University of Wollongong University of Wollongong

Research Online Research Online

University of Wollongong Thesis Collection 2017+ University of Wollongong Thesis Collections

2018

Surface and Interface Engineering in Semiconducting Electrocatalyst and Surface and Interface Engineering in Semiconducting Electrocatalyst and

Photo(electro)catalyst Photo(electro)catalyst

Haifeng Feng University of Wollongong

Follow this and additional works at: https://ro.uow.edu.au/theses1

University of Wollongong University of Wollongong

Copyright Warning Copyright Warning

You may print or download ONE copy of this document for the purpose of your own research or study. The University

does not authorise you to copy, communicate or otherwise make available electronically to any other person any

copyright material contained on this site.

You are reminded of the following: This work is copyright. Apart from any use permitted under the Copyright Act

1968, no part of this work may be reproduced by any process, nor may any other exclusive right be exercised,

without the permission of the author. Copyright owners are entitled to take legal action against persons who infringe

their copyright. A reproduction of material that is protected by copyright may be a copyright infringement. A court

may impose penalties and award damages in relation to offences and infringements relating to copyright material.

Higher penalties may apply, and higher damages may be awarded, for offences and infringements involving the

conversion of material into digital or electronic form.

Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily

represent the views of the University of Wollongong. represent the views of the University of Wollongong.

Recommended Citation Recommended Citation Feng, Haifeng, Surface and Interface Engineering in Semiconducting Electrocatalyst and Photo(electro)catalyst, Doctor of Philosophy thesis, Institute for Superconducting and Electronic Materials, University of Wollongong, 2018. https://ro.uow.edu.au/theses1/430

Research Online is the open access institutional repository for the University of Wollongong. For further information contact the UOW Library: research-pubs@uow.edu.au

Surface and Interface Engineering in Semiconducting

Electrocatalyst and Photo(electro)catalyst

This thesis is presented as part of the requirements for the

Award of the Degree of

Doctor of Philosophy

from the

University of Wollongong

by

HAIFENG FENG

B. Sc., M. Sc.

Supervisors:

Dr. Yi Du, Dr. Xun Xu, Prof. Shi Xue Dou

Institute for Superconducting and Electronic Materials

Faculty of Engineering

July 2018

i

Table of contents

Table of contents........................................................................................................... i

Acknowledgements ..................................................................................................... iv

Certification ................................................................................................................. v

List of abbreviations................................................................................................... vi

Abstract ....................................................................................................................... ix

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

1.1 General background ............................................................................................ 1

1.2 Literature review and research motivation .......................................................... 2

1.2.1 Semiconductor electrocatalysts .................................................................... 2

1.2.2 TiO2 based catalysts ..................................................................................... 4

1.2.3 Semiconductor photocatalyst and photo(electro)catalysts ......................... 11

1.2.4 BiOBr based catalysts ................................................................................ 14

1.3 Outline of chapters ............................................................................................ 17

1.4 References ......................................................................................................... 18

Chapter 2 Experimental Techniques ....................................................................... 32

2.1 Scanning tunneling microscopy ........................................................................ 32

2.1.1 Quantum tunneling ..................................................................................... 32

2.1.2 Working principle of STM ......................................................................... 34

2.1.3 Ultra-high vacuum system and molecular beam epitaxy ........................... 36

2.2 Atomic force microscopy .................................................................................. 38

2.2.1 Working principles of AFM ....................................................................... 38

2.2.2 Working modes of AFM ............................................................................ 39

2.3 Other facilities ................................................................................................... 43

2.3.1 Scanning electron microscopy and focused ion beam ............................... 43

2.3.2 Transmission electron microscopy ............................................................. 44

2.3.3 Photoelectron spectroscopy ........................................................................ 47

2.4 References ......................................................................................................... 49

Chapter 3 Characterization of TiO2(110) surface by STM ................................... 51

3.1 Introduction ....................................................................................................... 51

3.2 Experimental section ......................................................................................... 52

3.3 Results and discussion ....................................................................................... 53

3.3.1 Rutile(110) surface with (1×1) reconstruction ........................................... 53

3.3.2 OV point defects on (1×1) surface ............................................................. 54

3.3.3 Rutile(110) surface with (1 × 2) reconstruction ......................................... 56

ii

3.4 Summary ........................................................................................................... 58

3.5 References ......................................................................................................... 58

Chapter 4 Activating titania for electrocatalysis by surface vacancy engineering

..................................................................................................................................... 62

4.1 Introduction ....................................................................................................... 62

4.2 Experimental sections ....................................................................................... 63

4.3 Characterization of OV by STEM ..................................................................... 65

4.4 Characterization the electronic structure by in-situ synchrotron XPS and UPS68

4.5 Electrochemical HER activity of OV-TiO2 ....................................................... 71

4.6 Determination of the origin of the electrochemical HER activity of OV-TiO2 73

4.6.1 Electrical conductivity................................................................................ 73

4.6.2 Electrochemical activate sites .................................................................... 75

4.6.3 DFT calculation of the Gibbs free energy .................................................. 80

4.7 Summary ........................................................................................................... 87

4.8 References ......................................................................................................... 87

Chapter 5 Tuning electronic structure of BiOBr 2D nanosheets by strain for

photocatalysis ............................................................................................................. 93

5.1 Introduction ........................................................................................................... 93

5.2 Experimental section ......................................................................................... 94

5.3 Morphology and structure characterization ....................................................... 96

5.4 Characterization of strain in BiOBr 2D nanosheets .......................................... 97

5.5 Characterization of photocatalytic activity of BiOBr 2D nanosheets ............. 100

5.6 DFT of the strain effect on the electronic structure of BiOBr ........................ 102

5.7 Summary ......................................................................................................... 105

5.8 References ....................................................................................................... 105

Chapter 6 Construction of 2D lateral pseudo-heterostructures by strain

engineering ............................................................................................................... 109

6.1 Introduction ..................................................................................................... 109

6.2 Experimental section ....................................................................................... 111

6.3 Structure characterization of BiOBr nanosheets ............................................. 113

6.4 Strain induced pseudo-heterostructure ............................................................ 116

6.5 Electronic properties characterization ............................................................. 122

6.6 DFT calculation of the strain effect on the band structure of BiOBr .............. 126

6.7 Performance of BiOBr nanosheets in photoelectronic devices ....................... 128

6.8 Summary ......................................................................................................... 132

6.9 References ....................................................................................................... 132

Chapter 7 Summary and outlooks ......................................................................... 137

iii

Appendix A: List of publications .............................................................................. 139

Appendix B: Conference contributions ..................................................................... 141

iv

Acknowledgements

This thesis was accomplished under the supervision of Dr. Yi Du, Dr. Xun Xu, and

Prof. Shi Xue Dou in Institute for Superconducting and Electronic Materials (ISEM)

in University of Wollongong.

Firstly, I want to acknowledge Dr. Du, my principle supervisor, for both his strong

supports in experiments and building my scientific thoughts. I am also grateful to my

co-supervisors, Dr. Xu and Prof. Dou, for their enthusiastic guidance and significant

contributions during the whole PhD course. Their meticulous scientific attitudes and

creative ways of thinking will be a wealth that benefit to my entire working career.

I would also like to thank all the colleagues, also my friends, in our research group.

I received continuous help from Dr. Jincheng Zhuang, Dr. Yundan Liu, Miss Yani Liu

and Dr. Zhi Li in STM operation, data analysis and daily discussions. Meanwhile, I am

grateful for other group members for their kindness in internal collaborations and

discussions, including Mr. Amar Al-Keisy, Miss Li Wang, Mr. Long Ren, Mr. Liang

Wang, Mrs. Nana Wang, Dr. Zhongchao Bai, Miss, Ningyan Cheng, Miss Nana Liu

and Miss Guyue Bo. Thanks Assoc. Prof. Michael Higgins and Dr. Tian Zheng from

IPRI for AFM training. Kind helps from Mr. Robert Morgan, Mr. Paul Hammersley,

Mr. John Wilton, Mr. Mat Davies for maintaining STM equipment are appreciated.

In addition, I need to thank my collaborators, Prof. Weichang Hao, Dr. Zhongfei Xu,

and Mrs. Dandan Cui in Beihang University and UOW-BUAA Joint Research Centre,

Prof. Jun Chen in IPRI, Prof. Zhenpeng Hu in Nankai University, Prof. Jijun Zhao and

Mr. Nan Gao in Dalian University of Technology.

At last, I want to take this opportunity of thanks my parents and my wife for their

endless love and support during this four years.

v

Certification

I, Haifeng Feng, declare that this thesis, submitted in fulfilment of the requirements for

the award of Doctor of Philosophy, in the Institute for Superconducting & Electronic

Materials, Faculty of Engineering, University of Wollongong, is wholly my own work

unless otherwise referenced or acknowledged. This document has not been submitted

for qualifications at any other academic institution.

vi

List of abbreviations

HER hydrogen evolution reaction

OER oxygen evolution reaction

LDOS local density of states

TMO transition metal oxide

XAS x-ray absorption spectroscopy

STM scanning tunneling microscopy

STS scanning tunneling spectroscopy

DFT density functional theory

AFM atomic force microscopy

KPFM kelvin probe force microscopy

SKPM scanning kelvin probe microscopy

CPD contact potential difference

UHV ultra-high vacuum

TMP turbo molecular pump

TSP titanium sublimation pump

MBE molecular beam epitaxy

LT low temperature

GV gate valve

SEM scanning electron microscopy

FIB focused ion beam

EDX energy-dispersive x-ray spectroscopy

TEM transmission electron microscopy

vii

SAED selected area electron diffraction

HRTEM high resolution transmission electron microscopy

STEM scanning transmission electron microscopy

ABF annular bright-field

HAADF high angle annular dark-field

EELS electron energy loss spectroscopy

GPA geometric phase analysis

XPS x-ray photoelectron spectroscopy

UPS ultraviolet photoelectron spectroscopy

UV ultraviolet

UV-vis ultraviolet-visible

CB conduction band

VB valence band

CBM conduction band maximum

VBM valence band maximum

OV oxygen vacancy

OHP hydroxyl group pairs

CNT carbon nanotube

MO methyl orange

Rh B rhodamine B

CTAB cetyl trimethylammonium bromide

FFT fast Fourier transform

XRD x-ray diffraction

PPMS physical property measurement system

VASP vienna Ab initio simulation package

viii

GGA generalized gradient approximation

PBE Perdew-Burke-Ernzerhof

PAW projector augmented wave

ZPE zero point energies

LSV linear sweep voltammetric

EASA electrochemically active surface area

RHE eversible hydrogen electrode

ix

Abstract

Surface and interface engineering is one of the most effective approaches in tuning

semiconductors for their chemistry (energy, environment and catalysis) and device

applications (electronic devices, and optical-electronic devices).

In this thesis, we show several approaches of modifying the catalytic and electronic

properties of several semiconductors through manipulating their surface and interface.

Techniques including scanning probe microscopies (SPM), electron microscopies, and

electron spectroscopies, and, X-ray spectroscopies, were taken used to characterize the

effect of surface and interface engineering, and the effect on their electronic properties.

This thesis includes:

1. Oxygen vacancies (OV) engineering on TiO2 rutile(110) single crystal. By carefully

controlling the annealing temperature and annealing time, OV defects can be precisely

introduced into the single crystal. The evolution of the surface structure was

investigated by in-situ low temperature STM. It was found that at 900 K, OV point

defects isolating with each other exist on the (1 × 1) surface. In addition, OVs on the

surface tended to be filled either water molecule or OHs. When the annealed

temperature increased to 1300 K, (1 × 2) surface reconstruction dominated on the

surface, in which cross-links defects caused by more oxygen deficiency were found.

2. Activating TiO2 for electrocatalysis by surface vacancy engineering. STEM

techniques clarified that the good crystalline nature of reduced TiO2 with and the spatial

distribution of OV in the surface and near-surface region. A mid-gap defect electronic

state was created by the OVs, which can effective enhance the electric conductivity of

the reduced TiO2. The reduced TiO2 exhibit tremendous enhancement in hydrogen

x

evolution reaction (HER) due to the increased electric conductivity and amounts of

active sites. Combing the in-situ observation of the water splitting reaction by STM

and DFT calculations, subsurface oxygen vacancies and low coordinated Ti ions (Ti3+)

demonstrated their roles in enhancing the electrical conductivity and promoting

electron transfer and hydrogen desorption, which activate reduced TiO2 single crystal

in the hydrogen evolution reaction in alkaline media. This study offers a rational route

for developing reduced transition metal oxide for low-cost and highly active hydrogen

evolution reaction catalysts, to realize over all water splitting in alkaline media.

3. The band structure of BiOBr 2D nanosheets was tuned by the strain for

photocatalysis. The inner strain in the BiOBr nanosheets has been tuned continuously

by controlled manipulating their shapes. The photocatalytic performance of BiOBr in

dye degradation can be manipulated by the strain effect. The low-strain BiOBr

nanosheets show improved photocatalytic activity. DFT suggest that strain can modify

the band structure and symmetry in BiOBr. The enhanced photocatalytic activity in

low-strain BiOBr nanosheets is due to improved charge separation attributable to a

highly dispersive band structure with an indirect band gap.

4. Constructing two dimensional (2D) lateral pseudo-heterointerface by strain

engineering in BiOBr nanosheets. Taking advantage of their strain-sensitive layer

structure, 2D lateral pseudo-heterogeneous interface are realized in the single-

component BiOBr nanosheets by finely tune the pH value in synthesis. Due to the

proper band alignment cross the interface, charge separation under visible light

irradiation was enhanced, which was reflected by the photo-current measurements and

the degradation experiments of pollutions. The strain engineering was demonstrated to

xi

be an effective way to tune the electronic structure of BiOBr and promote its efficiency

in photoenergy conversion applications. In addition, the construction of the lateral

pseudo-heterostructure through strain exhibit promising applications in building

unprecedented 2D systems with exciting properties.

1

Chapter 1 Introduction

1.1 General background

Semiconducting electronic materials can be defined as the semiconductors which are

applied for their electrical properties. According to band theory, a semiconductor

material with small but non-zero band gap, has an electric conductivity between that of

an insulator and that of most metals. The behaviours of electrons in semiconductors are

determined by their band structure, which has been successfully implemented in

describing many physical properties of solids. They have been utilized in a variety of

application fields mainly based on their different electronic properties, including

chemistry (energy conversion and storage, environmental remediation and catalysis)

and device applications (electronic devices, and optical-electronic devices).[1-2]

Surface and interface plays a critical role in determining the electronic properties of

semiconductor materials, especially for those in nanostructures with very large specific

surface area.[3-5] Abundant surface mismatches, low coordinated ions or atoms, and

surface defects can strongly alter their electronic structures, which could play a vital

role in the corresponding performances in their applications.[6-8] However, effectively

characterizing conditions of surface or interface, and modifying the electronic

properties through controllably manipulating surface conditions are challenging, due

to the lack of approaches in monitoring fine surface structure and localized variation of

electronic structure in micro or nano-scale. Effectively control the species,

concentration, and distributions of surface defects and their effects on the electronic

structure, hence, are of great significance in developing their applications.

2

1.2 Literature review and research motivation

1.2.1 Semiconductor electrocatalysts

In the research filed of energy conversion, including fuel cells, solar cells and

batteries, electrocatalysts are crucial in the reaction rate, efficiency, and selectivity.[9]

Pursuing efficient and low-cost electrocatalysts is one of the key topics in this filed. In

this thesis, we will largely limit the introduction to water electrolysis that split water

into hydrogen and oxygen. For both the hydrogen evolution reaction (HER) and oxygen

evolution reaction (OER), noble metal (Pt, Ir, Au) or noble metal based electrocatalysts

with high activity and stability are still dominating in these applications, which are not

desirable due to their high cost and low-abundance.

Varieties of earth-abundant semiconductor materials have been extensively explored

as alternatives of noble metal based electrocatalysts, in which the most efficient ones

are designed based on transition metal elements (Fe, Co, Ni, Cu, Mo, W, Mn, Ti, Zn)

and p-block elements (S, Se, C, O, N, P).[10-16] Among them, transitional metal oxides

(TMOs), which are being widely applied either as catalysts or supports in industrial

processes, have been recognized as appealing alternatives to noble-metal based

electrocatalysts for the OER, especially in alkaline media.[17-19]

To reach a high electrocatalytic activity close to noble metals, two requirements are

needed to be met in the first place. The first one is a high conductivity, which enable

the transport of electrons between electrodes and active sites. To overcome the poor

conductivity of common semiconductor materials, conductive materials, for examples

carbon based materials, such as, carbon based materials and Ni foams, are need to be

used as conducting substrate to fabricate a heterogeneous structure.[20-22]

3

Figure 1.1 (a) Trend in overpotential for the OER is shown as a function of the 3d

transition elements. (b) Trend in overpotential for the HER is shown as a function of

the 3d transition elements. In both reactions, the trends of Mn < Fe < Co < Ni was

determined.[23] Reproduced with permission from Springer Nature.

The second important aspect is the active site on the surface of an electrocatalyst, in

which the catalytic charge transfer happens. Identifying the active site of high active

electrocatalysts are important in further improving their activity, revealing the catalytic

mechanism and rationally developing new electrocatalysts. For example, for an early

work shown in Figure 1.1, combined with X-ray absorption spectroscopy (XAS),

4

scanning tunneling microscopy (STM) and electrochemical measurements,

Subbaraman et al. proposed that the strength of OH-M2+δ (0 ≤ δ ≤ 1.5) energetic (M,

Ni < Co < Fe < Mn) play the determining role in the reactivity of both OER and HER

in alkaline media (Mn < Fe < Co < Ni), which was helpful for guiding further designing

electrocatalysts by tuning active sites.[23]

Inspired by these works, intense attempts have been made in developing TMOs for

HER and overall water splitting in recent years. Nanoscale NiO/Ni-carbon nanotube

(CNT) material[24], Ni/CeO2−CNT[25], 3D crystalline/amorphous Co/Co3O4 core/shell

nanostructure on Ni foam[26], oxygen vacancy (OV)-CoO/carbon[27] and TiO2

nanodots/Co nanotubes on carbon fibers[28] have exhibited competitive HER activities

toward commercial Pt/C catalysts in alkaline media. Hollow Co3O4 microtube arrays

on nickel skeleton,[29] porous MoO2 nickel foam,[30] Ni−Co−A (A = P, Se, O)

nanosheets on nickel foam[31] and have been reported to be highly active bifunctional

electrocatalysts for overall water splitting.

1.2.2 TiO2 based catalysts

Figure 1.2 The crystal structure of the three phases of TiO2. (a) anatase (tetragonal, a =

0.3785 nm, c = 0.9513 nm), (b) rutile (tetragonal, a = 0.4593 nm, c = 0.2959 nm), (c)

brookite (orthorhombic, a = 0.5455 nm, b = 0.9181 nm, c = 0.5142 nm).

TiO2 is found exist in nature in three crystal structures.[38] The crystal structure of

the three phases is shown in Figure 1.2. Rutile, a tetragonal structure, is the most

5

common one in nature with the most stable structure. Due this reason, rutile is the most

studied oxide surface although in the two decades.[32] Beside rutile, anatase is also

tetragonal structure, while brookite is orthorhombic. In recent years, the study of

anatase in fields of both surface characterization and surface catalysis have also got

huge progresses.[33-37]

In the study of catalytic properties of TiO2, rutile(110) face is mostly selected,

because it is the most thermodynamically stable crystal face of TiO2. The information

and knowledge we got from this surface are also useful for the understanding of other

crystal faces and phases of TiO2, as well as other oxide surfaces. As shown in Figure

1.2 (a), rutile TiO2 presents a tetragonal structure, with one Ti atom surrounded by six

O atoms distributed in a distorted octahedral. In the bulk structure, Ti atoms are six-

fold (6f) and oxygen atoms are threefold coordinated. While, on the rutile(110) surface

as shown in Figure 1.3, there are two types Ti atoms, five-fold (5f) and 6f coordinated

Ti atoms, as well as two-fold (2f) coordinated oxygen atoms (bridging oxygen atoms)

and three-fold (3f) coordinated oxygen atoms. On the perfect stoichiometric rutile TiO2

(110) surface, 5f Ti atom row, 6f atom rows and 3f oxygen atom rows lie in a plane of

along the [001] direction, while bridging oxygen atom rows locate above 6f Ti rows in

the same direction. The surface of (110) cleaved rutile has been proved to be auto-

compensated and charge neutral, as well as breaks the minimum number of bonds and

the longest bonds in the crystal structure, which consequently is the lowest-energy

surface structure, by means like first-principles calculations and STM.[39,40] For the

freshly cleaved surface, the first few layers on the TiO2 rutile(110) structure will relax

to saturate the dangling bonds. As a result, a rumpling structure is created by two

different types of Ti (5f Ti and 6f Ti) atoms moving in opposite direction on the top

several atomic layers. The unit cell of TiO2 rutile(110)-(1×1), therefore, is defined as

6

two adjacent 5f Ti atoms in [001] direction is with a distance 2.96 Å, and two adjacent

5f Ti atoms in [1-10] direction with a distance of 6.5 Å.[41-44]

Figure 1.3 Diagram of the crystal structure of TiO2 rutile(110) surface, with 5f Ti, Obr

and surface OV marked.[44]

Figure 1.4 STM images of TiO2(110)-(1×1) surfaces with different surface species. (a)

a reduced surface with OVs, (b) partially hydroxylated surface with OH groups, and

(c) fully hydroxylated surface with OHPs. (d) Height profiles of the labeled sites in (b)

for OV, OH, and OHP, respectively. All the STM images have a size of 11 nm × 11

nm, and were obtained at V = 1.6 V, I = 10 pA. [45] Reprinted with permission from

American Chemical Society.

7

Plenty of STM images have been acquired from TiO2 rutile(110)-(1×1) surface

which are consistent with the theoretical structure in Figure 1.3. Most images are

obtained on empty-state mode, because normally on filled-state mode the tip will

scrape on the surface. Limited images on filled-state mode were achieved by some

groups.[46,47] The bright rows along [001] direction are composed of 5f Ti atoms, while

the dark rows are assigned to bridging oxygen atoms on empty-state imaging mode.

This contrast is dominated by the distribution of density of states despite bridging

oxygen atoms are higher than 5f Ti atoms in physical geometry structure.

Figure 1.5 STS measurements on TiO2 rutile(110)-(1×1) surfaces. (a) dI/dV of the

occupied state of different site at 78 K: a, at the center of a lobe in the occupied STM

image between the two 5f Ti sites indicated in the inset STM image. b, at a 6f Ti site.

c, at the OVbr site. Insets are the STM image and the I-V curves corresponding to the

three dI/dV curves.[47] Reproduced with permission from AIP Publishing. (b) STS

measured above OVbr and above regular 5f Ti surface atoms on TiO2 rutile(110) surface

at 78 K, with the measurement range from -3 V (occupied state) to 1.5 V (empty

state).[48] Reproduced with permission from American Physical Society.

Many types of surface defects have been observed on TiO2 rutile(110)-(1×1)

surface. Three types of intrinsic point defects in the reduced TiO2 rutile(110), including

OV defect marked as I, hydroxyl group (OH group) marked as II and hydroxyl group

8

pairs (OHP), are shown in Figure 1.4. Normally, in ultra-high vacuum (UHV) OVs are

created by the reduction of rutile by heating or ion sputtering, while OH groups are

created by the dissociation of water at the oxygen vacancies. Although these defects

are all bright by STM and locate at the middle of two adjacent 5f Ti atoms rows, the

heights at same scanning conditions by STM are different. For example, as shown in

Figure 1.4, the height of OHP (1.6 V, 10 pA) is around 1.2 Å and 1.0 Å for OH at the

same scanning conditions. While for OV the height is much less than OH and OHP. It

is also found that the relative apparent height of OH groups are sensitive to the tip-

sample distance in STM measurements, which were not observed on OV and 5f Ti

atoms.[45]

Figure 1.6 (a)-(d) Dynamic processes of the dissociation of water at an OV site on

TiO2(110) at about 187 K. Protrusions are labeled as follows: OV, OHbr groups and

water on 5f Ti sites are marked as open white circles, filled white circles and filled

black squares, respectively. (e) Schematic diagram of the dynamic processes of water

dissociation on the rutile(110) surface.[50] Reproduced with permission from American

Physical Society.

9

Besides getting the surface morphology with atomic resolution of TiO2, the

electronic band structure TiO2 has been studied by STM, which is strong relevant to its

performance in many applications.[45,46,48] As shown in Figure 1.5 (a), Minato et al. use

scanning tunneling spectroscopy (STS) to identify the occupied Ti 3d defect state of

TiO2. While in the work of Setvin et al. (Figure 1.5 (b)) no clear differences between

the LODS of 5f Ti atoms and OV were observed.

Figure 1.7 Two sets, (a)-(b) and (c)-(e) show two reaction processes between Oa and

water at 300 K, with the interpretation illustrated by the diagram below.[57] Reproduced

with permission from American Physical Society

10

Since the Fujishima and Honda found the photocatalytic water splitting by TiO2, the

study of the interaction of water on TiO2 prospers in last several decades, particularly

detailed understandings of water adsorption, diffusion, and dissociation on prototypical

TiO2 rutile(110).[49] Besides that, the study of the interaction of water with TiO2 surface

is inevitable to fully understand the chemical and catalytic properties of TiO2, because

water always be present either as part of an aqueous environment or as water vapor,

even in UHV conditions.

Many research demonstrate that dissociative adsorption of water happens at OV sites

on reduced TiO2 surface, followed by the appearance of OHP located on two

neighboring Obr sites, with the details of the dissociation of water shown in Figure

1.6.[50] The dissociative adsorption of water molecules and the tip assisted dissociation

of water molecules and OH have also been reported by many groups.[51-56] While on

oxide TiO2 surface, not only the dissociative adsorption of a water molecular is

observed at an OV site, but also the formation of a new water molecule at 300 K, as

revealed in Figure 1.7 by STM.[57]

Other TiO2 surfaces have also draw abundant attentions, for examples, rutile(011),

(100) and (001), as well as anatase(101) and (001), because the phase and facet are

recognized to be closely related to the (photo)catalytic performances.[58,59] It is also

known that the mixed TiO2 of rutile and anatase even exhibit much better

photocatalytic activity than the individual polymorphs, due to the proper band

alignment could effectively promote the separation of photoexcited charge

carriers.[60,61] Among these studies, anatase(101), which is the most frequently exposed

surface of this high active TiO2 polymorph, has been widely studied.[35, 62-68] As shown

in Figure 1.8, water monomer appears as isolated black spot surrounded by two bright

features (whit-black-white, w-b-w). In addition, through STM it was found that at 190

11

K, water monomer appearing as w-b-w can hop on the surface, as shown in Figure 1.8

(c)-(f). With more water on the surface (0.24 ML), they tend to form a (2 × 2)

superstructure on the surface, induced by the charge rearrangement at the molecule-

anatase interface.[62]

Figure 1.8 (a) Schematic diagram of the crystal structure of anatase(101) surface. (b)

STM image of anatase(101) surface at empty state, with unit cell marked by the black

rectangle containing two equivalent Ti5c/O2c surface atoms, and a water monomer

marked by the black arrow. (c) and (d) two consecutive STM images of anatase(101)

surface with 0.11 ML water on the surface, showing the hopping of water monomer

during STM scanning. (e) Difference image of c and d. (f) Height profiles crossing a

water molecule, and two adsorbed water molecules along the blue and red line in d,

respectively. (g) Ordered water overlayer in (2 × 2) on anatase(101), as indicate by the

red unit cell.[62] Reproduced with permission from Springer Nature.

1.2.3 Semiconductor photocatalyst and photo(electro)catalysts

Photocatalysis, through which one can convert solar energy into chemical energy,

has been regarded as one of the most promising strategies to solve crisis of energy

shortage and environmental pollution. It has been demonstrated to exhibit great

potential in several applications, including photocatalytic water splitting,

photosynthesis, and the treatment of pollutants in aqueous or gas phase.[69-72]

Unfortunately, reliable photocatalysts are scarce, although the appeal of the direct

12

conversion of solar into chemical energy via semiconductor compounds has been

recognized for a long time. This is mainly due to their low solar-energy conversion

efficiency, especially in the visible-light spectrum.

Generally, a typical photocatalytic process involves three steps: (i) absorption of

solar light (with the efficiency denoted as ηA), (ii) separation of photoexcited charge

carriers (with the efficiency denoted as ηS), and (iii) reduction or oxidation reactions at

the surface to complete the solar energy conversion (with the efficiency denoted as ηC).

The overall efficiency (η) is therefore determined by multiplying the efficiencies of the

individual steps: η = ηA × ηS × ηC. Thus, several criteria are essential for the design and

development of novel photocatalysts that possess high solar-energy conversion

efficiency. First of all, the band gaps of the photocatalysts should be narrow enough

(Eg < 3.0 eV), which allow absorption of both ultraviolet (UV) and visible light in the

solar spectrum. This is because UV and visible light account for 5% and 43% of the

solar spectrum, respectively. Secondly, a high separation rate and high mobility of the

photoexcited charge carriers (electrons and holes) are essential for high-efficiency

photocatalysts. The former can lead to high quantum conversion efficiency, while the

latter increases the effective charge-carrier diffusion length and thus enhances the

photocatalytic activity. For semiconductors, the mobility of electrons and holes is

determined by their effective mass (m*). As shown in Equation 1.1, the m* is estimated

to the second order derivative of energy (E) with respect to the wave vector (k), which

is reflected by the curvature of the band edges.

1

𝑚∗ =1

ħ2 × 𝑑2𝐸

𝑑𝑘2 (1.1)

As illustrated in Figure 1.9 (a), a larger curvature of the band leads to a small m*

(light effective mass) of the charge carriers. This means the dispersion of the electronic

band determines m*, that is, the more dispersive the band the smaller m* will be, and

13

consequently, the higher the mobility it will have. In contrast, a dispersion-less band

leads to large m* (heavy effective mass) of charge carriers with low mobility. Thirdly,

the positions of the valence and conduction bands (VB and CB) are of importance to

determine the chemical potential of photoexcited electrons and holes, which have

significant impact on the efficiency of different photocatalytic reactions (Figure 1.9

(b)). The overall solar-energy conversion during the photocatalysis process is,

therefore, determined by the electronic structures of photocatalysts.

Figure 1.9 (a) Diagram of the relationship between the band shape and the effective

mass of electrons and holes. (b) Diagram of the band structure of direct-gap

semiconductors, illustrating its role in determining the catalytic properties of

photocatalysts.

Surface and interface properties of semiconductors play a vital role in determining

their photocatalytic activity. This is not only because nanostructure or microstructure

with very large specific surface area are desired, but also due to electronic structure of

photocatalysts and separation of photo-excited electron-hole pairs can be effectively

tuned through surface and interface engineering.[73]

14

1.2.4 BiOBr based catalysts

BiOX (X = Cl, Br, I and F) belongs to the family of bismuth oxyhalides (BixOyXz).

As a heavy metal, bismuth atoms tend to form very weak bonds with non-metal

elements, which generally tend to have small molecular orbital overlap and favors the

establishment of narrow band gaps. Along with a narrow band gap, a small effective

mass may to be acquired for semiconductors with similar crystal and band structures,

according to the k·p perturbation theory. Various bismuth oxyhalides compounds

exhibit high visible-light activity towards photocatalytic degradation of organic and

inorganic toxic substances, water splitting, CO2 reduction and N2 fixation.[74-84]

Their VB maximum is mainly composed of O 2p and X np states (n = 3, 4, and 5 for

Cl, Br, and I, respectively). Their CB minimum in most cases is constructed from Bi

6p states.[85-87] Thus, the band structures of bismuth oxyhalides are expected to be

tunable via the halogen species and the ratios of Bi:O:X. Meanwhile, due to the

dispersive properties of the p and s-p hybridization states, high-mobility charge carriers

can be obtained in many bismuth oxyhalides compounds.

As shown in Figure 2.0 (a) and (b), both experimental and theoretical works have

suggested that the band gap of BiOX can be gradually narrowed from 3.4 eV for BiOCl

to 2.8 eV for BiOBr and 1.9 eV for BiOI, due to the increasing participation of the X

np states.[88,89] As a result, Zhang et al. proved that BiOI shows a wider visible-light

absorption range than BiOCl and BiOBr, and higher visible-light photocatalytic

activity towards degradation of methyl orange (MO).[90] On the other hand, due to the

appropriate VB edge, BiOBr and BiOCl exhibited much higher oxygen evolution

efficiency and better photocatalytic degradation of Rhodamine B (RhB) and phenol

under visible light or simulated sunlight irradiation than BiOI.[88,91] As shown in Figure

2.0 (d) and (e), Bhachu et al. demonstrated that BiOBr also had much superior water

15

oxidation activity compared to BiOI and BiOCl under simulated sunlight irradiation,

due to its more positive VB edge.[92] It should be noted that for BixOyXz photocatalysts,

the nature of their internal electric field contributes to the efficiency of separating

photoexcited charge carriers, which is induced by their layered structure, consisting of

interleaving positive [Bi-O] layers and negative X-layers.[93,94]

Figure 2.0 (a) Ultraviolet-visible (UV-Vis) diffuse reflectance spectra of BiOX (X =

Cl, Br, and I).[85] (b) Calculated band gap and the band alignment of the BiOX (X = Cl,

Br, I and F) compounds by density functional theory (DFT).[89] (c) Oxygen evolution

of BiOX under simulated sunlight irradiation.[29] (d) Photoanodic activity measurement

of BiOBr film under simulated sunlight irradiation. (e) Stability of the BiOBr film at

an applied voltage of 1.0 V vs. Ag/AgCl.[92]

Adjusting the ratio of Bi:O:X in bismuth oxyhalides has also been demonstrated to

be an effective approach to obtain visible-light photocatalysts with the desired

electronic structure.[76,77,95,96] Shang et al. reported that Bi24O31Br10 (Eg ≈ 2.8 eV,

16

similar to BiOBr) exhibited considerable photocatalytic activity towards Cr(VI) ion

reduction and H2 evolution through water splitting. BiOBr cannot split water to release

H2 through photocatalytic reactions, however, because its conduction band minimum

(CBM) is more positive than the electrode potential of H+/H2. While in the case of

Bi24O31Br10, the CBM, mainly consisting of hybridized Bi 6p and Br 4s orbitals, is

uplifted to be more negative than the electrode potential of H+/H2, which enables

Bi24O31Br10 to reduce water into H2 under visible-light irradiation. In addition, the

hybridization s-p orbitals lead to a dispersive band, which is expected to increase its

efficiency through promoting the separation of the photoexcited electrons and holes.

Surface engineering has successfully been implemented in promoting the

photocatalytic activity of bismuth oxyhalides. Manipulations of exposed facets have

been achieved in a variety of BiOX photocatalysts, in which the tunable exposed facets

can efficiently adjust their photocatalytic activities.[75,97-101] Morphology and thickness

engineering are also efficient approaches for modulating the photocatalytic activities

of bismuth oxyhalides through adjusting their the exposed surfaces, specific surface

area and electronic properties, which are closely relevant to their layered structures.[102-

104]

In addition, OVs have been reported to be particularly efficient in promoting the

photocatalytic activity of ultra-thin bismuth oxyhalides nanosheets. For examples,

enhanced photocatalytic performances in dye degradation by BiOCl,[105-107] and CO2

reduction by BiOBr,[74, 108] through introducing OVs in their ultra-thin nanosheets. It

demonstrates that effect of defect engineering could be more effective for

semiconductor photocatalysts in 2D, which has a very high specific surface area and

more sensitive electronic properties compared with bulk materials.

17

1.3 Outline of chapters

Chapter 2 introduces the experimental instruments used in this thesis. SPM

techniques, comprising AFM and STM, are illustrated in details of their working

principles and applications. Other facilities, including electron microscopies (SEM,

TEM and STEM) and photoelectron spectroscopy (XPS and UPS), which are widely

applied in the study of surface science and technology, are also introduced.

Chapter 3 presents STM study of the evolution of the TiO2 rutile(110) surface with

increasing oxygen deficiency, which was created by annealing in vacuum condition.

At 900 K, the surface exhibit a (1×1) structure with isolated OV point defects. While

at 1300 K, the surface changed to (1×2) reconstruction with cross-links defects

dominating the surface.

Chapter 4 presents the surface study of OV-TiO2(110) surfaces by STM and STEM.

The effect of OVs on the electronic properties were further revealed by photoelectron

spectroscopy. More importantly, the vital roles of OVs on the greatly enhanced

electrocatalytic activity toward HER were revealed under the combination of surface

studies and DFT calculations. The promoted electrical conductivity and hydrogen

desorption capability induced by OVs were suggested to be decisive in the enhanced

electrocatalytic activities.

Chapter 5 presents the study of modification of inner strain in 2D {001} facets

exposed BiOBr nanosheets. The effect of strain on the crystal structure and electronic

structure were revealed by TEM and DFT calculations. It is also demonstrated that the

photocatalytic performance of BiOBr in dye degradation can be manipulated by the

strain effect. The improved charge separation attributable to a highly dispersive band

structure in low-strain BiOBr nanosheets contributed to the promoted photocatalytic

activities.

18

Chapter 6 presents the strain induced 2D lateral pseudoheterogeneous structures in

2D BiOBr nanosheets. AFM, TEM and STEM characterizations revealed that the

pseudoheterogeneous interface without atomic mismatch can be feasibly modulated by

local strain distribution, which exhibits similar local electronic band structure of

corresponding heterostructures. Significant enhancement in charge separation at the

pseudoheterostructure was demonstrated under visible light irradiation of individual

single nanoplates, which is given rise to the controllable electronic band alignment

across the interface. The construction of the lateral pseudoheterostructure offers a

feasible and promising way to build unprecedented 2D systems with exciting

properties.

1.4 References

(1) H. Morkoc, S. Strite, G. B. Gao, M. E. Lin, B. Sverdlov, M. Burns, Large-band-gap

SiC, III-V nitride, and II-VI ZnSe-based semiconductor device technologies, J. Appl.

Phys. 1994, 76, 1363.

(2) A. P. Alivisatos, Perspectives on the physical chemistry of semiconductor

nanocrystals, J. Phys. Chem. 1996, 100, 13226.

(3) X. Hu, G. Li, J. C. Yu, Design, fabrication, and modification of nanostructured

semiconductor materials for environmental and energy applications, Langmuir 2010,

26, 3031.

(4) J. Mannhart, D. G. Schlom, Oxide interfaces-an opportunity for electronics, Science

2010, 327, 1607.

(5) H. Tong, S. Ouyang, Y. Bi, N. Umezawa, M. Oshikiri, J. Ye, Nano-photocatalytic

materials: possibility and challenges, Adv. Mater. 2012, 24, 229.

19

(6) E. G. Seebauer, K. W. Noh, Trends in semiconductor defect engineering at the

nanoscale, Mater. Sci. Eng. R Rep. 2010, 70, 151.

(7) F. E. Osterloh, Inorganic nanostructures for photoelectrochemical and

photocatalytic water splitting, Chem. Soc. Rev. 2013, 42, 2294.

(8) Q. Zhang, E. Uchaker, S. L. Candelaria, G. Cao, Nanomaterials for energy

conversion and storage, Chem. Soc. Rev. 2013, 42, 3127.

(9) Z. W. Seh, J. Kibsgaard, C. F. Dickens, I. Chorkendorff, J. K. Nørskov, T. F.

Jaramillo, Combining theory and experiment in electrocatalysis: insights into materials

design, Science 2017, 355, eaad4998.

(10) X. Zou, Y. Zhang, Noble metal-free hydrogen evolution catalysts for water

splitting, Chem. Soc. Rev. 2015, 44, 5148.

(11) I. Roger, M. A. Shipman, M. D. Symes, Earth-abundant catalysts for

electrochemical and photoelectrochemical water splitting, Nat. Rev. Chem. 2017, 1,

0003.

(12) Y. Yan, B. Y. Xia, B. Zhao, X. Wang, A review on noble-metal-free bifunctional

heterogenous catalysts for overall electrochemical water splitting, J. Mater. Chem. A

2016, 4, 17587.

(13) J. R. McKone, S. C. Marinescu, B. S. Brunschwig, J. R. Winkler, H. B. Gray,

Earth-abundant hydrogen evolution electrocatalysts, Chem. Sci. 2014, 5, 865.

(14) M. S. Faber, S. Jin, Earth-abundant inorganic electrocatalysts and their

nanostructures for energy conversion applications, Energy Environ. Sci. 2014, 7, 3519.

(15) P. Du, R. Eisenberg, Catalysts made of earth-abundant elements (Co, Ni, Fe) for

water splitting: recent progress and future challenges, Energy Environ. Sci. 2012, 5,

6012.

20

(16) M. Wang, L. Chen, L. Sun, Recent progress in electrochemical hydrogen

production with earth-abundant metal complexes as catalysts, Energy Environ. Sci.

2012, 5, 6763.

(17) N. Danilovic, R. Subbaraman, D. Strmcnik, K.-C. Chang, A. P. Paulikas,

V. R. Stamenkovic, N. M. Markovic, Enhancing the alkaline hydrogen evolution

reaction activity through the bifunctionality of Ni(OH)2/metal catalysts, Angew. Chem.

2012, 124, 12663.

(18) L. Trotochaud, J. K. Ranney, K. N. Williams, S. W. Boettcher, Solution-cast metal

oxide thin film electrocatalysts for oxygen evolution, J. Am. Chem. Soc. 2012, 134,

17253.

(19) B. Zhang, X. Zheng, O. Voznyy, R. Comin, M. Bajdich, M. García-Melchor, L.

Han, J. Xu, M. Liu, L. Zheng, F. P. G. de, Arquer, C. T. Dinh, F. Fan, M. Yuan, E.

Yassitepe, N. Chen, T. Regier, P. Liu, Y. Li, P. D. Luna, A. Janmohamed, H. L. Xin,

H. Yang, A. Vojvodic, E. H. Sargent, Homogeneously dispersed multimetal oxygen-

evolving catalysts, Science 2016, 352, 333.

(20) Z. Pei, Z. Tang, Z. Liu, Y. Huang, Y. Wang, H. Li, Q. Xue, M. Zhu, D. Tang, C.

Zhi, Construction of a hierarchical 3D Co/N-carbon electrocatalyst for efficient oxygen

reduction and overall water splitting, J. Mater. Chem. A 2018, 6, 489.

(21) J. R. Swierk, T. D. Tilley, Electrocatalytic water oxidation by single site and small

nuclearity clusters of cobalt, J. Electrochem. Soc. 2018, 165, H3028.

(22) H. Wang, H.-W. Lee, Y. Deng, Z. Lu, P.-C. Hsu, Y. Liu, D. Lin, Y. Cui,

Bifunctional non-noble metal oxide nanoparticle electrocatalysts through lithium-

induced conversion for overall water splitting, Nat. Commun. 2015, 6, 7261.

(23) R. Subbaraman, D. Tripkovic, K.-C. Chang, D. Strmcnik, A. P. Paulikas,

21

P. Hirunsit, M. Chan, J. Greeley, V. Stamenkovic, N. M. Markovic, Trends in activity

for the water electrolyser reactions on 3d M(Ni,Co,Fe,Mn) hydr(oxy)oxide catalysts,

Nat. Mater. 2012, 11, 550.

(24) M. Gong, W. Zhou, M.-C. Tsai, J. Zhou, M. Guan, M.-C. Lin, B. Zhang, Y. Hu,

D. Y. Wang, J. Yang, S. J. Pennycook, B.-J. Hwang, H. Dai, Nanoscale nickel

oxide/nickel heterostructures for active hydrogen evolution electrocatalysis, Nat.

Commun. 2014, 5, 4695.

(25) Z. Weng, W. Liu, L.-C. Yin, R. Fang, M. Li, E. I. Altman, Q. Fan, F. Li, H.-M.

Cheng, H. Wang, Metal/oxide interface nanostructures generated by surface

segregation for electrocatalysis, Nano Lett. 2015, 15, 7704.

(26) X. Yan, L. Tian, M. He, X. Chen, Three-dimensional crystalline/amorphous

Co/Co3O4 core/shell nanosheets as efficient electrocatalysts for the hydrogen evolution

reaction, Nano Lett. 2015, 15, 6015.

(27) T. Ling, D.-Y. Yan, H. Wang, Y. Jiao, Z. Hu, Y. Zheng, L. Zheng, J. Miao, H.

Liu, X.-W. Du, M. Jar oniec, S.-Z. Qiao, Activating cobalt(II) oxide nanorods for

efficient electrocatalysis by strain engineering, Nat. Commun. 2017, 8, 1509.

(28) J.-X. Feng, H. Xu, Y.-T. Dong, X.-F. Lu, Y.-X. Tong, G.-R. Li, Efficient hydrogen

evolution electrocatalysis using cobalt nanotubes decorated with titanium dioxide

nanodots, Angew. Chem. Int. Ed. 2017, 56, 2960.

(29) Y. P. Zhu, T. Y. Ma, M. Jaroniec, S. Z. Qiao, Self-templating synthesis of hollow

Co3O4 microtube arrays for highly efficient water electrolysis, Angew. Chem. Int. Ed.

2017, 56, 1324.

(30) Y. Jin, H. Wang, J. Li, X. Yue, Y. Han, P. K. Shen, Y. Cui, Porous MoO2

nanosheets as non-noble bifunctional electrocatalysts for overall water splitting, Adv.

Mater. 2016, 28, 3785.

22

(31) Z. Fang, L. Peng, Y. Qian, X. Zhang, Y. Xie, J. J. Cha, G. Yu, Dual tuning of Ni-

Co-A (A = P, Se, O) nanosheets by anion substitution and holey engineering for

efficient hydrogen evolution, J. Am. Chem. Soc. 2018, 140, 5241.

(32) U. Diebold, The surface science of titanium dioxide, Surf. Sci. Rep. 2003, 48, 53.

(33) O. Stetsovych, M. Todorovicć, T. K. Shimizu, C. Moreno, J. W. Ryan, C. P. León,

K. Sagisaka, E. Palomares, V. Matolín, D. Fujita, R. Perez, O. Custance, Atomic

species identification at the (101) anatase surface by simultaneous scanning tunnelling

and atomic force microscopy, Nat. Commun. 2015, 6, 7265.

(34) Y. Wang, H. Sun, S. Tian, H. Feng, Z. Cheng, J. Zhao, A. Zhao, B. Wang, Y. Luo,

J. Yang, J. G. Hou, Role of point defects on the reactivity of reconstructed anatase

titanium dioxide (001) surface, Nat. Commun. 2013, 4, 2214.

(35) M. Setvin, X. Shi, J. Hulva, T. Simschitz, G. S. Parkinson, M. Schmid, C. D.

Valentin, A. Sellon, U. Diebold, Methanol on anatase TiO2 (101): mechanistic insights

into photocatalysis, ACS Catal. 2017, 7, 7081.

(36) J. Balajka, M. A. Hines, W. J. I. DeBenedetti, M. Komora, J. Pavelec, M. Schmid,

U. Diebold, High-affinity adsorption leads to molecularly ordered interfaces on TiO2

in air and solution, Science, 2018, 361, 786.

(37) C. M. Yim, J. Chen, Y. Zhang, B.-J. Shaw, C. L. Pang, D. C. Grinter, H. Bluhm,

M. Salmeron, C. A. Muryn, A. Michaelides, G. Thornton, Visualization of water-

induced surface segregation of polarons on rutile TiO2(110), J. Phys. Chem. Lett. 2018,

9, 4865.

(38) Q. Guo, C.-Y. Zhou, Z.-B. Ma, Z. F. Ren, H.-J. Fan, X.-M. Yang, Fundamental

processes in surface photocatalysis on TiO2, Acta Phys.-Chim. Sin. 2016, 32, 28.

(39) M. Ramamoorthy, D. Vanderbilt, R. D. King-Smith, First-principles calculations

of the energetics of stoichiometric TiO2 surfaces, Phys. Rev. B 1994, 49, 16721.

23

(40) P. W. Murray, N. G. Condon, G. Thornton, Effect of stoichiometry on the structure

of TiO2(110), Phys. Rev. B 1995, 51, 10989.

(41) G. Charlton, P. B. Howes, C. L. Nicklin, P. Steadman, J. S. G. Taylor, C. A.

Muryn, S. P. Harte, J. Mercer, R. McGrath, D. Norman, T. S. Turner, G. Thornton,

Relaxation of TiO2(110)-(1×1) using surface X-ray diffraction, Phys. Rev. Lett. 1997,

78, 495.

(42) V. Swamy, J. Muscat, J. D. Gale, N. M. Harrison, Surf. Sci. 2002, 504, 115.

(43) R. Lindsay, A. Wander, A. Ernst, B. Montanari, G. Thornton, N. M. Harrison,

Phys. Rev. Lett. 2005, 94, 246102.

(44) S. Wendt, R. Schaub, J. Matthiesen, E.K. Vestergaard, E. Wahlström, M. D.

Rasmussen, P. Thostrup, L.M. Molina, E. Lægsgaard, I. Stensgaard, B. Hammer, F.

Besenbacher, Oxygen vacancies on TiO2(110) and their interaction with H2O and O2:

a combined high-resolution STM and DFT study, Surf. Sci. 2005, 598, 226.

(45) X. Cui, Z. Wang, S. Tan, B. Wang, J. Yang, J. G. Hou, Identifying hydroxyls on

the TiO2 (110)-1×1 surface with scanning tunneling microscopy, J. Phys. Chem. C

2009, 113, 13204.

(46) A. C. Papageorgiou, N. S. Beglitis, C. L. Pang, G. Teobaldi, G. Cabailh, Q. Chen,

A. J. Fisher, W. A. Hofer, G. Thornton, Electron traps and their effect on the surface

chemistry of TiO2 (110), Proc. Natl. Acad. Sci. USA 2010, 107, 2391.

(47) T. Minato, Y. Sainoo, Y. Kim, H. S. Kato, K. Aika, M. Kawai, J. Zhao, H. Petek,

T. Huang, W. He, B. Wang, Z. Wang, Y. Zhao, J. Yang, J. G. Hou. The electronic

structure of oxygen atom vacancy and hydroxyl impurity defects on titanium dioxide

(110) surface, J. Chem. Phys. 2009, 130, 124502.

24

(48) M. Setvin, C. Franchini, X. Hao, M. Schmid, A. Janotti, M. Kaltak, C. G. V. de

Walle, G. Kresse, U. Diebold, Direct view at excess electrons in TiO2 rutile and

anatase, Phys. Rev. Lett. 2014, 113, 086402.

(49) A. Fujishima, A. Honda, Electrochemical photolysis of water at a semiconductor

electrode, Nature 1972, 238, 37.

(50) S. Wendt, J. Matthiesen, R. Schaub, E. K. Vestergaard, E. Lægsgaard, F.

Besenbacher, B. Hammer, Formation and splitting of paired hydroxyl groups on

reduced TiO2(110), Phys. Rev. Lett. 2006, 96, 066107.

(51) O. Bikondoa, C. L. Pang, R. Ithnin, C. A. Muryn, H. Onishi, G. Thornton, Direct

visualization of defect-mediated dissociation of water on TiO2(110), Nat. Mater. 2006,

5, 189.

(52) D. P. Acharya, C. V. Ciobanu, N. Camillone III, P. Sutter, Mechanism of electron-

induced hydrogen desorption from hydroxylated rutile TiO2(110), J. Phys. Chem. C

2010, 114, 21510.

(53) C. L. Pang, O. Bikondoa, D. S. Humphrey, A. C. Papageorgiou, G. Cabailh, R.

Ithnin, Q. Chen, C. A. Muryn, H. Onishi, G. Thornton, Tailored TiO2(110) surfaces

and their reactivity, Nanotechnology 2006, 17, 5397.

(54) Z. Zhang, O. Bondarchuk, B. D. Kay, J. M. White, Z. Dohnàlek, Imaging water

dissociation on TiO2(110): evidence for inequivalent geminate OH groups, J. Phys.

Chem. B 2006, 110, 21840.

(55) S.-C. Li, Z. Zhang, D. Sheppard, B. D. Kay, J. M. White, Y. Du, I. Lyubinetsky,

G. Henkelman, Z. Dohnàlek, Intrinsic diffusion of hydrogen on rutile TiO2(110), J.

Am. Chem. Soc. 2008, 130, 9080.

25

(56) Z.-T. Wang, Y.-G. Wang, R. Mu, Y. Yoon, A. Dahal, G. K. Schenter, V.-A.

Glezakou, R. Rousseau, I. Lyubinetsky, Z. Dohnálek, Probing equilibrium of molecular

and deprotonated water on TiO2(110), Proc. Natl. Acad. Sci. USA 2017, 114, 1801.

(57) Y. Du, N. A. Deskins, Z. Zhang, Z. Dohnàlek, M. Dupuis, I. Lyubinetsky, Two

pathways for water interaction with oxygen adatoms on TiO2(110), Phys. Rev. Lett.

2009, 102, 096102.

(58) D. O. Scanlon, C. W. Dunnill, J. Buckeridge, S. A. Shevlin, A. J. Logsdail, S. M.

Woodley, C. R. A. Catlow, M. J. Powell, R. G. Palgrave, I. P. Parkin, G. W. Watson,

T. W. Keal, P. Sherwood, A. Walsh, A. A. Sokol, Band alignment of rutile and anatase

TiO2, Nat. Mater. 2013, 12, 798.

(59) S. Selcuk, A. Selloni, Facet-dependent trapping and dynamics of excess electrons

at anatase TiO2 surfaces and aqueous interfaces, Nat. Mater. 2016, 15, 1107.

(60) T. Ohno, K. Sarukawa, K. Tokieda, M. Matsumura, Morphology of a TiO2

photocatalyst (Degussa, P-25) consisting of anatase and rutile crystalline phases, J.

Catal. 2001, 203, 82.

(61) D. C. Hurum, A. G. Agrios, K. A. Gray, T. Rajh, M. C. Thurnauer, Explaining the

enhanced photocatalytic activity of degussa P25 mixed-phase TiO2 Using EPR, J.

Phys. Chem. B 2003, 107, 4545.

(62) Y. He, A. Tilocca, O. Dulub, A. Selloni, U. Diebold, Local ordering and electronic

signatures of submonolayer water on anatase TiO2(101), Nat. Mater. 2009, 8, 585.

(63) M. Setvín, U. Aschauer, P. Scheiber, Y.-F. Li, W. Hou, M. Schmid, A. Selloni, U.

Diebold, Reaction of O2 with subsurface oxygen vacancies on TiO2 anatase (101),

Science 2013, 341, 988.

26

(64) M. Setvín, X. Hao, B. Daniel, J. Pavelec, Z. Novotny, G. S. Parkinson, M. Schmid,

G. Kresse, C. Franchini, U. Diebold, Charge trapping at the step edges of TiO2 anatase

(101), Angew. Chem. Int. Ed. 2014, 53, 4714.

(65) M. Setvín, J. Hulva, G. S. Parkinson, M. Schmid, U. Diebold, Electron transfer

between anatase TiO2 and an O2 molecule directly observed by atomic force

microscopy, Proc. Natl. Acad. Sci. USA 2017, 114, E2556.

(66) M. Setvín, U. Aschauer, J. Hulva, T Simschitz, B. Daniel, M. Schmid, A. Selloni,

U. Diebold, Following the reduction of oxygen on TiO2 anatase (101) step by step, J.

Am. Chem. Soc. 2016, 138, 9565.

(67) Y. Wang, H. Sun, S. Tan, H. Feng, Z. Cheng, J. Zhao, A. Zhao, B. Wang, Y. Luo,

J. Yang, J. G. Hou, Role of point defects on the reactivity of reconstructed anatase

titanium dioxide (001) surface, Nat. Commun. 2013, 4. 2214.

(68) C. Dette, M. A. Perez-Osorio, S. Mangel, F. Giustino, S. J. Jung, K. Kern, Atomic

structure of water monolayer on anatase TiO2(101) surface, J. Phys. Chem. C 2018,

122, 11954.

(69) Q. Wang, T. Hisatomi, Q. Jia, H. Tokudome, M. Zhong, C. Wang, Z. Pan, T.

Takata, M. Nakabayashi, N. Shibata, Y. Li, I. D. Sharp, A. Kudo, T. Yamada, K.

Domen, Scalable water splitting on particulate photocatalyst sheets with a solar-to-

hydrogen energy conversion efficiency exceeding 1%, Nat. Mater. 2016, 15, 611.

(70) J. Liu, M. S. Kelley, W. Wu, A. Banerjee, A. P. Douvalis, J. Wu, Y. Zhang, G. C.

Schatz, M. G. Kanatzidis, Nitrogenase-mimic iron-containing chalcogels for

photochemical reduction of dinitrogen to ammonia, Proc. Natl. Acad. Sci. USA 2016,

113, 5530.

27

(71) H. Park, H.-H. Ou, U. Kang, J. Choi, M. R. Hoffmann, Photocatalytic conversion

of carbon dioxide to methane on TiO2/CdS in aqueous isopropanol solution, Catal.

Today 2016, 266, 153.

(72) Y. Li, S. Ouyang, H. Xu, X. Wang, Y. Bi, Y. Zhang, J. Ye, Constructing solid-

gas-interfacial fenton reaction over alkalinized-C3N4 photocatalyst to achieve apparent

quantum yield of 49% at 420 nm, J. Am. Chem. Soc. 2016, 138, 13289.

(73) S. Bai, W. Jiang, Z. Li, Y. Xiong, Surface and interface engineering in

photocatalysis, Chem. Nano Mat. 2015, 1, 223.

(74) Y. Sun, J. Wu, X. Li, W. Shi, P. Ling, X. Jiao, S. Gao, L. Liang, J. Xu, W. Yan,

C. Wang, Y. Xie, Efficient visible-light-driven CO2 reduction realized by defect-

mediated BiOBr atom layers, Angew. Chem. Int. Ed. 2018, 57, 1.

(75) H. Li, J. Shang, Z. Ai, L. Zhang, Efficient visible light nitrogen fixation with

BiOBr nanosheets of oxygen vacancies on the exposed {001} facets, J. Am. Chem. Soc.

2015, 137, 6393.

(76) S. Wang, X. Hai, X. Ding, K. Chang, Y. Xiang, X. Meng, Z. Yang, H. Chen, J.

Ye, Light-switchable oxygen vacancies in ultrafine Bi5O7Br nanotues for boosting

solar-driven nitrogen fixation in pure water, Adv. Mater. 2017, 29, 1701774.

(77) J. Shang, W. Hao, X. Lv, T. Wang, X. Wang, Y. Du, S. X. Dou, T. Xie. D. Wang,

J. Wang, Bismuth oxybromide with reasonable photocatalytic reduction activity under

visible light, ACS Catal. 2014, 4, 954.

(78) J. Di, J. Xia, M. Ji, S. Yin, H. Li, H. Xu, Q. Zhang, H. Li, Controllable synthesis

of Bi4O5Br2 ultrathin nanosheets for photocatalytic removal of ciprofloxacin and

mechanism insight, J. Mater. Chem. A 2015, 3, 15108.

(79) W. W. Lee, C.-S. Lu, C.-W. Chuang, Y.-J. Chen, J.-Y. Fu, C.-W. Siao, C.-C.

28

Chen, Synthesis of bismuth oxyiodides and their composites: characterization,

photocatalytic activity, and degradation mechanisms, RSC Adv. 2015, 5, 23450.

(80) Z.-Y. Zhao, W.-W. Dai, Electronic structure and optical properties of BiOI

ultrathin films for photocatalytic water splitting, Inorg. Chem. 2015, 54, 10732.

(81) J. Han, G. Zhu, M. Hojamberdiev, J. Peng, X. Zhang, Y. Liu, B. Ge, P. Liu, Rapid

adsorption and photocatalytic activity for Rhodamine B and Cr(VI) by ultrathin BiOI

nanosheets with highly exposed {001} facets, New J. Chem. 2015, 39, 1874.

(82) C.-Y. Wang, X. Zhang, X.-N. Song, W.-K. Wang, H.-Q. Yu, Novel Bi12O15Cl6

photocatalyst for the degradation of bisphenol A under visible-light irradiation, ACS

Appl. Mater. Interfaces 2016, 8, 5320.

(83) Z. H. Ai, W. K. Ho, S. C. Lee, L. Z. Zhang, Efficient photocatalytic removal of

NO in indoor air with hierarchical bismuth oxybromide nanoplate microspheres under

visible light, Environ. Sci. Technol. 2009, 43, 4143.

(84) L. Wang, J. Shang, W. Hao, S. Jiang, S. Huang, T. Wang, Z. Sun, Y. Du, S. X.

Dou, T. Xie, D. Wang, J. Wang, A dye-sensitized visible light photocatalyst

Bi24O31Cl10, Sci. Rep. 2014, 4, 7384.

(85) X. Liu, Y. Su, Q. Zhao, C. Du, Z. Liu, Constructing Bi24O31Cl10/BiOCl

heterojunction via a simple thermal annealing route for achieving enhanced

photocatalytic activity and selectivity, Sci. Rep. 2016, 6, 28689.

(86) S. Jiang, L. Wang, W. Hao, W. Li, H. Xin, W. Wang, T. Wang, Visible-light

photocatalytic activity of S-doped α-Bi2O3, J. Phys. Chem. C 2015, 119, 14094.

(87) Z. Xu, W. Hao, Q. Zhang, Z. Fu, H. Feng, Y. Du, S. X. Dou, Indirect-direct band

transformation of few-layer BiOCl under biaxial strain, J. Phys. Chem. C 2016, 120

8589.

29

(88) H. An, Y. Du, T. Wang, C. Wang, W. Hao, J. Zhang, Photocatalytic properties of

BiOX (X = Cl, Br, and I), Rare Metals 2008, 27, 243.

(89) A. M. Ganose, M. Cuff, K. T. Butler, A. Walsh, D. O. Scanlon, Interplay of orbital

and relativistic effects in bismuth oxyhalides: BiOF, BiOCl, BiOBr, and BiOI, Chem.

Mater. 2016, 28, 1980.

(90) X. Zhang, Z. Ai, F. Jia, L. Zhang, Generalized one-pot synthesis, characterization,

and photocatalytic activity of hierarchical BiOX (X = Cl, Br, I) nanoplate

microspheres, J. Phys. Chem. C 2008, 112, 747.

(91) H. Huang, X. Han, X. Li, S. Wang, P. K. Chu, Y. Zhang, Fabrication of multiple

heterojunctions with tunable visible light-active photocatalytic reactivity in BiOBr-

BiOI full-range composites based on microstructure modulation and band structures,

ACS Appl. Mater. Interfaces 2015, 7, 482.

(92) D. S. Bhachu, S. J. A. Moniz, S. Sathasivam, D. O. Scanlon, A. Walsh, S. M.

Bawaked, M. Mokhtar, A. Y. Obaid, I. P. Parkin, J. Tang, C. J. Carmalt, Bismuth

oxyhalides: synthesis, structure and photoelectrochemical activity, Chem. Sci. 2016, 7,

4832.

(93) J. Li, H. Li, G. Zhan, L. Zhang, Solar water splitting and nitrogen fixation with

layered bismuth oxyhalides, Acc. Chem. Res. 2017, 50, 112.

(94) J. Li, G. Zhan, Y. Yu, L. Zhang, Superior visible light hydrogen evolution of Janus

bilayer junctions via atomic-level charge flow steering, Nat. Commun. 2016, 7, 11480.

(95) Y. Bai, T. Chen, P. Wang, L. Wang, L. Ye, Bismuth-rich Bi4O5X2 (X = Br, and I)

nanosheets with dominant {101} facets exposure for photocatalytic H2 evolution,

Chem. Eng. J. 2016, 304, 454.

30

(96) R. He, S. Cao, J. Yu, Y. Yang, Microwave-assisted solvothermal synthesis of

Bi4O5I2 hierarchical architectures with high photocatalytic performance, Catal. Today

2016, 264, 221.

(97) J. Jiang, K. Zhao, X. Xiao, X. L. Zhang, Synthesis and facet-dependent

photoreactivity of BiOCl single-crystalline nanosheets. J. Am. Chem. Soc. 2012, 134,

4473.

(98) H. Zhang, Y. Yang, Z. Zhou, Y. Zhao, L. Liu, Enhanced photocatalytic properties

in BiOBr nanosheets with dominantly exposed (102) facets, J. Phys. Chem. C 2014,

118, 14662.

(99) Y. Xu, S. Xu, S. Wang, Y. Zhang, G. Li, Citric and modulated electrochemical

synthesis and photocatalytic behavior of BiOCl nanoplates with exposed {001} facets,

Dalton Trans. 2014, 43, 479.

(100) X. Xiong, L. Ding, Q. Wang, Y. Li, Q. Jiang, J. Hu, Synthesis and photocatalytic

activity of BiOBr nanosheets with tunable exposed {010} facets, Appl. Catal. B 2016,

188, 283.

(101) M. Pan, H. Zhang, G. Gao, L. Liu, W. Chen, Facet-dependent catalytic activity

of nanosheet-assembled bismuth oxyiodide microspheres in degradation of bisphenol

A, Environ. Sci. Technol. 2015, 49, 6240.

(102) J.-M. Song, C.-J. Mao, H.-L. Niu, Y.-H. Shen, S.-Y. Zang, Hierarchical

structured bismuth oxychlorides: self-assembly from nanoplates to nanoflowers via a

solvothermal route and their photocatalytic properties, CrystEngComm 2010, 12, 3875.

(103) L. Chen, R. Huang, M. Xiong, Q. Yuan, J. He, J. Ma, M.-Y. Yao, S.-L. Luo, C.-

T. Au, S.-F. Yin, Room-temperature synthesis of flower-like BiOX (X=Cl, Br, I)

hierarchical structures and their visible-light photocatalytic activity, Inorg. Chem.

2013, 52, 11118.

31

(104) X. Zhang, X.-B. Wang, L.-W. Wang, W.-K. Wang, L.-L. Long, H.-Q. Yu,

Synthesis of a highly efficient BiOCl single-crystal nanodisk photocatalyst with

exposing {001} facets, ACS Appl. Mater. Interfaces 2014, 6, 7766.

(105) H. Li, T. Hu, J. Liu, S. Song, N. Du, R. Zhang, W. Hou, Thickness-dependent

photocatalytic activity of bismuth oxybromide nanosheets with highly exposed (010)

facets, Appl. Catal. B 2016, 182, 431.

(106) M. Guan, C. Xiao, J. Zhang, S. Fan, R. An, Q. Cheng, J. Xie, M. Zhou, B. Ye,

Y. Xie, Vacancy associates promoting solar-driven photocatalytic activity of ultrathin

bismuth oxychloride nanosheets, J. Am. Chem. Soc. 2013, 135, 10411.

(107) D. Cui, L. Wang, K. Xu, L. Ren, L. Wang, Y. Yu, Y. Du, W. Hao, Band-gap

engineering of BiOCl with oxygen vacancies for efficient photooxidation properties

under visible-light irradiation, J. Mater. Chem. A 2018, 6, 2193.

(108) X. Y. Kong, B.-J. Ng, K. H. Tan, X. Chen, H. Wang, A. R. Mohamed, S.-P. Chai,

Simultaneous generation of oxygen vacancies on ultrathin BiOBr nanosheets during

visible-light-driven CO2 photoreduction evoked superior activity and long-term

stability, Catal. Today 2018, 314, 20.

32

Chapter 2 Experimental Techniques

2.1 Scanning tunneling microscopy

STM was firstly demonstrated in 1982 by Gerd Binning and Heinrich Rohrer, who

were later awarded the Nobel Prize in physics in1986 for their design of STM.[1,2] STM

can achieve atomic resolution on the conductive surface of semiconductors and metals,

and is able to acquire the local density of states (LDOS) as function of position on the

surface. In addition, STM has also demonstrate its strong capability of manipulating

and transferring atoms on the surface. As a consequence, STM is applied in a wide

range of research and application fields, including physical, chemical, biological,

material, and nano science and technology.

2.1.1 Quantum tunneling

The quantum tunneling effect is a quantum phenomenon that has no counterpart in

classical physics. A particle has a probability to tunnel cross a barrier, even though the

energy of the particle is less than the barrier height. These barriers can either be

physically impassable medium, including insulators or vacuum, or a region of high

potential energy.

33

Figure 2.1 Quantum mechanical tunneling through a potential barrier of height (V) and

width (D).

The quantum tunneling probability of a particle can be determined through solving

Schrödinger equation. As shown in Figure 2.1, the incident particle from the left of the

energy barrier (width D) has a free particle wavefunction of an energy smaller than the

potential barrier of height (V). When the incident particle approaches the barrier,

The Schrodinger equation can be described as:

−ħ2

2𝑚 𝜕2Ψ(x)

𝜕x2 = (𝐸 − 𝑉)Ψ(x) (2.1)

The solution is:

Ψ = A𝑒−𝛼𝑥 (2.2)

where the decay constant α: 𝛼 = √2𝑚 (E−V)

ħ2 (2.3)

Therefore, the decay constant depends not only on the energy barrier but also the

mass (m) of the particles and the reduced Planck constant (ħ).

34

2.1.2 Working principle of STM

STM is designed based on the principle of quantum tunneling of electrons between

a sharp metallic tip and the conductive surface, when they have a close distance within

several angstroms. Meanwhile, in order to achieve the scanning of the surface, a 3-

dimisonial piezoelectric scanner with the sub-angstrom precision in both x-y plane and

z direction are need. When a voltage between the tip and the sample is given, electrons

can tunnel between them. For example, if a positive voltage is applied on the sample,

electrons will tunnel from the filled state of the tip to the empty states of the sample

when they are in the tunneling region, and vice versa. The tunneling current can be

calculated using the time-dependent perturbation theory. If a positive V is applied to

the sample, the Fermi level of the sample shifts down with respect to the Fermi level

of the tip, and electrons tunnel from the occupied states of the tip into the empty states

of the sample. The net electric current, which is in the range of picoamperes to

nanoamperes, then can be detected. The intensity of the tunneling current is

proportional to the overlap between the wavefunctions of tip and sample, which

depends strongly on their distance. A typical setup of STM system is illustrated in

Figure 2.2.

Figure 2.2 Diagram of the setup of STM system.

35

The total tunneling current can be estimated by the perturbation theory[3-5]:

𝐼(𝑉) ∝ ∫ 𝜌𝑠 (𝐸)𝜌𝑡(𝐸 − 𝑒𝑉)|𝑀𝑡,𝑠(𝐸, 𝑉, 𝑧)|2

(𝑓(𝐸 − 𝑒𝑉, 𝑇) − 𝑓(𝐸, 𝑇))𝑑𝐸 (2.4)

where ρs is the DOS of sample, ρt is the DOS of tip, E is the energy of electrons, T

is temperature, f is Fermi function, V is the applied voltage, 𝑀𝑡,𝑠(𝐸, 𝑉, 𝑧) is the tunnel

matrix.

According to Tersoff-Haman approach, ρs and ρt can be treated as constant under

small bias voltage. Considering in STM studies, low temperature and metallic tips

(such as W or PtIr) with flat DOS around Fermi level are applied, the expression for

the tunneling current can be simplified as:

𝐼(𝑉) ∝ ∫ 𝜌𝑠 (𝐸)𝑑𝐸 (2.5)

Therefore, the tunneling current under certain bias V is proportional to the integral

of the DOS of the sample from the Fermi level to eV.

For the widely adopted constant-current mode, a feedback loop is used to regulate

the distance between STM tip and the sample in the z-direction to maintain the set-

point current. Then the change of the z position of the tip can be obtained to produce

the surface topography.

Meanwhile, tunneling current is dependent on distance between STM tip and sample,

and the integral of the DOS from Fermi level to eV. For a sample with a homogenous

DOS, the topography is entirely contributed by the geometric surface profiles.

Whereas, considering that most materials exhibit a spatially inhomogeneous DOS, the

obtained STM images are dominated by both the geometric surface profile and the local

DOS.

As the measured tunneling current is proportional to the integral of the DOS of the

sample from the Fermi level to eV, STM can also acquire the local DOS (LDOS) of a

36

selected energy range on the surface by fixing distance between STM tip and the

sample. This means the LDOS can be directly measured by sweeping the applied

voltage at a fixed position on the surface, which can be explained by Equation 1.6

(Derived from Equation 2.5). Instead of numerical calculation of the obtained tunneling

current and the applied voltage that is easily be affected by the noise, lock-in amplifier

technique is widely employed to directly record the dI/dV signal, which works through

applying a small bias voltage modulation dV to the applied V and record the change in

the tunneling current dI.

𝑑𝐼

𝑑𝑉∝ 𝜌𝑠(𝑒𝑉) (2.6)

2.1.3 Ultra-high vacuum system and molecular beam epitaxy

In the LT-STM and MBE combination system, UHV condition is necessary for

acquiring highly clean sample surface, and high resolution and stable images. To

achieve UHV (less than 1× 10-10 Torr) in our system, a serious of supporting structures

and vacuum pumps are used, as illustrated in the setup diagram in Figure 2.3.

Figure 2.3 The relative schematic vacuum layout of our STM/MBE system.

A load-lock is designed for loading and transferring samples and STM tips into the

UHV preparation chamber without venting the chamber to atmosphere. The vacuum of

37

load-lock chamber is achieved by to 10-9 Torr. Load-lock chamber is separated from

preparation chamber and observation chamber by a gate value (GV). To obtain the a

UHV condition for preparation chamber and observation chamber, the system needs to

be firstly pumped by a combination of rotary pump and turbo molecular pumps (TMP)

to the vacuum level around 1 × 10-9 Torr after baking (generally 100 oC to 250 oC).

Then under the pumping of ion pump and titanium sublimation pump (TSP), the system

can reach an UHV condition of 5 × 10-11 Torr.

Figure 2.4 The relative schematic of preparation chamber with MBE system.

Molecular beam epitaxy (MBE) is an advanced UHV epitaxy method with high

precision of sub monolayer and purity. The key aspect of MBE is the slow deposition

rate, which enable the precisely control of the amount of dopants or thickness of crystal

film. In the UHV environment, atom or molecular beams are thermally evaporated and

deposit on a heated substrate placed in the line of sight of the beam line, as shown in

38

the relative schematic of our preparation chamber. The widely adopted evaporation

source in commercial MBE is Knudsen Effusion Cell (K-cells), which consists

mechanical shutters, boron nitride or aluminum oxide crucible, tungsten filament

heater, tantalum heat shielding, thermocouple, and cooling water system. Other

facilities, for examples the heating stage and the argon sputtering gun, are also installed

in the preparation chamber for both sample deposition, treatment and cleaning.

2.2 Atomic force microscopy

Atomic force microscopy (AFM), which is another important SPM technique

beyond STM, was developed by Gerd Binning et al. in 1986.[6,7] Different with STM

which was designed based on the quantum tunneling effect between STM tip and

sample surface, AFM works through measuring the force interaction between AFM tip

and sample surface. Thus, the application scopes of AFM are not limited for conductive

samples, but also can be extended to almost any measurable force interaction, including

van der Waals, electrical, chemical bonding, magnetic, thermal, and capillary forces,

which enable AFM to be the most versatile SPM technique.[8] Thus, combining the

high-resolution for 3D topography, capability of operating in ambient conditions and

liquid conditions, and integrating with variety of optical microscopy and spectroscopy

techniques, AFM has been widely employed in measuring surface topography and

detecting force interaction in many research fields, such as material science, condensed

matter physics, nanoscience and nanotechnology, chemistry, biology, and medicine.

2.2.1 Working principles of AFM

AFM, typically, mainly consists of a sharp tip fixed on a cantilever, laser beam and

optical system, laser detector and feedback system, display and processing system, and

piezo-based scanning stage, as shown in Figure 2.5.

39

Figure 2.5 Diagram of the setup of a basic AFM in laser beam deflection system.

𝐹𝑡𝑠 = −𝜕𝑉𝑡𝑠

𝜕𝑧 (2.7)

where Fts is the tip-sample force and Vts is the potential.

When a sharp AFM tip approaches close a sample surface, a force will arise between

AFM tip and the sample due to the potential energy, which is illustrated in Equation

2.7. Then the force lead to a deflection of the cantilever according to Hooke's law, with

the spring constant kts determined as:

𝑘𝑡𝑠 = −𝜕𝐹𝑡𝑠

𝜕𝑧 (2.8)

Fts can be divided into long-range and short-range forces. For the ambient AFM

system employed in this thesis, long-range force typically represents the attractive

forces, including Van der Waals force, magnetic force, capillary force, and electrostatic

force. While, short-range force can be measured in vacuum conditions when the tip-

sample distance can reach less than several angstroms), chemical force, covalent force,

electrostatic, magnetic and Van der Waals forces can be determined.

2.2.2 Working modes of AFM

There are basically three operation modes of AFM classified by the way of tip

motion, including contact mode, tapping mode and non-contact mode. These different

40

working modes can be adopted in different interaction forces, and corresponding

applications. Figure 2.6 illustrates the relationships between AFM working modes and

the force ranges. Normally, contact mode and tapping mode are used in ambient

environment, while non-contact mode is mostly used in vacuum. Therefore, only

contact mode and tapping mode are introduced in this thesis.

Figure 2.6 Classification of AFM working modes based on Van der Walls force curve.

Contact mode-AFM

In contact mode, the AFM tip directly contact with the sample surface under

repulsive force. The constant force mode is mostly adopted for most applications.

During the scanning, the deflection of the cantilever is kept constant, which enables the

force to be constant. The constant deflection of the cantilever is controlled by an active

feedback loop that can adjust the tip and sample surface distance corresponding to the

topography. For constant height mode, the spatial variation of the cantilever deflection,

in contrast, is directly recorded to yield the topography of sample surface. Constant

41

height mode is mostly employed only for acquiring atomic resolution of sample

surface. In addition, based on contact mode AFM, a serious of functions have been

developed, for examples, force spectroscopy and mapping, lithography and

nanomanipulation, conductive AFM, and so on.

Tapping mode-AFM

In tapping mode, the AFM tip oscillates above sample surface under a driving

frequency very close to the resonance frequency of the cantilever. During the scanning,

the oscillation amplitude of the cantilever is kept constant through the feedback loop,

when the tip periodically contacts with sample surface. As the force between the tip

and sample surface changes, the feedback loop will adjust the height to maintain the

set oscillation amplitude of the cantilever, which reflects the force distribution or the

topography of the sample surface. In tapping mode, as the tip gently interact with

sample surface, both the tip and sample are less destructive compared with contact

mode. In addition, in tapping mode, the phase lag of the cantilever oscillation relative

to the signal sent to the driving piezo can be simultaneously recorded. The phase

channel reflects the energy dissipated by the cantilever during the scanning, which can

reveal the stiffness or adhesion information of sample surface.

Kelvin probe force microscopy (KPFM) combined with Tapping mode

KPFM, also known as surface potential imaging, has been well developed combined

with tapping mode AFM, in which the difference between the potential of the tip and

that of the sample can be determined.[9,10] The data obtained in this mode is a

combination of three contributing factors: the work function difference, trapped charge,

and any permanent or applied voltage between the tip and the sample. As a consequent,

42

KPFM is generally considered a pseudo-quantitative technique, in which the acquired

accurate contact potential difference (CPD) mostly consists of more than one physical

quantity. In KPFM measurement, an AC bias (VAC) is usually applied on a conductive

tip to oscillate the electrostatic force between tip and sample. The feedback loop then

can provide a DC bias (VDC) to the tip to minimize the electrostatic force between the

tip and the sample.

The electrostatic force between the tip and the sample under the VAC can be described

as Equation 2.9, when they are modeled as a parallel plate capacitor.

𝐹 =1

2

𝜕𝐶

𝜕𝑧𝑉2 (2.9)

The total potential difference (V) between the tip and the sample is the sum of VAC,

VCPD, and the VDC, as shown in Equation 2.10.

𝑉 = 𝑉𝐶𝑃𝐷 + 𝑉𝐷𝐶 + 𝑉𝐴𝐶𝑠𝑖𝑛(⍵𝑡) (2.10)

Therefore, the electrostatic force can be described as:

𝐹 =1

2

𝜕𝐶

𝜕𝑧([(𝑉𝐷𝐶 − 𝑉𝐶𝑃𝐷)2 +

1

2𝑉𝐴𝐶

2 ] + 2[(𝑉𝐷𝐶 − 𝑉𝐶𝑃𝐷)𝑉𝐴𝐶sin (⍵𝑡)] −

[1

2𝑉𝐴𝐶

2 cos (2⍵𝑡)]) (2.11)

The first part represents force is static and not frequency dependent. The second part

occurs at the drive frequency ⍵. The third part reacts at twice the drive frequency. As

a result, the most important term here as far as surface potential is concerned is the

second, since this depends not on the square of the voltage, but rather on the potential

difference between the tip and the sample, multiplied by the magnitude of the applied

VAC. This means that if there is a VCPD between the tip and the sample, then when an

VAC voltage is applied, there will be an oscillatory force at the frequency of the drive

and proportional of the magnitude of the applied voltage, and also proportional to the

CPD. Further, if we make VDC = VCPD, with a potential feedback loop between the tip

43

and the sample, then the oscillations at ⍵ will be nulled. Therefore, the CPD can be

recorded during the scanning.

2.3 Other facilities

Besides STM and AFM, which can directly probe the surface morphology, and in

some extent, visit the surface electronic states and catalytic reactions, other facilities

have also demonstrated their capabilities in this research filed. Here, several facilities

of electron microscopies, X-ray and electron spectroscopies involved in this work are

briefly introduced.

2.3.1 Scanning electron microscopy and focused ion beam

Scanning electron microscopy (SEM) is a very useful and robust facility for imaging

the surface morphology of samples with a resolution higher than 1 nanometer, which

belongs to the family of the electron microscopes. It works through scanning the

electrically conductive sample surface with a focused electron beam. For samples that

are not conducting, coating with very thin noble metal film (Au or Pt) enable their

surface morphologies to be imaged by SEM. During the interactions between the

focused electron beam and sample surface, main products are secondary electrons,

backscattered electrons, auger electrons, and characteristic x-rays. These products can

be collected by different detectors, which contribute the SEM images. For examples,

secondary electrons come from the inelastic interactions between the primary electron

beam and the sample, which mostly contributed by the atoms in top surface and near

surface regions. It is the most common used SEM mode in imaging the surface

morphology. The characteristic x-rays signal can be collected for the energy-dispersive

x-ray spectroscopy (EDX), which enables elemental analysis or chemical

characterization of sample surface.

44

Besides SEM, focused ion beam (FIB) is another important technique in the fields

of materials science and industry. Different from SEM which is designed based on

focused electron beam, FIB commonly uses a focused ion beam, or integrate focused

electron beam together. Therefore, FIB is widely used for lithography and etching in

micro or nano-scale by the focused high energy ion beam (Gallium ion is most widely

used ion source). In addition, FIB has also been used in locally depositing films,

modifying and fabricating materials, and preparing transmission electron microscopy

(TEM) specimen.

2.3.2 Transmission electron microscopy

TEM is another very important electron microscopy, which is well known as its

capability of imaging real-space atomic-resolution of nanostructures. TEM uses high

energy focused electron beam to transmit through and interact with the specimen as it

passes through it to image the structure, shape, morphology, size and composition of

the specimen. The specimen used in TEM are required to be have a thickness less than

100 nm or in nanostructures that can be suspension on a grid, which guarantee effective

transmission of electron beam. TEM appears in three different forms of high-resolution

transmission electron microscopy (HRTEM), scanning transmission electron

microscopy (STEM) and analytical electron microscope (AEM).[11]

The working principle of TEM can be described using a single lens microscope, with

the difference in using electrons rather than photons as the source. Figure 2.7 shows

the simplified model, in which only the objective lens that determining the resolution

are included and the intermediate lenses and projection lenses are omitted. The

resolution of TEM can be estimated by the Abbe’s equation modified by the by using

DeBroglie’s formula.

45

𝑑 =0.753

𝛼𝑉12

(2.12)

where d is the resolution in nanometer, α is the half aperture angle, V is the

accelerating velocity of the electron beam. For a TEM system with accelerating voltage

of 100 kV, the resolution d is valued to be 0.24 nm.

Figure 2.7 The simplified model of a one-lens TEM based on the Abbe’s theory.[12]

HRTEM is one of the most well-developed tools to in material science with its

atomic resolution capability. HRTEM uses both the scattered and transmitted electrons,

which is an interference pattern between the forward-scattered and diffracted electron

46

waves from the specimen. Therefore, it is a phase contrast imaging mode, which can

achieve a resolution as high as 0.1 nm. The acquired diffraction pattern with the

information of the specimen, then can be converted to the Fourier transform analysis.

In recent years, STEM has experienced revolutionary development due to the

aberration correction technique. Sub-angstrom resolution has been achieved in

STEM,[13] which greatly benefits many research filed, such as the solid state physics,

materials sciences, catalysis and chemistry sciences. In addition, the successfully

implementation of combining high angle annular dark-field (HAADF) mode and

electron energy loss spectroscopy (EELS) which belongs to AEM have been achieved.

This provides a very powerful approach to coupling high resolution images with the

local electronic structure of the specimen, which are valuable for revealing fundamental

mechanism for many scientific challenges, especially in the field involving surface and

interface.

Different with HRTEM, STEM uses the focused electron to scan over the specimen.

The electron beam is focused with the assistance of the aberration correction to reach

a fine spot in the order of sub-angstrom. The scattered electrons undergone different

interactions are then collected by the different detectors. The most common detectors

in commercial STEM include HAADF, annular dark-field (ADF), bright-field (BF),

EELS and EDX. The HAADF detector is, in particular, important in acquiring atomic

resolution images. It collects the electrons scattered out to high angles, in which

electrons are not Bragg scattered. As a result, HAADF images show little or no

diffraction effects. The detected intensity of the incoherently scattered electron is

highly sensitive with the atomic number and the image contrast has a proportional

relationship Z2 according to Rutherford scattering model. Therefore, the extremely high

47

atomic resolution is determined by the size of the focused electron spot, rather than

HRTEM which depended on electron diffractions.

2.3.3 Photoelectron spectroscopy

Photoelectron spectroscopy is the most widely used quantitative spectroscopic

technique in analyzing the surface chemical and elemental state, and electronic

structure of materials. Photoelectron spectroscopy is designed based on the

photoelectric effect, using photo-ionization and measuring the kinetic energy

distribution of the emitted photoelectrons. The recorded photoelectron spectrum then

can be converted to the binding energy of electrons which represents the energy

difference between the ionized and neutral atoms. The relationship is described by the

Einstein relationship:

𝐸𝑏 = ℎν − 𝐸𝑘 − (2.13)

where Eb is the electron binding energy, hν is the photon energy of the radiation

source, Ek is the kinetic energy of photoelectron, is the work function induced by the

analyzer, which dependent on both the analyzer and the material. This equation is

essentially a conservation of energy equation. The work function, , usually can be

regarded as constant in the measurement.

According to the different exciting radiation source, photoelectron spectroscopy can

be divided into x-ray photoelectron spectroscopy (XPS) and ultraviolet photoelectron

spectroscopy (UPS). As shown in the diagram in Figure 2.8, XPS normally uses soft

x-rays (200-2000 eV) as the radiation source and can examine the core-level electrons

with a penetration depth of 1-10 nm. While UPS applies vacuum UV radiation (10-45

eV) to examine valence elections of samples with a penetration depth even smaller than

XPS due to its lower incident photon energies.

48

Figure 2.8 Schematic diagram of the electron excitation progress of XPS and UPS.

It has to be mentioned that, the implementation of synchrotron radiation as the

excitation source for photoelectron spectroscopy has greatly broaden its

applications.[14,15] Compared with the normal x-ray generator, and He lamp for UV

light, synchrotron radiation has advantages of high intensity, very broad and continuous

spectral range, high flux and brightness, and high degree of polarization, which has

been playing a very important role in many advanced research fields, including

materials science, physical, biological and chemical sciences, geosciences,

environmental sciences, and medical and pharmaceutical sciences.

49

2.4 References

(1) G. Binning, H. Rohrer, C. Gerber, E. Weibel, Surface studies by scanning tunneling

microscopy, Phys. Rev. Lett. 1982, 49, 57.

(2) G. Binning, H. Rohrer, Scanning tunneling microscopy from birth to adolescence,

Rev. Mod. Phys. 1987, 59, 615.

(3) C. J. Chen, Introduction to Scanning tunneling microscopy, Oxford Univ. Press,

New York, 1993.

(4) J. Tersoff, D. R. Hamann, Theory and application for the scanning tunneling

microscope, Phys. Rev. Lett. 1983, 50, 1998.

(5) J. Tersoff, D. R. Hamann, Theory of the scanning tunneling microscope, Phys. Rev.

B 1985, 31, 805.

(6) G. Binnig, Atomic force microscope and method for imaging surfaces with atomic

resolution, 1986, US Patent No. 4, 724, 318.

(7) G. Binning, C. F. Quate, C. Gerber, Atomic force microscope, Phys. Rev. Lett. 1986,

56, 930.

(8) F. J. Giessibl, Advances in atomic force microscopy, Rev. Mod. Phys. 2003, 75,

949.

(9) M. Nonnenmacher, M. P. O’Boyle, H. K. Wickramasinghe, Kelvin probe force

microscopy, Appl. Phys. Lett. 1991, 58, 2921.

(10) V. Palermo, M. Palma, P. Samorì, Electronic characterization of organic thin films

by kelvin probe force microscopy, Adv. Mater. 2006, 18, 145.

(11) D. B. Williams, C. B. Carter, Transmission electron microscopy, Plenum press,

New York, 1996.

(12) Z. L. Wang, Transmission electron microscopy of shape-controlled nanocrystals

and their assemblies, J. Phys. Chem. B 2000, 104, 1153.

50

(13) O. L. Krivanek, M. F. Chisholm, V. Nicolosi, T. J. Pennycook, G. J. Corbin,

N. Dellby, M. F. Murfitt, C. S. Own, Z. S. Szilagyi, M. P. Oxley, S. T. Pantelides, S. J.

Pennycook, Atom-by-atom structural and chemical analysis by annular dark-field

electron microscopy, Nature 2010, 464, 571.

(14) C. S. Fadley, X-ray photoelectron spectroscopy: progress and perspectives, J

Electron Spectrosc. 2010, 178-179, 2.

(15) S. Günther, B. Kaulich, L. Gregoratti, M. Kiskinov, Photoelectron microscopy and

applications in surface and materials science, Prog. Surf. Sci. 2002, 70, 187.

51

Chapter 3 Characterization of TiO2(110) surface by STM

3.1 Introduction

TiO2, as a typical TMO, have been widely applied in the field of energy conversion,

surface pigments, and photocatalytic wastewater and hazardous gas remediation, due

to its catalytic activity and semiconducting properties as well as its earth abundance,

low toxicity, chemical and thermal stability.[1-2] Surface and interface engineering has

been recognized as an effective approach in promoting the performances or tuning

reactivity of TiO2 in catalysis or energy conversion, for examples facet engineering,[3,4]

curved surfaces,[5] surface defects,[6-8] surface adsorbates and constructing interface

heterostructures[9-14].

Among these strategies, OV is regarded as one of the most important and prevalent

defects in TiO2, as well as a large number of metal oxides, which has been widely

investigated both by theoretical calculations and experimental characterizations.[15-20]

Compared with defects created by doping, OVs are the kinds of defect which do not

disturb the intrinsic crystal structure and involve other impurity elements. At the same

time, OVs are thought could contribute excess electrons that will occupy 3d orbitals of

the neighboring Ti atoms and forms Ti3+ ion, which could enhance the electronic

conductivity of TiO2. As a result, OVs are expected to significantly affect the physical,

chemical and catalytic properties of TiO2, for examples, absorbing and desorbing

behaviors toward molecules, electron transfer in the catalyst and between the surface

of catalyst and adsorbed molecules, as well as the ability of light absorption.

Therefore, acquiring the detailed knowledges of OVs about their location,

distribution, as well as effect on electronic structure and adsorption/desorption of

52

molecules, especially on the surface region, are of great importance toward rationally

promote the activity of TiO2 and other TMOs in their applications in catalysis and

energy conversions. In this chapter, thermally created OV based defects were

introduced into (110) single crystal in vacuum conditions. By carefully controlling the

annealing temperature and time, the defect species and concentration on the surface

were examined by in-situ STM observations. It was found that at the surface of

rutile(110) experienced a reconstruction evolution from (1×1) with point OV defect to

(1×2) with cross-links defects, with the rutile(110) single crystal heated from 900 K to

1300 K.

3.2 Experimental section

TiO2 rutile(110) single crystal (5 × 5 × 1 mm) was purchased from Mateck, GmbH.

To obtain the reduced TiO2, the rutile single crystals were annealed in UHV at 900 K

to 1300 K. The concentration of OVs in the single crystal were controlled by the

annealing time and annealing temperature. STM images were acquired by using a low-

temperature STM (LT-STM) (USM 1500-M, Unisoku Co.) at 78 K. STM images can

be obtained in both constant current mode and constant height mode. Here, STM

images were acquired in constant current mode unless noted otherwise. The samples in

STM measurements were treated by several cycles of Ar+ sputtering (1 kV, 20 min)

and annealing.

53

3.3 Results and discussion

3.3.1 Rutile(110) surface with (1×1) reconstruction

Figure 3.1 Large scale STM image of the reduced TiO2 surface (1.3 V, 30 pA).

Figure 3.1 shows the large scale STM image of the rutile(110) surface after several

sputtering and annealing at 900 K, in which flat terraces were observed. When we

zoomed in to the very small scale shown in the empty-state images in Figure 3.2, atomic

resolution STM of the surface were acquired in both constant current mode and

constant height mode. As the empty-state of TiO2 is dominated by the Ti-3d electronic

states, 5f Ti atoms and Obr atoms appear as bright and dark rows in the empty-state

STM image, respectively, which is reverse-contrast of their real topographies. The

54

lattice constant of was measured to be 0.63 nm by 0.28 nm, with each bright spot

representing one 5f Ti atom, which agrees well with the crystal structure in Figure 1.3.

Figure 3.2 Atomic resolution of the stoichiometric rutile(110) surface. (a) constant

current mode, 0.5 V, 100 pA, scan size is 6 nm × 6 nm, (b) constant height mode, 0.5

V, 100 pA, scan size is 6 nm × 6 nm.

3.3.2 OV point defects on (1×1) surface

Figure 3.3 The reduced rutile(110) surface with OVs on the surface, 1.2 V, 20 pA.

55

In the STM image of reduced (sputtered and annealed at 900 K) rutile(110) surface,

bright protrusions between two 5f Ti rows can be observed, as indicated by the arrows

in Figure 3.3. The OVs trends to individually locate between 5f Ti rows. While as the

surface stored in the UHV chamber for several hours, other two kinds of bright

protrusions appeared on the surface, as shown in Figure 3.4 (a). These three types of

bright protrusions (marked as I, II, and III) exhibit different brightness and apparent

heights (indicated in Figure 3.4 (b)), which can be used not only in identifying different

species, but also in monitoring their dynamic reaction processes.[21-23]

Figure 3.4 (a) STM image of the surface of rutile(110) obtained several hours after

annealed in the UHV (10 nm × 10 nm, 1.6 V, 30 pA). (b) Apparent height profiles of

OV, OH group and H2O in STM images. OV, OH, and H2O are bright protrusions

between two adjacent Ti rows.

They can also be identified by their different responses toward voltage bias plus,

which a voltage higher than a thresh-old voltage (around 2.0 V in our measurements),

either through giving pulses or scanning, was applied. As shown in Figure 3.5, when

three 2.5 V pulses were applied at the selected OHs marked in Figure 3.5 (a), the

hydrogen atoms were removed, allowing the OV to be healed. In the case of OVs, they

are always inactive towards 2.5 V pulses or scanning. Therefore, by a combination of

their apparent heights and their different behavior under a bias higher the threshold

56

voltage, we can identify them. This phenomenon indicates that OVs on the surface of

rutile(110) are reactive toward water molecules even in UHV conditions and favorable

to be hydroxylated on the surface.

Figure 3.5 (a) STM image of the reduced rutile(110) surface with OVs and OHs, 1.3

V, 30 pA. (b) STM image in a same area in (a), in which 2.5 V pulses were applied at

the marked OHs, 1.3 V, 30 pA.

3.3.3 Rutile(110) surface with (1 × 2) reconstruction

Figure 3.6 Large scale STM image of the reduced TiO2 surface acquired after annealed

at 1300 K, 3 V, 50 pA, with the inset showing the side view of (1 × 2) reconstruction

surface.

57

When the annealed temperature was increased to 1300 K, the flat terraces on (1 × 1)

surface evolved to chain structures along Ti rows, as presented in Figure 3.6. This

surface has been explained by the (1 × 2) reconstruction due to the more oxygen

deficiency at higher annealing temperature.[24-26] In addition some short chains

perpendicular to the (1 × 2) chains can be observed, which are marked by the white

solid line as links in Figure 3.6. As shown in the zoomed-in images in Figure 3.7, the

unit cell of (1 × 2) reconstruction is marked by the black rectangle. Two cross-links of

single-link and double-link structure are indicated by the yellow and white arrows.

Meanwhile, it notices that the high quality STM image in filled state can be obtained

on (1 × 2) surface, which is not applicable on (1 × 1) surface, as shown in Figure 3.7

(b). It indicates that the electronic structures of rutile(110) also undergo significant

change with the increasing oxygen deficiency.

Figure 3.7 STM images of rutile(1 × 2) surface. (a) empty state mode, scan size is 10

nm × 10 nm, 1 V, 50 pA, (b) filled state mode, scan size is 10 nm× 10 nm, -1.8 V, 50

pA.

58

3.4 Summary

OV based defects were introduced into (110) single crystal through annealling in

vacuum conditions. By carefully controlling the annealing temperature and time, the

defect species and concentration on the surface were examined by in-situ STM

observations. On the rutile(110) (1×1) surfaces obtained at 900 K, isolated OV point

defects were found, which could be occupied by water molecule or convert to OH. On

the (1×2) surfaces obtained at 1300 K, cross-links defects dominated the surfaces.

3.5 References

(1) X. Chen, S. S. Mao, Titanium dioxide nanomaterials: synthesis, properties,

modifications, and applications, Chem. Sci. 2007, 107, 2891.

(2) B. O’Regan, M. Grätzel, A low-cost, high-efficiency solar cell based on dye-

sensitized colloidal TiO2 film, Nature 1991, 353, 737.

(3) H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zhou, G. Liu, S. C. Smith, H. M. Cheng, G.

Q. Lu, Anatase TiO2 single crystals with a large percentage of reactive facets, Nature

2008, 453, 638.

(4) J. S. Chen, Y. L. Tan, C. M. Li, Y. L. Cheah, D. Luan, S. Madhavi, F. Yin, C. Boey,

L. A. Archer, X. W. Lou, Constructing hierarchical spheres from large ultrathin anatase

TiO2 nanosheets with nearly 100% exposed (001) facets for fast reversible lithium

storage, J. Am. Chem. Soc. 2010, 132, 6124.

(5) K. Shirai, G. Fazio, T. Sugimoto, D. Selli, L. Ferraro, K. Watanabe, M. Ηaruta, B.

Ohtani, H. Kurata, C. D. Valentin, Y. Matsumoto, Water-assisted hole trapping at the

highly curved surface of nanoTiO2 photocatalyst, J. Am. Chem. Soc. 2018, 140, 1415.

59

(6) M. Kong, Y. Li, X. Chen, T. Tian, P. Fang, F. Zheng, X. Zhao, Tuning the relative

concentration ratio of bulk defects to surface defects in TiO2 nanocrystals leads to high

photocatalytic efficiency, J. Am. Chem. Soc. 2011, 133, 16414.

(7) M. Batzill, E. H. Morales, U. Diebold, Influence of nitrogen doping on the defect

formation and surface properties of TiO2 rutile and anatase, Phys. Rev. Lett. 2006, 96,

026103.

(8) X. Yu, B. Kim, Y. K. Kim, Highly enhanced photoactivity of anatase TiO2

nanocrystals by controlled hydrogenation-induced surface defects, ACS Catal. 2013, 3,

2479.

(9) G. Liu, L. Wang, H. G. Yang, H.-M. Cheng, G. Q. Lu, Titania-based photocatalysts-

crystal growth, doping, and heterostructuring, J. Mater. Chem. 2010, 20, 831.

(10) C. L. Pang, R. Lindsay, G. Thornton, Structure of clean and adsorbate-covered

single-crystal rutile TiO2 Surfaces, Chem. Rev. 2013, 113, 3887.

(11) J. Zhang, J. H. Bang, C. Tang, P. V. Kamat, Tailored TiO2−SrTiO3 heterostructure

nanotube arrays for improved photoelectrochemical performance, ACS Nano 2010, 4,

387.

(12) J. Tian, P. Hao, N. Wei, H. Cui, H. Liu, 3D Bi2MoO6 nanosheet/TiO2 nanobelt

heterostructure: enhanced photocatalytic activities and photoelectochemistry

performance, ACS Catal. 2015, 5, 4530.

(13) J. Resasco, H. Zhang, N. Kornienko, N. Becknell, H. Lee, J. Guo, A. L. Briseno,

P. Yang, TiO2/BiVO4 nanowire heterostructure photoanodes based on type II band

alignment, ACS Cent. Sci. 2016, 2, 80.

(14) V. Subramanian, E. E. Wolf, P. V. Kamat, Catalysis with TiO2/gold

nanocomposites. effect of metal particle size on the fermi level equilibration, J. Am.

Chem. Soc. 2004, 126, 4943.

60

(15) T. R. Gordon, M. Cargnello, T. Paik, F. Mangolini, R. T. Weber, P. Fornasiero,

C. B. Murray, Nonaqueous synthesis of TiO2 nanocrystals using TiF4 to engineer

morphology, oxygen vacancy concentration, and photocatalytic activity, J. Am. Chem.

Soc. 2012, 134, 6751.

(16) A. Janotti, J. B. Varley, P. Rinke, N. Umezawa, G. Kresse, C. G. Van de Walle,

Hybrid functional studies of the oxygen vacancy in TiO2, Phys. Rev. B 2010, 81,

085212.

(17) C. D. Valentin, G. Pacchioni, A. Selloni, Reduced and n-type doped TiO2: nature

of Ti3+ species, J. Phys. Chem. C 2009, 113, 20543.

(18) Q. Kang, J. Cao, Y. Zhang, L. Liu, H. Xu, J. Ye, Reduced TiO2 nanotube arrays

for photoelectrochemical water splitting, J. Mater. Chem. A 2013, 1, 5766.

(19) M. Setvín, U. Aschauer, P. Scheiber, Y.-F. Li, W. Hou, M. Schmid, A. Selloni, U.

Diebold, Reaction of O2 with subsurface oxygen vacancies on TiO2 anatase (101),

Science 2013, 341, 988.

(20) X. Pan, M.-Q. Yang, X. Fu, N. Zhang, Y.-J. Xu, Defective TiO2 with oxygen

vacancies: synthesis, properties and photocatalytic applications, Nanoscale 2013, 5,

3601.

(21) O. Bikondoa, C. L. Pang, R. Ithnin, C. A. Muryn, H. Onishi, G. Thornton, Direct

visualization of defect-mediated dissociation of water on TiO2(110), Nat. Mater. 2006,

5, 189.

(22) X. Cui, Z. Wang, S. Tan, B. Wang, J. Yang, J. G. Hou, Identifying hydroxyls on

the TiO2 (110)-1×1 surface with scanning tunneling microscopy, J. Phys. Chem. C

2009, 113, 13204.

(23) N. G. Petrik, Z. Zhang, Y. Du, Z. Dohnálek, I. Lyubinetsky, G. A. Kimmel,

Chemical reactivity of reduced TiO2(110): the dominant role of surface defects in

oxygen chemisorption, J. Phys. Chem. C 2009, 113, 12407.

61

(24) R. A. Bennett, P. Stone, N. J. Price, M. Bowker, Two (1 × 2) reconstructions of

TiO2(110): surface eearrangement and reactivity studied using elevated temperature

scanning tunneling microscopy, Phys. Rev. Lett. 1999, 82, 3831.

(25) M. Blanco-Rey, J. Abad, C. Rogero, J. Mendez, M. F. Lopez, J. A. Martin-Gago,

P. L. de Andres, Structure of rutile TiO2 (110)-(1 × 2): formation of Ti2O3 quasi-1D

metallic chains, Phys. Rev. Lett. 2006, 96, 055502.

(26) K. T. Park, M. H. Pan, V. Meunier, E.W. Plummer, Surface reconstruction of TiO2

(110) driven by suboxides, Phys. Rev. Lett. 2006, 96, 226105.

62

Chapter 4 Activating titania for electrocatalysis by surface vacancy engineering

4.1 Introduction

As an alternative to fossil fuels, hydrogen is regarded as one of the most promising

key energy carriers for a global-scale sustainable energy system.[1] In particular, water-

alkali electrolyzers for overall water splitting exhibit tremendous potential for the

evolution of high-purity hydrogen and oxygen gases, with the additional values of

simple processes, zero CO2 emission and low pollutant. Finding efficient and low-cost

noble-metal-free electrocatalysts towards HER and OER is urgently required for their

large-scale commercial application.[2-6] Thanks to extensive efforts, earth-abundant

transition TMOs have recently been reported to be promising candidates for

electrochemical water splitting in alkaline conditions through defect engineering.[7-11]

These defects are expected to effectively improve the intrinsic electrical conductivity

of defective TMOs, act as catalytic active cites and enhance the catalytic activity.

Among various defects, surface OV is regarded as one of the most important, and

supposed to be the prevalent defect in many TMOs.[12-15] Previous studies have

demonstrated that when oxygen atoms are removed from the TMOs, the geometric,

physical and chemical properties can be profoundly modified. For example, the

coexistence of different oxidation states of metals, such as Ti3+-Ti4+ in TiO2, Co2+-Co3+

in Co2O3 and Ce3+-Ce4+ in CeO2, are essential in the electron translocations and

extremely important for many of their catalytic applications.[9,15,16] Additionally, the

surface OV has been well-recognized as favorable adsorption for adsorbates (such as

H2O, O2, metal nanocluster, CO2, etc) for the defective TMOs in catalytic processes.[17-

22] Obviously, elucidating the role of OV, as well as other defects, in the modified

electrocatalytic properties of defective TMOs is of immense scientific and

technological importance towards an in-depth understanding and optimizing catalysts.

63

However, it remains challenging owing to the lack of detailed knowledges of catalytic

active sites and the electron translocations at the atomic level. Meanwhile, the precise

control of defects is very difficult, because complicated defect compositions along with

a wide diversity of sample conditions, such as morphology, crystallinity, dimension

and diameters are common in defective TMOs.

Here, in this paper, surface OVs were introduced into the highly pure and

stoichiometric TiO2 rutile(110) single crystal, which is a prototypal and widely studied

TMO, through thermal treatment under UHV condition.[23,24] STEM and STM

techniques found that the as-treated single crystal kept a good crystalline structure. In

addition, OVs were found mainly exist in the surface and subsurface region with a

depth of around 100 nm rather than the inner bulk region. The simple sample conditions

allow us to have an in-depth visit on the effect of OV on the physical and chemical

properties of TiO2, as well as its roles in surface electrocatalytic processes. In-situ

synchrotron XPS confirmed the appearance of mid-gap defect state in the as-treated

single crystal, which was aroused by the associating Ti3+ ions with OVs. The catalytic

activities of TiO2 were investigated by electrochemical HER in alkaline conditions,

suggesting a tremendous enhancement for the as-treated single crystal with OVs.

Further STM and DFT calculations suggested Ti3+ ions neighboring at OV could be

helpful for the proton reduction process and promote the electrons translocation. Gibbs

free energy calculations verified that the existence of OV on the surface or subsurface

could effectively low the energy of hydrogen adsorption and benefit the HER.

4.2 Experimental sections

The cross-sectional TEM images were acquired with a STEM (JEOL JEM-

ARM200f) both in HAADF mode and annular bright-field (ABF) mode. In-situ EELS

64

was used to characterize the OV concentration of the cross-sectional sample in TEM

measurements. The cross-sectional sample was prepared by the FIB (Zeiss Auriga FIB-

SEM) technique.

In-situ XPS characterizations were carried out at Beamline 4B9B in the Beijing

Synchrotron Radiation Facility (BSRF), and variable photon energies were referenced

to a fresh Au polycrystalline film. The spot size of incident light in XPS was about 1

mm in diameter. All the data were recorded in UHV at room temperature. The Hall

coefficient and magnetoresistance were measured by the five-probe technique using a

Quantum Design Physical Property Measurement System (PPMS)-14T. The electrolyte

was 1 M KOH solution. The electrochemical HER experiments were performed by a

typical three-electrode method, in which a Pt plate and Hg/HgO (0.923 V versus the

standard hydrogen electrode) were used as the counter and reference electrodes,

respectively. All electrochemical measurements were performed with a Bio Logic

Science Instruments VSP-300 electrochemistry workstation. The linear portions of

Tafel plots were fitted to the Tafel equation: η = blog|J| + a, where η is the overpotential,

a is the exchange current density, and b is the Tafel slope.

All DFT calculations were performed using the Vienna Ab Initio Simulation

Package (VASP). The generalized gradient approximation (GGA) was applied to treat

the exchange correlation energy with the Perdew-Burke-Ernzerhof (PBE) functional.

The projector augmented wave (PAW) method was employed to describe electron−ion

interactions, with the cut-off energy of 400 eV. The structural model of the TiO2(110)

surface was constructed with four Ti-O layers as a 4 × 2 periodic supercell comprising

192 atoms with a vacuum spacing of 20 Å to avoid interaction between adjacent

surfaces. Spin-polarized local density approximation plus on-site Coulomb self-

interaction potential (LDA+U) calculations were performed for the Hubbard

65

correction, and an effective U (Ueff) value of 4.2 eV was applied in all calculations. All

structures in the calculations were relaxed until the convergence tolerance of the force

on each atom was smaller than 0.02 eV. The energy convergence criterion was set to

be 1 × 10-4 eV for self-consistent calculations, and k-point sampling was restricted to

the Gamma point only because of the large size of the supercell.

4.3 Characterization of OV by STEM

Figure 4.1 Crystal structure mode of the TiO2 rutile(110) surface, surf-OV can be

created through removing Obr atoms.

Figure 4.2 SEM image of reduced TiO2 single crystal sample cut by a focused ion beam

(FIB); inset is an enlarged image of the selected area.

66

First of all, pure and stoichiometric TiO2(110) single crystal can be well described

by the structure model shown in Figure 4.1, with alternating rows of bridging oxygen

(Obr) atoms and five-coordinate titanium (Ti5c) atoms lying in a plane along the [001]

direction. The reduced single crystal was obtained through annealing in UHV

conditions with base pressures in the low 1 × 10-10 torr regime at 900 K. The cross-

section of the reduced TiO2 single crystal was characterized by STEM, with the sample

prepared by FIB shown in Figure 4.2. The top surface, near-surface and inner bulk

region all exhibited high crystallinity, shown in Figure 4.3 (a) and (b), indicating that

the formation of OVs did not interrupt its well-crystalline structure of the reduced TiO2.

It should be emphasized that the good crystallinity of the surface area of the reduced

sample could exclude the existence of disorder or amorphous surface layer and foreign

atoms, which are widely studied in many other cases.[25-27] Moreover, the gradually

darkening contrast from inner region to top surface with a thickness of around 80 nm

in the STEM image implied the increasing oxygen deficiency and the displacement of

Ti atoms, because the contrast is proportional to the average atomic number of atomic

columns in HAADF mode.[28]

67

Figure 4.3 Cross-sectional TEM image of reduced TiO2(110) single crystal in ABF,

showing a high crystalline structure both in the surface and bulk regions. (b) Large-

region cross-sectional STEM image of reduced TiO2(110) single crystal in HAADF

mode, in which three areas with different depth from the surface (10 nm, 50 nm, and

100 nm) are marked. (c) and (d) Corresponding in-situ EELS Ti-L edge and low-level

spectra of the three regions marked in (b).

To further reveal the effect of OVs on the reduced TiO2 single crystal, in-situ EELS

were acquired followed the STEM measurement at three different regions (10 nm, 50

nm and 100 nm from top the surface, marked in Figure 4.3 (b). In the Ti-L edge EELS

spectra, the intensity ratio of dz2/dx2-y2 which is sensitive towards the Ti-O bonding

lengths[29] in the top surface region apparently increased compared with those in the

inner regions, agreeing well with the OV induced lattice distortions recognized in

above experiments. As shown in the low-level spectrum in Figure 4.3 (d), the three

68

peaks (6.4 eV, 11.3 eV and 14.6 eV), originating from O 2p orbitals to Ti 3d orbitals,

are marked in the region A. Compared with inner region, the peaks at the top surface

region (10 nm) exhibited obvious broadening, which were caused by the OV induced

lattice distortion and indicated a higher OV concertation near the top surface area.[30]

Meanwhile, the peaks at 25.2 eV (peak B) and 48.5 eV (peak C) in the inner region

shifted towards lower energy to 24.6 eV and 48.2 eV in the top surface region, due to

the emerging of more OVs. Peak B was assigned to transitions between O 2p state and

Ti 4sp states. Peak C was assigned to excitations from the Ti 3p core level to 3d excited

states.[31] Therefore, the existences of OVs are expected to modify the valence state of

their surrounding Ti atoms.

4.4 Characterization the electronic structure by in-situ synchrotron XPS and

UPS

To investigate the effect of OV on the electronic structure of reduced TiO2 single

crystal, in-situ synchrotron XPS spectra and valance band spectra, was carried out. As

shown in Figure 4.4 (a), after annealing the stoichiometric pure single crystal at 900 K

in HUV for 5 h, a broad peak at around 457 eV belongs to Ti3+ ions appeared in the Ti

2p XPS spectra.[32] It means the excess electrons from the removed oxygen atoms were

trapped by Ti atoms and formed Ti3+ ions. In the valence band (VB) spectra shown in

Figure 4.4 (b), a peak, locating about 0.8 eV below the Fermi Level appeared after

annealing, which is known as the Ti 3d defect state. Meanwhile, a broad peak locating

at 12 eV can be seen below the Fermi Level for both samples, which can be assigned

to the OHs.[33] The broad peaks reflected the dissociation of water at the surface, which

is consistent with the STM observation.

69

Figure 4.4 (a) In-situ Ti 2p XPS spectra of the pristine and OV-TiO2 single crystal. (b)

VB spectra of pristine and OV-TiO2 single crystal.

Figure 4.5 DFT calculations of the mid-gap states caused by OVs in TiO2(110). (a)

Calculation of the excess electron distribution of the reduced TiO2(110) surface with

an OV on the top surface. (b) The calculated band structure (left) and DOS (right) of

the TiO2(110) with a surface OV.

In addition, DFT calculation was used to verify the OV-induced changes of

electronic structure in the reduced TiO2. As shown in Figure 4.5, after introducing an

OV on the surface of TiO2(110), the two excess electrons belonged to the removed

oxygen atom will bond with two neighboring Ti atoms and form Ti3+ ions. The other

two cases, including OVs at inner region and surface OHs from water dissociation at

70

surface OV sites were also included in the DFT calculations, which demonstrated that

they exhibited similar results of creating Ti3+ ions (Figure 4.6). As a sequence, defect

states in the gap, can be observed in the calculated band structure and DOS, which are

mainly contributed to the Ti 3d orbitals. Above XPS and DFT results indicate that the

band gap Ti 3d defect states originate from these intrinsic defects and associating Ti3+

ions in the reduced TiO2.

Figure 4.6 Calculation of the excess electron distribution on the TiO2(110) surface with

an OV on the sublayer and with two surface Ad-H. (a) and (b) The calculated band

structure and DOS of TiO2(110) with sublayer OVs, respectively. (c) and (d) The

calculated band structure and DOS of TiO2(110) with surface Ad-Hs, respectively.

71

4.5 Electrochemical HER activity of OV-TiO2

Figure 4.7 (a) LSV data on different electrocatalysts for the HER at the rate of 10 mV∙s-

1, inset shows the pristine TiO2(110) single crystal (P), OV-low TiO2(110) single

crystal (L), OV-high TiO2(110) single crystal (H), and the Nb-doped TiO2(110) single

crystal (Nb). The OV-high TiO2 was tested for 1000 cycles with the LVS curve shown

as the red solid and green dashed lines for before and after cycling, respectively. (b)

Cycling stability over 18 h of the OV-high sample at a potential of –0.7 V. (c) Tafel

plot of the OV-high TiO2. (d) Cdl determined from the linear fitting of the capacitive

current vs. scan rate, measured from the cyclic voltammetry (CV) curves of OV-high

TiO2, OV-low TiO2 and Nb-doped TiO2.

To evaluate the electrochemical HER activity of the reduced TiO2 single crystal, two

reduced TiO2 single crystal annealed in HUV at 900 K for 5 h (OV-low) and 50 h (OV-

high), and a reference sample (Nb-doped (0.43 at%)) single crystal were selected. The

electrochemical behavior for the HER of these samples were then tested at 1 M KOH

aqueous solution with a scan rate of 10 mV/s. As shown in Figure 4.7, OV-high TiO2

72

exhibited considerable HER activity with a current density of 22.8 mA.cm-2 at -0.80 V,

while Nb-doped TiO2 exhibited very weak activity and OV-low TiO2 was almost not

active under the same experimental conditions (IR compensation based on resistance

test was applied for all linear sweep voltammetric (LSV) curves, Figure 4.8). Long-

term stability of the OV-high TiO2 was demonstrated by the cycling test of 1000 times

(red dash line in Figure 4.7 (a)), and a continuous test at a potential of –0.7 V for more

than 18 h (Figure 4.7 (b)).

The linear portions of Tafel plots were fitted to the Tafel equation: η = blog|J| + a,

where η is the overpotential, a is the exchange current density, and b is the Tafel slope.

The Tafel slope of OV-high TiO2 near the substantial cathodic current region from the

HER was plotted to be 187.5 mV.dec-1, as shown in Figure 4.7 (c), which represents

the hydrogen generation rate with the applied overpotential. Figure 4.7 (d) shows the

double layer capacitance (Cdl) of the three samples, which were valued by fitting the

slope of the capacitive current vs. scan rate measured from the cyclic voltammetry (CV)

curves (Figure 4.9). The Cdl value has a proportional relationship with the

electrochemically active surface area (EASA) of catalysts. Therefore, the Cdl value of

OV-high TiO2 at 72 mF.cm-2 demonstrated an apparent increase in the EASA compared

with Nb-doped TiO2 with a Cdl value of 8.8 mF.cm-2, which match well with their HER

performances. It should also be noticed that despite the Cdl value of OV-low (16.0

mF.cm-2) is higher than that of Nb-doped TiO2, its HER performance was severely

suppressed by its low conductivity, which will be presented in Figure 4.11. Therefore,

it is suggested that the elevated electrochemical HER of TiO2 not only originate from

the increased intrinsic conductivity and the charge carrier density, but also the modified

intrinsic electrocatalytic properties that needed to be further clarified.

73

Figure 4.8 Electrochemical impedance spectra of different electrodes at −0.3 V versus

eversible hydrogen electrode (RHE) (inset is the full range measurement).

Figure 4.9 CV curves of different samples under scanning rate ranging from 20 mV/s

to 180 mV/s.

4.6 Determination of the origin of the electrochemical HER activity of OV-TiO2

4.6.1 Electrical conductivity

Electrical conductivity and the number of electrocatalytic sites are considered to be

of great significance for the catalytic performance of electrocatalysts. First of all, five-

probe Hall effect measurements were carried out to reveal the conductivities of all the

74

TiO2 single crystals, as shown in the five probes configuration used in the PPMS

measurement in Figure 4.10.

Figure 4.10 The holder used in PPMS measurement and the five probes configuration

used in the PPMS measurement.

The electron density of n-type semiconductors has an inversely proportional

relationship with the Hall coefficient (RH) as n = ‒1/(e∙RH). As shown in Figure 4.11,

OV-high TiO2 had an electrical resistivity of 80.0 × 10-3 Ω∙m that is lower than that of

the OV-low TiO2 (94.0 × 10-3 Ω∙m). The charge carrier density in OV-high TiO2 (1.9

× 1017 cm-3) is about 40 times higher than that in the OV-low TiO2 (4.8 × 1015 cm-3).

The conductivities of both reduced TiO2 samples are believed to arise from the mid-

gap state induced by OVs. The reference Nb-doped TiO2 single crystal exhibits the

highest electrical conductivity of 40.6 × 10-3 Ω∙m and the highest charge carrier density

75

of 3.2 × 1017 cm-3 among all the samples. The weak electrocatalytic activity of the

reference Nb-doped TiO2 single crystal, however, demonstrates that good electrical

conductivity is inadequate to enable TiO2 activity towards the HER in alkaline media.

Figure 4. 11 Hall resistivity measurements of different samples. The Hall coefficient

RH was determined by fitting the slope of the curve of the Hall resistivity vs. magnetic

field.

4.6.2 Electrochemical activate sites

In order to reveal the catalytic process associated with active sites at the atomic level,

STM investigations were carried out on the OV-high TiO2(110) single crystal. As

shown in Figure 4.12, all the adsorbed H2O molecules reside on OV sites. as confirmed

by the adsorption dynamics observed in Figure 4.12 (a) and (b). OHs also appeared on

the OV sites after dissociation of H2O molecules, as observed in the dynamic processes

in Figure 4.13. This phenomenon suggests that surface OVs are highly active towards

adsorbing and dissociating residual H2O molecules in UHV.

76

Figure 4.12 In-situ STM studies on electrocatalytic dynamics occurring on TiO2(110)

surface associated with oxygen vacancies. (a) STM image of the partially hydroxylated

TiO2 surface, with the OVs indicated by light blue arrows (15 nm × 15 nm, 1.2 V, 20

pA). (b) STM image of the same region of (a), with the OVs mostly filled by H2O (15

nm × 15 nm, 1.2 V, 20 pA).

77

Figure 4.13 (a) STM image of an individual Ad-H2O at an OV site, with another OV

included as a reference point (3 nm × 6 nm, 1.2 V, 10 pA). (b) STM image of the two

OHs from the dissociation of Ad-H2O at the OV site (1.2 V, 10 pA). (c) STM image of

the same area in a and b, in which an OH was removed through a 2.5 V pulse by the

STM tip (1.2 V, 10 pA).

78

Figure 4.14 (a) STM image of the reduced TiO2 surface with two individual OVs in the

empty-state (1.2 V, 20 pA). (b) STM image of the same region of (a), but in the filled-

state, with Ti3+ ions indicated by red arrows (-2.3 V, 10 pA).

In addition, each OV induces two surrounding Ti3+ ions, which was revealed in STM

images taken with negative sample bias (filled-state), as shown in Figure 4.14. Recent

investigations pointed out that the Ti3+ ions in reduced TiO2 exhibit polaron

behavior.[34-36] This enables Ti3+ ions to rapidly hop across the nearby lattice in rutile,

and thus, leads to the increase of the conductivity of the reduced rutile TiO2. Since the

electrical conductivity of reduced TiO2 is proportional to the concentration of Ti3+ ions,

this supports the proposition that the OV-high TiO2(110) single crystal possesses

higher conductivity than the OV-low sample. In order to simulate electrocatalytic HER

processes, we applied a sample bias to allow the STM tip to act as an electron donator

to the OV-high TiO2(110) single crystal surface. Both Ad-H2O and OHs disappeared

from the sample surface when the bias was higher than the threshold of 2.0 V, as shown

in Figure 4.15 (a) and (b). Meanwhile, the OVs at corresponding sites were also healed.

This indicates that hydrogen desorption of Ad-H2O and OHs occurred on the OV-high

TiO2(110) surface at the cost of consumption of OVs. In this case, the HER activity of

79

reduced TiO2(110) single crystal is expected to be depressed gradually with a

decreasing concentration of OVs. Nevertheless, this contradicts our cycling stability

results, in which the OV-high TiO2(110) surface exhibited excellent stability and

durability, suggesting that OVs in the surface are not adequate to drive the

electrocatalytic HER constantly and steadily, although they are active towards

adsorbing H2O molecules and OHs.

Figure 4.15 STM image of the partially hydroxylated TiO2 surface, with OVs, OHs,

and Ad-H2O appearing on the surface (1.2 V, 20 pA). (f) STM image of the same region

of e after the OHs and Ad-H2O were removed by the last scan with a tip bias of 2.5 V.

(1.2 V, 20 pA).

80

4.6.3 DFT calculation of the Gibbs free energy

The origins of the electrocatalytic HER activity of the reduced TiO2(110) surface in

alkaline media was further revealed by DFT calculations, in terms of thermodynamics

and kinetics. The surface structures were modeled according to STEM and STM results

obtained on TiO2(110) surface with OVs.

The calculation methods for the hydrogen evolution reaction (HER) in alkaline

solutions are summarized as follows:

By considering the standard hydrogen electrode as the reference potential and

hydrolysis of water in the solution, the free energy of reactions (4.1) and (4.2) is set to

zero because they are in equilibrium. Therefore, the free energy of (H+ + e-)

corresponds to that of ½ H2 (1 bar, 298 K) and the free energy of (OH- ‒ e-) is calculated

according to the free energy of H2O and (H+ + e-),

(H+ + e-) → ½ H2 (4.1)

H2O → (H+ + e-) + (OH- ‒ e-) (4.2)

We used gas-phase H2O at 0.035 bar as the reference state, because at this pressure,

gas-phase H2O is in equilibrium with liquid water at 300 K. The calculated free

energies of the H2O, (H+ + e-), and (OH- ‒ e-) are listed in Table 4.1.

The Gibbs free energy of the intermediates were calculated as[37]

ΔG = ΔE + ΔZPE ‒ TΔS (4.3)

where ΔE is the binding energy of intermediates which is defined as the reaction

energies of the reactions

H2O + * → OH* + H* (4.4)

H2O + * → OH* + ½H2 (4.5)

½H2 + * → H* (4.6)

81

ΔZPE and ΔS can be obtained from the vibrational frequency ʋi, which are the

changes in zero point energies (ZPE) and entropies due to the reaction, respectively.

All the parameters have been taken from DFT calculations.

Zero point energies are calculated as follows,

ZPE = ∑i ½hʋi (4.7)

i = 3n‒5 (for linear molecule)

i = 3n‒6 (for non-linear molecule)

where i is the degree of freedom for the molecule, and n is the number of atoms in

the molecule.

Entropies are calculated from the sum of the translational entropy St, the rotational

entropy Sr, and the vibrational entropy Sv as follows:

S = St + Sr + Sv (4.8)

(4.9)

(for linear molecule) (4.10-1)

(for non-linear molecule) (4.10-2)

(4.11)

where kB, NA, and h are the Boltzmann constant, Avogadro constant, and Planck

constant, respectively. N, m, V, T, I, and σ are the number of particles, and the mass,

volume, temperature, moment of inertia, and symmetry number of the molecules,

respectively. Otherwise, the entropies of intermediates are calculated by the vibrational

entropy , because no translational or rotational behaviors can be found for an

adsorbed molecule. The calculated ZPE and S of free H2O, H2, and (OH- - e-) are listed

in Table 4.1.

82

At a pH different from 0, we can correct the free energy of H+ ions by the

concentration dependence of the entropy: ΔGpH = ‒kT∙ln[H+] = kT∙ln10∙pH. In our

work, pH is 14 and ΔGpH = 0.83 eV.

The effect of a bias ΔGU was imposed on each step by including an electron in the

electrode as an ‒eU term, where U is the electrode potential relative to the standard

hydrogen electrode. Therefore, the reaction free energy of processes was calculated as:

ΔG(U, pH) = ΔG + ΔGpH + ΔGU (4.12)

As is shown in the reactions of the HER in alkaline solution,

Volmer H2O + e- → H* + OH- (4.13)

Heyrovsky H2O + e- +H* → H2 + OH- (4.14)

Tafel 2H* → H2 (4.15)

We include the effects of pH and U on steps involving (OH- - e-). The free energies

of possible steps (such as water splitting, hydroxide adsorption, hydroxide desorption,

and hydrogen production) were calculated and compared to find out the most optimal

path in our calculation. As a result, the reactions in HER and corresponding free

energies under applied potential U and pH can be written as:

Volmer 1 - water splitting

H2O + M → H* + OH*

ΔG1 = ΔG (H* + OH*)

Volmer 2 - OH* desorption

OH* → (OH- ‒ e-) + M

ΔG2 = ‒ΔG(OH*) + eU + 0.83

Heyrovsky

H2O + H* → H2↑ + (OH- ‒ e-) + M

ΔG3 = ‒ΔG(H*) + eU + 0.83

83

Tafel

2H* + M → H2↑

ΔG4 = ‒ΔG(2H*)

Here, M represents the catalyst, which can be a TiO2(110) surface with surfOV and

subOV, only subOV, or subOV with Obr-H surface. The calculation results of main

reaction pathway are listed in Table 4.1.

Table 4.1. Binding energies, entropies, and zero point energies contribution to Gibbs

free energy of main reaction pathway. subOV + Obr-H* is regard as M.

subOV + Obr-H* ΔE ZPE ΔZPE TS TΔS

ΔZPE

- TΔS

ΔG

ΔG

(pH)

H2O 0.566 0 0.534 0 0 0 0

½ H2 0.133 0.201

OH- ‒ e- 0.433 0.333

H2O* -0.932 0.631 0.065 0.020 -0.514 0.579 -0.353 -0.353

H* + OH* -0.735 0.538 -0.028 0.009 -0.525 0.497 -0.238 -0.238

H*+ (OH- ‒ e-) -0.600 0.671 0.105 0.333 -0.201 0.306 -0.294 0.536

½H2 + (OH- ‒ e-) 0.566 0 0.534 0 0 0 0.83

84

Figure 4.16 DFT calculations reveal roles of oxygen vacancies in electrocatalysis on

TiO2(110) surface. (a) Free energy pathways of the relevant reaction intermediates in

alkaline media on the reduced TiO2 with both surfOV and subOV, with the OVs

marked as solid black circles. (b) Compare of HER free energy of pristine TiO2 and

reduced TiO2 with subOV and Obr-H* (an onset potential of URHE = ‒1.15 V was applied

for both catalysts).

Therefore, based on our calculated, the Free energy pathways of the relevant reaction

intermediates in alkaline media on the OV-TiO2 with both surfOV and subOV, is

shown in Figure 4.16. It was found that water molecules prefer to be dissociated at

surface OV (surfOV) site in initial stage (step 1). The produced hydrogen atoms are

favorable towards combing in pairs and forming hydrogen molecules (step 2 to 3). All

the reaction steps are exothermic processes. This supports our hypothesis that the

surface OVs are easily to be healed during the electrochemical reaction. Hence, it

indicates that surfOVs were not sustainable which agrees with STM. HER pathway on

the reduced TiO2 with sublayer OV (subOV), as well as the pristine TiO2 were further

calculated, as shown in Fig. 4.16 (b). Hydrogen desorption reaction (Heyrovsky

85

reaction) step between intermediate 2 and intermediate 3 is deter determined to be the

rate-limiting step with the largest free energy difference for both pristine TiO2 and

reduced TiO2. The pristine TiO2 needs a theoretical potential of URHE = ‒1.33 eV to

overcome this energy barrier. In contrast, the potential barrier can be effectively

lowered on the reduced TiO2 to enable all HER reaction steps to be exothermic and

energetically favorable, due to the presence of subOV. This has been confirmed

experimentally in our electrocatalytic HER process, in which onset potential for HER

in reduced TiO2 was significantly decreased. Therefore, the theoretical calculations

suggest that the electrocatalytic HER activity of reduced TiO2(110) single crystal

mainly originates from subOVs, which can effectively promote the hydrogen

desorption capability and lower the overpotential of HER in alkaline media.

Figure 4.17 Optimized pathway of the HER in alkaline media on the reduced TiO2

with surfOv and subOV based on the free energy calculation.

Figure 4.17 gives the detailed optimized overall pathway of the HER in alkaline

media on the OV-TiO2 with surfOv and subOV based on the free energy calculation,

with all the intermediates listed below:

Intermediates

1: TiO2-surfOV-subOV + H2O TiO2-subOV + surfOV-H2O*

2: TiO2-subOV + surfOV-H2O*

86

3: TiO2-subOV + Obr-H* + Ti-H*

4: TiO2-subOV + H2

5: TiO2-subOV + Ti5C-H2O*

6: TiO2-subOV + Ti5C-OH* + Obr-H*

7 and 11: TiO2-subOV + Obr-H* + OH-

8 and 12: TiO2-subOV + 2Obr-H* + Ti5C-OH*

9 and 13: TiO2-subOV + 2Obr-H* + OH-

10 and 14: TiO2-subOV + Obr-H* + OH- + H2

15: TiO2-subOV + H2

All reactions are written as:

a. TiO2-surfOV-subOV:

Volmer (1-2-3)

TiO2-surfOV-subOV + H2O → TiO2-subOV + surfOV-H2O*

TiO2-subOV + surfOV-H2O* → TiO2-subOV + Obr-H

* + Ti5C-H*

Tafel (3-4)

TiO2-subOV + Obr-H* + Ti5C-H*→ TiO2-subOV + H2↑

b. TiO2-subOV:

Volmer (4-5-6)

TiO2-subOV+ H2O → TiO2-subOV + Ti5C-H2O*

TiO2-subOV + Ti5C-H2O* → TiO2-subOV + Ti5C-OH* + Obr-H

*

Heyrovsky (6-11)

TiO2-subOV + Ti5C-OH* + Obr-H* → TiO2-subOV + Obr-H

* + (OH- ‒ e-)

c. TiO2-subOV + Obr-H*:

Volmer (11-12-13)

TiO2-subOV + Obr-H* +H2O → TiO2-subOV + 2Obr-H

* + Ti5C-OH*

87

TiO2-subOV + 2Obr-H* + Ti5C-OH* → TiO2-subOV + 2Obr-H

* + (OH- ‒ e-)

Heyrovsky (13-14)

TiO2-subOV + 2Obr-H* + H2O → TiO2-subOV + Obr-H

* + H2↑ + (OH- ‒ e-)

Tafel (13-15)

TiO2-subOV + 2Obr-H* → TiO2-subOV + H2↑

4.7 Summary

Our study shows that inactive pure TiO2 rutile(110) single crystal can be activated

towards the HER in alkaline media through creating OVs and accompanying Ti3+ ions

by annealing in UHV. OVs and Ti3+ ions in the surface region dominate the electrical

conductivity of reduced TiO2 and the amount of electrocatalytic active sites.

Combining the well-characterized atomic surface structure and theoretical

calculations, we conclude that subOVs can promote the electron transfer and hydrogen

desorption in the electrocatalytic HER on reduced TiO2 in alkaline media. Considering

that all the electrochemical characterizations were performed on a single crystal sample

with atomically flat surface and low specific surface area, the overpotential is expected

to be significantly decreased in reduced TiO2 nanoparticles with a higher density of

active sites. Our work helps to elucidate the fundamental mechanism of the

electrocatalytic activity of reduced oxides towards the HER, which is of immense

fundamental and practical importance towards an in-depth understanding and rational

optimization of TMOs as electrocatalysts in alkaline media.

4.8 References

(1) K. Zeng, D. Zhang, Recent progress in alkaline water electrolysis for hydrogen

production and applications, Prog. Energy Combust. Sci. 2010, 36, 307.

88

(2) L. Trotochaud, J. K. Ranney, K. N. Williams, S. W. Boettcher, Solution-cast metal

oxide thin film electrocatalysts for oxygen evolution, J. Am. Chem. Soc. 2012, 134,

17253.

(3) M. W. Louie, A. T. Bell, An investigation of thin-film Ni–Fe oxide catalysts for the

electrochemical evolution of oxygen, J. Am. Chem. Soc. 2013, 135, 12329.

(4) C. C. L. McCrory, S. Jung, J. C. Peters, T. F. Jaramillo, Benchmarking

heterogeneous electrocatalysts for the oxygen evolution reaction, J. Am. Chem. Soc.

2013, 135, 16977.

(5) R. D. L. Smith, M. S. Prévot, R. D. Fagan, S. Trudel, C. P. Berlinguette, Water

oxidation catalysis: electrocatalytic response to metal stoichiometry in amorphous

metal oxide films containing iron, cobalt, and nickel, J. Am. Chem. Soc. 2013, 135,

11580.

(6) B. Zhang, X. Zheng, O. Voznyy, R. Comin, M. Bajdich, M. García-Melchor, L.

Han, J. Xu, M. Liu, L. Zheng, F. P. G. de Arquer, C. T. Dinh, F. Fan, M. Yuan, E.

Yassitepe, N. Chen, T. Regier, P. Liu, Y. Li, P. D. Luna, A. Janmohamed, H. L. Xin,

H. Yang, A. Vojvodic, E. H. Sargent, Homogeneously dispersed multimetal oxygen-

evolving catalysts, Science 2016, 352, 333.

(7) Y. Meng, W. Song, H. Huang, Z. Ren, S.-Y. Chen, S. L. Suib, Structure–property

relationship of bifunctional MnO2 nanostructures: highly efficient, ultra-stable

electrochemical water oxidation and oxygen reduction reaction catalysts identified in

alkaline media, J. Am. Chem. Soc. 2014, 136, 11452.

(8) R. Subbaraman, D. Tripkovic, K.-C. Chang, D. Strmcnik, A. P. Paulikas, P.

Hirunsit, M. Chan, J. Greeley, V. Stamenkovic, N. M. Markovic, Trends in

activity for the water electrolyser reactions on 3d M(Ni,Co,Fe,Mn) hydr(oxy)oxide

catalysts, Nat. Mater. 2012, 11, 550.

89

(9) R. Subbaraman, D. Tripkovic, D. Strmcnik, K.-C. Chang, M. Uchimura, A. P.

Paulikas, V. Stamenkovic, N. M. Markovic, Enhancing hydrogen evolution activity in

water splitting by tailoring Li+-Ni(OH)2-Pt interfaces, Science 2011, 334, 1256.

(10) V. R. Stamenkovic, D. Strmcnik, P. P. Lopes, N. M. Markovic, Energy and fuels

from electrochemical interfaces, Nat. Mater. 2017, 16, 57.

(11) M. Gong, W. Zhou, M.-C. Tsai, J. Zhou, M. Guan, M.-C. Lin, B. Zhang, Y. Hu,

D.-Y. Wang, J. Yang, S. J. Pennycook, B.-J. Hwang, H. Dai, Nanoscale nickel

oxide/nickel heterostructures for active hydrogen evolution electrocatalysis, Nat.

Commun. 2014, 5, 4695.

(12) J.-X. Feng, H. Xu, Y.-T. Dong, X.-F. Lu, Y.-X. Tong, G.-R. Li, Efficient hydrogen

evolution electrocatalysis using cobalt nanotubes decorated with titanium dioxide

nanodots, Angew. Chem. Int. Ed. 2017, 129, 3006.

(13) M. K. Bates, Q. Jia, N. Ramaswamy, R. J. Allen, S. Mukerjee, Composite Ni/NiO-

Cr2O3 catalyst for alkaline hydrogen evolution reaction, J. Phys. Chem. C 2015, 119,

5467.

(14) Y. P. Zhu, T. Y. Ma, M. Jaroniec, S. Z. Qiao, Self-templating synthesis of hollow

Co3O4 microtube arrays for highly efficient water electrolysis, Angew. Chem. Int. Ed.

2017, 56, 1324.

(15) Z. Weng, W. Liu, L.-C. Yin, R. Fang, M. Li, E. I. Altman, Q. Fan, F. Li, H.-M.

Cheng, H. Wang, Metal/oxide interface nanostructures generated by surface

segregation for electrocatalysis, Nano Lett. 2015, 15, 7704.

(16) X. Gao, H. Zhang, Q. Li, X. Yu, Z. Hong, X. Zhang, C. Liang, Z. Lin, Hierarchical

NiCo2O4 hollow microcuboids as bifunctional electrocatalysts for overall water-

splitting, Angew. Chem. Int. Ed. 2016, 55, 6290.

90

(17) M. Sun, Y. Chen, G. Tian, A. Wu, H. Yan, H. Fu, Stable mesoporous ZnFe2O4 as

an efficient electrocatalyst for hydrogen evolution reaction, Electrochimica Acta 2016,

190, 186.

(18) R. Venkatkarthick, D. J. Davidson, S. Ravichandran, S. Vengatesan, G. Sozhan,

S. Vasudevan, Eco-friendly and facilely prepared silica modified amorphous titania

(TiO2–SiO2) electrocatalyst for the O2 and H2 evolution reactions, Catal. Sci. Technol.

2015, 5, 5016.

(19) D. Chanda, J. Hnát, M. Paidar, J. Schauer, K. Bouzek, Synthesis and

characterization of NiFe2O4 electrocatalyst for the hydrogen evolution reaction in

alkaline water electrolysis using different polymer binders, J. Power Sources 2015,

285, 217-226.

(20) H. Wang, H.-W. Lee, Y. Deng, Z. Lu, P.-C. Hsu, Y. Liu, D. Lin, Y. Cui,

Bifunctional non-noble metal oxide nanoparticle electrocatalysts through lithium-

induced conversion for overall water splitting, Nat. Commun. 2015, 6, 7261.

(21) X. Yan, L. Tian, M. He, X. Chen, Three-dimensional crystalline/amorphous

Co/Co3O4 Core/Shell nanosheets as efficient electrocatalysts for the hydrogen

evolution reaction, Nano Lett. 2015, 15, 6015.

(22) Z. Luo, R. Miao, T. D. Huan, I. M. Mosa, A. S. Poyraz, W. Zhong, J. E. Cloud, D.

A. Kriz, S. Thanneeru, J. He, Y. Zhang, R. Ramprasad, S. L. Suib, Mesoporous MoO3–

x material as an efficient electrocatalyst for hydrogen evolution reactions, Adv. Energy

Mater. 2016, 6, 1600528.

(23) Y. Jin, H. Wang, J. Li, X. Yue, Y. Han, P. K. Shen, Y. Cui, Porous MoO2

nanosheets as non-noble bifunctional electrocatalysts for overall water splitting, Adv.

Mater. 2016, 28, 3785.

91

(24) C. Shu, S. Kang, Y. Jin, X. Yue, P. K. Shen, Bifunctional porous non-precious

metal WO2 hexahedral networks as an electrocatalyst for full water splitting, J. Mater.

Chem. A 2017, 5, 9655.

(25) X. Chen, L. Liu, P. Y. Yu, S. S. Mao, Increasing solar absorption for

photocatalysis with black hydrogenated titanium dioxide nanocrystals, Science 2011,

331, 746.

(26) M. Tian, M. Mahjouri-Samani, G. Eres, R. Sachan, M. Yoon, M. F. Chisholm, K.

Wang, A. A. Puretzky, C. M. Rouleau, D. B. Geohegan, G. Duscher, Structure and

formation mechanism of black TiO2 nanoparticles, ACS Nano 2015, 9, 10482.

(27) Y. Liu, H. Feng, X. Yan, J. Wang, H. Yang, Y. Du, W. Hao, The origin of

enhanced photocatalytic activities of hydrogenated TiO2 nanoparticles, Dalton Trans.

2017, 46, 10694.

(28) S. Stemmer, A. Sane1, N. D Browning, T. J Mazane, Characterization of oxygen-

deficient SrCoO3−δ by electron energy-loss spectroscopy and Z-contrast imaging, Solid

State Ionics 2000, 130, 71.

(29) G. G. Fuentes, I. G. Mancheno, F. Balbas, C. Quiros, J. F. Trigo, F. Yubero, E.

Elizalde, J. M. Sanz, Dielectric properties of Ti, TiO2 and TiN from 1.5 to 60 eV

determined by reflection electron energy loss spectroscopy (REELS) and ellipsometry,

Phys. Stat. Sol. (a) 1999, 175, 429.

(30) M. Fusi, E. Maccallini, T. Caruso, C. S. Casari, A. Li Bassi, C. E. Bottani, P.

Rudolf, K.C. Prince, R.G. Agostino, Surface electronic and structural properties of

nanostructured titanium oxide grown by pulsed laser deposition, Surf. Sci. 2011, 605,

333.

(31) M. H. Mohamed, H. R. Sadeghi, V. E. Henrich, Electron-energy-loss study of the

TiO2(110) surface, Phys. Rev. B 1988, 37, 8417.

92

(32) M. Batzill, E. H. Morales, U. Diebold, Influence of nitrogen doping on the defect

formation and surface properties of TiO2 rutile and anatase, Phys. Rev. Lett. 2006, 96,

026103.

(33) Z. Wang, X. Hao, S. Gerhold, Z. Novotny, C. Franchini, E. McDermott, K.

Schulte, M. Schmid, U. Diebold, Water adsorption at the tetrahedral titania surface

layer of SrTiO3(110)-(4 × 1), J. Phys. Chem. C 2013, 117, 26060.

(34) M. Setvin, C. Franchini, X. Hao, M. Schmid, A. Janotti, M. Kaltak, C. G. Walle,

Van de, G. Kresse, U. Diebold, Direct view at excess electrons in TiO2 rutile and

anatase, Phys. Rev. Lett. 2014, 113, 086402.

(35) T. Minato, Y. Sainoo, Y. Kim, H. S. Kato, K.-I. Aika, M. Kawai, J. Zhao, H. Petek,

T. Huang, W. He, B. Wang, Z. Wang, Y. Zhao, J. Yang, J. G. Hou, The electronic

structure of oxygen atom vacancy and hydroxyl impurity defects on titanium dioxide

(110) surface, J. Chem. Phys. 2009, 130,124502.

(36) C. M. Yim, M. B. Watkins, M. J. Wolf, C. L. Pang, K. Hermansson, G. Thornton,

Engineering polarons at a metal oxide surface, Phys. Rev. Lett. 2016, 117, 116402.

(37) J. K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, Origin of the

overpotential for oxygen reduction at a fuel-cell cathode, J. Phys. Chem. B 2004, 108,

17886.

93

Chapter 5 Tuning electronic structure of BiOBr 2D nanosheets by strain for

photocatalysis

5.1 Introduction

The ability to continuously control the electronic structures in photocatalysts is

highly desirable for a wide range of energy and environmental applications, including

H2 production by water splitting, carbon fixation, and the elimination of pollution.[1,2]

For example, the absorption spectrum and the quantum conversion efficiency of a

photocatalyst, which determine its performance, can be modulated by tuning the band

gap (Eg), the positions of the VB and the conduction band (CB), and the band dispersion

of photocatalysts.[3] In a similar way to chemical composition, strain is a continuous

variable that is capable of altering electronic structure. Although strain engineering is

a straightforward method, its potential in photocatalysis remains largely under-

exploited.

Bismuth oxyhalides BiOX (X = Cl, Br, I) are p-block photocatalysts that have

attracted considerable attention due to their unique 2D layered structure and excellent

photocatalytic properties under visible light.[4-9] In BiOX, [Bi2O2]2+ slabs are

interleaved with double halogen atoms slabs by strong electrovalent bonds along the

[001] direction, while two closely adjacent slabs of halogen atoms are connected by

van de Waals interactions.[10,11] The dispersive VB and CB induced by sp hybridization

give rise to high mobility of the photo-induced charge carriers. In addition, the internal

electric field resulting from the asymmetric charge distribution between [Bi2O2]2+ and

the halogen layers facilitates the effective separation of these photo-induced charge

carriers, and hence, enables the photocatalytic activity of BiOX.[12-14] Due to their 2D

94

layered structure, the electronic structure of BiOX compounds is highly sensitive to

even a subtle inner strain variation.[15] It is, however, a practical challenge to modulate

the photocatalytic properties through fine-tuning the inner strain in nanoscale BiOX.

In this work, we illustrate experimentally and theoretically that the photocatalytic

performance of BiOBr nanosheets can be tuned by the inner strain effect. The

characterizations of photocatalytic degradation and geometric phase analysis (GPA) of

the TEM images indicate that the distribution and intensity of the inner strain dominate

the photocatalytic activity of BiOBr nanosheets. DFT calculations demonstrate that the

strain-modulated photocatalytic properties in BiOBr originate from variation of the

intrinsic electronic structure of this photocatalyst.

5.2 Experimental section

In the synthesis procedure, 5 mmol Bi(NO3)3∙5H2O and 5 mmol cetyl

trimethylammonium bromide (CTAB) were added into 100 mL distilled water at room

temperature with stirring for 20 min, and then, 1 M NaOH solution was added to the

solution to adjust the pH value to 7, 5, 3, and 2 (with the resulting samples denoted as

BiOBr-1, BiOBr-2, BiOBr-3, and BiOBr-4, respectively). The mixed solution was

stirred for 1 h, and then poured into a 100 mL Teflon-lined stainless autoclave up to

80% of the total volume. The autoclave was heated at 170 °C for 17 h, and then cooled

to room temperature in air. The resulting precipitates were collected, washed with

ethanol and deionized water several times, and dried at 80 °C for 10 h.

The morphologies and microstructures of the as-prepared samples were

characterized by SEM (Hitachi CS 3400) and TEM (JEOL JEM-3010, operated at 300

kV). In the photo-degradation experiments, BiOBr powder (100 mg) was added to an

aqueous solution of Rh B (0.02 mmol/L, 100 mL) in a 150 mL quartz reactor.

95

Degradation experiments on Rh. B using N-doped P25 TiO2 nanopowders were also

carried out under the same conditions as a reference. For the photo-degradation of MO

(10 mg/L, 50 mL), 50 mg catalyst was added to an aqueous solution of MO (10 mg/L,

50 mL) in a 150 mL quartz reactor. All photocatalytic experiments were carried out

under irradiation by a 300 W Xe lamp with a filter glass (λ ≥ 420 nm) to remove

ultraviolet (UV) light after dark adsorption experiments. The absorption spectra of Rh.

B and MO were collected on a Hitachi U3010 UV-Visible (UV-Vis)

spectrophotometer. In the photocurrent-time response system, a 300 W Xe lamp with

a monochromator and a cut-off filter (λ ≥ 420 nm) was used as the light source. UV-

Vis diffuse reflectance spectra were collected on a Cintra-10e spectrometer. The

distributions of in-plane strain across the nanosheets were identified by GPA based on

the HRTEM images. In this paper, STEM-CELL was used to simulate the distribution

of the strain, based on the HRTEM images, by calculating and analysing the fast

Fourier transform (FFT) and inverse FFT of the entire images. In the analysis process,

the displacement of lattice parameters (u) are determined by calculating and analysing

the Fourier transform of the selected image. The strain is then obtained from the

derivative of the displacement in the picked direction, and visualized by inverse Fourier

transform.

The first-principles calculations were performed using the VASP code. The GGA

was applied to treat the exchange correlation energy with the PBE functional. The PAW

method was employed to describe the electron-ion interactions. A k-point sampling of

9×9×6 was generated with original Gamma meshes. The cut-off energy for the plane

wave basis was 550 eV. The biaxial strain simulations were realized by fixing the x

and y axes and optimizing the z axis. Equilibrium geometries were obtained by the

minimum energy principle.

96

5.3 Morphology and structure characterization

Figure 5.1 SEM images of (a) BiOBr-square and (d) BiOBr-circle (scale bar is 1 m).

Cross-sectional TEM images of (b) BiOBr-square and (e) BiOBr-circle (scale bar is

200 nm); insets are the HRTEM images of the cross-sections, where both samples show

layer structure along the c-axis (scale bar is 10 nm). HRTEM images of (c) BiOBr-

square and (f) BiOBr-circle, with the corresponding SAED patterns as the insets,

indicating that both samples have the (001) face exposed (scale bar is 10 nm-1 and 5

nm-1 for the insets).

Figure 5.1 shows two typical morphologies of BiOBr nanosheets, square-shaped

(also assigned as BiOBr-1) and circle-shaped BiOBr (BiOBr-4), which were fabricated

by hydrothermal reaction with different pH values. We found that the BiOBr

nanosheets underwent a morphology transition from square-like to circle-like when the

pH value was decreased. The BiOBr nanosheets are several micrometres in size. The

thicknesses of the BiOBr-square and BiOBr-circle nanosheets are 31 nm and 22 nm,

respectively, as revealed by TEM in Figure 5.1 (a) and (b). The insets demonstrate the

layered structure of the BiOBr nanosheets. The interlayer distance is approximate 0.8

nm for both samples. As shown in Figure 5.1 (c)-(f), the d-spacing of 0.27 nm indicates

97

that the (110) face is along the in-plane BiOBr nanosheets. The corresponding insets

show the selected area electron diffraction (SAED) patterns, in which the marked spots

can be indexed as (200) face, (110) face and (1-10) face. Hence, the exposed surfaces

can be identified as {001} facets for both the BiOBr-square and the BiOBr-circle

nanosheets.

5.4 Characterization of strain in BiOBr 2D nanosheets

By carefully examining the X-ray diffraction (XRD) patterns, obvious diffraction

peak shifts could be identified. Figure 5.2 shows the XRD patterns of the as-prepared

BiOBr nanosheets. All the samples (labelled as BiOBr-1 to BiOBr-4) prepared by the

hydrothermal method with different pH values are well-crystallized single-phase

nanopowders. The diffraction patterns can be indexed to the tetragonal structure

(P4/nmm(129)) according to the standard data (PDF card #09-0393). By precisely

controlling the concentration of NaOH, the diffraction peaks can be shifted gradually

in accordance with the pH value of the precursors, as shown in Figure 4.2 (a). In

particular, the (110) peak and the (004) peak demonstrate a clear upshifting with

increasing pH value, as shown in Figure 5.2 (b)-(d). BiOBr possesses a typical 2D

layered crystal structure, in which [Bi2O2]2+ slabs are interleaved with double bromine

atom slabs by strong electrovalent bonds along the [001] direction, as shown in the

insets of Figure 5.2 (c) and (d). The diffraction peak shift indicates the presence of in-

plane strain in BiOBr. The change in the corresponding interplanar crystal spacing,

identified by the refined cell parameters obtained using MDI Jade 5.0 (Materials data,

Inc., Livermore, CA, 1999), was used to estimate the degree of in-plane strain in BiOBr

by comparing it with the theoretical d-value. As shown in Figure 5.2 (c) and (d), in

contrast to the BiOBr-4 sample, which exhibits tiny strain, BiOBr-1 shows a much

98

higher degree of compressive strain for both the {110} and the {001} facets. More

importantly, our results indicate that by carefully controlling the pH value during

sample preparation, the crystal strains were gradually changed for the BiOBr

nanosheets by a facile chemical method.

Figure 5.2 (a) XRD patterns of BiOBr samples fabricated with different pH values.

(b) Local areas of (110) and (004) peaks of BiOBr, where obvious shifts of the peaks

are observed, which are indicated for different BiOBr samples. (c) and (d) Plots of the

changes in the peak positions and the relative tensile strain compared with the

theoretical crystal structure of the (110) and (004) faces, respectively; insets are

schematic illustrations of the BiOBr crystal structure.

99

Figure 5.3 TEM images of (a) BiOBr-square and (c) BiOBr-circle (scale bars are 1000

nm). Strain simulation of (b) BiOBr-square and (d) BiOBr-circle based on HRTEM

(scale bars are 10 nm). The internal strain distributions are in the xy-direction (Exy), the

x-direction (Exx) and the y-direction (Eyy), with the scale for the whole image area

In order to reveal the details of the morphology dependence of the inner strain, we

carried out TEM characterization and GPA simulation based on the HRTEM images,

as shown in Figure 5.3.[16-18] The in-plane wrinkles observed in the TEM images of the

BiOBr-square nanosheets reflect the existence of a large inner strain, while the BiOBr-

circle sample exhibits a relatively strain-free character, as shown in Figure 5.3 (a) and

(c). Figure 5.3 (b) and (d) show the inner strain distribution maps of the BiOBr

nanosheets in the xy-direction (Exy), the x-direction (Exx), and the y-direction (Eyy),

respectively, as obtained by strain simulation. The inhomogeneous compressive strain

100

distribution in the BiOBr-square nanosheets is reflected by the severe local lattice

distortions across the whole surface in Figure 5.3 (b). In contrast, the BiOBr-circle

nanosheets exhibit a quite uniform strain distribution, and this sample shows much less

lattice distortion in the strain maps in Figure 5.3 (d). It should be noted that the strain

difference across the BiOBr-square nanosheets is higher than that across the BiOBr-

circle nanosheets. For example, the strain in the BiOBr-square sample varies from 0.73

to 1.29 (Exy), from 0.78 to 1.21 (Exx), and from 0.80 to 1.16 (Eyy), while in BiOBr-

circle, it varies from 0.88 to 1.17 (Exy), from 0.87 to 1.10 (Exx), and from 0.93 to 1.08

(Eyy). These results confirm that the inner strain of the BiOBr nanosheets can be varied

by the morphology, which can be precisely tuned by the fabrication conditions.

5.5 Characterization of photocatalytic activity of BiOBr 2D nanosheets

Firstly, UV-Vis diffuse reflectance spectra of the BiOBr nanosheets with different

strains were carried out to determine their band gap, which has a significant impact on

their light absorption ability. The band gap of semiconductor can be determined by the

Tauc formula,

(5.1)

(5.2)

Here, α, ν, Eg, A, n, and λ are the absorption coefficient, the incident light frequency,

the band gap, a constant, an integer, and the absorption edge, respectively. Firstly, the

approximate Eg can be calculated by Equation 5.2, and then ln(αhν) vs. ln(hν-Eg) is

plotted; thus, by refining the slope of the straightest line near the band edge we can

obtain the value of n. Secondly, (αhν)2/n vs. hν is plotted, and then the band gap Eg is

evaluated by drawing an extension line to the hν axis intercept. Finally, the accurate

101

value of Eg of BiOBr-square is calculated as 2.68 eV, and 2.82 eV is calculated for

BiOBr-circle. This indicates that the inner strain has a significant effect on the band

gap of BiOBr.

As shown in Figure 5.4, the BiOBr samples exhibit excellent photocatalytic

performance in Rh B degradation. Their photocatalytic activities are higher than for N-

doped P25 TiO2 nanopowders. The low-strain BiOBr-circle sample shows the highest

activity among all the samples, and it demonstrates almost 100 % degradation of Rh.

B within 30 min. As demonstrated in Figure 5.4 (c), the apparent first-order rate

constant (k) for BiOBr-circle is almost twice those for BiOBr-square and the N-doped

P25 TiO2 nanopowders. Similar results were also observed for degradation of other

dyes, such as MO, as shown in Figure 5.4 (d). As shown in Figure 5.4 (e), both the

BiOBr-square and the BiOBr-circle samples can degrade phenol under visible light.

Again, the BiOBr-circle nanosheets show better visible-light photocatalytic

degradation performance on phenol than the BiOBr-square nanosheets. The

photocurrent response of the BiOBr samples were measured for several On-Off cycles

under visible light irradiation. As shown in Figure 5.4 (f), the photocurrent of the

BiOBr samples exhibits a quick response to light irradiation, with the photocurrent

sharply decreasing to zero as soon as the light is turned off, while the photocurrent

quickly reaches stable values when the light is turned on. The stable photocurrents

measured on the BiOBr-square and BiOBr-circle samples under visible light are 0.7

µA and 1.5 µA, respectively. The higher photocurrent of BiOBr-circle than BiOBr-

square under visible light suggests that more efficient photoexcited charge carrier

separation and less recombination of electron-hole pairs were possibly achieved by

decreasing the inner strain in the BiOBr nanosheets.

102

Figure 5.4 (a) UV-Vis diffuse reflectance spectra of BiOBr-square and BiOBr-circle;

inset shows the derivation of the band-gap values for BiOBr. (b) Degradation

experiments on Rh. B by P25, BiOBr-square, and BiOBr-circle under visible light

irradiation, and (c) the corresponding apparent rate constants. (d) Degradation

experiments on MO by BiOBr-square and BiOBr-circle under visible light irradiation,

and (e) Degradation experiments of phenol by BiOBr-square and BiOBr-circle

nanosheets under visible-light irradiation. (f) Current density transient with light on/off

for BiOBr powders under light irradiation.

5.6 DFT of the strain effect on the electronic structure of BiOBr

We carried out DFT calculations in order to the reveal the strain effect (both

compressive and tensile strains) on the electronic structure of BiOBr, which dominates

the photocatalytic properties.[19-21] The strain-free and compressive-strained BiOBr

show the typical electronic features of an indirect-band-gap semiconductor. It changes

to a direct band-gap semiconductor, however, if tensile strain is present in BiOBr. With

a 9.1% tensile strain in the BiOBr lattice, the valence band maximum (VBM) of BiOBr

moves from the R point to the Z point, while the CBM is still located at the same high

symmetry point (Z point) in k-space. It is found that the Eg of BiOBr can be modulated

by the strain effect, as shown in Figure 5.5 (a). For example, the band gap varies

103

between 1.94 eV, 2.12 eV, and 1.27 eV in BiOBr which has 8.8% compressive strain,

is free of strain, and has 9.1% tensile strain, respectively. Both compressive and tensile

strains lead to a narrowed band gap, which is also observed in the other indirect-band-

gap semiconductors.[22] Figure 5.5 (b) shows the calculated DOS. It is found that the

bottom of the CB of strain-free BiOBr is mainly contributed by Bi 6p, and the top of

the VB is dominated by Br 3p, O 2p, and Bi 6s orbitals. While under tensile strain, the

contribution of Bi 6s to the VBM is suppressed.

Figure 5.5 DFT calculations of the band structure of BiOBr with biaxial strain. (a) The

left panel models compressive strain, the central panel is for a strain-free sample, and

the right is for tensile strain. (b) Calculations of the DOS of BiOBr with different kinds

of strain.

Several key factors can affect the photocatalytic activity of BiOBr nanosheets in

photocatalytic reactions, which include photon absorption, separation of photoexcited

carriers, and surface area. Our UV-Vis diffuse reflectance spectra results demonstrate

that the band gap of BiOBr-square (Eg = 2.68 eV) is smaller than that of BiOBr-circle

(Eg = 2.82 eV). BiOBr-square is, therefore, expected to have a broader range of light

absorption, in contrast to BiOBr-circle. BET measurements reveal that the BiOBr-

square sample exhibits a larger surface area of 7.03 m3/g compared to BiOBr-circle

(4.49 m3/g) (Table 5.1). It is interesting, however, that the photocatalytic measurements

indicate that the photocatalytic activity of BiOBr-sqaure is much lower than that of

BiOBr-circle. This is because the photoexcited charge separation in BiOBr determines

104

the overall photocatalytic activity in the photocatalytic process, which was also

reported in previous works.[23,24] As shown in Figure 5.4 (f), the photocurrent

measurements suggest that the charge separation is indeed more efficient in BiOBr-

circle nanosheets than in BiOBr-square nanosheets. Based on the DFT calculation

results, we believe that the electronic structure of the BiOBr-circle sample may

facilitate the separation of photoexcited charge carriers. It was found that the

interactions between Br atoms and [Bi2O2]2+ slabs mediated by weak van de Waals

coupling are tunable by the strain effect. In other words, the band symmetry of BiOBr

can be modulated by strain through tuning the electronic interaction between the Br

atoms and the [Bi2O2]2+ slabs. The dramatic changes in the band symmetry, e.g. from

direct to indirect band gap or the change of energy dispersion due to the strain, would

affect the separation of photoexcited charge carriers. The electron-hole recombination

occurring in indirect semiconductors typically requires the emission of multiple

phonons to accommodate the energy and momentum differences between the CB and

the VB. An appropriate rearrangement of the electronic symmetry, for instance, in the

BiOBr-circle sample, may tune the momentum mismatch and improve electron-hole

separation. Moreover, the VB and CB in strain-free BiOBr are more dispersive, which

is expected to lead to a high mobility of photoexcited charge carriers. In the case of the

BiOBr-square nanosheets, their electronic structure modifications by the strain effect

might depress the separation of photoexcited charge carriers, and consequently,

weaken the photocatalytic performance. In addition, strain also leads to the formation

of structural defects in 2D materials. In BiOBr-square nanosheets, a large strain of

1.8 % is verified by the XRD and GPA results. It is believed that strain-induced defects

will act as recombination centres for photoexcited electrons and holes, and also depress

the quantum conversion efficiency of strained BiOBr-square nanosheets.

105

Table 5.1 BET test of the surface area of different BiOBr samples.

Sample BET (m3/g)

BiOBr-1

BiOBr-2

BiOBr-3

BiOBr-4

7.03

7.85

5.47

4.49

5.7 Summary

In summary, the strain effect on the photocatalytic activity of BiOBr nanosheets with

highly reactive {001} facets exposed was studied. The XRD, TEM, and strain tensor

simulation results reveal that the intensity and distribution of inner strain in the BiOBr

nanosheets can be modulated by adjusting the pH value in the synthesis reaction. It is

found that the strain effect can effectively tune the photocatalytic activity of BiOBr

nanosheets in dye degradation. Our work suggests that strain engineering could be an

effective approach to controlling the electronic structure of semiconductors for further

enhancement of their efficiency in converting light into other forms of energy.

5.8 References

(1) X. Chen, L. Liu, P. Y. Yu, S. S. Mao, Increasing solar absorption for photocatalysis

with black hydrogenated titanium dioxide nanocrystals, Science 2011, 331, 746.

106

(2) J. Zhang, J. Sun, J. K. Maeda, K. Domen, P. Liu, M. Antonietti, X. Fu, X. Wang,

Sulfur-mediated synthesis of carbon nitride: band-gap engineering and improved

functions for photocatalysis, Energy Environ. Sci. 2011, 4, 675.

(3) H. Tong, S. Ouyang, Y. Bi, N. Umezawa, M. Oshikiri, J. Ye, Nano-photocatalytic

materials: possibilities and challenges, Adv. Mater. 2012, 24, 229.

(4) H. Cheng, B. Huang, Y. Dai, Engineering BiOX (X = Cl, Br, I) nanostructures for

highly efficient photocatalytic applications, Nanoscale 2014, 6, 2009.

(5) Li, J.; Yu, Y.; Zhang, L. Bismuth oxyhalide nanomaterials: layered structure meet

photocatalysis, Nanoscale 2014, 6, 8473.

(6) L. Ye, Y. Su, X. Jin, H. Xie, C. Zhang, Recent advances in BiOX (X = Cl, Br, I)

photocatalysts: synthesis, modification, facet effect and mechanisms, Environ. Sci.:

Nano 2014, 1, 90.

(7) H. Zhang, L. Liu, Z. Zhou, Towards better photocatalysts: first-principles studies of

the alloying effects on the photocatalytic activities of bismuth oxyhalides under visible

light, Phys. Chem. Chem. Phys. 2012, 14, 1286.

(8) H. Zhang, L. Liu, Z. Zhou, First-principles studies on facet-dependent

photocatalytic properties of bismuth oxyhalides (BiOXs), RSC Adv. 2012, 2, 9224.

(9) H. Zhang, Y. Yang, Z. Zhou, Y. Zhao, L. Liu, L. Enhanced photocatalytic properties

in BiOBr nanosheets with dominantly exposed (102) facets, J. Phys. Chem. C. 2014,

118, 14662.

(10) H. Deng, J. Wang, Q. Peng, X. Wang, Y. Li, Controlled hydrothermal synthesis of

bismuth oxyhalide nanobelts and nanotubes, Chem. Eur. J. 2005, 11, 6519.

(11) H. An, Y. Du, T. Wang, C. Wang, W. Hao, J. Zhang, Photocatalytic properties of

BiOX (X = Cl, Br, and I), Rare Metals 2008, 27, 243.

107

(12) J. Jiang, K. Zhao, X. Xiao, X. L. Zhang, Synthesis and facet-dependent

photoreactivity of BiOCl single-crystalline nanosheets, J. Am. Chem. Soc. 2012, 134,

4473.

(13) M. Shang, W. Wang, L. Zhang, Preparation of BiOBr lamellar structure with high

photocatalytic activity by CTAB as Br source and template, J. Hazard. Mater. 2009,

167, 803.

(14) H. Li, L. Liu, X. Liang, W. Hou, X. Tao, Enhanced visible light photocatalytic

activity of bismuth oxybromide lamellas with decreasing lamella thicknesses, J. Mater.

Chem. A 2014, 2, 8926.

(15) J. Feng, X. Qian, C.-W. Huang, J. Li, Strain-engineered artificial atom as a broad-

spectrum solar energy funnel, Nat. Photonics 2012, 6, 866.

(16) M. J. Hÿtch, E. Snoeck, R. Kilaas, Quantitative measurement of displacement and

strain fields from HREM micrographs, Ultramicroscopy 1998, 74, 131.

(17) V. Grillo, E. Rotunno, STEM_CELL: a software tool for electron microscopy: part

I-simulations, Ultramicroscopy 2013, 125, 97.

(18) V. Grillo, F. Rossi, STEM_CELL: a software tool for electron microscopy. Part 2

analysis of crystalline materials, Ultramicroscopy 2013, 112.

(19) G. Kresse, J. Hafner, Ab initio molecular dynamics for open-shell transition

metals, Phys. Rev. B 1993, 48, 13115.

(20) J. P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made

simple, Phys. Rev. Lett. 1996, 77, 3865.

(21) P. E. Blöchl, Projector augmented-wave method, Phys. Rev. B 1994, 50, 17953

(22) D. Shiri, Y. Kong, A. Buin, M. P. Anantram, Strain induced changes of bandgap

and effective mass in silicon nanowires, Appl. Phys. Lett. 2008, 93, 073114.

108

(23) W. Li, X. Wang, Y. Wang, J. Zhang, Z. Lin, B. Zhang, F, Huang, Synthesis and

facet-dependent photocatalytic activity of BiOBr single-crystalline nanosheets, Chem.

Comm. 2014, DOI: 10.1039/C3CC41498A.

(24) J. Chen, M. Guan, W. Cai, J. Guo, C. Xiao, G. Zhang, The dominant {001} facet-

dependent enhanced visible-light photoactivity of ultrathin BiOBr nanosheets, Phys.

Chem. Chem. Phys. 2014, 16, 20909.

109

Chapter 6 Construction of 2D lateral pseudo-heterostructures by strain

engineering

6.1 Introduction

Discontinuities at interfaces in heterogeneous structures can lead to exciting and

possibly non-trivial properties, due to broken symmetries at the interfaces. Through

constructing heterostructures, one can feasibly engineer and manipulate electronic,

optical, and magnetic phases at so-called heterointerfaces, and thus, generate unusual

properties and new phenomena.[1,2] Numerous breakthroughs have been achieved by

taking advantage of vertical heterostructures in previous studies. For examples,

naturally formed heterostructures give birth to high-temperature superconductivity

(High-TC) and topological states in copper-oxide-based superconductors and

topological insulators, respectively.[3,4] Artificial heterostructures were also

constructed in laminated LaAlO3/SrTiO3 films, in which magnetic order and

superconductivity surprisingly coexist.[5,6] Very recently, fabrication of vertical van de

Waals (vdW) heterogeneous structures were successfully achieved by using atomically

thin layer materials, including graphene and h-BN, which drew immediate attention.[7-

10] Besides vertical heterostructures, lateral heterostructures with two materials joined

laterally have also drawn great attention.[11] The research in this field was boosted soon

after single-layer-thick 2D lateral heterostructures were successfully fabricated, as they

possess controllable band-offset tuning and can be used in electronics, optoelectronics,

and catalysis. Lateral interfaces in 2D lateral heterostructures are constructed from

covalent bonds and not linked by vdW forces. Less electron and phonon scattering

centers are expected across the interface. Electron hopping and band alignment in such

110

lateral heterostructures are therefore less affected by the interfaces, which, in turn,

promotes charge carrier transport across the lateral interfacial junctions. These

advantages of 2D lateral heterostructures have been observed and verified in

graphene/h-BN lateral heterogeneous structures. In these systems, high field-effect

mobility of charge carriers and low hysteresis behavior have been demonstrated.[12,13]

2D lateral heterostructures based on two different transition-metal dichalcogenides

(TMD) were also successfully fabricated, exhibiting lateral p-n diodes behaviors or

staggered band alignment. A broad range of applications based on 2D lateral TMD

heterostructures, therefore, have been proposed, such as logic circuits, field-effect

transistors, and photodetectors.[14-16]

Despite their appealing advantages, great challenges remain in constructing 2D

lateral heterostructures with the desired properties. A clean and atomically sharp

interface is the essential requirement to achieve exotic characteristics. To achieve this

goal, sophisticated and costly manufacturing methods such as MBE, lithography, and

chemical vapor deposition are necessary for fabricating these high-quality

interfaces.[11,17] It is still extremely difficult to achieve 2D lateral heterostructures with

a large area, instead, most epitaxially-grown samples possess small heterogeneous

domains. The other limitation originates from lattice mismatch at the lateral

heterointerface. The candidate materials for 2D lateral heterostructures must have

similar crystal structures and lattice constants. This is the reason why very few 2D

lateral heterostructures have been reported so far.[18] An alternative strategy to

overcome these limitations is to develop a “lateral pseudo-heterostructure”.[19,20]

Generally, the pseudo-heterostructures have a single chemical component but show

spatial variation in their physical properties, which offers a promising way to build 2D

111

systems that show the functionalities of 2D lateral heterostructures but overcome the

problems of lattice mismatch.

Here, we report a 2D lateral pseudo-heterostructure with an electronic heterogeneous

interface constructed from single-component BiOBr 2D nanosheets by strain

engineering. As a typical bismuth oxyhalide (BiOX, where X = Cl, Br, I),[21,22] BiOBr

has great potential for use in photoenergy conversion applications, owing to its indirect

band gap of 2.8 eV.[23-26] Taking advantage of their strain-sensitive electronic structure,

few-layer-thick BiOBr nanosheets were successfully prepared with controlled spatial

distributions of local electronic structures by manipulating the strain distribution. The

position and characteristics of the electronic heterogeneous interface can also be tuned

by adjusting the strain distribution in the nanosheets. Effective separation of charge

carriers at the electronic heterointerface is then demonstrated in this lateral pseudo-

heterostructure, owing to the appropriate band alignment at the electronic

heterogeneous interface. Its excellent photoresponse and enhanced photocurrent

suggest that such lateral pseudo-heterostructures are potential candidates for superior

optoelectronic devices.

6.2 Experimental section

To synthesize BiOBr nanosheets, 5 mmol Bismuth nitrate pentahydrate and 5 mmol

CTAB were, respectively, added into 100 mL distilled water at room temperature.

Sodium hydroxide solution (1 M) was added to the mixed solution to adjust the pH

value. The mixed solution was then stirred for 1 h, and poured into a 100 mL Teflon-

lined stainless autoclave up to 80% of the total volume. In the hydrothermal reaction,

the sealed autoclave was heated at 170 °C for 17 h, and then cooled in air. The resulting

precipitates were collected, washed with ethanol and deionized water several times,

112

and dried at 80 °C for 10 h. Bismuth nitrate pentahydrate, Sodium hydroxide, and

CTAB were purchased from Sinopharm Chemical Reagent Co., Ltd. (SCRC). All

reagents used in this work were of analytical grade and were used as received without

any further purification

The morphologies and microstructures of the as-prepared samples were characterized

by SEM (JEOL JSM-7500FA) and HAADF images were obtained by STEM (JEOL

JEM-ARM200F). A commercial AFM (Asylum Research MFP-3D) was used to

measure the morphology and surface potential of the BiOBr nanosheets by scanning

Kelvin probe microscopy (SKPM). A Pt/Ir coated n-silicon probe with resonance

frequency of 45-115 kHz and force constant of 0.5-0.95 M/m was used in the AFM

measurements. BiOBr powders were distributed on gold (50 nm) coated silicon

substrates. Before characterization of the surface potential of BiOBr samples, the

surface potential of the Pt/Ir tip was calibrated on a standard Si wafer coated with Au

film. Raman mapping investigations were carried out with a confocal Raman

spectrometer (Horiba Xplora) using a 532 nm wavelength laser as the excitation source.

The first-principles calculations were performed using the VASP code. The GGA was

applied to treat the exchange correlation energy with the PBE functional. The PAW

method was employed to describe the electron-ion interactions. A k-point sampling of

a 9 × 9 × 6 grid was generated with original Gamma-centred meshes. The cut-off energy

for the plane wave basis was 550 eV. The biaxial strain simulations were realized by

fixing the x and y axes and optimizing the z axis. Equilibrium geometries were obtained

by the minimum energy principle.

A pair of electrical contacts (Ti/Au, 5/50 nm thick) was fabricated on the a SiO2/Si

surface, by standard electron beam lithography, using a poly(methyl methacrylate)

resist and the lift-off method. BiOBr nanosheets were dispersed in ethanol and then

113

dropped on the as-prepared surface followed by air-drying at 60 °C. Flakes suitable for

electrical characterization were identified by an optical microscope. Photocurrent

measurements were performed in vacuum at room temperature, using a probe station

connected to a semiconductor parameter analyser (Agilent B1500). The drain bias of

10 V and a gate bias of 80 V have been applied. The wavelength of the light source

was 450 nm. The power of the light was measured to be 50 mW·cm-2.

6.3 Structure characterization of BiOBr nanosheets

Figure 6.1 (a) AFM images of BiOBr-circle nanosheets. (b) and (c) AFM images of

BiOBr nanosheets synthesized with pH values between 7 and 2. (d) AFM images of

BiOBr-circle nanosheets. Scale bars are 1μm.

The morphologies of the BiOBr nanosheets were revealed by SEM and AFM. Figure

6.1 shows the evolution of the morphology of BiOBr nanosheet, which undergo a shape

transition from square to circle. They also indicate that BiOBr nanosheets are

approximately several micrometers in size. The concentration of hydroxide is thus

expected to be the key factor that determines the shape of the 2D BiOBr nanosheets.

Despite their different shapes, all the nanosheets display a layered structure, which

is confirmed in the cross-section TEM images in Figure 6.2. The thickness of the

BiOBr-square and BiOBr-circle show similar thickness of 15-40 nm, as shown in

114

Figure 6.3. The large area-to-thickness ratio reflects their 2D nature. The interlayer

distances for the BiOBr-square and BiOBr-circle nanosheets are ~0.76 nm and ~0.78

nm, respectively, which correspond to the spacing of BiOBr (001) face. This indicates

that the BiOBr nanosheets have a preferred growth orientation normal along the [001]

direction.

Figure 6.2 (a) SEM image of BiOBr-square nanosheets. (b) and (c) Cross-sectional

TEM images of a single BiOBr-square nanosheet. (d) SEM image of BiOBr-circle

nanosheets. (e) and (f) Cross-sectional TEM of a single BiOBr-circle nanosheet.

115

Figure 6.3. (a) AFM image of BiOBr-square nanosheets. (b) The height profile

corresponding to the black line in (a). (c) AFM image of BiOBr-circle nanosheets. (d)

The height profile corresponding to the black line in (c).

Figure 6.4 XRD patterns of BiOBr-square and BiOBr-circle powders.

116

Figure 6.5 EDS mapping of elements (Bi, O, Br) distribution of BiOBr-square and

BiOBr-circle nanosheets. All elements uniformly distribute across the surface.

Our XRD, SEM, TEM, and EDS measurements verify that all BiOBr nanosheets

are pure phase without any detectable impurities or structural defects, as shown in

Figure 6.4 and Figure 6.5.

6.4 Strain induced pseudo-heterostructure

Figure 6.6 shows STEM images of BiOBr nanosheets with both square and circular

shape that were collected in HAADF mode. The atomic arrangements of the nanosheets

are clearly revealed and demonstrate their 2D single-crystal nature. Interestingly, these

BiOBr nanosheets have distinct local atomic features associated with their shapes,

especially in the areas close to the edges of the nanosheets. A homogeneous atomic

arrangement is observed across the whole BiOBr-square nanosheet, including edge

areas, as shown in Figure 6.6 (b). No obvious difference has been found in the atomic

fringes across the whole square nanosheet. The corresponding FFT pattern is shown in

Figure 6.6 (c) (pattern A across the whole nanosheet). Only one set of patterns can be

observed. In contrast, the BiOBr-circle sample exhibits an inhomogeneous atomic

arrangement across the nanosheet in Figure 6.6 (e). The areas close to the edge exhibit

relaxation in their atomic structures, although the central area retains a similar atomic

117

arrangement to that in the square nanosheet. The FFT pattern, shown in Figure 6.6 (g),

exhibits a group of symmetric spots (pattern B, labelled in green) in addition to the

intrinsic spots (pattern A, labelled in yellow). These spots of pattern B with a shorter

nearest-neighbor spot distance are attributed to the area close to the edge (Area B) in

Figure 6.6 (f). Their presence reflects the considerable lattice relaxation exists in the

areas close to the edges, as illustrated in Figures 6.7.

Figure 6.6. (a) HAADF image of BiOBr-square, showing a homogeneous crystal

structure (inset: lower magnification to show whole nanosheet). (b) Enlarged image

from the selected area in (a). (c) Corresponding FFT pattern of the area in (b). (d) Strain

mapping of BiOBr-square by GPA. (e) HAADF image of BiOBr-circle, showing an

interface of the pseudo-heterostructure (inset: lower magnification to show whole

nanosheet). (f) Enlarged image from the selected area in (e). (g) Corresponding FFT

pattern of the area in (f). (h) Strain mapping of BiOBr-square by GPA. In the image of

GPA, the blue zones are under compressive strain, the red zones are under tensile strain,

and the green zones are not strained.

118

Figure 6.7 Schematic diagram of the transition of the shape in (001) face exposed

BiOBr nanosheets. (a) BiBOr-square with most area of the edge of BiOBr-square are

terminated by {110} facets and the other small edge area terminated by {100} facets.

(b) BiOBr-circle nanosheet, without any specific terminated faces. Bi atoms and O

atoms are shown as purple and red balls respectively.

These relaxed areas induce a significantly inhomogeneous distribution of inner strain

in the BiOBr-circle nanosheets. It should be noted that a sharp interface can be

observed between Area A and Area B in the BiOBr-circle nanosheets, as indicated by

the dashed lines in Figure 6.6 (e) and (f). This interface can be regarded as a lateral

pseudo-heterointerface, because both areas have identical chemical component

composition but distinctly different inner strains. (see more TEM images in Figure 6.8

and Figure 6.9) In order to revealing the detailed local strain distribution, we used the

GPA method to study both square and circle nanosheets.[27,28] Figure 6.6 (d) and (h)

show GPA strain mapping results for the BiOBr-square and the BiOBr-circle

nanosheets, respectively. As shown in Figure 6.6 (d), the distribution of local strain is

uniform despite the subtle fluctuation in the BiOBr-square nanosheets. Whereas,

Figure 6.6 (h) shows that the inner strains of Area A and Area B in the BiOBr-circle

nanosheet are different, with a clear interface between these two areas (see more results

119

in Figure 6.10). Our high resolution TEM (HRTEM) images and GPA suggest that the

strain distribution in the BiOBr nanosheets can be manipulated through controlling the

concentration of hydroxide (pH value) in the hydrothermal reactions. Lateral pseudo-

heterostructures based on inhomogeneous strain distribution can be obtained in a single

BiOBr nanosheet.

Figure 6.8 (a) HRTEM images of BiOBr-circle nanosheets, showing clear interfaces

between edge area and central area. (b) HRTEM images of BiOBr-square nanosheets.

No interfaces are observed in BiOBr-square nanosheets.

Figure 6.9 (a)-(d) TEM images of BiOBr nanosheets showing the change of shape from

square to circle with the pH value change from 7 to 2. (e)-(f) HRTEM of BiOBr

nanosheets with intermediate shapes between square and circle.

120

Figure 6.10 Strain distribution of BiOBr nanosheets obtained by GPA. (a) HAADF

image of BiOBr-square nanosheet. (b)-(d) Homogeneous strain distribution of BiOBr-

square. (e) HAADF image of BiOBr-circle nanosheet. (f)-(h) Inhomogeneous strain

distribution of BiOBr-circle, showing a clear interface at the pseudo-heterointerface.

Finally, based on the above experimental results, we propose a possible formation

mechanism of the lateral pseudo-heterostructure by reliving the inner strain in BiOBr-

circle nanosheets. In this synthesis method, the reactions could be described as follows

which were proposed by M. Shang et al.[29]

Bi3+ + C16H33(CH3)3N-Br → C16H33(CH3)3N-Br-Bi3+ (6.1)

C16H33(CH3)3N-Br-Bi3+ + OH- → BiOBr + C16H33(CH3)3N+ + H+ (6.2)

CTAB is proposed to act as both the template where BiOBr microplates nucleate and

grow, and the Br source. In the aqueous solution, as shown in the schematic diagram

121

of reaction procedures in Figure 6.11, CTAB is lamellar structure with the surface

terminated by Br- anions. Mixed Bi3+ cations combine with CTAB at the initial

positions of the Br- anions, which will further react with OH- anions and form

multilamellar BiOBr precursors, as illustrated in Equation 6.2. Through adjusting the

concentration of OH-, the reaction speed can be controlled. A higher concentration of

OH- will result in a higher reaction speed, and vice versa. Then with the assistance of

the energy (high temperature and pressure) in the hydrothermal reaction, BiOBr crystal

could nucleate and grow on templates.

Figure 6.11 The schematic diagram of the reaction process of BiOBr nanosheets.

122

In this process, BiOBr single crystal seed should take a tetragonal shape (square

shape) to maximize the expression of {001} facets and minimize the total surface

energy according to Wulff theorem, because {001} facets have the lowest surface

energy than other facets.[30-32] When the reaction speed is quick (high concentration of

OH- anions), as observed in TEM images, square nanosheets (BiOBr-square) can be

harvested, accompanying with large inner strain in BiOBr crystal. However, when the

reaction speed is slow (small concentration of OH- anions), the shape of BiOBr

nanosheets change to circle (BiOBr-circle), because in stress-free and initially defect-

free nanocrystalline materials edge relaxations and reorientations are effective ways to

release the inner strain. Therefore, even though the BiOBr nanosheets in two pH values

both have a layer structure with same crystal structure and exposed facets, but with

different growing speeds of the crystal, we could obtain BiOBr nanosheets with

different degrees of inner strain. Moreover, relieving the inner strain by relaxation near

the edge could create a lateral pseudo-heterostructure between the central and edge area

in BiOBr-circle nanosheets.

6.5 Electronic properties characterization

The local electronic structures in 2D materials can be significantly affected by local

strain.[33,34] To characterize to effect of the pseudo-heterointerface on the local

electronic properties, we carried out the KPFM measurement, which can

simultaneously acquire the surface morphologies and local work functions of BiOBr

nanosheets.

123

Figure 6.12 (a) AFM image of BiOBr-square. (b) KPFM image of BiOBr-square. (c)

Corresponding line profiles of the lines in a and b, with the height as a black line and

the work function as a red line. (d) AFM image of BiOBr-circle. (e) KPFM image of

BiOBr-circle. (f) The corresponding line profiles of the lines in d and e, with the height

as a black line and work function as a red line.

As shown in Figure 6.12 (a) and (d), both square and circle BiOBr nanosheets present

a flat surface. The morphologies are consistent with TEM results. Figure 6.12 (b) and

(e) show the local work function mapping images acquired simultaneously with AFM

measurements on square and circle BiOBr nanosheets, respectively. By measuring the

CPD between the conductive tip with a constant work function and the surface, KPFM

can precisely determine the work function of materials with high resolution.[35] The

BiOBr-square nanosheet shows a homogeneous work function distribution across its

entire surface, as demonstrated by the line profiles in Figure 6.12 (c). In contrast, the

work function on the BiOBr-circle nanosheet exhibits an apparent variation between

the area near the edge and the central area, as shown in Figure 6.12 (e) and the line

profile in Figure 6.12 (f). We conjecture that the variation in the work function might

be attributable to distinct local electronic structures induced by the inhomogeneous

124

distribution of local strains beside the lateral pseudo-heterointerface in BiOBr-circle

nanosheets. This was also observed in the BiOBr nanosheets with intermediate shapes,

as shown in Figure 6.13.

Figure 6.13 (a) The AFM image of the BiOBr-circle nanosheet with the shape between

square and circle. (b) The corresponding KPFM image of the BiOBr nanosheet with

only the circle area (indicated by green arrows) showing anisotropic charge

distribution. Scale bars are 1 μm.

In order to verify this hypothesis, confocal Raman spectroscopy was employed

(incident laser wavelength λ = 532 nm) to identify the correlation between the work

function and the local strain in BiOBr nanosheets, as shown in Figure 6.14. Two strong

phonon modes can be observed in the Raman spectra of the BiOBr samples. The Raman

peaks of BiOBr-square are located at 118.3 cm-1 and 160.5 cm-1, while the Raman

peaks of BiOBr-circle are located at 113.6 cm-1 and 160.5 cm-1. According to previous

reports, the peak around 113.6 cm-1 can be assigned to the A1g mode (internal Bi-Br

stretching), and the peak at 160.5 cm-1 can be assigned to the Eg mode (internal Bi-Br

stretching).[36] Although the Eg mode remains at 160.5 cm-1 for both samples, the A1g

peak of BiOBr-square possesses a blue shift of ~5 cm-1 from 113.6 cm-1 to 118.3 cm-1,

which suggests a large inner compressive strain. Raman mapping of the A1g peak on

125

both square and circle nanosheets were carried out to provide a straightforward way to

identify the strain distributions. As shown in the insets in Figure 6.14, the intensity of

the peak at 118.3 cm-1 for BiOBr-square was homogenously recorded across the whole

nanosheet, indicating the homogeneous distribution of strain. In contrast, for BiOBr-

circle, the intensity of the peak at 113.6 cm-1 is much stronger near the edge area than

in the central area.

Figure 6.14 Raman spectra of BiOBr nanosheets, with BiOBr-square shown by the

black line and BiOBr-circle by the red line. Insets are the Raman mappings at 118.3

cm-1 of BiOBr-square (top) and 113.6 cm-1 of BiOBr-circle (bottom).

This observation is supported by the fact that the relaxation of strain sprang from the

edge area to the inner central area in BiOBr-circle nanosheets. It should be notice that

during the Raman mapping, the resolution of is limited by the size of the laser spot,

126

which is around 361 nm in this work. The spatial variation in the A1g peak is on a

smaller scale than the resolution of the instrument, and so there is some averaging

between each analysis spot, which will border the Raman activate area in the

measurement. These results demonstrate that the variation of the local electronic

structure can be directly correlated with its inhomogeneous strain distribution beside

the lateral pseudo-heterointerface in the BiOBr-circle nanosheet.

6.6 DFT calculation of the strain effect on the band structure of BiOBr

We carried out DFT calculations to simulate the strain dependence of the electronic

structure in BiOBr. Figure 6.15 (a) and (b) show DFT calculation results for the pristine

and strain-applied BiOBr. When an in-plane strain (for example, the compressive strain

here) is applied, the band structure and DOS of BiOBr are significantly modulated. The

band gap is decreased by 0.2 eV in BiOBr under compressive strain (2.12 eV for

pristine BiOBr). The CBM shifts from the Z point to the M point in the irreducible

Brillouin zone, and the VBM moves from its position between the R and Z points to a

position between the Z and A points. The most attractive feature in the electronic

structure is that the band structure around the CBM is more dispersive in strained

BiOBr than in pristine BiOBr. This suggests that the excited electrons in strained

BiOBr have a small effective mass and high mobility. As a result, the total DOS of

BiOBr can also be tuned by strain, as shown in Figure 6.15 (b), in which the partial

DOS contributed by the O 2p orbitals is significantly changed. The change in the

electronic structure can be ascribed to the symmetry transition in the BiOBr crystal

under strain, which is a common phenomenon in many other semiconductors.[37] The

DFT calculations indicate that the distinct electronic properties flanking the pseudo-

127

heterointerface in the BiOBr-circle nanosheets are indeed controlled by the local strain

distribution, which supports our experimental findings.

Figure 6.15 (a) Calculated band structure of BiOBr, with the pristine BiOBr shown by

the black solid line and the BiOBr with compressive strain shown by the red dashed

line. (b) Calculated DOS of BiOBr, with the pristine BiOBr shown by the solid line

and BiOBr with compressive strain shown by the dashed line.

The quantitative evaluation of strain effect on the electronic structures of BiOBr

were estimated based on the DFT calculations. As shown in Figure 6.16, the value of

Eg of BiOBr, which determines its ability of light absorption and photoresponse

threshold, varies remarkably under both compressive strain and tensile strain (vary

from 1.27 eV to 2.67 eV). Meanwhile, the effect mass of charge carries near the CBM

and VBM also exhibit strain-sensitive properties, according the result of DFT

calculations, which will greatly affect the transport properties of charge carriers. The

strain-sensitive electronic structure of BiOBr provides diversity in designing and

creating hetrostructures through strain engineering.

128

Figure 6. 16 (a) The Eg values of BiOBr under different degree of strain. (b) The effect

mass of the electron at CBM and hole at VBM under different degree of strain, which

is calculated from the second order of derivative of energy with respect to wave-vector.

6.7 Performance of BiOBr nanosheets in photoelectronic devices

Figure 6.17 (a) Photocurrent of individual BiOBr nanosheets. (b) Acquiring rise and

decay time of BiOBr-square by fitting the on/off curve, inset is an image of the

nanodevice. (c) Acquiring rise and decay time of BiOBr-circle by fitting the on/off

curve, inset is the image of the device.

To verify the ability of separating photoexcited charge carriers in BiOBr nanosheets,

photodetectors were fabricated on BiOBr-square and BiOBr-circle nanosheets,

respectively, via electron-beam lithography. We measured the photocurrent and

photoresponse of both BiOBr nanosheets. As shown in Figure 6.16 (a), under visible

light irradiation (450 nm, 50 mW∙cm-2), photocurrent was generated in both BiOBr-

square and BiOBr-circle nanosheets. The photocurrent generated in the BiOBr-circle

129

nanosheet is about 2 nA, which is nearly one order of magnitude larger than that of the

BiOBr-square nanosheet under the same experimental conditions.

In addition to the contrasting photocurrents in these two BiOBr nanosheets, a

difference in the photoresponse between BiOBr-square and BiOBr-circle nanosheets

can be observed. Generally, the dynamic response of the rise (Equation 6.3) and decay

(Equations 6.4 and 6.5) of photocurrent in BiOBr nanosheets can be fitted by the

following stretched exponential functions:

I = I0 - I0e(

-t

τγ)γ

(6.3)

I = I0e(

-t

τd)γ

(6.4)

I = AI0e(

-t

τd)γ

+ BI0𝑒(

−𝑡

𝜏′𝑑)𝛾′

(6.5)

where τr and τd are the relaxation time constants of the rise and decay, respectively.

The parameter γ falls into the range of 0 to 1. As γ approaches 1, the function

approaches classic single-exponential behavior without stretching. We have fitted the

photoresponse curves of the BiOBr nanosheets to reveal the dynamic responses under

visible light irradiation. As shown in Figure 6.17 (b), the τr for BiOBr-square is

calculated to be 0.24 s. The value of γ is calculated as 1, which suggests that the

generation of photocurrent in BiOBr-square under light irradiation is dominated by the

separation of photoexcited electron-hole pairs. The decay of photocurrent for BiOBr-

square can be described by a stretched exponential function (Equation 6.4), where the

value of τd is 0.22 s and the value of γ is 0.16, suggesting that the recombination of

photocurrent involves multiple energy processes, which may origin from the strain-

induced distortion of the crystal lattice. For BiOBr-circle, as shown in Figure 6.17 (c),

the rise of photocurrent could also be described using single exponential functions with

τr of 0.23 s, which is almost the same as for BiOBr-square. To describe the decay of

130

photocurrent in BiOBr-circle, however, at least two components are necessary to fit the

persistent photocurrent decay, where two exponential functions (Equation 6.5) are used

to separately analyze two different photo-relaxation processes. The τd of the fast

process was determined to be 0.12 s with γ as 0.29, and the τ′d of the slow process was

calculated to be 0.72 s with γ as 0.05. During the photocurrent decay, it is well known

that photoexcited electron-hole recombination dominates the fast decay process in the

photocurrent. Hence, the photocurrent decreases very rapidly in the initial stage. The

slow process is caused by a mechanism in which the photoexcited carriers are trapped

and spatially separated by local potential fluctuations, which can suppress the electron-

hole recombination. For our samples, because both the BiOBr-square and BiOBr-circle

nanosheets exhibit defect-free characteristics, supported by the homogeneous element

distribution across the whole nanosheets. Therefore, it is believed that the slow

relaxation process in BiOBr-circle, which reflects the longer lifetime of photoexcited

charge carriers, is governed by the strain-induced pseudo-heterointerface.

The responsivity (Rλ, Equation 6.6), and the external quantum efficiency (EQE,

Equation 5.7) are two important measures for photoconductors and photodiodes.

(6.6)

(6.7)

where Iλ represents the photocurrent (Iillumination − Idark), Pλ is the light intensity, S is the

effective device area under illumination, and λ, h, c, and e are the wavelength of

incident light, the Planck constant, the speed of light, and the charge of the electron,

respectively. Based on photocurrent results, the values of Rλ are 0.28 A/W and 1.29

A/W for BiOBr-square and BiOBr-circle, respectively. The corresponding EQE are,

therefore, calculated as 0.77 × 102 % and 3.56 × 102 % (all parameters listed in Table

6.1), for BiOBr-square and BiOBr-circle, respectively.

131

Table 6.1 Calculated R and EQE for BiOBr nanosheets.

Sample (nm) P (mW∙cm-2) I (nA) S (m2) R (A/W) EQE (%)

BiOBr-square 450 50 0.26 1.83 0.28 0.77 102

BiOBr-circle 450 50 1.80 2.80 1.29 3.56 102

Figure 6.18 Schematic diagram of the band alignment at the pseudo-heterointerface in

BiOBr-circle, which promotes the separation of photoexcited carriers.

We attribute the significant enhancement of the photocurrent in the BiOBr-circle

nanosheets to improved separation efficiency of photoexcited charge carriers, which is

illustrated in Figure 6.18. In the BiOBr-circle nanosheets, the inhomogeneous work

function () distribution beside the lateral pseudo-heterointerface induces band

bending across the interface.[38] Therefore, under light irradiation, photoexcited

electrons and holes will be spatially driven by this electric field and separate to two

different regions in the nanosheet, which will prolong their lifetime and consequently

132

enhance the light harvesting efficiency in the BiOBr-circle nanosheets. In contrast,

there are not any pseudo-heterointerface in BiOBr-square nanosheets. Therefore, we

do not expect the same scenario occurring in square nanosheets.

6.8 Summary

To summarize, the lateral pseudo-heterostructure on individual BiOBr nanosheets

was fabricated by strain engineering. The local strain distribution has been revealed by

TEM and Raman mapping. The dependence of local strain on electronic structures has

been revealed by DFT calculations. An inhomogeneous work function distribution is

observed across the lateral pseudo-heterostructure, which favors the separation of

photoexcited charge carriers under visible light irradiation. Enhanced photocurrent (by

one order of magnitude) has been achieved in BiOBr pseudo-heterostructure-based

photodetectors, which is attributed to the appropriate band alignment across the

pseudo-heterointerface. The growth of such lateral pseudo-heterostructures, thus,

offers a promising way to build systems that show the functionalities of 2D lateral

heterostructures but overcome the problems of lattice mismatch.

6.9 References

(1) H. Y. Hwang, Y. Iwasa, M. Kawasaki, B. Keimer, N. Nagaosa, Y. Tokura,

Emergent phenomena at oxide interfaces, Nat. Mater. 2012, 11, 103.

(2) Z. I. Alferov, Nobel lecture: the double heterostructure concept and its applications

in physics, electronics, and technology, Rev. Mod. Phys. 2001, 73, 767.

(3) B. Keimer, S. A. Kivelson, M. R. Norman, S. Uchida, J. Zaanen, From

quantummatter to high-temperature superconductivity in copper oxides, Nature 2015,

518, 179.

133

(4) K. Nakayama, K. Eto, Y. Tanaka, T. Sato, S. Souma, T. Takahashi, K. Segawa, Y.

Ando, Manipulation of topological states and the bulk band gap using natural

heterostructures of a topological insulator, Phys. Rev. Lett. 2012, 109, 236804.

(5) A. Ohtomo, H. Y. Hwang, A high-mobility electron gas at the LaAlO3/SrTiO3

heterointerface, Nature 2004, 427, 423.

(6) L. Li, C. Richter, J. Mannhart, R. C. Ashoori, Coexistence of magnetic order and

two-dimensional superconductivity at LaAlO3/SrTiO3 interfaces, Nat. Phys. 2011, 7,

762.

(7) C. R. Dean, A. F. Young, I. Meric, C. Lee, L. Wang, S. Sorgenfrei, K. Watanabe,

T. Taniguchi, P. Kim, K. L. Shepard, J. Hone, Boron nitride substrates for high-quality

graphene electronics, Nat. Nanotechnol. 2010, 5, 722.

(8) M. Wang, S. K. Jang, W.-J. Jang, M. Kim, S.-Y. Park, S.-W. Kim, S.-J. Kahng, J.-

Y. Choi, R. S. Ruoff, Y. J. Song, S. A. Lee, Platform for large-scale graphene

electronics-CVD growth of single-layer graphene on CVD-grown hexagonal boron

nitride, Adv. Mater. 2013, 25, 2746.

(9) A. K. Geim, I. V. Grigorieva, Van der Waals heterostructures, Nature 2013, 499,

419.

(10) S. J. Haigh, A. Gholinia, R. Jalil, S. Romani, L. Britnell, D. C. Elias, K. S.

Novoselov, L. A. Ponomarenko, A. K. Geim, R. Gorbachev, Cross-sectional imaging

of individual layers and buried interfaces of graphene-based heterostructures and

superlattices, Nat. Mater. 2012, 11, 764.

(11) Y. Liu, N. O. Weiss, X. Duan, H.-C. Cheng, Y. Huang, X. Duan, Van der waals

heterostructures and devices, Nat. Rev. Mater. 2016, 1, 16042.

134

(12) L. Liu, J. Park, D. A. Siegel, K. F. McCarty, K. W. Clark, W. Deng, L. Basile, J.

C. Idrobo, A.-P. Li, G. Gu, Heteroepitaxial growth of two-dimensional hexagonal

boron nitride templated by graphene edges, Science 2014, 343, 163.

(13) M. P. Levendorf, C.-J. Kim, L. Brown, P. Y. Huang, R. W. Havener, D. A. Muller,

J. Park, Graphene and boron nitride lateral heterostructures for atomically thin

circuitry, Nature 2012, 488, 627.

(14) X. Duan, C. Wang, J. C. Shaw, R. Cheng, Y. Chen, H. Li, X. Wu, Y. Tang, Q.

Zhang, A. Pan, J. Jiang, R. Yu, Y. Huang, X. Duan, Lateral epitaxial growth of two-

dimensional layered semiconductor heterojunctions, Nat. Nanotechnol. 2014, 9, 1024.

(15) C. Huang, S. Wu, A. M. Sanchez, J. J. P. Peters, R. Beanland, J. S. Ross, P. Rivera,

W. Yao, D. H. Cobden, X. Xu, Lateral heterojunctions within monolayer MoSe2-WSe2

semiconductors, Nat. Mater. 2014, 13, 1096.

(16) X.-Q. Zhang, C.-H. Lin, Y.-W. Tseng, K.-H. Huang, Y.-H. Lee, Synthesis of

lateral heterostructures of semiconducting atomic layers, Nano Lett. 2015, 15, 410.

(17) K. S. Novoselov, A. Mishchenko, A. Carvalho, A. H. C. Neto, 2D materials and

van der Waals heterostructures, Science 2016, 353, 461.

(18) H. Wang, F. Liu, W. Fu, Z. Fang, W. Zhou, Z. Liu, Two-dimensional

heterostructures: fabrication, characterization, and application, Nanoscale 2014, 6,

12250.

(19) D. Nam, D. S. Sukhdeo, J.-H. Kang, J. Petykiewicz, J. H. Lee, W. S. Jung, J.

Vučković, M. L. Brongersma, K. C. Saraswat, Strain-induced pseudoheterostructure

nanowires confining carriers at room temperature with nanoscale-tunable band

profiles, Nano Lett. 2013, 13, 3118.

(20) J. Feng, X. Qian, C.-W. Huang, J. Li, Strain-engineered artificial atom as a broad-

spectrum solar energy funnel, Nat. Photonics 2012, 6, 866.

135

(21) H. An, Y. Du, T. Wang, C. Wang, W. Hao, J. Zhang, Photocatalytic properties of

BiOX (X = Cl, Br, and I), Rare Metals 2008, 27, 243.

(22) K.-L. Zhang, C.-M. Liu, F.-Q. Huang, C. Zheng, W.-D. Wang, Study of the

electronic structure and photocatalytic activity of the BiOCl photocatalyst, Appl. Catal.

B 2006, 68, 125.

(23) H. Cheng, B. Huang, Y. Dai, Engineering BiOX (X = Cl, Br, I) nanostructures for

highly efficient photocatalytic applications, Nanoscale 2014, 6, 2009.

(24) H. Li, J. Shang, Z. Ai, L. Zhang, Efficient visible light nitrogen fixation with

BiOBr nanosheets of oxygen vacancies on the exposed {001} facets, J. Am. Chem. Soc.

2015, 137, 6393.

(25) Z. Xu, L. Han, B. Lou, X. Zhang, S. Dong, High-performance BiOBr ultraviolet

photodetector fabricated by a green and facile interfacial self-assembly strategy,

Nanoscale 2014, 6, 145.

(26) H. Li, T. Hu, J. Liu, S. Song, N. Du, R. Zhang, W. Hou, Thickness-dependent

photocatalytic activity of bismuth oxybromide nanosheets with highly exposed (010)

facets, Appl. Catal. B 2016, 182, 431.

(27) V. Grillo, E. Rotunno, STEM_CELL: a software tool for electron microscopy: part

I simulations, Ultramicroscopy 2013, 125, 97.

(28) V. Grillo, F. Rossi, STEM_CELL: a software tool for electron microscopy. part 2

analysis of crystalline materials, Ultramicroscopy 2013, 125, 112.

(29) M. Shang, W. Wang, L. Zhang L, Preparation of BiOBr lamellar structure with

high photocatalytic activity by CTAB as Br source and template, J. Hazard. Mater.

2009, 167, 803.

(30) H. Zhang, L. Liu, Z. Zhou, First-principles studies on facet-dependent

Photocatalytic Properties of Bismuth Oxyhalides (BiOXs), RSC Adv. 2012, 2, 9224.

136

(31) J. Andrés, L. Gracia, A. F. Gouveia, M. M. Ferrer, E. Longo, Effects of surface

stability on the morphological transformation of metals and metal oxides as

investigated by first-principles calculations, Nanotechnology 2015, 26, 405703.

(32) P. Dandekar, Z. B. Kuvadia, M. F. Doherty, Engineering crystal morphology,

Annu. Rev. Mater. Res. 2013, 43, 359.

(33) F. Guinea, M. I. Katsnelson, A. K. Geim, Energy gaps and a zero-field quantum

hall effect in graphene by strain engineering, Nat. Phys. 2010, 6, 30.

(34) Z. Xu, W. Hao, Q. Zhang, Z. Fu, H. Feng, Y. Du, S. Dou, Indirect-direct band

transformation of few-layer BiOCl under biaxial strain, J. Phys. Chem. C 2016, 120,

8589.

(35) V. Palermo, M. Palma, P. Samorì, Electronic characterization of organic thin films

by kelvin probe force microscopy, Adv. Mater. 2006, 18, 145.

(36) J. E. D. Davies, Solid state vibrational spectroscopy - III [1] the infrared and raman

spectra of the bismuth (III) oxide halides, J. Inorg, Nucl. Chem. 1973, 35, 1531.

(37) K.-H. Hong, J. Kim, S.-H. Lee, J. K. Shin, Strain-driven electronic band structure

modulation of Si nanowires, Nano Lett. 2008, 8, 1135.

(38) A. Kahn, Fermi level, work function and vacuum level, Mater. Horiz. 2016, 3, 7.

137

Chapter 7 Summary and outlooks

OV defects were carefully introduced into rutile(110) single crystal by annealing in

vacuum conditions in the temperature range from 900 K to 1300 K, which were

checked by in-situ LT-STM. Isolated OV point defects were found on the rutile(110)

(1×1) surfaces, which are favorable to be occupied by water molecule or OH. At 1300

K, the surface experienced the evolution from (1×2) to (1×2) surfaces, which were

dominated by cross-links defects.

In further work, we demonstrate that inactive pure TiO2 rutile(110) single crystal can

be activated towards the HER by creating OV point defects. OVs and Ti3+ ions in the

surface region dominate the electrical conductivity of reduced TiO2 and the amount of

electrocatalytic active sites. Combining the well-characterized atomic surface structure

and theoretical calculations, we conclude that subOVs can promote the electron transfer

and hydrogen desorption in the electrocatalytic HER on reduced TiO2 in alkaline

media. It helps to elucidate the fundamental mechanism of the electrocatalytic activity

of reduced oxides towards the HER, which is of immense fundamental and practical

importance towards an in-depth understanding and rational optimization of TMOs as

electrocatalysts in alkaline media. Considering that all the electrocatalytic

characterization of TiO2 were measured using the single crystals, which could benefit

in controlling defect species and studying their roles in electrocatalytic reactions,

synthesis TiO2 materials in nano-scale will largely enhance the number of surface

active sites and improve their performances. In addition, TiO2(110) surface with (1×2)

reconstruction has a complicated surface structure and cross-links defects, which could

alter the surface electrocatalytic properties and electrical conductivity, providing an

opportunity to further tune and enhance the electrocatalytic activities of TiO2.

138

Another focus was by the mean of precisely tunning the growth conditions, the strain

can be controllably introduced into 2D BiOBr nanosheets with highly reactive {001}

facets exposed. It is found that the strain effect can effectively tune the photocatalytic

activity of BiOBr nanosheets in dye degradation. Our work suggests that strain

engineering could be an effective approach to controlling the electronic structure of

semiconductors for further enhancement of their efficiency in converting light into

other forms of energy.

By characterizing single nanosheets with STEM, lateral pseudo-heterostructure was

confirmed in BiOBr-circle nanosheets. The local strain distribution has been revealed

by STEM and GPA analysis. The dependence of local strain on electronic structures

has been revealed by DFT calculations. An inhomogeneous work function distribution

is observed across the lateral pseudo-heterostructure, which favors the separation of

photoexcited charge carriers under visible light irradiation. Enhanced photocurrent (by

one order of magnitude) has been achieved in BiOBr pseudo-heterostructure-based

photodetectors, which is attributed to the appropriate band alignment across the

pseudo-heterointerface. The growth of such lateral pseudo-heterostructures, thus,

offers a promising way to build systems that show the functionalities of 2D lateral

heterostructures but overcome the problems of lattice mismatch.

139

Appendix A: List of publications

(1) J. Zhuang, C. Liu, Q. Gao, Y. Liu, H. Feng, X. Xu, J. Wang, J. Zhao, S. X. Dou, Z.

Hu, Y. Du, Bandgap modulated by electronic superlattice in blue phosphorene, ACS

Nano 2018, 12, 5059.

(2) H. Feng, Z. Xu, L. Ren, C. Liu, J. Zhuang, Z. Hu, X. Xu, J. Chen, J. Wang, W.

Hao, Y. Du, S. X. Dou, Activating titania for efficient electrocatalysis by vacancy

engineering, ACS Catal. 2018, 8, 4288.

(3) J. Zhuang, C. Liu, Z. Zhou, G. Casillas, H. Feng, X. Xu, J. Wang, W. Hao, X.

Wang, S. X. Dou, Z. Hu, Y. Du, Dirac signature in germanene on semiconducting

substrate, Adv. Sci. 2018, 1800207.

(4) Z. Xu, K. Xu, H. Feng, Y. Du, W. Hao, Sp orbital hybridization: a strategy for

developing efficient photocatalysts with high carrier mobility, Sci. Bull. 2018, 63, 465.

(5) P. De Padova, H. Feng, J. Zhuang, Z. Li, A. Generosi, B. Paci, C. Ottaviani, C.

Quaresima, B. Olivieri, M. Krawiec, Y. Du, Synthesis of multilayer silicene on

Si(111)√3 × √3-Ag, J. Phys. Chem. C 2017, 121, 27182.

(6) H. Feng, J. Zhuang, A. D. Slattery, L. Wang, Z. Xu, X. Xu, D. Mitchell, T. Zheng,

S. Li, M. Higgins, L. Ren, Z. Sun, S. X. Dou, Y. Du, W. Hao, Construction of two-

dimensional lateral pseudoheterostructures by strain engineering, 2D Mater. 2017, 4,

025102.

(7) H. Feng, Y. Du, C. Wang, W. Hao, Efficient visible-light photocatalysts by

constructing dispersive energy band with nisotropic p and s-p hybridization states,

Curr. Opin. Green Sustain. Chem. 2017, 6, 93.

140

(8) Y. Liu, H. Feng, X. Yan, J. Wang, H. Yang, Y. Du, W. Hao, The origin of enhanced

photocatalytic activities of hydrogenated TiO2 nanoparticles, Dalton Trans. 2017, 46,

10694.

(9) X. Lou, J. Shang, L. Wang, H. Feng, W. Hao, T. Wang, Y. Du, Enhanced

photocatalytic activity of Bi24O31Br10: constructing heterojunction with BiOI, J. Mater.

Sci. Technol. 2017, 33, 281.

(10) Z. Li, H. Feng, J. Zhuang, N. Pu, L. Wang, X. Xu, W. Hao, Y. Du, Metal-silicene

interaction studied by scanning tunneling microscopy, J. Phys.: Condens. Matter 2016,

28, 034002.

(11) H. Feng, X. Xu, W. Hao, Y. Du, D. Tian, L. Jiang, Magnetic field actuated

manipulation and transfer of oil droplets on a stable underwater superoleophobic

surface, Phys. Chem. Chem. Phys. 2016, 18, 16202.

(12) H. Liu, H. Feng, Y. Du, J. Chen, K. Wu, J. Zhao, Point defects in epitaxial silicene

on Ag(111) surface, 2D Mater. 2016, 3, 025034.

(13) Y. Du, J. Zhuang, J. Wang, Z. Li, H. Liu, J. Zhao, X. Xu, H. Feng, L. Chen, K.

Wu, X. Wang, S. X. Dou, Quasi-freestanding epitaxial silicene on Ag(111) by oxygen

intercalation, Sci. Adv. 2016, 2, e1600067.

(14) L. Ren, J. Zhuang, G. Casillas, H. Feng, Y. Liu, X. Xu, Y. Liu, J. Chen, Y. Du, L.

Jiang, S. X. Dou, Nanodroplets for stretchable superconducting circuits, Adv. Funct.

Mater. 2016, 26, 8111.

(15) R. J. Wong, J. Scott, G. K. C. Low, H. Feng, Y. Du, J. N. Hart, R. Amal,

Investigating the effect of UV light pre-treatment on the oxygen activation capacity of

Au/TiO2, Catal. Sci. Technol. 2016, 6, 8188.

141

(16) Z. Xu, W. Hao, Q. Zhang, Z. Fu, H. Feng, Y. Du, S. Dou, Indirect-direct band

transformation of few-layer BiOCl under biaxial strain, J. Phys. Chem. C 2016, 120,

8589.

(17) H. Feng, Z. Xu, L. Wang, Y. Yu, D. Mitchell, D. Cui, X. Xu, J. Shi, T. Sannomiya,

Y. Du, W. Hao, S. Dou, Modulation of photocatalytic properties by strain in 2D BiOBr

nanosheets, ACS Appl. Mater. Interfaces 2015, 7, 27592.

(18) J. Zhuang, X. Xu, H. Feng, Z. Li, X. Wang, Y. Du, Honeycomb silicon: A review

of silicene, Sci. Bull. 2015, 60, 1551.

(19) X. Xu, J. Zhuang, Y. Du, H. Feng, N. Zhang, C. Liu, T. Lei, J. Wang, M. Spencer,

T. Morishita, X. Wang, SX. Dou, Effects of oxygen adsorption on the surface state of

epitaxial silicene on Ag(111), Sci. Rep. 2014, 4, 7543.

Appendix B: Conference contributions

1. 2018 international symposium on advanced materials and sustainable technology,

Brisbane, Oral presentation: H. Feng, W. Hao, Y. Du, S. X. Dou, Surface defect

engineering in semiconducting (photo)electrocatalyst, 2018.

2. The international conference on nanoscience and nanotechnology, Wollongong,

Poster section: H. Feng, Z. Xu, Z. Hu, W. Hao, S. X. Dou, Y. Du, Vacuum reduced

TiO2 for electrochemical HER in alkaline condition, 2018.

3. The 3rd international conference on 2D materials and technology, Singapore, Poster

section: H. Feng, L. Wang, Z. Xu, J. Zhuang, X. Xu, Y. Du, W. Hao, S. X. Dou,

Construction of 2D lateral pseudoheterostructures by strain engineering, 2017.

ICON-2DMAT Best poster award.

4. The 2nd international symposium on renewable energy technology, Sydney. Poster

section: H. Feng, Z. Xu, L. Wang, W. Hao, S. X. Dou, Y. Du, Modulation of

photocatalytic properties by strain in 2D BiOBr nanosheets, 2016. Best poster reward.