-temperature ferroelectricity in MoTe2 down to the

atomic monolayer limit, Nature Communications 10, 1775 (2019).

2. Shuoguo Yuan, Weng Fu Io, Jianfeng Mao, Yancong Chen, Xin Luo, and Jianhua

Hao*. Large out-of-plane piezoelectricity in two-dimensional layered material

indium selenide, 2020, to be submitted.

3. Shuoguo Yuan, Sin-Yi Pang, and Jianhua Hao*. 2D transition metal

dichalcogenides, carbides, nitrides and their applications in supercapacitors and

electrocatalytic hydrogen evolution reaction, Applied Physics Reviews 7, 021304

(2020).

4. Shuoguo Yuan, Zhibin Yang, Chao Xie, Feng Yan, Jiyan Dai, Shu Ping Lau,

Helen L.W. Chan, and Jianhua Hao*. Ferroelectric-driven performance

enhancement of graphene field-effect transistor based on vertical tunnelling

heterostructures, Advanced Materials 28, 10048 (2016).

5. Weng-Fu Io, Shuoguo Yuan, Lok-Wing Wong, Sin-Yi Pang, and Jianhua Hao*.

Temperature-and thickness-dependence of robust out-of-plane ferroelectricity in

fffIV YUAN Shuoguo



CVD grown ultrathin van der Waals α-In2Se3 layers, Nano Research

https://doi.org/10.1007/s12274-020-2640-0 (2020).

6. MeilinTu, Haipeng Lu, Songwen Luo, Hao Peng, Shangdong Li, Zhengdong Ke,

Shuoguo Yuan, Wen Huang, Wenjing Jie*, and Jianhua Hao*. Reversible

transition between bipolar memory switching and bidirectional threshold switching

in 2D layered K-birnessite nanosheets, ACS Applied Materials & Interfaces

https://doi.org/10.1021/acsami.0c04872 (2020).

7. Ran Ding, Chun-Ki Liu, Zehan Wu, Feng Guo, Sin-Yi Pang, Lok Wing Wong,

Weng Fu Io, Shuoguo Yuan, Man-Chung Wong, Michal Bartlomiej Jedrzejczyk,

Jiong Zhao, Feng Yan, and Jianhua Hao. A general wet transferring approach for

diffusion-facilitated space-confined grown perovskite single-crystalline

optoelectronic thin films, Nano Letters 20, 2747 (2020).

8. Sin-Yi Pang, Yuen-Ting Wong, Shuoguo Yuan, Yan Liu, Ming-Kiu Tsang, Zhibin

Yang, Haitao Huang, Wing-Tak Wong, and Jianhua Hao*. Universal strategy for

HF-free facile and rapid synthesis of two-dimensional MXenes as multifunctional

energy materials, Journal of the American Chemical Society 141, 9610 (2019).

9. Gongxun Bai, Shuoguo Yuan, Zhibin Yang, Yuda Zhao, Sin Yuk Choi, Yang Chai,

Siu Fung Yu, Shu Ping Lau, and Jianhua Hao*. 2D layered laterials of rare-earth

fffV YUAN Shuoguo



Er-doped MoS2 with NIR-to-NIR down- and up-conversion photoluminescence,

Advanced Materials 28, 7472 (2016).

10. Hung-Lit Chan, Shuoguo Yuan, and Jianhua Hao*. Vertical graphene tunneling

heterostructure with ultrathin ferroelectric BaTiO3 film as a tunnel barrier, Physica

Status Solidi RRL 12, 1800205 (2018).

11. Huiying Du, Meilin Tu, Songwen Luo, Shangdong Li, Shuoguo Yuan, Tianxun

Gong, Wen Huang, Wenjing Jie*, and Jianhua Hao*. Transition from nonvolatile

bipolar memory switching to bidirectional threshold switching in layered MoO3

nanobelt, Journal of Materials Chemistry C 7, 12160 (2019).

fffVI YUAN Shuoguo



Presentations in International Conferences

1. Shuoguo Yuan, Hung-Lit Chan, Jianhua Hao*. Piezoelectricity and

ferroelectricity in 2D layered materials for electronic devices, poster presentation,

2019 MRS Spring Meeting and Exhibit (April 22-26, 2019), Phoenix, Arizona.

(The Best Poster Award in the MRS Symposium)

2. Shuoguo Yuan, Zhibin Yang, Jianhua Hao*. Ferroelectric control of vertical

graphene field-effect transistor based on tunnelling heterostructures, poster

presentation, The 20th Physical Society of Hong Kong Conference (June 16-17,

2017), The Hong Kong University of Science and Technology, Hong Kong.

fffVII YUAN Shuoguo


Acknowledgements

Acknowledgements

First and foremost, I would like to express my highest respect and sincere

appreciation to my supervisor, Prof. Jianhua Hao for his valuable suggestion, patient

guidance and continuous encouragement throughout the whole period of my

postgraduate study. His profound scientific insight and enthusiastic attitude towards

science and research has deeply inspired me to accomplish my PhD study.

I would like to express my great gratitude to Prof. Xin Luo, Prof. Maohai Xie,

Prof. Jiyan Dai, Prof. Feng Yan, Dr. Yang Chai for their generous assistance on my

experimental research or course study. I want to thank Dr. Vincent Chan and Dr.

Hardy Lui for their help in facility utilization in our department and materials research

centre in UMF.

I gratefully also appreciate our groupmates Dr. Wenjing Jie, Dr. Zhibin Yang, Dr.

Gongxun Bai, Dr. Wei Xu, Dr. Li Chen, Dr. Ming-Kiu Tsang, Dr. Ran Ding, Dr.

Beining Zheng, Mr. Man-Chung Wong, Mr. Hung-Lit Chan, Ms. Weng Fu Io, Ms.

Yongxin Lyu, Ms. Sin-Yi Pang, Mr. Feng Guo, Mr. Menglin Song, Mr. Jianfeng Mao,

Mr. Zehan Wu, Ms. Yuen-Ting Wong, Ms. Yuqian Zhao for their kind support to my

experiments. Besides, I would like to give my thanks to my friends and classmates for

their kind help and friendship.

fffVIII YUAN Shuoguo


Acknowledgements

Lastly, I would like to thank my family for their endless love, support and

encouragement throughout my study and life.

fffIX YUAN Shuoguo


Table of Content

Table of Content

Abstract .......................................................................................................................... I

List of Publications ...................................................................................................... III

Acknowledgements .................................................................................................... VII

Table of Content ........................................................................................................... IX

List of Figures ........................................................................................................... XIII

List of Tables ........................................................................................................... XXII

List of Acronyms.................................................................................................... XXIII

Chapter 1 Introduction ................................................................................................... 1

1.1 Background of two-dimensional materials ...................................................... 1

1.1.1 Graphene ................................................................................................... 1

1.1.2 Transition metal dichalcogenides .............................................................. 4

1.1.3 Group III-VI compounds .......................................................................... 5

1.1.4 Graphene analogous .................................................................................. 7

1.2 Ferroelectric and piezoelectric properties of nanoscale films .......................... 9

1.2.1 Typical ferroelectric materials ................................................................... 9

1.2.2 Typical piezoelectric materials ................................................................ 16

1.3 Significance of research ................................................................................. 19

fffX YUAN Shuoguo


Table of Content

1.4 Structure of thesis .......................................................................................... 22

Chapter 2 Experimental Methods ................................................................................ 24

2.1 Fabrication methods of 2D materials ............................................................. 25

2.1.1 Exfoliation ............................................................................................... 25

2.1.2 Physical vapor deposition ....................................................................... 28

2.1.3 Chemical vapor deposition ...................................................................... 30

2.2 Measurement methods ................................................................................. 32

2.2.1 Optical characterization .......................................................................... 32

2.2.2 Structural characterization ...................................................................... 33

2.2.3 Piezoresponse force microscopy characterization .................................. 38

2.2.4 Transport characteristic measurement ..................................................... 40

Chapter 3 Ferroelectricity in distorted 1T phase MoTe2 .............................................. 42

3.1 Introduction .................................................................................................... 42

3.2 Experiment ..................................................................................................... 51

3.2.1 Fabrication of d1T-MoTe2 samples ......................................................... 51

3.2.2 Thickness characterization of d1T-MoTe2 .............................................. 51

3.2.3 Raman characterization of d1T-MoTe2 ................................................... 53

3.2.4 Structural characterization of d1T-MoTe2 ............................................... 56

3.3 Ferroelectric characterization of d1T-MoTe2 ................................................. 59

fffXI YUAN Shuoguo


Table of Content

3.4 Ferroelectric mechanism of d1T-MoTe2 ........................................................ 65

3.5 Ferroelectric tunnel junction devices ............................................................. 69

3.6 Summary ........................................................................................................ 75

Chapter 4 Out-of-plane Piezoelectricity in the Exfoliated In2Se3 ................................ 77

4.1 Introduction .................................................................................................... 77

4.2 Experiment ..................................................................................................... 83

4.2.1 Preparation and measurement of In2Se3 samples .................................... 83

4.2.2 Phase characterization of In2Se3 samples ................................................ 84

4.3 Out-of-plane piezoelectric properties of In2Se3 ............................................. 87

4.4 Out-of-plane ferroelectric properties of In2Se3 .............................................. 93

4.5 Simulated piezoelectric actuator devices ....................................................... 96

4.6 Summary ........................................................................................................ 98

Chapter 5 Out-of-plane Piezoelectricity in Van der Waals Heterostructure .............. 100

5.1 Introduction .................................................................................................. 100

5.2 Experiment ................................................................................................... 101

5.2.1 Fabrication of In2Se3, MoS2 and WS2 samples ..................................... 101

5.2.2 Raman and PL characterizations of In2Se3/MoS2 heterostructure ......... 103

5.2.3 Raman and PL characterizations of In2Se3/WS2 heterostructure........... 104

5.2.4 Strcutural characterization of In2Se3 ..................................................... 105

fffXII YUAN Shuoguo


Table of Content

5.3 Vertical piezoelectric properties of vdW heterostructure ............................. 107

5.3.1 Piezoelectricity in In2Se3 ....................................................................... 107

5.3.2 Piezoelectricity in monolayer MoS2 ..................................................... 109

5.3.3 Piezoelectricity in In2Se3/MoS2 vdW heterostructure ........................... 111

5.3.4 Piezoelectricity in In2Se3/WS2 vdW heterostructure ............................. 113

5.4 Physical mechanism of piezoelectric properties of vdW heterostructure .... 115

5.5 Summary ...................................................................................................... 121

Chapter 6 Conclusion and Future Prospect ................................................................ 122

6.1 Conclusion ................................................................................................... 122

6.2 Future prospect............................................................................................. 124

References .................................................................................................................. 127

fffXIII YUAN Shuoguo


List of Figures

List of Figures

Figure 1.1 a) Schematic of monolayer graphene lattice.7 b) The electronic band

structure of graphene8. ................................................................................................... 4

Figure 1.2 Crystal structures of MoS2 with different polymorphisms9.......................... 5

Figure 1.3 Crystal structure of monolayer a) and bilayer b) α-In2Se3. .......................... 7

Figure 1.4 The hysteresis loop of typical ferroelectric materials. ................................ 10

Figure 1.5 The relationship between dielectrics, piezoelectrics, pyroelectrics and

ferroelectrics. ............................................................................................................... 11

Figure 1.6 The energy barrier diagram of the FTJ for two different polarization

directions, M1 stands for the metal 1, FE stands for ferroelectric barrier and M2

respresents the metal 249

. ............................................................................................. 14

Figure 1.7 Schematic of typical piezoelectric effect. ................................................... 17

Figure 2.1 A typical method to fabricate the 2D flakes by mechanical exfoliation75

. . 26

Figure 2.2 Liquid exfoliation based on sonication. The bulk crystal is sonicated in a

solvent system, which leads to exfoliate into the 2D nanosheets76

.............................. 28

Figure 2.3 Schematic of the PLD equipment78

............................................................ 29

Figure 2.4 Schematic of CVD growth ......................................................................... 32

Figure 2.5 Basic principle of Raman scattering ........................................................... 34

fffXIV YUAN Shuoguo


List of Figures

Figure 2.6 Schematic of the second harmonic generation in non-centrosymmetric

materials, in which the excitation frequency is half of the generated light86

............... 35

Figure 2.7 Schematic of working principle of PFM .................................................... 40

Figure 2.8 Schematic of the experimental setup of CAFM95

....................................... 41

Figure 3.1 a) OM image of d1T-MoTe2, scale-bar, 3 μm. b) AFM image and

corresponding height profile of monolayer d1T-MoTe2, scale-bar, 0.5 μm. c) AFM

image of d1T-MoTe2 nanosheets onto Pt substrate, scale bar, 2 μm. d) Cross-sectional

TEM image of d1T-MoTe2, scale bar, 6 nm ................................................................. 52

Figure 3.2 Experimental and calculated Raman spectra of 2H and d1T-MoTe2 .......... 54

Figure 3.3 a) XPS spectra of d1T-MoTe2. b,c) HRTEM images of 2H and d1T-MoTe2,

arrows denote Te defects, scale-bar, 2 nm. d,e) EDS curves of 2H-MoTe2 and

d1T-MoTe2. .................................................................................................................. 57

Figure 3.4 a) OM image of MoTe2 nanosheets onto monolayer graphene. b) AFM

image and height curve of 5 layers d1T-MoTe2 nanosheets onto monolayer graphene.

c) Cross-sectional TEM image of 5 layers d1T-MoTe2 on monolayer graphene. d)

Raman spectrum of monolayer graphene layer used. e,f) Raman spectra of 5 layers

2H-MoTe2 and d1T-MoTe2 nanosheets onto monolayer graphene. ............................. 58

Figure 3.5 a) PFM phase hysteretic and butterfly loops of monolayer d1T-MoTe2. b,c)

PFM phase hysteretic and butterfly loops of d1T-MoTe2 before and after 30 days. d)

fffXV YUAN Shuoguo


List of Figures

PFM phase image of monolayer d1T-MoTe2, where the electrical poling was applied

by writing two square patterns with ± 8 V, scale bar, 1 μm ......................................... 61

Figure 3.6 a) Phase signals of d1T-MoTe2 nanosheets with different layer number onto

Pt substrate (layer number n=2, 3, 4, 9). b) The PFM phase hysteretic loop of 5 layers

d1T-MoTe2 nanosheets onto monolayer graphene. c,d) PFM phase hysteretic loops of

2H and d1T-MoTe2 taken at random locations, the inset shows the randomly selected

locations. e,f) In-plane PFM amplitude (top) and phase images (bottom) of

d1T-MoTe2 with 5 layers. ............................................................................................. 62

Figure 3.7 a) Raman spectra of d1T-MoTe2 film by MBE. b) Surface topography of

MoTe2 film by MBE. c) STM image of monolayer MoTe2 films by MBE. d,e) PFM

phase loops of monolayer MoTe2 films deposited onto HOPG and SiC substrate. f,g)

PFM amplitude loops of monolayer MoTe2 films deposited onto HOPG and SiC

substrate. h) Second harmonic generation spectrum from monolayer d1T-MoTe2 ..... 64

Figure 3.8 a) Top-view HRTEM image and intensity profile, the atomic structure of

d1T-MoTe2 is placed on top, scale-bar, 0.5 nm. b) Atomic structure image of

monolayer d1T-MoTe2 and the inset shows atomic structure model (cyan and orange

colors represent Mo and Te atoms, respectively), scale-bar, 0.2 nm. .......................... 66

Figure 3.9 a) Phonon dispersion of d1T-MoTe2. b) Phonon dispersion of the 1T-MoTe2,

with the structure of 1T shown in the middle. The unstable modes in M and K points

fffXVI YUAN Shuoguo


List of Figures

lead to the 1T′ and d1T-MoTe2, respectively. c) Energy barrier between d1T and

1T-MoTe2 (lable the 1T below the image at the lower middle, and the d1T at the upper

image). d) The top view of the trimerized/d1T-MoTe2, the structures show no

difference in trimerized and d1T-MoTe2 in the top view. The side view shows the

difference between the e) trimerized and f) d1T structure. g) The change of the dipole

moment in the d1T phase and the trimerized structure. h,i) Top-view and side-view of

charge density difference between ferroelectric d1T and paraelectric 1T phases (green,

purple, cyan, orange and pink colors denote negative charge, positive charge, Mo

atom, Te atom and polarization, respectively) ............................................................. 68

Figure 3.10 Layer-dependent Tc of d1T-MoTe2. .......................................................... 71

Figure 3.11 a) Temperature dependence of PFM phase loops of 5 layers d1T-MoTe2. b)

PFM phase loops of 16 layers d1T-MoTe2 at room temperature. c,d) AFM image and

height profile of MoTe2 with 16 layers ........................................................................ 71

Figure 3.12 Energy difference of the d1T phase and 1T phase with increasing layer

thicknesses. .................................................................................................................. 72

Figure 3.13 I-V characteristic of monolayer d1T-MoTe2 on Pt and inset shows the

energy diagram of FTJ devices .................................................................................... 73

Figure 3.14 I-V characteristic of few-layer d1T-MoTe2 on graphene and inset shows

the schematic of device structure ................................................................................. 74

fffXVII YUAN Shuoguo


List of Figures

Figure 3.15 a) Structure illustration of the d1T-MoTe2 and graphene heterostructure,

as well as the bandstructure of the heterostructures. b) The characteristic band of the

graphene is highlighted with the red ball, with the size proportional to the weights of

the electronic states on to the C atom .......................................................................... 75

Figure 4.1 Raman characterizations of α-In2Se3 nanosheets. a) OM image of In2Se3

nanosheets, scale bar, 4 μm. b) AFM images in the red square of OM image and

corrsponding height profile of 3 nm In2Se3 nanosheets, scale bar, 0.3 μm. c) Thickness

dependent Raman spectra of In2Se3 nanosheets. d) Raman spectrum of In2Se3

nanosheets .................................................................................................................... 85

Figure 4.2 a,b) XPS spectra of In2Se3 nanosheets. c,d) Elemental mapping of In and

Se. e,f) high-resolution TEM image of In2Se3 (scale bar, 3 nm) and the inset shows

SAED pattern of In2Se3 nanosheets ............................................................................. 86

Figure 4.3 EDS of In2Se3 nanosheets. .......................................................................... 87

Figure 4.4 Piezoelectrcity in α-In2Se3 nanosheets onto Pt substrate. Amplitude images

of 20 layers In2Se3 with different drive voltages (a: 1 V; b: 2 V; c: 3 V; d: 4 V; scale

bar, 0.3 μm) ................................................................................................................ 88

Figure 4.5 SHG spectrum of In2Se3. SH spectral intensity at 399 nm generated in

reflection of In2Se3 ....................................................................................................... 89

Figure 4.6 Amplitude images of In2Se3 onto SiO2/Si substrate with different drive

fffXVII

I YUAN Shuoguo


List of Figures

voltages (a: 1 V; b: 2 V; c: 3 V; d: 4 V; scale bar, 2.5 μm) ........................................... 90

Figure 4.7 Surface topography a) and PFM amplitude image b) of 20 nm In2Se3 onto

SiO2/Si substrate, scale bar, 2 μm). Surface topography c) and PFM amplitude image

d) of 20 nm In2Se3 onto SiO2/Si substrate, scale bar, 15 μm. ...................................... 91

Figure 4.8 Piezoelectric properties d33 of In2Se3 onto SiO2/Si substrate with number of

layers. ........................................................................................................................... 91

Figure 4.9 Fifty phase hysteresis a) and amplitude loops b) of In2Se3 nanosheets ...... 95

Figure 4.10 Surface topography a) and PFM phase image b) of 20 layers In2Se3

nanosheet...................................................................................................................... 95

Figure 4.11 Surface topography a) and PFM phase image b) of 6 layers In2Se3

nanosheet...................................................................................................................... 96

Figure 4.12 Simulation of the atomical deformation actuator based on out-of-plane

piezoelectricity. a) The thickness dependent amplitude loops of In2Se3. b) The

thickness dependence of out-of-plane piezoelectric response of In2Se3 nanosheets. c)

Deformation of 11 nm In2Se3 with an applied voltage of −8 V. d) Deformation of

upper surface of In2Se3 at different applied voltage. e) Surface deformation of In2Se3

at different voltages. f) The thickness dependence of surface deformation for the

In2Se3............................................................................................................................ 98

Figure 5.1 a) Raman spectrum of MoS2. b) Raman spectrum of In2Se3. c) PL spectrum

fffXIX YUAN Shuoguo


List of Figures

of In2Se3 ..................................................................................................................... 103

Figure 5.2 a) Raman spectrum of In2Se3. b) Raman spectrum of WS2. c) PL spectrum

of WS2 ........................................................................................................................ 105

Figure 5.3 Structural characterizations of α-In2Se3. a) Atomic structure of In2Se3. b)

Thickness dependent Raman spectra of In2Se3. c) XPS spectra of In2Se3. d,e)

Elemental mapping of In and Se. f) high-resolution TEM image of In2Se3 (scale bar, 3

nm) and the inset shows SAED pattern of In2Se3 ...................................................... 107

Figure 5.4 Amplitude images of In2Se3 onto SiO2/Si substrate with different drive

voltages. a) The surface topography and height profile of In2Se3 exfoliated onto

SiO2/Si substrate, scale bar, 2.5 μm. b) 1 V; c) 2 V; d) 3 V; e) 4 V; f) Amplitude versus

drive voltage loop. The linearly fitted result exhibits that the out-of-plane

piezoelectric coefficient d33 is ~7.6 pm/V ................................................................. 109

Figure 5.5 Piezoelectricity in monolayer MoS2. a) Optical microscope image and b)

PL mapping of monolayer CVD grown MoS2 onto SiO2/Si substrate (scale bar, 10

μm). c) Raman spectrum of monolayer CVD grown MoS2 onto SiO2/Si substrate. d)

AFM image and e) PFM amplitude image of monolayer CVD grown MoS2. f) Optical

microscope image and g) Raman spectrum of monolayer exfoliated MoS2 onto

SiO2/Si substrate (scale bar, 2.5 μm). h) Surface topography and i) PFM amplitude

images of monolayer exfoliated MoS2 onto SiO2/Si substrate .................................. 110

fffXX YUAN Shuoguo


List of Figures

Figure 5.6 Piezoelectricity in In2Se3/MoS2 heterostructure. a) Surface topography and

the inset shows the height profile of 40 nm In2Se3 fabricated onto monolayer MoS2

(scale bar, 1.5 μm). The enlarged PFM amplitude images of In2Se3/MoS2, b) 1 V; c) 2

V; d) 3 V. The larger scale PFM amplitude images of In2Se3/MoS2, e) 1 V; f) 2V; g) 3V,

exhibiting remarkable piezoelectricity. h) Amplitude vs drive voltage loop of

In2Se3/MoS2 heterostructure ...................................................................................... 112

Figure 5.7 a-c) The quantitative piezoresponse profiles of the extracted out-of-plane

PFM amplitude of In2Se3/MoS2 with different drive voltages. d) Thickness

dependence of the piezoelectric coefficient d33 of In2Se3/MoS2 ................................ 113

Figure 5.8 a) OM image of monolayer CVD grown WS2 onto SiO2/Si substrate, scale

bar, 5 μm. b) PL mapping of monolayer WS2 onto SiO2/Si substrate. c) AFM image

and d) PFM amplitude image of monolayer WS2 ...................................................... 114

Figure 5.9 The PFM amplitude images of In2Se3/WS2 vdW heterostructure (scale bar,

4 μm). a) 1 V; b) 2 V; c) 3 V. d-f) The quantitative piezoresponse profiles of the

extracted PFM amplitude of In2Se3/WS2 with different drive voltages ..................... 115

Figure 5.10 a) Atomic structure of In2Se3/WS2. b) Schematic band structure of

In2Se3/MoS2. c,d) UPS spectra of In2Se3. e,f) UPS spectra of MoS2 ......................... 117

Figure 5.11 a) Atomic structure of In2Se3/WS2. b) Schematic band structure of

In2Se3/WS2. c,d) UPS spectra of WS2 ........................................................................ 118

fffXXI YUAN Shuoguo


List of Figures

Figure 5.12 a) Atomic structure and band structure of the In2Se3/MoS2 heterostructure

when the polarization of In2Se3 is pointing downward. b) The atomic structure and

band structure of the In2Se3/MoS2 heterostrucutre with polarization pointing upward.

The charge density difference between the heterostructure and its components is

plotted overlap above the atomic structure, Δn=nIn2Se3/MoS2-nIn2Se3-nMoS2, the red and

the blue colours denote the charge accumulation and deplection, respetviely. c) and d)

is the plane-averaged differential charge density of a) and b), respetively ................ 120

fffXXII YUAN Shuoguo


List of Tables

List of Tables

Table 3.1 LDA calculated Gamma point Phonon frequency (cm−1

) and the relative

irreducible representation for the d1T, 2H and 1T phase MoTe2, the Raman (R) and

Infrared (I) activity are also indicated. Due to the absentence of the inversion center,

all the modes in d1T-MoTe2 are both Raman and infrared active. ………………… 55

Table 3.2 Comparison of the critical or thinnest thickness and Curie temperature (Tc)

with the previously reported common ferroelectric materials possessing out-of-plane

ferroelectricity by experiments (RT represents room temperature)... .......................... 63

Table 4.1 The calculated elastic stiffness coefficients (Cml, unit: GPa) and the

piezoelectric coefficients of bulk In2Se3 under different compressive strains (eij, unit:

10-10

C/m; dij, unit: pm/V).. .......................................................................................... 93

fffXXII

I YUAN Shuoguo


List of Acronyms

List of Acronyms

Acronyms Description

2D Two-dimensional

vdW Van der Waals

FTJ Ferroelectric tunnel junction

FeRAM Ferroelectric random access memory

TER Tunnel electroresistance

CVD Chemical vapor deposition

PLD Pulsed laser deposition

MBE Molecular beam epitaxy

PFM Piezoresponse force microscopy

AFM Atomic force microscopy

CAFM Conductive atomic force microscopy

STM Scanning tunnelling microscopy

XPS X-ray photoemission spectroscopy

HRTEM High-resolution transmission electron microscopy

STEM Scanning transmission electron microscopy

fffXXI

V YUAN Shuoguo


List of Acronyms

EDS Energy dispersive spectroscopy

SHG Second-harmonic generation

OM Optical microscopy

UPS Ultraviolet photoelectron spectroscopy

VBM Valence band maximum

CBM Conduction band minimum

PL Photoluminescence

d1T Distorted 1T

DFT Density functional theory

TMDs Transition metal dichalcogenides

h-BN Hexagonal boron nitride

BP Black phosphorus

MXenes Transition metal carbides and/or nitride

fff1 YUAN Shuoguo




Chapter 1

Chapter 1 Introduction

1.1 Background of two-dimensional materials

1.1.1 Graphene

The demand for more compact and powerful devices in electronic applications

has been growing since silicon has reached its limit. The most promising application

of graphene is to become a substitute for silicon, making ultrathin transistors for the

production of future supercomputers. Traditional semiconductors and conductors,

such as silicon and copper, have released some energy in the form of heat from

traditional semiconductors and conductors due to collision of electrons and atoms, and

the computer chips waste much electrical energy in this way. In the past decade, 2D

layered materials have drawn much more attention in fundamental research and

various applications.1-5

When the materials size approached atomic scale, 2D

materials exhibit attractive characteristics different from their bulk materials. In 2004,

graphene is the first 2D layered material discovered using mechanical exfoliation

method.6 Each carbon atom of graphene has an unbonded p-electron perpendicular to

the σ orbital of the plane of the carbon atom, forming a highly parabolic large π bond

on both sides of the lattice plane, which can be free and efficient in the crystal. The

speed of movement is up to 1/300 of the speed of light and the electron energy is not

fff2 YUAN Shuoguo




Chapter 1

lost, giving graphene good conductivity. One of the most striking properties of

graphene is its extraordinary electronic properties, which offers great potential for its

essential properties and electronic devices applications. The large π bond electrons

that are highly crucible on both sides of the lattice plane have the light-absorbing and

nonlinear optical properties, Dirac carrier characteristics, and quantum Hall effect at

room temperature. Figure 1.1a shows the schematic strcuture of monolayer graphene.

There are two interpenetrating carbon atoms per unit cell in the lattice.7 The bond

length between two nearest neighboring carbon atoms is around 1.42 Å and the lattice

parameter is around 2.46 Å. Due to its unique two-dimensional structure and

electronic band structure, electrons in graphene can be compared to relativistic

particles, which means that these electrons can be regarded as massless Dirac-Ferimi,8

as shown in Figure 1.1b. This unique electronic property provides an ideal platform

for studying the relativistic effects. Another unique property is that when graphene is

thicker than three layers, electrons are allowed to pass a few microns without

scattering. This property ensures that graphene has an extremely high charge carrier

mobility and conductivity at room temperature. The graphene electron mobility at

room temperature exceeds 15000 cm2 /V·s, which is higher than that of carbon

nanotubes or silicon crystals and the resistivity is only about 10-6

Ω·cm, which is

lower than that of copper or silver. By replacing silicon with graphene, computer

fff3 YUAN Shuoguo




Chapter 1

processors can run hundreds of times faster. These excellent properties make graphene

possess promising applications in high frequency devices, field effect transistors, solar

cells, touch screens, spin devices, field emission materials, sensitive sensors, high

performance batteries, supercapacitors, micro-nano electromechanical devices, and

composite materials. Graphene is also an important type of transparent conductive

materials. In the visible region, the transmittance of four layers graphene is

comparable to that of the conventional ITO film. On the other band, the transmittance

of monolayer graphene is higher than that of the ITO film. Graphene is almost

completely transparent and has a light transmittance of up to 97.7%. The thermal

conductivity of graphene is as high as 5300 W/m·K, which is higher than that of

carbon nanotubes and diamond. These features make it ideal for use as a raw material

for transparent electronic products such as transparent touch displays, illuminators

and solar panels. Based on the excellent physical properties of graphene, graphene has

extraordinary potential in a wide variety of applications.1-7

fff4 YUAN Shuoguo




Chapter 1

Figure 1.1 a) Schematic of monolayer graphene lattice.7 b) The electronic band

structure of graphene.8

1.1.2 Transition metal dichalcogenides

The transition metal dichalcogenides (TMDs) are a chemically diverse class of

layered compounds with a general formula of MX2, where M represents a transition

metal element (e.g. Mo, Te etc) and X is a chalcogen (e.g. S, Se or Te). The

monolayer TMDs consist of a transition metal layer sandwiched by two chalcogen

layers and they are specially featured with the formation of different crystal

polytypes,9 as displayed in Figure 1.2. TMDs exist in different phases, including the

fff5 YUAN Shuoguo




Chapter 1

semiconducting 2H phase, the metallic octahedral (1T) and distorted octahedral (1T′)

phases. The common semiconducting 2H phase MX2 is the most stable phase. The

2H-type MoS2 is dominant because it is thermodynamically stable in nature. The 2D

TMDs materials have wide applications in the fields ranging from electronic, optical,

optoelectronic and quantum devices to energy storage and catalysis.10-14

Figure 1.2 Crystal structures of MoS2 with different polymorphisms.9

1.1.3 Group III-VI compounds

Group III-VI compounds semiconductors M2X3 (M = Ga, In; X = S, Se, Te) are

an vital type of 2D layered materials because of the excellent electronic and

optoelectronic performance.15-18

Similar to 2D TMDs, M2X3 have the significant

advantages of atomic layer thickness, ultrahigh specific surface area and high

compatibility. Compared to TMDs, the M2X3 demonstrate other excellent

fff6 YUAN Shuoguo




Chapter 1

characteristics, for example high carrier mobility, high charge density, etc.18

The most

common study with M2X3 stoichiometry is In2Se3. Typically, it has five known crystal

strcutures (α, β, γ, δ, and κ) at various temperatures. Among these phases, γ-In2Se3 has

a distorted wurtzite-like structure and δ-In2Se3 has a monoclinic structure. The

α-phase, β-phase and κ-phase In2Se3 are layered structures. It is generally believed

that the α-phase is thermodynamically stable. In each single-layer In2Se3, every

Se-In-Se-In-Se atomic layers form a stable 2D unit by covalent bonds, and the

different units are combined by a weak vdW interaction. The monolayer In2Se3

thickness is around 1 nm. The crystal structure of monolayer and bilayer α-In2Se3 is

displayed in Figure 1.3. The difference between the two common α-phase and β-phase

is the atomic arrangement and the atomic bond length, but their phases can be

transformed by means of temperature treatment.18

fff7 YUAN Shuoguo




Chapter 1

Figure 1.3 Crystal structure of monolayer a) and bilayer b) α-In2Se3.

1.1.4 Graphene analogous

Besides the above 2D layered materials, there are some typical layered graphene

analogous materials, such as hexagonal boron nitride (h-BN), black phosphorus (BP),

Group III-VI layered semiconductors (MX), transition metal carbides and/or nitride

(MXenes), etc.

2D h-BN has honeycomb structure formed by hybridization of B and N atoms by

sp2 bonding, which is also called white graphene.

19,20 It consists of atomic-level planar

layers of alternating hexagonal B atoms and N atoms, which are joined together by

interlayer vdW interactions. The h-BN has excellent physical and chemical properties

such as wide band gap, high thermal conductivity, stability and oxidation resistance.

fff8 YUAN Shuoguo




Chapter 1

Unlike the excellent conductivity of graphene, h-BN has extremely excellent

insulating properties, which makes it play an important role in various basic sciences

and technologies, such as charge fluctuation, contact resistance, gate dielectric,

passivation layer and atomic tunneling layer.

The bulk black phosphorus has a layered orthorhombic crystal structure. The

spacing between adjacent layers is 5.4 Å, and the layers are also connected by vdW

force. A single layer black phosphorus consists of a pleated honeycomb structure in

which the phosphorus atoms are connected to the other three atoms. Among four

phosphorus atoms, three of the atoms are in the same plane and the fourth atom is in

the adjacent parallel layer.21-23

Group III-VI layered semiconductors are a class of layered metal chalcogenides

having the general formula MX (M = Ga, In; X =S, Se, Te).24,25

GaSe is an example

of 2D layered materials from the Group III-VI family, which is connected by a

vertical stack of Se-Ga-Ga-Se layers by vdW forces. Each layer has a D3h

symmetrical hexagonal structure. The adjacent layer spacing is around 0.84 nm and

the lattice constant along the axis is around 0.40 nm.

MXenes, an emerging class of 2D layered transition metal carbides and/or nitride

materials, were discovered by Gogotsi group in 2011.26

MXenes were produced by

selective etching of the raw MAX phase materials, presented by a chemical formula

fff9 YUAN Shuoguo




Chapter 1

of Mn+1AXn (n = 1, 2, or 3), where M2AX, M3AX2, or M4AX3, where M represents

the transition metal (Ti, V, Nb, etc), A stands for an element from group IIIA or IVA,

and X is carbon and/or nitrogen. Since M-A bonding (metallic bond) is weaker than

M-X (mixed metallic-ionic covalent bond), the A layer can be selectively etched from

the MAX phase by HF to become a Mn+1Xn (n = 1, 2 or 3) structure. The MXenes

have excellent physical and chemical properties, which have wide application

prospects in energy storage and catalysts.27-31

1.2 Ferroelectric and piezoelectric properties of nanoscale

films

1.2.1 Typical ferroelectric materials

Ferroelectrics are a class of dielectric crystals with spontaneous polarization

characteristics in which the spontaneous polarization orientation changes with the

applied electric field.32-35

When the external electric field is removed, its polarization

state can still be maintained. The most basic characteristic of ferroelectrics is

spontaneous polarization, which is mainly caused by the fact that the positive and

negative charges do not coincide in the ferroelectric unit cell to produce an electric

dipole moment. The property of ferroelectrics is called ferroelectricity and can be

macroscopically characterized by the hysteresis loop characteristics, as shown in

fff10 YUAN Shuoguo




Chapter 1

Figure 1.4.

Figure 1.4 The hysteresis loop of typical ferroelectric materials.

With the rapid development of microelectronics technology, the miniaturization

and integration of electronic devices have become the current research trend.36

Integrated ferroelectrics thin films and semiconductor CMOS processes have broad

application prospects in many fields such as microelectronics, integrated optics,

optoelectronics, and microelectromechanics. In addition to excellent ferroelectric

properties, ferroelectric thin films also have piezoelectric properties, dielectric

properties, pyroelectric properties and other important features, as shown in Figure

1.5. All ferroelectric materials have piezoelectric properties. Piezoelectric materials

can provide polarized charges via applying external stress. Since the discovery in

1880, many materials have been reported to have the piezoelectric response, which

are widely used in actuators, transducers, sensors and energy applications. Moreover,

the inverse piezoelectric effect is also widely used in the generation of sound waves

fff11 YUAN Shuoguo




Chapter 1

and scanning probe microscopy systems.37,38

Figure 1.5 The relationship between dielectrics, piezoelectrics, pyroelectrics and

ferroelectrics.

Typical perovskite structure ABO3 ferroelectrics are the most widely studied and

used ferroelectric materials. In the ABO3 structure, the A and B sites are occupied by

atoms with different valence states and radii. The B site is in the oxygen octahedron,

and different properties can be obtained by doping with different atoms.

BaTiO3-based materials are the first type of the discovered perovskite structure

ferroelectric materials. This type of material has the advantages of large residual

polarization, large dielectric constant and good fatigue performance, but the leakage

current density is high under high electric field condition, limiting the

functionalization of such materials for devices applications.39

The Pb(Zr,Ti)O3-based

ferroelectric material is another type of perovskite structure after the discovery of

fff12 YUAN Shuoguo




Chapter 1

BaTiO3 series.40-42

The most common ferroelectric material is Pb(Zr,Ti)O3 thin film

which has been widely used due to its large residual polarization and low deposition

temperature. However, due to the presence of toxic lead, the environmental pollution

caused by the volatilization of lead has resulted in serious pollution to nature and

human living environment. BiFeO3 has a very large polarization, and its polarization

is as high as 100 μC/cm2. It has been widely studied as a typical single-phase

multiferroic material.43,44

Although the polarization of such material is very large, a

large leakage current is caused owing to the volatilization of Bi element and the

change of Fe element, which makes a gap with the practical application requirements

of the ferroelectric memory.

Since Bell Labs first proposed the concept of ferroelectric memory, it has

attracted great interest.45

Due to its advantages of non-volatility, fast read/write speed,

high storage capacity, low power consumption, fatigue and excellent radiation

performance, ferroelectric random access memory (FeRAM) is recognized as the

most promising next-generation memory, as it has a very large market application

prospect in the field of electronic information products in the future.46

Typical

FeRAM device is based on the capacitor structure, and its readout is destructive.

However, the ferroelectric tunnel junction (FTJ) is able to read using a change in the

quantum tunneling current through the ferroelectric ultrathin film, and thus its readout

fff13 YUAN Shuoguo




Chapter 1

is non-destructive.47-49

A typical FTJ consists of two metal electrodes and an ultrathin

ferroelectric film, as shown in Figure 1.6. According to the basic principles of

quantum mechanics, if the barrier is thin enough, the charge carriers can pass through

the barrier, which means that the carrier has a probability to appear on another side of

the barrier. The charge carrier can tunnel through the ultrathin ferroelectric film,

depending on the barrier height and width. When the ultrathin ferroelectric film acts

as a barrier layer, the polarization of the ferroelectric film will cause a corresponding

polarization charge at the interface and the screening charge will be generated at the

interface. When the lengths of the Thomas-Fermi screening charge of the left and

right electrodes are different, the barrier height of the ferroelectric film will be

different when the polarization direction of the ferroelectric film is left or right, and

therefore the tunneling electro-resistance of the tunnel junction will also be different.

For FTJ, different ferroelectric polarization values are obtained for different

ferroelectric tunneling directions, and thus non-destructive reading of the memory can

be achieved. The tunnel electroresistance (TER) is an important indicator to evaluate

the performance of the FTJ. The tunnel transmission is exponentially related to the

square root of the average height of tunnel barrier. By applying a voltage to the tunnel

barrier, two spontaneous polarization directions result in two distinct resistance states.

The TER ratio is used to estimate the efficiency of ON state and OFF state resistance,

fff14 YUAN Shuoguo




Chapter 1

and the equation of TER is expressed as follows, TER = (ROFF – RON)/RON, where RON

and ROFF stand for the resistances of ON state and OFF state, respectively.

Figure 1.6 The energy barrier diagram of the FTJ for two different polarization

directions, M1 stands for the metal 1, FE stands for ferroelectric barrier and M2

respresents the metal 2.49

Since the tunneling effect is not exhibited when the barrier layer is thick in the

FTJ, the thickness of the tunnel barrier layer is relatively thin. For ferroelectric

materials, the phenomenon of ferroelectric polarization is a collective effect, which is

a phenomenon exhibited by many dipoles, rather than a phenomenon exhibited by a

single dipole. Therefore, whether the ferroelectric material has a critical thickness,

that is, the ferroelectricity disappears when the thickness of the ferroelectric thin film

is less than a certain critical value.50-52

This problem is very important for the study of

fff15 YUAN Shuoguo




Chapter 1

nanoscale ferroelectrics, which is also important for FTJ. Theoretical calculations and

experimental results have demonstrated that when the thickness of the ferroelectric

thin film is reduced to several unit cells, the ferroelectricity of the thin film can still

maintain.53-55

For a long time, it is generally believed that the critical thickness of

ferroelectricity is hundreds of nanometers, which hinders the application of

ferroelectric materials as a barrier layer in FTJ. Until 1999, Tybell et al. reported that

the epitaxially grown 4 nm Pb(Zr0.2Ti0.8)O3 film had a stable ferroelectric phase.56

In

2003, Ghosez et al. first used density functional theory (DFT) to verify that

ferroelectricity was still present in 2.4 nm BaTiO3 films.57

In the same year, Contreras

et al. reported the heterostructure of Pt/Pb(Zr0.52Ti0.48)O3/SrRuO3. The I-V curve was

studied and the change of electrical resistance was observed.58

In 2004, Fong et al.

found that the ferroelectricity of the 1.2 nm PbTiO3 film epitaxially grown on the

SrTiO3 substrate remained stable by experimental study.59

Until 2005, Tsymbal and

Kohlstedt et al. proposed the concept of FTJ, that is, the upper and lower electrodes

plus the ferroelectric barrier layer.60

In 2006, Tenne et al. confirmed that the 0.4 nm

BaTO3 film can still maintain its ferroelectricity.61

In 2009, Garcia et al. deposited

PbTiO3/La0.67Sr0.33MnO3 on a (001) NdGaO3 substrate with pulsed laser deposition,

and a 30 nm thick La0.67Sr0.33MnO3 film was used as a buffer layer and a bottom

fff16 YUAN Shuoguo




Chapter 1

electrode. Under the condition of 1~3 nm thick BaTO3, when the direction of

ferroelectric polarization is reversed, a large change of electroresistance is observed.

By detecting the tunnel current passing through the thin layer, the polarization state of

the material can be read. In addition, these devices can achieve high storage densities

(about 16 Gb per square inch) for application in high-density storage.62

1.2.2 Typical piezoelectric materials

Unlike piezoelectric materials, ferroelectric materials have an electric

polarization under external electric filed and have residual polarization even after

removing the external electric field. Piezoelectric materials are a class of functional

materials that have a piezoelectric effect, as shown in Figure 1.7. The piezoelectric

effect refers to the effect of the material generating an electrical signal under external

force or the deformation of the material under applied electric field. Specifically,

when the material is deformed by external force, the relative displacement of positive

and negative ions in the unit cell makes the positive and negative charge centers no

longer coincide, resulting in macroscopic polarization of the material. The surface

charge density of the material is equal to the projection of the polarization in the

surface direction, a different electric charge will appear on both sides of the material

when the material is deformed by the external force. In contrast, when a piezoelectric

fff17 YUAN Shuoguo




Chapter 1

material is polarized by applying an electric field, the material is deformed owing to

the displacement of the charge center. The piezoelectricity of a material is mainly

determined by the piezoelectric constant. Typically, all ferroelectric materials have

piezoelectric and inverse piezoelectric characteristics, making them widely used in

transducers, actuators, sensing, and energy applications.63,64

Since the piezoelectric

effect was discovered in 1880, the traditional piezoelectric materials mainly include

inorganic and organic piezoelectric materials.65-67

Figure 1.7 Schematic of typical piezoelectric effect.

The first type is inorganic piezoelectric material, which is divided into

piezoelectric crystal and piezoelectric ceramic.68

Piezoelectric crystal typically

corresponds to piezoelectric single crystals and piezoelectric ceramic generally refers

to piezoelectric polycrystals. Piezoelectric ceramic is typically synthesized by using

solid phase reaction. The piezoelectric ceramic has ferroelectric domains among the

fff18 YUAN Shuoguo




Chapter 1

crystal grains of the ceramic, and the ferroelectric domains are composed of 180°

domains which are antiparallel in the spontaneous polarization direction and 90°

domains which are perpendicular to each other in the spontaneous polarization

direction. These domains are artificially polarized under the applied electric field, the

spontaneous polarization is sufficiently aligned according to the direction of the

external electric field and the remanent polarization is maintained after the external

electric field is cancelled, hence it has piezoelectricity. Such piezoelectric materials

mainly include BaTiO3, Pb(Zr,Ti)O3, PbTiO3, etc.69,70

The piezoelectric ceramics have

high piezoelectricity, high dielectric constant, low mechanical quality factor, large

electrical loss and poor stability, thus they are suitable for high-power transducers,

broadband filtering and piezoelectric sensors. Piezoelectric crystals generally

correspond to piezoelectric single crystals, which refer to crystals grown in a

long-range order according to the crystal lattice space. The piezoelectric crystal has

non-centrosymmetric structure and therefore is piezoelectric, such as quartz, LiNbO3,

TiNbO3, Pb(Mg1/3Nb2/3)O3, etc. The piezoelectric single crystals are used for standard

frequency-controlled oscillators and high selectivity filters.71

The second type is organic piezoelectric material, such as piezoelectric polymer

polyvinylidene fluoride (PVDF) and other related organic piezoelectric materials.72-74

These materials have the advantage of flexible, low-density and low-impedance,

fff19 YUAN Shuoguo




Chapter 1

which are used for acoustic ultrasonic measurement, pressure sensing, etc. The

disadvantage is that the low piezoelectric strain constant, limiting the application as

an active emission transducer.

Currently, the demands for miniaturization and multifunction of electronic

devices are growing. 2D piezoelectric materials have drawn significant attention due

to the ultrathin scale, excellent electromechanical response and intergrating with other

materials. Therfore, these characteristics make 2D piezoelectric materials have the

potential applications for designing and development of nanomechanical system,

self-adaptive energy harvesting system, piezotronics, and smart electronic devices,

etc.

1.3 Significance of research

Ferroelectric materials have attracted intensive interest due to a wide range of

applications in smart sensors, capacitors, energy harvesting devices and non-volatile

memories. Ferroelectricity at the nanoscale has emerged as fertile ground for novel

physics and applications. However, maintaining the ferroelectricity in oxide-based

films below a critical thickness has been hampered by the intrinsic depolarization

effects, and the Curie temperature significantly decreases when a ferroelectric material

becomes thinner, which greatly limits their usefulness in modern high-density or

fff20 YUAN Shuoguo




Chapter 1

nanoscale systems. Therefore, the critical thickness in the ferroelectric materials, such

as typical perovskite ABO3, is limited by the scale of a few unit cells. 2D layered

materials provide unprecedented possibilities to overcome the problems due to no

dangling bonds, no constraints of lattice mismatch, structure flexible characteristic and

compatibilities of other materials. It is noted that in-plane ferroelectricity is

experimentally reported in single-unit cell SnTe thin-films, whereas only thicker films

possess ferroelectric properties above room temperature. Ferroelectricity in single-unit

cell SnTe film is absent at room temperature and in-plane ferroelectricity is not suitable

for many promising applications. In addition, a few 2D materials are reported to

demonstrate few-layer ferroelectricity at room-temperature. Recently, several

monolayer 2D layered materials have been predicted to exhibit novel ferroelectric

phenomena by theoretical calculation, but still few experimental observations have

been reported from them so far. In this thesis, the experimental observation of

room-temperature ferroelectricty in an unexploited phase of MoTe2 down to the

monolayer limit was reported. The ferroelectricity mechanism is addressed by

combining structural characterizations and DFT calculations. Furthermore, an ultrathin

vdW heterostructure based FTJ device using the discovered 2D ferroelectrics is

constructed. The vdW heterostructure platform based on 2D ferroelectricity provides

fff21 YUAN Shuoguo




Chapter 1

the way of developing ultrathin ferroelectric devices, which will facilitate the

miniaturization of next generation atomic-scale logical memory devices.

On the other hand, piezoelectricity in 2D vdW materials has attracted enormous

interest owing to promising unique applications in nanoscale electromechanical

systems, ultrasensitive sensors, high-precision actuators, piezotronics,

piezo-phototronics and energy conversion. In ultrathin atomic scale, in-plane

piezoelectricity has been experimentally reported in monolayer MoS2, WSe2, g-C3N4

and Se, which may not be suitable for wide applications in the vertical structural

nano-electromechnical systems. Currently, piezoelectricity in 2D material family has

been intensively predicted by theoretical calculations, but still few experimental

reports from them. Recent studies have reported that out-of-plane piezoelectricity

exists in MoSSe and α-In2Se3, which exhibits low piezoelectric response. Therefore,

enhancing the piezoelectric performance in 2D layered materials is still challenging.

Recent studies have demonstrated that the vdW heterostructure engineering can be

employed to generate the novel physical phenomena and performance, providing an

important approach to enhance various physical properties of 2D layered materials.

There is still no experimental study on the enhanced out-of-plane piezoelectric

performance via vdW heterostructure. The remarkable out-of-plane piezoelectric

response is achieved in 2D vdW heterostructure for the first time, which is different

fff22 YUAN Shuoguo




Chapter 1

from in-plane piezoelectricity in MoS2 restricted by edge-effect and odd-even layers.

In addition, high-precision actuator device based on the piezoelectricity in 2D atomic

scale is shown. Moreover, the vertical switching polarization characteristic of In2Se3

is also investigated, which can retain the ferroelectric behaviour down to 6 nm. The

heterostructure fabricated onto silicon substrate is compatible with state-of-the-art

microfabrication technology, which will result in the unexpected applications in

nanoelectronics and atomic-scale multifunctional coupling devices.

1.4 Structure of thesis

The chapters of this thesis are organized as follows:

Chapter 1: Introduction. In this chapter, the crystal structure and fascinating

properties of typical 2D layered materials are first introduced. Then, the perovskite

oxides-based nanoscale ferroelectric and piezoelectric materials are illustrated.

Following that, the motivations and objectives of this thesis are presented.

Chapter 2: Experimental methods. This chapter introduces the important

experimental methods used in this research, including fabrication methods of 2D

materials, structural characterization methods, ferroelectric and piezoelectric

properties measurement and transport characteristic measurement systems.

Chapter 3: Ferroelectricity in distorted 1T phase MoTe2. This chapter explores

fff23 YUAN Shuoguo




Chapter 1

the fabrication of the samples, structure characterization, ferroelectric characterization

and physical mechanism in MoTe2. The devices based on ferroelectricity in MoTe2 are

demonstrated.

Chapter 4: Out-of-plane piezoelectricity in the exfoliated In2Se3. This chapter

demonstrates the preparation and structural characterization of In2Se3 samples, and

piezoelectric and ferroelectric properties of In2Se3 as well as its device applications.

Chapter 5: Out-of-plane piezoelectricity in van der Waals heterostructure. In

this chapter, the preparation and Raman characterization of vdW heterostructure are

shown. Piezoelectricity in various vdW heterostructures is presented. In addition, the

physical mechanism of piezoelectricity in vdW heterostructure is also discussed.

Chapter 6: Conclusion and future prospect. In this chapter, the results in this

thesis are summarized. In addition, future prospect of 2D ferroelectricity and

piezoelectricity and related applications is presented.

fff24 YUAN Shuoguo


Chapter 2

Chapter 2 Experimental Methods

This chapter will introduce the primary experimental methods which were used

in the thesis. 2D materials like MoTe2, In2Se3, MoS2, and WS2 were prepared by

mechanial exfoliation and chemical vapor deposition. The MoTe2 films were also

fabricated by molecular beam epitaxy (MBE). The Raman spectra of the samples were

measured by Raman spectroscopy. The chemical composition was measured using

X-ray photoelectron spectroscopy (XPS). A femtosecond laser was utilized to measure

the second-harmonic generation (SHG) signals of samples. The microstructures and

chemical compositions of samples were characterized by transmission electron

microscope (TEM) equipped with energy-dispersive X-ray spectroscopy (EDX). The

atomic structures of samples were measured by a scanning transmission electron

microscopy (STEM) aberration corrector operated. The surface topography of the

samples was characterized by using atomic force microscopy (AFM) and scanning

tunnelling microscopy (STM). The piezoresponse force microscopy (PFM)

measurements were performanced on a commercial atomic force microscope. The

PFM hysteresis loops, PFM phase and amplitude images were collected in the single

frequency or dual frequency resonance tracking mode. The vector PFM was also

performed by imaging the in-plane signal of samples. The I-V curves of the devices

fff25 YUAN Shuoguo


Chapter 2

were measured by conductive atomic force microscopy (CAFM) at room temperature

to characterize the electrical transport characteristics.

2.1 Fabrication methods of 2D materials

2.1.1 Exfoliation

Mechanical exfoliation is a typical method to fabricate the 2D thin flakes.19,75

The method is to apply a force to debilitate the vdW interaction between the bulk

crystal layers with a tape, meanwhile, the in-plane covalent bonds of each layer are

not broken, which leads to peel off monolayer or few-layer 2D crystals. An adhesive

Scotch tape is pressed close to the 2D bulk crystal peeled into a thin flake via using

another adhesive surface. This process can be repeated several times to obtain an

appropriately thin flake, and then the tape with layered material crystals is pressed

against the target substrate surface. Finally, after peeling off the tape, monolayer or

few-layer thin flakes left on the substrate can be obtained, as shown in Figure 2.1. The

exfoliated samples via an optical microscope are to observe and identify the location

and possible thickness when a suitable substrate is employed.

fff26 YUAN Shuoguo


Chapter 2

Figure 2.1 A typical method to fabricate the 2D flakesby mechanical exfoliation.75

Liquid exfoliation is also an important and simple approach of exfoliation to

fabricate the high yield 2D nanosheets.19,76,77

Currently, liquid exfoliation method has

been widely used to exfoliate a large number of the 2D bulk crystals (e.g., graphene,

TMDs, h-BN, BP, etc). Typically, sonication is the most common and simple method

to exfoliate the layered bulk crystals into 2D nanosheets (Figure 2.2). First, the 2D

bulk crystals were dispersed in a specific solvent and then treated with sonication for

a selective time. Then, the nanosheet suspension was purified through centrifugation

after sonication. The surface energy between the layered bulk crystal and liquid

solution plays an important role in performing efficient exfoliation of layered bulk

crystals. The thickness, size and concentration of the nanosheets can be generally

controlled by the sonication time, sonication temperature, solvent system, ultrasonic

power, etc. According to the experimental and theoretical results, the choice of the

liquid solution is a vital factor for the efficient exfoliation via minimizing the

fff27 YUAN Shuoguo


Chapter 2

exfoliation energy. The liquid solution is to stabilize the exfoliated nanosheets and

prohibit their restacking and aggregating. The mixture of water and ethanol is usually

effective for exfoliating and dispersing TMD nanosheets.

It is worth noting that pure water or ethanol is not efficient in removing TMD

due to its large surface energy. The ultrasonic-based liquid exfoliation with effective

spalling can be realized in solvents where the surface energy matches the bulk crystals.

The surface energy of materials is different, which is difficult to find a suitable

solvent system for each bulk crystal. Meanwhile, the addition of a polymer or

surfactant to an aqueous solution for sonicating layered bulk crystals is another

powerful method for stripping them into ultrathin 2D nanosheets. By adding a

polymer or a surfactant to an aqueous solution, the surface tension of the aqueous

solution can be easily adjusted to match the surface energy of the bulk crystal, and

effective stripping of the layered material can be achieved.

Though the sonication-based liquid exfoliation method is the most widely used

for the preparation of 2D materials, there are several disadvantages for the sonication-

based liquid exfoliation. Firstly, the yield of the monolayer nanosheets is low because

some special properties of 2D materials can only be observed in the monolayer scale.

Secondly, for the ultrasonic treatment in aqueous polymer or surfactant solution, the

residual polymer or surfactant absorbed on the prepared nanosheets is not desirable

fff28 YUAN Shuoguo


Chapter 2

for some device applications. Thirdly, the horiziontal size of the prepared nanosheets

is comparatively small because the ultrasound force decomposes the big flakes into

small fragments. Lastly, ultrasonic treatment may cause some defects on the prepared

nanosheets, which will change the properties of the nanosheets.

Figure 2.2 Liquid exfoliation based on sonication. The bulk crystal is sonicated in a

solvent system, which leads to exfoliate into the 2D nanosheets.76

2.1.2 Physical vapor deposition

Pulsed laser deposition (PLD) is an important technique to prepare the large area

high-quality thin film samples.78

PLD is a kind of physical vapor deposition method,

the typical PLD system includes three main sections: laser system, vacuum system

and deposition chamber, as shown in Figure 2.3. The fundamental working principle

of PLD includes the connection of the laser and the objective material in high vacuum

condition. During the thin film growth process, a pulsed laser is concentrated, and the

incident laser goes into the vacuum deposition chamber and hits on the surface of the

fff29 YUAN Shuoguo


Chapter 2

target. Typically, a KrF excimer laser with a laser wavelength of 248 nm is utilized.

When the laser energy is sufficiently high, the material will make a plasma plume,

which incorporates atoms, particles, ions and cluster. The substrate is typically placed

in front of the objective material, and the plasma plume moves to the substrate in a

direction perpendicular to the surface of the target, and finally transports to the

substrate to agglomerate and nucleate to form a thin film. The thin film thickness can

be controlled precisely by the numbers of laser pulses or deposition time. To control

the pulsed laser into the chamber, laser path ought to be balanced accurately. The

sample quality of the PLD growth thin film is modulated by a series of experimental

parameters, for example, laser energy, substrate temperature, fabrication time, vacuum

pressure, and the distance between the target and substrate.

Figure 2.3 Schematic of the PLD equipment.78

fff30 YUAN Shuoguo


Chapter 2

The MBE is another important type of physical vapor deposition method for the

preparation of high quality single-crystal films and nanostructures.79

The basic

principle is that in an ultra-high vacuum environment, molecules or atoms with

certain kinetic energy are deposited on the surface of a selective substrate by heating

the evaporation source, and then adsorption, migration or surface generation occurs.

Epitaxial growth of the material is achieved after the reaction. MBE is essentially a

non-equilibrium growth process, which is the process of vapor phase atomic

deposition onto the surface of the substrate to become a solid phase, as a result of

growth kinetics and thermodynamic interactions. The growth process of MBE is

carried out in an ultra-high vacuum environment to avoid interference of impurities.

The substrate is cleaned, and the epitaxial material is high in purity, and therefore

high-quality samples can be prepared. In addition, its evaporation rate is slow but

stable, and a single-crystal film with uniform composition and structure can be

obtained. The growth process has atomic scale controllability, and the film thickness

can be precisely controlled. A variety of 2D materials are currently available by this

method, such as graphene, silene, In2Se3, etc.

2.1.3 Chemical vapor deposition

The chemical vapor deposition (CVD) has been widely used for 2D layered

fff31 YUAN Shuoguo


Chapter 2

materials growth.80-82

The CVD method is able to fabricate high quality and limited

defects with scalable dimensions and controllable thickness. During CVD growth

process, a pre-selected substrate is placed in the furnace chamber and one or more

gaseous precursors are circulated within the furnace chamber, and then the precursor

is reacted and/or deposited on the surface of the substrate, as shown in Figure 2.4.

Based on this method, ultrathin 2D nanosheets can be obtained under suitable

experimental conditions. It is worth noting that the electronic properties of 2D

materials such as graphene and TMDs are approaching the electronic properties of

mechanically exfoliated samples. Therefore, the ultrathin 2D materials grown by

CVD are also expected to be candidates for the fabrication of high performance

electronic and optoelectronic devices. Unlike the low yield and productivity of

mechanical exfoliation method, CVD can produce industrial-scale materials, which

shows broad application prospects in the electronic and optoelectronic devices. So far,

many ultrathin 2D TMDs nanosheets have been grown under different conditions

including MoS2, MoSe2, MoTe2, WS2, WSe2, WTe2, etc. Apart from pure TMDs

nanosheets, the alloyed TMDs nanosheets have been prepared, such as MoxW1-xS2,

MoS2xSe2(1−x), etc. In addition, lateral and vertical growth also can be controlled in the

2D TMDs heterostructure growth. Due to the different band gaps of TMDs nanosheets,

the lateral and vertical heterostructures can be considered as natural p-n junctions,

fff32 YUAN Shuoguo


Chapter 2

making them ideal for building high performance electronic and optoelectric devices.

But the CVD strategy has a few shortcomings, for exmaple, the CVD-grown 2D

materials are constantly deposited on the selected substrates, which requires transfer

process to different substrates for further measurements. Typically, the CVD method

needs high temperature and inactive environment, which increases the fabrication

expense of samples.

Figure 2.4 Schematic of CVD growth.

2.2 Measurement methods

2.2.1 Optical characterization

The optical microscope provides a simple and rapid method to obtain

information about samples location, shape, and thickness.19,83

The working principle

of the optical microscope method is based on the interference of reflected light at

different substrate interfaces. When the sample introduces significant optical paths

and opacity perturbations, an optical contrast is produced between the bare substrate

and the 2D materials. Silicon coated with a silicon dioxide thin layer (SiO2/Si) is the

fff33 YUAN Shuoguo


Chapter 2

most commonly used substrate to observe the 2D materials, especially, 285 nm thick

SiO2/Si. A part of the incident light is transmitted while another part is reflected to

each interface. The intensity of the collective reflected light with respect to

interference varies with locations depending on the sample thickness, absorption

coefficient, and difference in refractive index. The difference between the

experimental and calculated results may be attributed to the uncertainty of the optical

parameters of these ultrathin 2D materials. The contrast studies are very helpful to

investigate the physical properties of emerging 2D materials depending on their

layer-number. The approximate thickness based on the color contrast of 2D materials

can be identified. But for accurate thickness, further characterizations such as AFM,

STEM, Raman and photoluminescence (PL) spectroscopy are required.

2.2.2 Structural characterization

Raman spectroscopy is a fast, non-destructive means of giving material structure,

electronic information, and high spatial resolution.84,85

Beginning with the successful

characterization of graphene, Raman spectroscopy has been used to demonstrate the

layers number, crystal strcuture orientation, doping effect, phase, strain, and vdW

coupling of various ultrathin 2D layered materials. The basic mechanism of the

Raman spectroscopy is shown in Figure 2.5. When light illuminates on the objective

fff34 YUAN Shuoguo


Chapter 2

material, both Rayleigh scattering and Raman scattering can be seen. Rayleigh

scattering is a sort of elastic scattering. Analysis of the frequency (wavelength) of the

scattered light reveals that

Documents

Copyright Undertaking · 2020. 7. 9. · enhancement of graphene field-effect transistor based on vertical tunnelling heterostructures, Advanced Materials. 28, 10048 (2016). 5. Weng-Fu