Fundamental Characterization Studies of Advanced Photocatalytic M PDF

Lehigh University
Lehigh Preserve
Theses and Dissertations
2013
Fundamental Characterization Studies of

Advanced Photocatalytic Materials
Somphonh Peter Phivilay
Lehigh University
Follow this and additional works at: http://preserve.lehigh.edu/etd

Recommended Citation
Phivilay, Somphonh Peter, "Fundamental Characterization Studies of Advanced Photocatalytic Materials" (2013). Theses and
Dissertations. Paper 1204.
This Dissertation is brought to you for free and open access by Lehigh Preserve. It has been accepted for inclusion in Theses and Dissertations by an
authorized administrator of Lehigh Preserve. For more information, please contact preserve@lehigh.edu.
Fundamental Characterization Studies

of Advanced Photocatalytic Materials
by
Somphonh Peter Phivilay
A Dissertation
Presented to the Graduate and Research Committee
of Lehigh University
in Candidacy for the Degree of
Doctor of Philosophy
in
Chemical Engineering
Lehigh University
January 2013
Copyright 2013 by Somphonh Peter Phivilay
ii
CERTIFICATE OF APPROVAL
Approved and recommended for acceptance as a dissertation in
partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
_______________________
Date
_______________________
Accepted Date
Dissertation Advisor:
_______________________
Dr. Israel E. Wachs
Lehigh University
Committee Members:
_______________________
Dr. Christopher J. Kiely
Lehigh University
_______________________
Dr. Bruce E. Koel
Princeton University
_______________________
Dr. Mark A. Snyder
Lehigh University
iii
Acknowledgements
I would like to give special thanks to my advisor Professor Israel Wachs for
giving me the opportunity to join his lab group and for his guidance throughout my
years in graduate school. His passionate attitude and knowledge about his research
field was inspirational and has opened my eyes to how research needs to be
approached and conducted. I am very fortunate and thankful that he gave me the
special chance to go travel to Japan since my dissertation would not have been
possible without that collaborative work and living in a different country was an
experience I will never forgot.
I would like to thank all the members in my lab group for their help and
support throughout the years: Liz Ross, Kamalakanta Routray, Chip Roberts, Kevin
Doura, Julie Molinari, Chris Keterakis, Jih Mirn Jehng, Mike Ford, Soe Lwin, Yadan
Tang and Mingui Zhu.
This dissertation wouldnt have been possible without the help from my
collaborators. I would like to thank Prof. Domen and his lab group for letting me
work in their lab and helping me to learn about photocatalysis research. I would like
to thank Alexander Puretsky at Oak Ridge National Laboratory for help in obtaining
photoluminescence spectroscopy data along with Chip Roberts. I would like to thank
Al Miller and Rob Pafchek at Lehigh University for help in obtaining HR-XPS and
HS-LEISS data.
iv
Most of all I would like to thank my friends and family for their loving
support, always encouraging and motivating me to do my best. Without my parents
and the rest of my family, I would not have accomplished as much as I have in my
life. This work is dedicated to them and I am thankful for everything that they have
given me.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
iv
vi
xi
xiii
ABSTRACT
CHAPTER 1
Overview of Heterogeneous Photocatalyst Literature for Water Splitting

1.1. Introduction
3
1.2. Basic Principle of Photocatalytic Water Splitting
5
1.3. Semiconductor Materials for Photocatalytic Water Splitting
7
1.4. Overview of Characterization Techniques Utilized for Heterogeneous
Mixed Oxide Photocatalyst Systems for Water Splitting
14
1.4.1 X-ray Diffraction (XRD)
15
1.4.2 X-ray Absorption Spectroscopy (XANES and EXAFS)
16
1.4.3 Electron Microscopy
17
1.4.4 Infrared (IR) spectroscopy
19
1.5 Approach and Outline of This Thesis
20
1.5.1 Raman Spectroscopy
21
1.5.2 Ultraviolet-visible (UV-vis) Diffuse Reflectance Spectroscopy
(DRS)
22
1.5.3 Photoluminescence (PL) Spectroscopy
23
1.5.4 High-Resolution X-ray Photoemission Spectroscopy
(HR-XPS)
24
1.5.5 High Sensitivity Low Energy Ion Scattering Spectroscopy
(HS-LEISS)
25
1.5.6 Photocatalytic Reactor for Water Splitting
27
1.5.7 Dissertation Outline
27
References
29
Figures
43
CHAPTER 2
46
Nature of Catalytic Active Sites Present on the Surface of Advanced Bulk

Tantalum Mixed Oxide Photocatalysts
vi
Abstract
2.1 Introduction
2.2. Experimental
2.2.1 Photocatalyst Synthesis
2.2.2 In Situ Raman Spectroscopy
2.2.3 In Situ UV-Vis Diffuse Reflectance Spectroscopy
2.2.4 Photoluminescence Spectroscopy and Lifetime Emissions
Decay
2.2.5 High Resolution X-ray Photoelectron Spectroscopy
(HR-XPS)
2.2.6 High Sensitivity Low Energy Ion Scattering (HS-LEIS)
Spectroscopy
2.3 Results
2.3.2 UV-Vis Diffuse Reflectance Spectroscopy
2.3.3 Photoluminescence Spectroscopy and PL lifetime decay
2.3.5 High Sensitivity-Low Energy Ion Scattering Spectroscopy
2.4 Discussion
2.4.1 Bulk Molecular and Electronic Structures of NaTaO3
Photocatalysts
2.4.2 Surface Composition of Promoted NaTaO3 Photocatalyst
2.4.3 Bulk and Surface Structures of the Promoted NaTaO3:La
Photocatalyst
2.4.4 Generation of Excited Electron/Hole Pairs and Their
Lifetimes
2.4.5 Structure-Activity Relationships for Splitting of H2O by
NaTaO3 Photocatalysts
2.5 Conclusions
Acknowledgements
References
Tables
Figures
CHAPTER 3
46
47
50
50
51
51
52
54
54
55
55
56
56
58
58
60
60
60
62
62
64
65
66
67
72
76
83
Fundamental Bulk/Surface Structure Photoactivity Relationships of

Supported (Rh2-yCryO3)/GaN Photocatalysts
Abstract
3.1 Introduction
3.2. Experimental
83
84
85
vii
3.2.1 Catalyst Synthesis

3.2.3 UV-Vis NIR Diffuse Reflectance Spectroscopy (DRS)
3.2.4 Photoluminescence Spectroscopy and Photoluminescent
Decay
(HR-XPS)
3.2.6 High Sensitivity Low Energy Ion Scattering (HS-LEIS)
Spectroscopy
3.3 Results and Discussion
3.3.1 Bulk Characteristics of GaN Photocatalysts
3.3.1.1 Bulk Molecular Structure
3.3.1.2 Bulk Electronic Structure
3.3.2 Surface Composition of GaN photocatalysts
3.3.2.1 Atomic Composition and Chemical State of the
Surface Region (~1-3 nm)
3.3.2.2 Outermost Atomic Layer Composition (~0.3 nm)
3.3.3 Electron/Hole Generation and Recombination
3.3.4 Bulk/Surface Photoactivity Relationships of GaN
Photocatalysts
3.4 Conclusions
Acknowledgements
References
Tables
Figures
CHAPTER 4
85
86
86
87
88
89
89
90
90
90
91
91
91
93
95
96
96
98
104
106
116
Nature of Catalytic Active Surface Sites Present on Bulk (Ga1-xZnx)(N1-xOx)

Photocatalysts
Abstract
4.1 Introduction
4.2. Experimental
4.2.4 High Resolution X-ray Photoelectron (HR-XPS)
Spectroscopy
4.2.5 High Sensitivity Low Energy Ion Scattering (HS-LEISS)
Spectroscopy
4.2.6 Photoluminescence Spectroscopy and Photoluminescent
Decay
viii
116
117
121
121
122
123
124
124
125
4.2.7 Photocatalytic Water Splitting

126
4.3 Results
127
4.3.1 Bulk Molecular Structures
127
4.3.2 Bulk Electronic Structures
128
4.3.3 Atomic Composition of Surface Region (~1-3nm)
129
4.3.4 Atomic Composition of Outermost Surface Layer (~0.3nm) 131
4.3.5 Dynamics of Photoexcited Electrons and Holes
134
136
4.4 Discussion
136
4.4.1 Bulk and Electronic Structures of the (Ga1xZnx)(N1xOx)
Oxynitride Support
137
4.4.2 Atomic Distribution of the Surface Region of Supported
(Ga1xZnx)(N1xOx) Photocatalysts
137
4.4.2.1 Model of the Supported (Rh2-yCryO3)/
(Ga1xZnx)(N1xOx) Mixed Oxide Photocatalyst
137
4.4.2.2 Model of the Supported (Rh/Cr2O3)/
(Ga1xZnx)(N1xOx) Core/Shell Photocatalyst
139
4.4.3 Generation of Excited Electron/Hole Pairs and Their
Lifetimes
140
4.4.4 Structure-Photoactivity Relationships for Splitting of H2O by
142
4.4.4.1 Supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx)
Photocatalyst
142
4.4.4.2 Supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx)
Core/Shell Photocatalyst
143
4.5 Conclusions
144
Acknowledgements
145
References
146
Tables
153
Figures
157
CHAPTER 5
177
Investigating the Surface Nature of TaON Photocatalysts

Abstract
5.1 Introduction
5.2. Experimental
5.2.4 High Sensitivity-Low Energy Ion Scattering (HS-LEISS)
Spectroscopy
ix
177
178
179
180
180
180
181
5.2.5 High Resolution X-ray Photoelectron (HR-XPS)

Spectroscopy
5.3.1 Bulk Characteristics of TaON Photocatalysts
5.3.2 Surface Atomic Composition of TaON photocatalysts
5.3.2.1 Outermost Surface Layer (~0.3 nm)
5.3.2.2 Surface Region (~1-3 nm)
5.3.2.3 Surface Nature of TaON Photocatalysts
5.3.3 Structure-Photoactivity Relationships
5.4 Conclusions
Acknowledgements
References
Tables
Figures
CHAPTER 6
182
182
182
182
183
184
184
185
186
187
188
188
189
195
197
205
Conclusions and Future Studies

6.1 Conclusions
6.2 Future Studies
205
208
LIST OF TABLES
Chapter 2
Table 2.1: Edge Energy Values from UV-Vis DRS.
72
Table 2.2: Photoluminescence decay fit parameters for catalysts at 267 nm

excitation.
73
Table 2.3: XPS surface region atomic concentration of promoted

0.2%NiO/NaTaO3:2%La photocatalyst.
74
Table 2.4: Photoactivity of Tantalum Oxide Photocatalysts for Water Splitting

with UV Excitation (>270 nm).
75
Chapter 3
Table 3.1: Eg values from UV-vis DRS, and peak emission wavelengths from
PL spectroscopy of the GaN photocatalysts at 400 nm excitation.
104
Table 3.2: HR-XPS surface region atomic composition (~1-3nm) of GaN

photocatalysts.
105
Chapter 4
Table 4.1: Eg values, peak emission wavelengths, and decay fit parameters of the
catalysts at 400 nm excitation. The decay parameters displayed are those found in
the PL spectra taken in the emission range of the peak value.
153
Table 4.2: XPS surface region atomic composition (~1-3nm) of
(Ga1xZnx)(N1xOx).
154
Table 4.3: XPS surface region atomic composition (~1-3nm) of fresh and used
(Rh2-yCryO3) /(Ga1xZnx)(N1xOx) mixed oxide photocatalysts.
155
Table 4.4: HR-XPS surface region atomic composition (~1-3nm) of fresh and
used (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalysts.
156
Chapter 5
Table 5.1: Uv-vis DRS Edge Energy Values for Tantalum-based Photocatalysts. 195
xi
Table 5.2: HR-XPS Surface Region Atomic Composition of TaON

Photocatalysts.
xii
196
LIST OF FIGURES
Chapter 1
Figure 1.1: Basic Principles of Photocatalytic Water Splitting.
43
Figure 1.2: Schematic Energy Diagram of photocatalytic water-splitting systems:

(a) conventional one-step system and (b) two-step Z-Scheme system.
44
Figure 1.3: Schematic of Photocatalytic Reactor Designs.
45
Chapter 2
Figure 2.1: Raman Spectra of Tantalum-based Photocatalysts.
76
Figure 2.2: Photoluminescence Spectra at 267 nm Excitation for Tantalumbased Photocatalysts.
77
Figure 2.3: Photoluminescence Emissions Decay Curves for Tantalum-based

Photocatalysts.
78
Figure 2.4: XPS survey spectrum of surface region for 0.2%NiO/NaTaO3:2%La

promoted photocatalyst.
79
Figure 2.5: HS-LEIS spectra for 0.2%NiO/NaTaO3:2%La promoted photocatalyst
using (a) He+ ion gas and (b) Ne+ ion gas.
80
Figure 2.6: HS-LEIS Depth Profile for 0.2%NiO/NaTaO3:2%La promoted
photocatalyst using (a) He+ ion gas and (b) Ne+ ion gas.
81
Figure 2.7: Schematic Diagram of the Bulk and Surface of the Promoted
0.2%NiO/NaTaO3:2%La Photocatalyst.
82
Chapter 3
Figure 3.1: Raman spectra for GaN photocatalysts (532 nm).
106
Figure 3.2: XPS survey spectra of (a) GaN and (b) supported (Rh2-yCryO3)/GaN
photocatalysts.
107
xiii
Figure 3.3: HR-XPS spectra of (a) Cr 2p and (b) Rh 3d regions of the supported
(Rh/Cr2O3)/GaN photocatalyst.
108
Figure 3.4: HS-LEIS Depth Profile for GaN photocatalyst using He+ ion gas.
109
Figure 3.5: HS-LEIS spectra of untreated and atomic O treated supported

(Rh2-yCryO3)/GaN photocatalyst.
110
Figure 3.6: HS-LEIS depth profile for the supported (Rh2-yCryO3)/GaN

111
Figure 3.7: HS-LEIS Depth Profile of Cr and Rh for supported (Rh2-yCryO3) /GaN
photocatalyst using Ne+ ion gas.
112
Figure 3.8: PL emissions spectra for GaN photocatalysts (400 nm excitation).
113
Figure 3.9: PL lifetimes spectra for (a) GaN and (b) (Rh2-yCryO3)/GaN
photocatalysts at different emissions wavelengths (400 nm excitation).
114
Figure 3.10: Schematic of (Rh2-yCryO3)/GaN photocatalysts.
115
Chapter 4
Figure 4.1: Raman Spectra of -Ga2O3 and ZnO precursors (532 nm) under
ambient conditions.
157
Figure 4.2: Raman Spectra of (Ga1-xZnx)(N1-xOx) and (Rh2-yCryO3)

/(Ga1xZnx)(N1xOx) (432 nm) under ambient conditions.
158
Figure 4.3: XPS Survey spectra of (a) fresh and (b) used (Rh2-yCryO3)
/(Ga1xZnx)(N1xOx) photocatalysts.
159
Figure 4.4: HR-XPS spectra of Cr 2p and Rh 3d regions of (a) fresh and

(b) used supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) mixed oxide photocatalysts. 160
Figure 4.5: HR-XPS spectra of Cr 2p and Rh 3d regions of (a) fresh and
(b) used (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) photocatalysts.
161
Figure 4.6: HS-LEIS spectra for (Ga1xZnx)(N1xOx) photocatalyst using

(a) He+ ion gas and (b) Ne+ ion gas.
162
xiv
Figure 4.7: HS-LEIS Depth Profile for the fresh (Rh2-yCryO3)/(Ga1xZnx)(N1xOx)

163
Figure 4.8: HS-LEIS depth profile intensity ratios of elements/oxygen for the fresh
(Rh2-yCryO3)/(Ga1xZnx)(N1xOx) photocatalyst using He+ ion gas.
164
Figure 4.9: HS-LEIS Depth Profile of Cr and Rh for the fresh (Rh2-yCryO3)
/(Ga1xZnx)(N1xOx) photocatalyst using Ne+ ion gas.
165
Figure 4.10: HS-LEIS Depth Profile for the used supported (Rh2-yCryO3)
/(Ga1xZnx)(N1xOx) photocatalyst using (a) He+ ion gas and (b) Ne+ ion gas.
166
Figure 4.11: HS-LEIS Depth Profile of Cr and Rh for used supported

(Rh2-yCryO3)/(Ga1xZnx)(N1xOx) photocatalyst using Ne+ ion gas.
167
Figure 4.12: HS-LEIS Depth Profile for the fresh supported

(Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst using (a) He+ ion gas and
(b) Ne+ ion gas.
168
Figure 4.13: HS-LEIS Depth Profiles of Cr and Rh for the fresh supported
(Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst using Ne+ ion gas.
169
Figure 4.14: HS-LEIS Depth Profile for the used supported

(Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst using (a) He+ ion gas and
(b) Ne+ ion gas.
170
Figure 4.15: HS-LEIS Depth Profiles for Cr and Rh for the used supported
171
Figure 4.16: PL spectra of bulk -Ga2O3, ZnO and oxynitride catalysts at 400 nm
excitation.
172
Figure 4.17: PL decay curves for the photocatalysts.
173
Figure 4.18: Comparison of Photocatalytic Water Splitting Activity

( > 400 nm).
174
Figure 4.19: Schematic Model of Rh-Cr NPs for Co-Impregnation Synthesis

Method.
175
Figure 4.20: Schematic Model of Rh-Cr NPs for Photodeposition Synthesis

Method.
176
xv
Chapter 5
Figure 5.1: Raman Spectra of Tantalum-based Photocatalysts (532 nm).
197
Figure 5.2: HS-LEIS Depth Profile of the Bulk TaON Photocatalyst.
198
Figure 5.3: HS-LEIS spectra of Bulk TaON Photocatalyst Before and After
Atomic O Pretreatment.
199
Figure 5.4: HS-LEIS Depth Profile of the Supported RuO2/TaON Photocatalyst. 200
Figure 5.5: XPS Survey Spectra of Bulk TaON.
201
Figure 5.6: HR-XPS surface analysis of Rh 3d region for supported

0.5%RuO2/TaON photocatalysts.
202
Figure 5.7: HR-XPS surface analysis of Ta 4f region for supported

203
Figure 5.8: Schematic Model for the Surface Region of the Supported
0.5%RuO2/TaON photocatalyst.
204
xvi
Abstract
Fundamental Characterization Studies of Advanced
Photocatalytic Materials
Solar powered photocatalytic water splitting has been proposed as a method for
the production of sustainable, non-carbon hydrogen fuel.
Although much
technological progress has been achieved in recent years in the discovery of advanced
photocatalytic materials, the progress in the fundamental scientific understanding of
such novel, complex mixed oxide and oxynitride photocatalysts has significantly
lagged. One of the major reasons for this slow scientific progress is the limited
number of reported surface characterization studies of the complex bulk mixed oxide
and oxynitride photocatalyst systems. Although photocatalytic splitting of water by
bulk mixed oxide and oxynitride materials involves both bulk (generation of excited
electrons and holes) and surface phenomena (reaction of H2O with excited electrons
and holes at the surface), the photocatalysis community has almost completely ignored
the surface characteristics of such complex bulk photocatalysts and correlates the
photocatalytic properties with bulk properties.
Some of the most promising photocatalyst systems (NaTaO3, GaN, (Ga1xZnx)(N1-xOx)
and TaON) were investigated to establish fundamental bulk/surface
structure photoactivity relationships. The bulk molecular and electronic structures of

1
the photocatalysts were determined with Raman and UV-vis spectroscopy.

Photoluminescence (PL) and transient PL spectroscopy were provided insight into
how recombination of photogenerated electrons is related to the photocatalysis
activity.
The chemical states and atomic compositions of the surface region of the
photocatalysts were determined with high resolution X-ray photoelectron spectroscopy

(~1-3 nm) and high sensitivity-low energy ion scattering spectroscopy (~0.3 nm).
The new insights obtained from surface characterization clarified the role of La
and Ni promoters species for the NaTaO3 photocatalyst system. The La2O3 additive
was found to be a structural promoter that stabilizes small NaTaO3 nanoparticles
(NPs) and increases the surface area, but not affecting the specific photoactivity. Only
the NiO additive was found to enhance the photoactivity due to the ability of surface
NiOx species to trap photogenerated electrons. The supported Rh-Cr NPs on GaN and
(Ga1-xZnx)(N1-xOx) photocatalysts consist of Rh+3 which is the catalytic photoactive
sites for H2 along with Cr+3, GaZnO or at their contact points being possible sites for
O2 production. The RuO2 promoted TaON photocatalyst system was found to consist
of dissolved Ru+4 cations in the TaOx thin film and not as RuO2 NPs as previously
proposed.
In summary, these current studies for the first time revealed the surface nature
of mixed oxide and oxynitride photocatalysts and stress the importance of establishing
fundamental
bulk/surface
structurephotoactivity
multicomponent photocatalyst systems.
relationships
for
complex,
CHAPTER 1
Overview of Heterogeneous Photocatalyst Literature for
Water Splitting
1.1 Introduction
One of the great challenges facing society today is finding alternative
renewable energy resources to help alleviate our reliance upon fossil fuels for energy.
Hydrogen from non-carbon sources is considered to be one of the potential candidates
for replacing fossil fuels for our energy needs. The use of sustainable hydrogen will
help to address the depletion of the fossil fuel supply and the environmental problems
associated with its use. However, there are still considerable barriers that need to be
overcome before hydrogen can be considered a viable energy source such as
constructing infrastructure for production, storage and use of hydrogen fuel and
finding ways to make sustainable hydrogen economically competitive against other
sources of energy [1]. There needs to be a strong emphasize on the research front so
that technology can be developed that can overcome these challenging entry barriers.
This critical research work needs to be accomplished before the dream of a hydrogen
economy can become a reality. Replacing fossil fuels with hydrogen for our energy
use is an ineffectual endeavor unless it can be derived from renewable resources.
Naturally abundant water and solar energy can be the renewable resources that
have the potential to produce hydrogen fuel. The sun produces over 1000 W/m2 on
the earths surface that far exceeds all of our energy needs [2, 3]. This vision for a
renewable hydrogen economy is inspired from biological processes found in nature
that have been using sunlight to convert water and carbon dioxide into glucose and
oxygen. The photosynthesis process is directly responsible for sustaining all forms of
life on Earth.
Mimicking this process for production of hydrogen fuel will help to
sustain our ever growing energy needs for our society. In this regard, photocatalytic
water splitting for hydrogen production is viewed as a form of artificial photosynthesis
where a catalyst material can utilize solar energy to convert water into H2 and O2.
Solar hydrogen fuel generated from water can be the ultimate in sustainable green
energy due to the use of inexpensive naturally abundant H2O reactants and lack of
environmentally damaging byproducts being formed in the reaction. Because of this it
is considered one of the Holy Grails of chemistry [4] and can be the basis of our
hydrogen economy in the future.
Photocatalytic water splitting was first brought to attention by the pioneering
work of Fujishima and Honda in 1972 which showed that UV light irradiation of a
TiO2 photoelectrode and a Pt photoelectrode in an aqueous solution led to the
production of H2 and O2 when a small voltage was applied [5]. The system in this
study is considered to be the first photoelectrochemical (PEC) cell designed for water
4
splitting. Since then, much progress has been made and many PEC cells have been
developed which can utilize visible light energy [6].
The designs of these PEC
systems are rather complicated due to photoelectrode instability that necessitates the
use of multi-component materials in the PEC system, i.e. integrated multilayers and
tandem systems [7-9]. The discovery of the Honda-Fujishima effect has provided the
inspiration for the utilization of heterogeneous semiconductor materials for
photocatalytic water splitting.
Unlike PEC cells, the use of heterogeneous
semiconductor materials allows for greater simplicity in design for photocatalytic

water splitting systems. These systems can be utilized using only the semiconductor
powder photocatalyst, water and sunlight which is advantageous since it does not
require an external bias or voltage source like PEC cells. The simplicity in design
makes photocatalytic water splitting by heterogeneous semiconductors viable for
implementation of solar hydrogen production on a large scale.
Although
photocatalytic water splitting is still in the research phase at this time, there is great
promise that this technology can be used to help solve our global energy problem.
1.2 Basic Principle of Photocatalytic Water Splitting
The basic reaction for photocatalytic water splitting given below in equation 1
is considered an uphill reaction due to the large Gibbs free energy involved.
G=237 kJ/mol
(1)
The schematic of the three step photocatalytic water splitting process is shown in
Figure 1. The first step of the process involves the absorption of photons in the bulk
of the photocatalyst system to form electron/hole pairs caused by the excitation of the
electrons from the valence band to the conduction band with the holes being left in the
valence band. The photo-generation of electron/hole pairs takes place on the time
scale of femtoseconds [10, 11]. The minimum theoretical energy needed is 1.23 eV
for the photocatalyst that requires light with wavelength of about 1010 nm, but greater
energy is needed to overcome the activation barrier. The bulk band gap energy of the
photocatalyst helps to determine the energy of the light source needed to form the
electron/hole pairs. The second step of the photocatalytic process deals with charge
separation of the generated electron and hole pairs followed by migration to the
surface reaction sites. A great challenge to the photocatalytic process is to minimize
the immediate recombination of electron/hole pairs before the charge separation
occurs. Transient photoluminescence spectroscopy has determined that about 90% or
more of the photogenerated electron/hole pairs recombine within 10 ns of excitation,
which limits the effective quantum yield to 10% or less on TiO2 photocatalysts [12].
Defects in the crystal structure can act as recombination centers in the photocatalyst so
highly ordered crystalline materials are desired [13]. Particle size also has a great
effect on charge transfer since smaller particles lead to shorter migration distances and
a lower probability of electron/hole recombination [13]. After the electron/hole pairs
migrate to the surface of the photocatalysts, the third and final step of the process is
the coupled surface chemical reactions at the catalytic active sites where the excited
6
electrons react with H+ to form H2 at one catalytic active site and excited holes help to
oxidize water into O2 at a different catalytic active site as depicted in Figure 1.
A co-catalyst, which is commonly used terminology in the literature, is
typically deposited on the photocatalyst that makes the photocatalyst active or
significantly increases the activity of the photocatalyst.
It is not always clear which
component in a multi-component photocatalyst system is the catalytic active site so it

is not appropriate to call one component the photocatalyst and another component
the co-catalyst. In order to discern the functions of the different components in the
photocatalyst system, it is more appropriate to view the bulk component as the
electron/hole generator and the active surface components as the catalytic active sites
in a photocatalytic system, which is analogous to supported metal oxide catalysts
where the catalytic active component is deposited on a high surface area support [14,
15]. This would help to clarify the role of each component in the photocatalyst
system. One of the main difficulties of finding appropriate component materials for
photocatalytic water splitting is the balance between finding suitable bulk
molecular/electronic properties and surface molecular/electronic properties of a
photocatalytic system.
1.3 Semiconductor Materials for Photocatalytic Water Splitting
Since TiO2 was first used by Fujishima and Honda for their PEC cell for water
splitting, it has become the most widely studied photocatalyst to date [16]. The main
difficulty with using TiO2 for photocatalytic water splitting is the low activity of H2
7
production and the undetectable amounts of O2 evolution attributed to adsorbed O2

species on the TiO2 [3, 17]. Only with the addition of another catalyst component
such as Pt [18] or in the presence of alkali carbonates [19] has stoichiometric amounts
of H2 and O2 been produced using TiO2 as the bulk electron/hole generator. TiO2
photocatalyst applications have focused more on environmental applications such as
photocatalytic oxidation of organic toxins in water and for air purification [16].
Over the years, many other metal oxide semiconductor materials have been
discovered (>130) which are much more efficient for photocatalytic water splitting
compared to TiO2 [3, 13, 20, 21].
From all of the discovered materials for
photocatalytic water splitting, only certain semiconductor materials possess suitable

bulk band gap structures which can generate electron/hole pairs that can be used for
photocatalytic water splitting. These semiconductor photocatalysts have been found to
have either empty d orbitals (defined as having d0 configuration) or filled d orbitals
(defined as having d10 configuration) [20]. Materials with partially filled d orbitals do
not possess a suitable electronic structure due to the partially filled d orbitals that act
as recombination centers for electron/hole pairs. Transition metal oxides that satisfy
the former criteria are those that contain Ti4+, Zr4+, Nb5+, Ta5+, and W6+ and main
group metal oxides that satisfy the latter criteria are those that contain Ga3+, In3+, Ge4+,
Sn4+, and Sb5+. Photocatalytic studies for water splitting under UV light excitation
have shown that trends in H2/O2 production follow the size of the transition metal
oxide band gap where the larger band gaps reflect an increasing thermodynamic
driving force for water splitting [3].
Tungstates (WO3, 2.8 eV) are less active than

8
titanates (TiO2, 3.0-3.1 eV) and are generally less active than niobates (Nb2O5, 3.1-3.5
eV), which in turn are generally less active than tantalates (Ta2O5, 4.0-4.6 eV). Pure
ZrO2 (5.0 eV) with its large band gap is able to photocatalytically split water [22-24]
but other zirconia-based photocatalysts have not been discovered like the other d0
transition metal oxides possibly due to poor absorption (< 248 nm) with commonly
used Xe and Hg light sources [3]. There have not been as many d10 main group metal
oxide photocatalysts which have been discovered compared to d0 transition metal
oxides and the d10 metal oxide photocatalyst are generally less active compared to d0
metal oxide photocatalyst.
Some of the effective photocatalysts consist of the bulk
mixed oxides of SrTiO3 [25], Ta2O5 [26], K2La2Ti3O10 [27], Sr2Nb2O7 [28],
K3Ta3Si2O13 [29], NaTaO3 [30], NaTaO3:La [31], RbTaWO6 [32], ZnGa2O4 [33],
CaIn2O4 [34], ZnGeO4 [35], Sr2SnO4 [36] and NaSbO3 [36].
Like TiO2-based
photocatalysts, the activity of these bulk mixed oxide photocatalysts have been shown
to be greatly enhanced by the presence of another catalyst component such as Pt,
Rh2O3, NiO and RuO2.
Out of all of these photocatalytic materials that are activated by UV
illumination, tantalum-based photocatalysts
have shown great
promise for
photocatalytic water splitting. Kudos group has demonstrated that tantalum-based

photocatalysts such as Ta2O5 [23, 26], NaTaO3 [30], K3Ta3Si2O13 [29], K2PrTa5O15
[37], Sr2Ta2O7 [38] and NiO/NaTaO3:La [31, 39] are among the most active
photocatalysts under UV-irradiation for the photocatalytic conversion of H2O into
H2/O2. The NiO/NaTaO3:La photocatalyst system is the most active photocatalyst
9
discovered to date with an apparent quantum efficiency (number of reacted electrons/

number of incident photons) of 56% using UV excitation with > 270 nm [31, 39].
The bulk NaTaO3 forms a perovskite crystal structure after calcination and is able to
photocatalytically split water into H2 and O2. The activity of the bare NaTaO3 was
improved by 2.6x by the addition of La, 12.8x by addition of NiO and 116.5x with the
simultaneous addition of both La and NiO. Transmission electron microscopy (TEM)
showed that the NaTaO3:La particles are smaller than the unmodified NaTaO3 and
contain characteristic ordered nanostep structures not found in the NaTaO3. It was
proposed that well-dispersed NiO deposited on the edges of the nanostep structures are
the catalytic active site for H2 production and O2 was produced at the grooves of the
nanostep structures. The proposed model, however, was established with only bulk
characterization techniques (XRD, TEM, XANES, and XAFS) and no surface
characterization methods were employed to directly confirm the conclusions on the
nature of the surface catalytic active sites.
Although there are a large number of semiconductor metal oxide materials that
can perform photocatalytic water splitting using UV-irradiation, the large band gap
energies of these semiconductors hinders their ability to utilize visible light energy.
Only NiO/InTaO4 (Q.E. = 0.7%) [40], RuO2/InTaO4 (Q.E. = 0.4%) [41] and
RuO2/YBiWO6 (Q.E. = 0.2%) [42] have been found to be able to perform the water
splitting reaction under visible light excitation, but with very low quantum
efficiencies.
Metal oxynitride catalysts ((Ga1-xZnx)(N1-xOx) and (Zn1+xGe)(N2Ox))
discovered by Domens group synthesized from metal oxides calcined under NH3 flow
10
have been shown to be able to overcome band gap limitations of semiconductor metal
oxides and can perform visible light water splitting [20, 43, 44]. DFT calculations
have theorized that there is a strong p-d repulsive force between N2p and Zn3d
orbitals in the valance band of these materials that shifts the valence band position
upwards without affecting the conduction band position [45, 46]. This narrowing of
the band gap is responsible for the visible light absorption for these metal oxynitride
photocatalysts. The (Rh2-yCryO3)/(Ga1-xZnx)(N1-xOx) catalyst is the most active of all
the oxynitride photocatalysts under visible light water splitting ( > 400nm) with an
optimized apparent quantum efficiency of 5.9% [47-50]. The bulk structure of the
(Ga1-xZnx)(N1-xOx) has been shown to have a wurtzite solid solution crystal structure.
The supported (Rh2-yCryO3) mixed oxide nanoparticles are proposed to be the active
site for H2 production while the (Ga1-xZnx)(N1-xOx) support provides the active sites
for O2 production.
As in the other studies on photocatalyst systems, only bulk
characterization techniques were applied and no surface characterization methods

were employed. Thus, further work needs to be performed to fully understand the
surface aspects of this novel photocatalyst system and especially the nature of the
catalytic active sites at the surface.
All of the photocatalyst systems mentioned above are conventional one-step
systems for photocatalytic water splitting where H2 and O2 are produced on the surface
of the photocatalyst. There is a new class of materials being developed which utilize a
two-step process for photocatalytic water splitting which are called Z-Scheme
photocatalyst systems. In the two-step Z-Scheme process, one catalyst is used for H2
11
evolution while another catalyst is used for O2 evolution. A shuttle redox couple
(Red/Ox) in solution is used to transfer electrons between the two catalysts.
schematic energy diagram comparing the two types of photocatalytic water splitting
systems is shown in figure 2. On the H2 evolution catalyst, excited electrons are used
to reduce water into H2 while holes in the valance band are used to oxidize the
reductant (Red) to an oxidant (Ox). On the O2 evolution catalyst, excited electrons are
used to reduce the Ox into the Red while holes in the valance band are used to oxidize
water into O2.
The advantage of using a Z-Scheme system is that the band gap
requirement for water splitting is less restrictive compared to a one-step system [51].
This allows for visible light irradiation to be more easily attainable since the band gap
positioning of the H2 evolution catalysts only needs to oxidize the reductant to an
oxidant and the O2 evolution catalyst only needs to be able to reduce the oxidant to the
reductant that are less energetic compared to the one-step system. The Z-scheme
systems which have been discovered involve combining Pt/SrTiO3:Rh with BiVO4,
Bi2MoO6, or WO3 using Fe3+/2+ as the shuttle redox couple [52], combining Pt/TaON
or Pt/ZrO2/TaON with RuO2/TaON or Pt/WO3 using IO3-/I- as the shuttle redox couple
[53-55].
Although Z-Scheme photocatalyst systems are a more recent discovery
compared to one-step photocatalyst systems, the most active Z-scheme photocatalyst

system (Pt/ZrO2/TaON with Pt/WO3) has a quantum yield of 6.3% at > 420 nm [55],
which is higher than the most active one-step system ((Rh2-yCryO3)/ (Ga1-xZnx)(N1xOx))
at 5.9% at > 400 nm. Not much research work has been devoted to Z-scheme
photocatalyst systems since it is a more recently developed system so there is potential

12
for new materials to be discovered that can greatly improve the photoactivity for Zscheme systems.
Metal sulfide materials have also attracted a good deal of attention for
photocatalytic water splitting due to their small band gap that allows for visible light
absorption [3, 13]. The valence band for these materials usually consist of S 3p
orbitals which are at a higher potential than O 2p orbitals typical of metal oxides. The
main disadvantage of using metal sulfides is that photocorrosion occurs readily and
the use of sacrificial reagents (Na2SO3, K2SO3, and Na2S) is required for stability and
to act as electron donors. In the presence of H2O, sulfide species rather than H2O are
oxidized by the photogenerated holes causing catalyst instability and preventing O2
evolution for these photocatalysts. With the use of sacrificial reagents the apparent
quantum efficiency of metal sulfide photocatalysts can reach 25% at 440 nm
(Rh/AgGaS2) [56] and 7.4% at 520 nm (Ru/Cu0.25Ag0.25In0.5ZnS2) [57, 58] for the
production of hydrogen without any oxygen products formed in the reaction. These
materials are currently able to absorb light up to 650 nm for hydrogen production.
Although metal sulfide photocatalysts act only as hydrogen evolution catalysts and do
not stoichiometrically produce H2 and O2 for overall water splitting, metal sulfide
photocatalysts have the potential to become important if abundant sulfur compounds
from chemical plants or nature can be found that can be used as electron donors for
this reaction.
This review of photocatalyst systems for water splitting has shown that many
advanced photocatalytic materials have been discovered over the past 40 years that
13
can perform the water splitting reaction.
The efficiency of these photocatalyst
systems, however, is still below the desired target of 30% quantum efficiency at 600
nm excitation that corresponds to about 5% solar energy conversion [13]. It is at this
point where photocatalyst systems start to become more economically viable for use
on a large-scale. Development of more efficient photocatalytic materials is currently
limited due to the lack of fundamental knowledge of their photocatalytic mechanisms.
Most of these studies have focused on using bulk characterization techniques to
propose photocatalyst models and are lacking in surface characterization of the
catalytically active sites, surface reaction intermediate species and interfacial
electronic mechanisms of the complex, coupled reactions involved in water splitting.
Combining both bulk and surface characterization is essential for determining the role
of each component in the system. This becomes even more critical since future
photocatalyst systems will most likely have an increased number of components and
verifying the role of each component in the system will help to establish fundamental
structure-photoactivity relationships for the complex photocatalyst systems.
1.4 Overview of Characterization Techniques Utilized for Heterogeneous Mixed
Oxide Photocatalyst Systems for Water Splitting
Most of the studies in the literature on the heterogeneous photocatalytic
systems for water splitting have primarily focused on correlating bulk properties with
photoactivity. In the past, the main focus of photocatalytic water splitting research has
been finding suitable bulk mixed oxide materials that can act as electron/hole
14
generators for the photocatalyst system.
Subsequently, bulk characterization
techniques have been primarily employed to investigate these active photocatalyst

systems.
The characterization techniques primarily focused on: X-ray Diffraction
(XRD), ultraviolet-visible diffuse reflectance spectroscopy (UV-vis DRS), X-ray

absorption near edge spectroscopy (XANES), extended X-ray absorption fine
spectroscopy (EXAFS), electron microscopy (EM), inductively coupled plasma
atomic emission spectroscopy (ICP-AES), photoluminescence (PL) spectroscopy,
Brunauer-Emmett-Teller (BET) surface area, attenuated total reflectance infrared
(ATR-IR) spectroscopy, and X-ray photoelectron spectroscopy (XPS) to give insights
into the properties of the photocatalyst systems. These characterization techniques,
however, provide only bulk structural and electronic information about the
photocatalyst systems, with the exception of BET, ATR-IR, and XPS that can give
some limited surface information about the photocatalyst systems.
The different
characterization techniques will be discussed in this section and the strengths and
weaknesses of each technique for studying photocatalytic systems will be highlighted.
1.4.1 X-ray Diffraction (XRD)
XRD involves focusing beams of X-rays which scatter from lattices in the
sample forming diffraction patterns. At certain angles, scattered X-rays are combined
constructively and are measured as a function of diffraction angle, 2. The diffraction
patterns formed can be used as identifiers for crystal lattices found in the material.
This allows for the determination of the crystalline phase, quantitative composition of
multi-component systems, lattice parameters, crystalline size and shape, bulk defect
15
types and concentration (micro-strains) and the presence of doping elements in the
crystal lattices of the photocatalyst system [59]. The disadvantages of XRD is that it
only provides analysis of bulk crystalline phases and is unable to provide surface
information, detect amorphous phases or detect crystalline particles smaller than 3 nm
due to peak broadening. XRD is, therefore, only limited to providing bulk structural
information about large particles (> 3 nm) that tend to be the supports that are
responsible for generating excited electron/hole of the photocatalyst system.
1.4.2 X-ray Absorption Spectroscopy (XANES and EXAFS)
The complimentary synchrotron-based X-ray absorption spectroscopic
techniques of X-ray Absorption Near Edge Spectroscopy (XANES) and Extended Xray Absorption Fine-Structure Spectroscopy (EXAFS) are based on the absorption of
X-rays by the sample that excites an electron from a core state to an empty state at
above the Fermi level. This process emits photons from the sample of interest that are
subsequently scattered by nearby atoms.
This interaction gives off fluorescence
emissions and transmissions modes that contain detailed local structural and electronic
information of the photocatalyst systems [60]. The XANES region focuses on the preedge, the edge and features up to 50 eV above the absorption edge of a specific
element in the photocatalyst, while the EXAFS region includes all features in the
spectrum above 50 eV from the edge of a specific element in the photocatalyst [61].
XANES is used to discriminate between different oxidation states and local bonding
geometry of a specific element. EXAFS is used to determine the bond length and
local geometry between different elements present in the photocatalyst. A major
16
drawback to theses characterization techniques is that they average their signal from
multiple sites of the same element and, therefore, are not well suited for materials
containing multiple components with different oxidation states or structures [62].
Furthermore, surface information is only provided for highly dispersed systems or
materials with high surface area, but the usually low surface areas of photocatalysts (<
10 m2/g) assures that the XAS signals will be dominated by the elements present in the
bulk phase of the photocatalysts.
1.4.3 Electron Microscopy
Electron microscopy provides the images of a photocatalyst with up to an
atomic scale resolution (<0.1 nm) that gives insight into the structure and morphology
of the photocatalyst system.
Electron microscopy encompasses many different
techniques, Transmission Electron Microscopy (TEM), Scanning Electron Microscopy

(SEM), Scanning Transmission Electron Microscopy (STEM), Energy-Dispersive Xray Spectroscopy (EDX), and Electron Energy-Loss Spectroscopy (EELS), which
have been used to characterize photocatalyst systems. TEM is the most commonly
used of these electron techniques for studying photocatalytic materials.
TEM
functions by generating a primary electron beam of high energy and high intensity that
passes through a condenser to produce parallel beams that irradiate the sample.
Magnified images of the sample are formed by combining the transmitted electrons
using an electromagnetic objective lens. TEM is primarily used to give information
on topography, morphology and crystal structure at with atomic resolution (0.1 nm)
[63] and bulk composition when coupled with EELS. SEM is carried out by rastering
17
a narrow electron beam over the surface of the sample and detects the yield of
secondary or backscattered electrons as a function of the primary beam. The main
difference between SEM and TEM is that SEM sees contrast due to topology and
composition of a surface, while the TEM projects all information on the mass it
encounters in a 2-D image of subnanometer resolution [62]. SEM instruments only
have a lateral resolution of ~ 5 nm and are useful for determination of surface
morphology and larger nanoparticles compared to TEM analysis. STEM combines
both modes mentioned above and utilize scanning coils to illuminate a small area of
the sample from which bright or dark field images are obtained.
STEM can be
combined with the high-angle annular dark field (HAADF) imaging technique to
provide Z-contrast images which can distinguish particles of different atomic number.
The image brightness is approximately dependent on the square of the atomic number
(Z2) of the atoms present and very useful for particles with a great difference in atomic
mass and not so useful for particles with similar atomic masses.
The elemental composition of a sample can be determined by combining EDX
and EELS analysis with the electron microscopic imaging techniques mentioned
above. EDX is the more commonly used technique and involves ionization of the
sample from the X-rays of the microscope. The X-ray emissions from this ionization
event are then analyzed with a solid-state energy dispersive detector. Each element
gives out a unique pattern of X-ray lines from the K, L, and M shells which allows for
straightforward analysis. EELS is complimentary to EDX in that the ionization event
that occurs in the sample leads to a characteristic loss in energy of the transmitted
18
electrons that can also be detected. EELS is very sensitive to lighter elements (Z< 11)
where the energy losses are small where EDX is more sensitive to heavier elements
(Z> 11) [63]. These techniques are also useful for constructing elemental maps to
characterize the microstructural composition and elemental distribution of
multicomponent catalysts on the nanometer scale.
All the electron microscopy
techniques, however, only provide structural information about crystalline components

(<1 nm) and do not provide structural information about amorphous or surface phases.
An exception occurs is when an element with a high Z number is dispersed as an
amorphous surface phase on a support with low Z number elements [64].
1.4.4. Infrared (IR) spectroscopy
Infrared spectroscopy provides vibrational information about chemically
bonded surface reaction intermediates and the changes in these molecular vibrations
give insight into the bond breaking/making process essential to understanding the
catalytic reaction mechanism. Infrared spectroscopy is a bulk technique that can
provide bulk structural information as well as the nature of the surface species if the
surface area is sufficient to accommodate significant number of surface species.
Conventional IR spectroscopy cannot be used to probe photocatalytic water splitting
reactions due to the strong absorption by water which requires the use of attenuated
total reflectance infrared (ATR-IR) spectroscopy [65, 66]. An advantage of ATR-IR
over conventional transmission IR and DRIFTS methods is that ATR-IR only probes
the molecules that are located near the surface (m) in the catalyst region instead of
the entire volume of the photoreactor. Thus, ATR-IR spectroscopy allows for in-situ
19
analysis of the solid/liquid interfaces during photocatalytic water splitting. It can be

used to obtain information about chemical and geometrical structures of surface
species and also give information on reaction mechanism and kinetics [67, 68]. Some
of the challenges with in-situ ATR-IR for photocatalytic water splitting include weak
signals from the low surface area photocatalytic materials, difficulty discriminating
between active and spectator species, diatomic molecules such as H2 and O2 not being
active in IR, and the strong water absorption leading to regions that cannot be used for
analysis.
1.5 Approach and Outline of This Thesis
The previous section has highlighted the characterization techniques that have
been employed to study photocatalytic materials. One of the major issues with the
previous studies has been the focus on correlating bulk structural properties of the
photocatalyst systems with photoactivity [26, 31, 47-49, 69, 70]. The lack of surface
characterization of the outermost surface layer has led to a catalysis paradigm for bulk
mixed metal oxide catalysis including photocatalysts, where bulk structural features
are correlated with catalytic activity [71, 72]. Catalytic reactions are a surface
phenomenon and surface characterization is needed to correlate surface properties
with catalytic activity and to better understand the fundamentals of photocatalysis.
Only when surface characterization is performed can appropriate models for catalytic
active sites be determined for photocatalytic systems.
20
This section will look at characterization techniques and the photocatalytic

reactor employed in this study for determination of bulk/surface structure
photoactivity relationships for photocatalyst systems for water splitting.
Raman
spectroscopy will be used to determine the bulk molecular structure of the

photocatalyst systems. The bulk electronic structure will be examined using UV-vis
diffuse reflectance spectroscopy. Photoluminescence spectroscopy will be used to
explore the state of the excited electrons and the lifetime of the excited electrons in the
photocatalyst system.
Since photocatalyst systems are complex multicomponent
systems, bulk characterization techniques are useful for determining the effect of each
component on the bulk properties of the photocatalyst system. The oxidation state of
surface species (~1-3 nm) will be determined using X-ray photoemission spectroscopy
while the composition of the outermost surface layer (~0.3 nm) of the photocatalyst
system will be revealed for the first time using high sensitivity low energy ion
scattering spectroscopy. Combining bulk and surface characterization will allow for
determining the role of each component in the photocatalyst system and allow for
more accurate models for the catalytic active sites and the photocatalytic reactions.
Raman spectroscopy is a powerful characterization technique which provides
vibrational information about the bulk mixed oxide molecular structures and can also
give information about surface structures in a photocatalyst system for high surface
area materials [73].
Raman spectroscopy is an ideal characterization technique
because it can operate in all phases (gas, solid, and liquid), and over a wide range of
21
temperatures and pressures that are useful for in situ studies during photocatalytic
water splitting reactions [73]. Like other bulk techniques, the Raman spectra are
dominated by the signal from bulk phase due to the greater Raman scattering from the
ordered crystalline lattice compared to disordered amorphous surface phases. The
ability to detect amorphous phases along with a greater sensitivity for detection of
smaller ordered nanoparticles (< 3 nm) makes Raman a more a better tool than the
comparable XRD characterization technique. The sharpness of the Raman peaks help
to qualitatively determine the relative crystallinity of the bulk phase of the
photocatalyst system.
Sample fluorescence, one of the limitations for Raman
spectroscopy, is not much of an issue when dealing with photocatalyst systems since
the bulk mixed oxide photocatalysts tend to give strong Raman signals. Additionally,
UV Raman has greater sensitivity towards the surface region of a solid sample than
conventional Raman spectroscopy [74].
1.5.2 Ultraviolet-visible (UV-vis) Diffuse Reflectance Spectroscopy (DRS)
UV-vis diffuse reflectance spectroscopy is a characterization technique used to
obtain information about the electronic structure of the photocatalyst system based on
the absorptive and light scattering properties of the sample under light excitation from
the 200800 nm range. The sample absorbance is then referenced against an MgO
standard and the Kubulka Monk function can be calculated for the sample that allows
determination of the optical band gap for photocatalyst materials [75-77]. The optical
band gap determines how much photon energy is required to generate excited
electron/hole pairs in the bulk of the photocatalyst material. It is desirable for a
22
photocatalyst system to have band gap values below 3.0 eV to be able to absorb
energy in the visible light region [20].
Since UV-vis DRS looks at electronic
transitions from ground state to excited states in the sample, it is also used to give
obtain information on the oxidation states, and coordination of metals from metalcentered and charge transfer transitions [78, 79]. One of the drawbacks to using UVvis DRS is that it is a bulk technique so the information obtained from it is averaged
from all the components of the photocatalysts system, bulk and surface, and cannot
give specific information about individual components and the surface region. An
advantage of UV-vis DRS is that it can performed under in situ conditions during
photocatalytic splitting of water
1.5.3 Photoluminescence (PL) Spectroscopy
Photoluminescence spectroscopy is a complimentary technique to UV-vis DRS
in that it also gives information on the electronic structure from light excitation.
Whereas UV-vis DRS looks at the transition from the ground state to the excited state,
PL spectroscopy differs by examining the transition from the excited state to the
ground state. The emission of photons from this phenomenon can be measured as
fluorescence. This process is relevant to the study of photocatalyst systems since
fluorescence occurs when electrons in the conduction band recombine with holes in
the valence band. This characterization technique can thus give direct information on
electron transfer kinetics during the photocatalytic process. The PL intensity can then
be used to compare recombination rates for different photocatalyst systems, determine
the band gap energy, and detect impurities and defects in the sample [80, 81]. PL
23
spectroscopy has the same drawback as UV-vis DRS in that it is a bulk technique and
cannot give specific information about individual components nor surface species.
However surface species can have a large effect on the PL spectra since surface sites
and the presence of adsorbed surface molecules can act as efficient electron traps
which help to decrease PL intensity preventing electron/hole recombination [74, 81].
Time-resolved picosecond in situ PL (TR-PL) spectroscopy will also be
utilized to examine the recombination lifetimes of electron/hole pairs.
Whereas
conventional or steady-state PL spectroscopy uses continuous excitation from a light

source,
TR-PL
spectroscopy
relies
on
pulsed
excitation
and
measures
photoluminescence at certain time intervals after the pulse excitation [80, 82]. The PL
emission intensity over time is then fitted to an exponential decay model to determine
the lifetimes of photo-generated electron/hole pairs [83, 84]. An increased lifetime of
photogenerated electron/hole pairs has been shown to correlate with TiO2 for
photocatalytic splitting [85, 86] and this relationship will be investigated for more
active photocatalyst systems.
1.5.4 High-Resolution X-ray Photoemission Spectroscopy (HR-XPS)
XPS is a surface technique that utilizes photons from an X-ray source to excite
the electronic states of atoms near the surface (~1-3 nm) of a solid. The electrons
ejected from the sample are collected by a hemispherical electron energy analyzer
which measures the kinetic energy of the ejected electrons.
The kinetic energy
measured can then be used to determine the binding energy of the electron from the
following equation:
24
(2)
Where h is Plancks constant, is the frequency of the exciting energy source, and is
the work function of the spectrometer [62]. Each ejected electron from an element has
a characteristic binding energy and the position of this binding energy is used to
identify the element and core level of the electron that can be found by consulting
binding energy tables [87].
The peak intensities can be used to quantitatively
determine the elemental composition (detection limit ~ 1000 ppm for most elements)
and the peak positions can be used to determine oxidation states for the elements. In
general, binding energies increases with increasing oxidation state typically shifting by
about 0-3 eV [62]. Some of the disadvantages of XPS are that it requires ultra-high
vacuum conditions, sample damage from the X-ray sources, not sensitive enough to
get information on only the outermost layer (~0.3 nm) and overlapping elemental
peaks that can make spectra analysis challenging.
1.5.5 High Sensitivity Low Energy Ion Scattering Spectroscopy (HSLEISS)
HS-LEIS is a unique surface characterization technique is a powerful tool for
analysis of the outermost surface layer (~0.3 nm) of the photocatalyst system that
gives insight into the nature of the catalytic active sites on the photocatalyst surface.
During HS-LEIS, the surface of the sample is bombarded by a noble gas ion with a
known energy and a binary collision occurs between the noble gas ion and a surface
25
atom.
The energy of the backscattered ion, Ef, can be determined by the
conservational laws of momentum and energy by the equation:
(3)
(
In this, k is the kinematic factor, m1 and m2 are the masses of the primary ion and the
scattered surface atom, is the back scattering angle, and E0 is the initial energy for
the primary ion [88]. All of the variables are known for the experiment except for m2
so that value is able to be back-calculated from the experimental data obtained for the
backscattered ions. HS-LEIS spectroscopy is used to quantitatively determine the
atomic composition of the outermost atomic layer and can detect all elements except
for H and He. HS-LEIS spectroscopy is orders of magnitude more sensitive than
conventional LEIS spectroscopy with elemental detection as low as 10 ppm and uses
lower doses of primary gas ions to obtain static spectra sputtering less than 0.5% of a
monolayer [89, 90]. Noble gas ions can penetrate the surface into deeper layers, but
are usually neutralized upon scattering; however some particles can become re-ionized
and show up as a low energy tail to the peak obtained from surface atoms. The shape
of this tail can give information on the distribution of the elements over deeper layers
up to 10 nm [89]. Alternatively, a secondary ion source can be used for sputter depth
profiling, which can also give information about deeper layers. The disadvantages of
using HS-LEIS spectroscopy are that it requires ultra-high vacuum conditions, it does
26
not give chemical state information (oxidation state) and it may be difficult to
distinguish between elements with similar atomic masses.
1.5.6 Photocatalytic Reactor for Water Splitting
There are two different reactor cell designs that are used for photocatalytic
water splitting and a schematic of the two reactors are presented in Figure 3. The
reactor system consists of a closed-gas circulation system connected to a vacuum
pump that removes ambient air from the system prior to the reaction so no
atmospheric oxygen is present in the system during photocatalysis. A suspension
consisting of the photocatalyst and water is created in the reaction cell with a magnetic
stirrer put in to continuously stir the suspension. An inner irradiation reactor cell is
used when a UV lamp (high pressure 450W Hg lamp) is used for irradiation while a
top down reactor cell is used when a Xe arc lamp is used. A solar simulator can also
be used if solar hydrogen production is desired.
Appropriate filters are employed
when the experiment requires visible light irradiation. Typically NaNO2 is utilized to
filter out UV light in the inner irradiation reactor while cut-off filters are employed for
top down reactors. These reactor cells are connected to a water cooling system to
maintain room temperature and to minimize the effects from heating due to the high
powered lamps. A Liebig condenser is utilized to prevent gas phase water from
escaping the reactor so only H2 and O2 can be found in the rest of the system. A gas
chromatograph is then used to quantify the amount of H2 and O2 products.
1.5.7 Dissertation Outline
27
Although there are over one hundred photocatalyst systems for water splitting
that have been discovered, only a few of the more relevant and most studied
photocatalyst systems will be investigated.
In chapter 2, Ta2O5 and NaTaO3 based
photocatalysts will be examined since these are the most active metal oxide
photocatalyst under UV irradiation. Chapter 3 will look at GaN which is an UV active
oxynitride photocatalyst. There are two different synthesis methods for the (Ga1xZnx)(N1-xOx)
photocatalysts which have led to two different proposed models for their
catalytic active sites. The validity of the models for these visible light oxynitride
photocatalysts will be the focus of investigation in Chapter 4. Chapter 5 will look at
TaON photocatalysts which are capable of absorbing light up to 600 nm and is used in
Z-scheme photocatalyst systems. Chapter 6 will give a summary of the conclusions
found in this dissertation on the photocatalyst systems studied. The future outlook
will look at how other characterization techniques can be used to further elucidate the
photoactivity of existing photocatalytic materials so fundamental structure-activity
relationships can be established for the design of advanced photocatalyst systems.
28
References
[1] U.S. Department of Energy (Ed.), Basic Research Needs for the Hydrogen
Economy; A Report from the Basic Energy Sciences Workshop on Hydrogen
Production,Storage, and Use, U.S. Department of Energy, Washington D.C., 2003.
[2] N.S. Lewis, D.G. Nocera, "Powering the planet: Chemical challenges in solar
energy utilization " Proceedings of the National Academy of Sciences, 103, 15729,
(2006).
[3] F.E. Osterloh, "Inorganic Materials as Catalysts for Photochemical Splitting of

Water." Chem. Mater., 20, 35, (2008).
[4] A.J. Bard, M.A. Fox, "Artificial Photosynthesis: Solar Splitting of Water to
Hydrogen and Oxygen " Acc. Chem. Res., 28, 141, (1995).
[5] A. Fujishima, K. Honda, "Electrochemical photolysis of water at a semiconductor

electrode." Nature (London), 238, 37, (1972).
[6] M.G. Walter, E.L. Warren, J.R. McKone, S.W. Boettcher, Q. Mi, E.A. Santori, and
N.S. Lewis, "Solar water splitting cells " Chem. Rev., 110, 6446, (2010).
[7] Y. Sakai, S. Sugahara, M. Matsumura, Y. Nakato, and H. Tsubomura,

"Photoelectrochemical water splitting by tandem type and heterojunction amorphous
silicon electrodes " Canadian Journal of Chemistry, 66, 1853, (1988).
29
[8] O. Khaselev, "A Monolithic Photovoltaic-Photoelectrochemical Device for

Hydrogen Production via Water Splitting " Science, 280, 425, (1998).
[9] S. Licht, B. Wang, S. Mukerji, T. Soga, M. Umeno, and H. Tributsch, "Efficient

Solar Water Splitting, Exemplified by RuO2-Catalyzed AlGaAs/Si Photoelectrolysis "
The Journal of Physical Chemistry B, 104, 8920, (2000).
[10] A. Mills, S. Le Hunte, "An overview of semiconductor photocatalysis " J.

Photochem. Photobiol. A., 108, 1, (1997).
[11] A. Kubacka, M. Fernandez-Garcia, and G. Colon, "Advanced Nanoarchitectures

for Solar Photocatalytic Applications," Chem. Rev., 112, 1555, (2012).
[12] N. Serpone, D. Lawless, R. Khairutdinov, and E. Pelizzetti, "Subnanosecond

Relaxation Dynamics in TiO2 Colloidal Sols (Particle Sizes Rp = 1.0-13.4 nm).
Relevance to Heterogeneous Photocatalysis." J. Phys. Chem., 99, 16655, (1995).
[13] A. Kudo, Y. Miseki, "Heterogeneous photocatalyst materials for water splitting."

Chem. Soc. Rev., 38, 253, (2009).
[14] I.E. Wachs, "Recent conceptual advances in the catalysis science of mixed metal
oxide catalytic materials " Catalysis Today, 100, 79, (2005).
[15] J.S.J. Hargreaves (Ed.), Metal Oxide Catalysis , Wiley-VCH.
30
[16] A.L. Linsebigler, G. Lu, and J.T. Yates, "Photocatalysis on TiO2 Surfaces:
Principles, Mechanisms, and Selected Results " Chem. Rev., 95, 735, (1995).
[17] T. Takata, A. Tanaka, M. Hara, J.N. Kondo, and K. Domen, "Recent progress of
photocatalysts for overall water splitting " Catalysis Today, 44, 17, (1998).
[18] S. Tabata, H. Nishida, Y. Masaki, and K. Tabata, "Stoichiometric photocatalytic

decomposition of pure water in Pt/TiO2 aqueous suspension system." Catal. Lett., 34,
245, (1995).
[19] K. Sayama, H. Arakawa, "Effect of carbonate salt addition on the photocatalytic

decomposition of liquid water over Pt-TiO2 catalyst." J. Chem. Soc. , Faraday Trans.,
93, 1647, (1997).
[20] K. Maeda, K. Domen, "New Non-Oxide Photocatalysts Designed for Overall

Water Splitting under Visible Light." J. Phys. Chem. C, 111, 7851, (2007).
[21] X. Chen, S. Shen, L. Guo, and S.S. Mao, "Semiconductor-based Photocatalytic

Hydrogen Generation " Chem. Rev., 110, 6503, (2010).
[22] K. Sayama, H. Arakawa, "Photocatalytic decomposition of water and

photocatalytic reduction of carbon dioxide over zirconia catalyst." J. Phys. Chem., 97,
531, (1993).
31
[23] K. Sayama, H. Arakawa, "Effect of Na2CO3 addition on photocatalytic

decomposition of liquid water over various semiconductor catalysts." J. Photochem.
Photobiol. , A, 77, 243, (1994).
[24] K. Sayama, H. Arakawa, "Effect of carbonate addition on the photocatalytic

decomposition of liquid water over a ZrO2 catalyst." J. Photochem. Photobiol. , A, 94,
67, (1996).
[25] K. Domen, S. Naito, T. Onishi, and K. Tamaru, "Photocatalytic decomposition of

liquid water on a nickel(II) oxide-strontium titanate (SrTiO3) catalyst." Chem. Phys.
Lett., 92, 433, (1982).
[26] H. Kato, A. Kudo, "New tantalate photocatalysts for water decomposition into H2
and O2." Chem. Phys. Lett., 295, 487, (1998).
[27] T. Takata, Y. Furumi, K. Shinohara, A. Tanaka, M. Hara, J.N. Kondo, and K.

Domen, "Photocatalytic Decomposition of Water on Spontaneously Hydrated Layered
Perovskites." Chem. Mater., 9, 1063, (1997).
[28] H.G. Kim, D.W. Hwang, J. Kim, Y.G. Kim, and J.S. Lee, "Highly donor-doped
(110) layered perovskite materials as novel photocatalysts for overall water splitting."
Chem. Commun. (Cambridge), 1077, (1999).
32
[29] A. Kudo, H. Kato, "Photocatalytic decomposition of water into H2 and O2 over

novel photocatalyst K3Ta3Si2O13 with pillared structure consisting of three TaO6
chains." Chem. Lett., 867, (1997).
[30] H. Kato, A. Kudo, "Water Splitting into H2 and O2 on Alkali Tantalate

Photocatalysts ATaO3 (A = Li, Na, and K)." J. Phys. Chem. B, 105, 4285, (2001).
[31] H. Kato, K. Asakura, and A. Kudo, "Highly Efficient Water Splitting into H2 and
O2 over Lanthanum-Doped NaTaO3 Photocatalysts with High Crystallinity and
Surface Nanostructure." J. Am. Chem. Soc., 125, 3082, (2003).
[32] S. Ikeda, T. Itani, K. Nango, and M. Matsumura, "Overall Water Splitting on

Tungsten-Based Photocatalysts with Defect Pyrochlore Structure." Catal. Lett., 98,
229, (2004).
[33] K. Ikarashi, J. Sato, H. Kobayashi, N. Saito, H. Nishiyama, and Y. Inoue,

"Photocatalysis for water decomposition by RuO2-dispersed ZnGa2O4 with d10
configuration." J. Phys. Chem. B, 106, 9048, (2002).
[34] J. Sato, N. Saito, H. Nishiyama, and Y. Inoue, "Photocatalytic Activity for Water
Decomposition of Indates with Octahedrally Coordinated d10 Configuration. I.
Influences of Preparation Conditions on Activity." J. Phys. Chem. B, 107, 7965,
(2003).
33
[35] J. Sato, H. Kobayashi, K. Ikarashi, N. Saito, H. Nishiyama, and Y. Inoue,

"Photocatalytic Activity for Water Decomposition of RuO2-Dispersed Zn2GeO4 with
d10 Configuration." J. Phys. Chem. B, 108, 4369, (2004).
[36] J. Sato, N. Saito, H. Nishiyama, and Y. Inoue, "New Photocatalyst Group for
Water Decomposition of RuO2-Loaded p-Block Metal (In, Sn, and Sb) Oxides with
d10 Configuration." J. Phys. Chem. B, 105, 6061, (2001).
[37] A. Kudo, H. Okutomi, and H. Kato, "Photocatalytic Water Splitting into H2 and
O2 over K2LnTa5O15 Powder. " Chem. Lett., 1212, (2000).
[38] K. Yoshioka, V. Petrykin, M. Kakihana, H. Kato, and A. Kudo, "The relationship

between photocatalytic activity and crystal structure in strontium tantalates." J. Catal.,
232, 102, (2005).
[39] A. Kudo, H. Kato, "Effect of lanthanide-doping into NaTaO3 photocatalysts for

efficient water splitting." Chem. Phys. Lett., 331, 373, (2000).
[40] Z. Zou, J. Ye, K. Sayama, and H. Arakawa, "Direct splitting of water under
visible light irradiation with an oxide semiconductor photocatalyst." Nature (London,
U. K. ), 414, 625, (2001).
34
[41] Z. Zou, H. Arakawa, "Direct water splitting into H2 and O2 under visible light
irradiation with a new series of mixed oxide semiconductor photocatalysts." J.
Photochem. Photobiol. , A, 158, 145, (2003).
[42] H. Liu, J. Yuan, W. Shangguan, and Y. Teraoka, "Visible-light-responding

BiYWO6 solid solution for stoichiometric photocatalytic water splitting." J. Phys.
Chem. C, 112, 8521, (2008).
[43] K. Maeda, K. Teramura, N. Saito, Y. Inoue, H. Kobayashi, and K. Domen,

"Overall water splitting using (oxy)nitride photocatalysts." Pure Appl. Chem., 78,
2267, (2006).
[44] Y. Lee, H. Terashima, Y. Shimodaira, K. Teramura, M. Hara, H. Kobayashi, K.

Domen, and M. Yashima, "Zinc germanium oxynitride as a photocatalyst for overall
water splitting under visible light." J. Phys. Chem. C, 111, 1042, (2007).
[45] W. Wei, Y. Dai, K. Yang, M. Guo, and B. Huang, "Origin of the Visible Light
Absorption of GaN-Rich Ga1-x ZnxN1-xOx (x = 0.125) Solid Solution " Journal of
Physical Chemistry C, 112, 15915, (2008).
[46] K. Maeda, K. Teramura, T. Takata, M. Hara, N. Saito, K. Toda, Y. Inoue, H.

Kobayashi, and K. Domen, "Overall Water Splitting on (Ga1-xZnx)(N1-xOx) Solid
Solution Photocatalyst: Relationship between Physical Properties and Photocatalytic
Activity." J. Phys. Chem. B, 109, 20504, (2005).
35
[47] K. Maeda, K. Teramura, D. Lu, T. Takata, N. Saito, Y. Inoue, and K. Domen,

"Characterization of Rh-Cr Mixed-Oxide Nanoparticles Dispersed on (Ga1-xZnx)(N1xOx) as a Cocatalyst for Visible-Light-Driven Overall Water Splitting." J. Phys.
Chem. B, 110, 13753, (2006).
[48] K. Maeda, K. Teramura, D. Lu, N. Saito, Y. Inoue, and K. Domen, "Noblemetal/Cr2O3 core/shell nanoparticles as a co-catalyst for photocatalytic overall water
splitting." Angew. Chem. , Int. Ed., 45, 7806, (2006).
[49] K. Maeda, K. Teramura, D. Lu, N. Saito, Y. Inoue, and K. Domen, "Roles of

Rh/Cr2O3 (Core/Shell) Nanoparticles Photodeposited on Visible-Light-Responsive
(Ga1-xZnx)(N1-xOx) Solid Solutions in Photocatalytic Overall Water Splitting." J.
Phys. Chem. C, 111, 7554, (2007).
[50] K. Maeda, K. Teramura, and K. Domen, "Effect of post-calcination on

photocatalytic activity of (Ga1-xZnx)(N1-xOx) solid solution for overall water
splitting under visible light." J. Catal., 254, 198, (2008).
[51] R. Abe, "Development of a New System for Photocatalytic Water Splitting into
H2 and O2 under Visible Light Irradiation," Bull. Chem. Soc. Jpn., 84, 1000, (2011).
[52] H. Kato, M. Hori, R. Konta, Y. Shimodaira, and A. Kudo, "Construction of Zscheme type heterogeneous photocatalysis systems for water splitting into H2 and O2
under visible light irradiation." Chem. Lett., 33, 1348, (2004).
36
[53] M. Higashi, R. Abe, A. Ishikawa, T. Takata, B. Ohtani, and K. Domen, "Zscheme overall water splitting on modified-TaON photocatalysts under visible light (l
< 500 nm)." Chem. Lett., 37, 138, (2008).
[54] R. Abe, T. Takata, H. Sugihara, and K. Domen, "Photocatalytic overall water

splitting under visible light by TaON and WO3 with an IO3-/I- shuttle redox
mediator." Chem. Commun. (Cambridge, U. K. ), 3829, (2005).
[55] K. Maeda, M. Higashi, D. Lu, R. Abe, and K. Domen, "Efficient Nonsacrificial

Water Splitting through Two-Step Photoexcitation by Visible Light using a Modified
Oxynitride as a Hydrogen Evolution Photocatalyst." Zhonghua Jianyan Yixue Zazhi,
32, 5858, (2009).
[56] A. Kudo, "Development of photocatalyst materials for water splitting." Int. J.

Hydrogen Energy, 31, 197, (2006).
[57] I. Tsuji, H. Kato, and A. Kudo, "Visible-Light-Induced H2 Evolution from an

Aqueous Solution Containing Sulfide and Sulfite over a ZnS-CuInS2-AgInS2 SolidSolution Photocatalyst " Angewandte Chemie International Edition, 44, 3565, (2005).
[58] I. Tsuji, H. Kato, and A. Kudo, "Photocatalytic Hydrogen Evolution on ZnSCuInS2-AgInS2Solid Solution Photocatalysts with Wide Visible Light Absorption
Bands " Chemistry of Materials, 18, 1969, (2006).
37
[59] D. Herein, "X-Ray Powder Diffraction" in: , Handbook of Heterogeneous

Catalysis, 2008, p. 765-774.
[60] D.C. Koningsberger, D.E. Ramaker, "Applications of X-ray Absorption

Spectroscopy in Heterogeneous Catalysis: EXAFS, Atomic XAFS, and Delta
XANES" in: , Handbook of Heterogeneous Catalysis, 2008, p. 774-803.
[61] S.R. Bare, T. Ressler, "Advances in Catalysis Volume 52; Chapter 6

Characterization of Catalysts in Reactive Atmospheres by Xray Absorption
Spectroscopy " 52, 339, (2009).
[62] J.W. Niemantsverdriet (Ed.), Spectroscopy in Catalysis , Wiley-VCH:Weinheim,

2000.
[63] A.K. Datye, P.L. Hansen, S. Helveg, "Electron Microscopy and Diffraction" in: ,
Handbook of Heterogeneous Catalysis, 2008, p. 803-833.
[64] W. Zhou, E.I. Ross-Medgaarden, W.V. Knowles, M.S. Wong, I.E. Wachs, and
C.J. Kiely, "Identification of active Zr-WOx clusters on a ZrO2 support for solid acid
catalysts." Nat. Chem., 1, 722, (2009).
[65] T. Burgi, A. Baiker, "Attenuated total reflection infrared spectroscopy of solid

catalysts functioning in the presence of liquid-phase reactants." Adv. Catal., 50, 227,
(2006).
38
[66] B.L. Mojet, S.D. Ebbesen, and L. Lefferts, "Light at the interface: the potential of
attenuated total reflection infrared spectroscopy for understanding heterogeneous
catalysis in water " Chem. Soc. Rev., 39, 4643, (2010).
[67] J. Andanson, A. Baiker, "Exploring catalytic solid/liquid interfaces by in situ

attenuated total reflection infrared spectroscopy " Chem. Soc. Rev., 39, 4571, (2010).
[68] R. Nakamura, Y. Nakato, "Primary Intermediates of Oxygen Photoevolution

Reaction on TiO2 (Rutile) Particles, Revealed by in Situ FTIR Absorption and
Photoluminescence Measurements." J. Am. Chem. Soc., 126, 1290, (2004).
[69] H. Kato, A. Kudo, "Photocatalytic water splitting into H2 and O2 over various
tantalate photocatalysts." Catal. Today, 78, 561, (2003).
[70] H. Kato, A. Kudo, "Highly efficient decomposition of pure water into H2 and O2
over NaTaO3 photocatalysts." Catal. Lett., 58, 153, (1999).
[71] I.E. Wachs, K. Routray, "Catalysis Science of Bulk Mixed Oxides" ACS
Catalysis, 2, 1235, (2012).
[72] S.V. Merzlikin, N.N. Tolkachev, L.E. Briand, T. Strunskus, C. Woll, I.E. Wachs,
and W. Grunert, "Anomalous Surface Compositions of Stoichiometric Mixed Oxide
Compounds" Angew. Chem. Int. Ed Engl., 49, 8037-8041, (2010).
39
[73] I.E. Wachs, C.A. Roberts, "Monitoring surface metal oxide catalytic active sites
with Raman spectroscopy" Chem. Soc. Rev., 39, 5002-5017, (2010).
[74] M. Anpo, P.V. Kamat (Eds.), Environmentally Benign Photocatalysts:

Applications of Titanium Oxide-based Materials, Springer:New York, 2010.
[75] P. Kubelka, F. Munk, Z. Tech. Phys., 12, 593, (1931).
[76] X. Gao, I.E. Wachs, "Investigation of Surface Structures of Supported Vanadium

Oxide Catalysts by UV-vis-NIR Diffuse Reflectance Spectroscopy." J. Phys. Chem. B,
104, 1261, (2000).
[77] W.N. Delgass (Ed.), Spectroscopy in heterogeneous catalysis, Academic

Press:New York, 1979.
[78] Z. Sojka, F. Bozon-Verduraz, M. Che, "UV-Vis-NIR and EPR Spectroscopies"

in: , Handbook of Heterogeneous Catalysis, 2008, p. 1039-1065.
[79] F.C. Jentoft, "Advances in Catalysis Volume 52; Chapter 3 UltravioletVisible

Near Infrared Spectroscopy in Catalysis " 52, 129 , (2009).
[80] G.D. Gilliland, "Photoluminescence spectroscopy of crystalline semiconductors"

Materials Science and Engineering: R: Reports, 18, 99, (1997).
40
[81] M. Anpo, S. Dzwigaj, and M. Che, "Advances in Catalysis Volume 52; Chapter 1
Applications of Photoluminescence Spectroscopy to the Investigation of OxideContaining Catalysts in the Working State" 52, 1, (2009).
[82] M. Matsuoka, T. Kamegawa, M. Anpo, "Photoluminescence Spectroscopy and Its

Application to the Characterization of Active Sites and Reaction Dynamics in
Catalysis" in: , Handbook of Heterogeneous Catalysis, 2008, p. 1065-1073.
[83] B.J. Selby, T.I. Quickenden, and C.G. Freeman, "The Fitting of Luminescence
Rises and Decays," Kinetics and Catalysis, 44, 5-15, (2003).
[84] T.I. Quickenden, T.A. Green, and D. Lennon, "Luminescence from UV-Irradiated
Amorphous H2O Ice " J. Phys. Chem., 100, 16801, (1996).
[85] J. Tang, J.R. Durrant, and D.R. Klug, "Mechanism of photocatalytic water
splitting in TiO2. Reaction of water with photoholes, importance of charge carrier
dynamics, and evidence for four-hole chemistry " J. Am. Chem. Soc., 130, 13885,
(2008).
[86] C.A. Roberts, A.A. Puretsky, S.P. Phivilay, and I.E. Wachs, "Determination of a
Structure-Photocatalytic Relationship for Water Splitting on Well-Defined TiO2
Nanodomains," In preparation for publication.
41
[87] J.F. Moulder (Ed.), Handbook of x-ray photoelectron spectroscopy : a reference

book of standard spectra for identification and interpretation of XPS data, PerkinElmer Corporation:Eden Prairie, 1992.
[88] H. BRONGERSMA, M. DRAXLER, M. DERIDDER, and P. BAUER, "Surface

composition analysis by low-energy ion scattering " Surface Science Reports, 62, 63,
(2007).
[89] H.R.J. ter Veen, T. Kim, I.E. Wachs, and H.H. Brongersma, "Applications of
High Sensitivity-Low Energy Ion Scattering (HS-LEIS) in heterogeneous catalysis."
Catal. Today, 140, 197, (2009).
[90] J.A. Kilner, S.J. Skinner, and H.H. Brongersma, "The isotope exchange depth
profiling (IEDP) technique using SIMS and LEIS " Journal of Solid State
Electrochemistry, 15, 861, (2011).
42
Figure 1.1: Basic Principles of Photocatalytic Water Splitting.
43
Figure 1.2: Schematic Energy Diagram of photocatalytic water-splitting systems: (a)

conventional one-step system and (b) two-step Z-Scheme system obtained from
[51].
44
Figure 1.3: Schematic of Photocatalytic Reactor Designs obtained from [13].
45
CHAPTER 2
Nature of Catalytic Active Sites Present on the Surface of
Advanced Bulk Tantalum Mixed Oxide Photocatalysts
Abstract
The most active photocatalyst system for water splitting under UV irradiation (270
nm) is the promoted 0.2%NiO/NaTaO3:2%La photocatalyst with optimized apparent
quantum efficiency (Q.E.) of 56%.
The promoted NaTaO3:2%La phase was
synthesized by solid-state reaction between Ta2O5, NaCO3 and La2O3 at extremely

high temperatures (1120-1470 K) and the NiO promoter was subsequently
impregnated from an aqueous Ni(NO3)2*6H2O solution and mildly calcined at 540 K.
Raman spectroscopy revealed the bulk molecular structure of NaTaO3 was not
modified by the addition of the La2O3 and NiO promoters. UV-vis spectroscopy
showed that the bulk band gap energy was likewise not significantly affected by the
addition of La2O3 and NiO in agreement with the same bulk structure for all the
photocatalysts. The surface area of the NaTaO3 phase is enhanced by ~10 for the Lacontaining catalysts and not altered for the NiO containing catalysts. HR-XPS and
HS-LEIS spectroscopy indicate that the NiO and La2O3 promoters are surface
46
segregated on the bulk NaTaO3 phase. Photoluminescence (PL) spectroscopy reveals

that the addition of La2O3 and NiO act as electron traps resulting in the suppression of
the PL emission intensity. The greater number of electron traps for the La-containing
NaTaO3 is strictly related to the enhanced surface area of this photocatalyst and the
enhanced number of electron traps for the NiO-containing NaTaO3 is an electronic
effect since the surface area of this material was not altered by the addition of NiO.
Consequently, the specific photoactivity of the NiO-containing NaTaO3 catalysts,
when normalized per unit surface area, is enhanced by a factor of ~101-102 because of
the electronic promotion of NiO.
These insights provide new fundamental
molecular/electronic structure-photoactivity relationships about the promoted NaTaO3

photocatalysts and the important role of surface catalytic active sites. These new
findings also bring into question the validity of the previously proposed model for the
catalytic active sites for the promoted 0.2%NiO/NaTaO3:2%La photocatalyst system.
2.1 Introduction
Photocatalytic water splitting is a thermodynamically challenging reaction
requiring a large positive change in Gibbs free energy (238 kJ/mol) to produce
hydrogen fuel and oxygen. This phenomenon was first brought to attention by the
pioneering work by Fujishima and Honda in 1972 [1] and research efforts since then
have focused on finding highly active metal oxide semiconductor materials for
photocatalytic hydrogen production by water splitting. Hydrogen is considered to be
one of the potential candidates to replace fossil fuel for our sustainable energy needs
47
especially if it can be generated from the photocatalytic conversion of cheap abundant

water into clean non-carbon hydrogen from solar energy resources. Development of
this clean, renewable form of energy will help to address our reliance on depleted
fossil fuel supplies and the environmental problems accompanying its use. Many
metal oxide semiconductor catalysts (>130) have been found which are able to
photocatalytically convert water into hydrogen and oxygen [2-5]. These semiconductor
catalysts are based on d0 (Ti, Zr, Nb, Ta, and W) transition metal oxides and based on
d10 (Ga, In, Ge, Sn and Sb) main group metal oxides have emerged as candidates for
use in heterogeneous photocatalytic systems because of their advantageous electronic
configuration. Although photocatalytic water splitting has garnered much interest in
academia, there has not been much industrial interest for photocatalytic water splitting
due to the low photocatalytic activity and lack of extensive studies on industrial scale
up for the process [5]. Among the discovered semiconductor photocatalyst systems,
tantalum-based photocatalysts such as Ta2O5 [6, 7], NaTaO3 [8], K3Ta3Si2O13 [9],
SrTa2O6 [10] and NaTaO3:La [11, 12] have been found to be among the most
promising for photocatalytic water splitting due to their high photoactivity under UV
irradiation.
Some of the strategies that are used to increase the activity of the
photocatalysts include the addition of a co-catalyst such as NiO, Pt, Rh2O3, and RuO2
and doping of the photocatalysts with metal ions to induce changes in morphology
[13]. These catalyst design strategies led towards the discovery of lanthanum-doped
NaTaO3 loaded with a NiO co-catalyst (NiO/NaTaO3:La) which is currently the most
48
active photocatalyst for water splitting with an optimized apparent quantum efficiency
(Q.E.) of 56% under UV irradiation at 270 nm with a stability of more than 400 hours
[11]. Bulk characterization techniques (XRD, TEM, XANES and EXAFS) suggested
that LaOx and NiO species are present as highly dispersed species on the surface of the
catalysts because separate crystalline phases of NiO and LaOx were not detected.
Electron microscopy revealed that the particle size of NaTaO3:La (0.1-0.7 m) were
smaller than undoped, La-free NaTaO3 (2-3 m) and ordered nanostep structures were
found only on NaTaO3:La [11]. It was proposed that the active site for H2 evolution
was at highly dispersed NiO sites selectively deposited on the edges of the nanostep
structures while the active site for O2 evolution was at the groove of the nanostep
structures. The high photoactivity of the catalysts was attributed to the separation
between the oxidative and reductive active sites. Deriving a model for the catalytic
active surface sites based only on bulk techniques is highly problematic since the
surface of bulk mixed metal oxides can be surface enriched or depleted of one or more
of its constituent components [14, 15]. Thus, it is necessary to understand the nature
of the outermost surface layer of a photocatalyst since the catalytic processes
producing the H2 and O2 proceed through catalytic surface phenomena.
This
fundamental piece of information is critical to develop solid fundamental

photocatalytic models involving the catalytic active surface sites.
The current study will utilize in situ optical spectroscopic characterization
methods (Raman, UV-vis, photoluminescence (PL) and time-resolved picosecond PL49
Raman) to give further insight into the bulk molecular and electronic structure of
NaTaO3 photocatalysts and how the addition of NiO and La2O3 phases affect those
bulk properties. The use of the surface science characterization methods of highresolution X-ray photoelectron spectroscopy (HR-XPS) and high-sensitivity low
energy ion scattering (HS-LEIS) spectroscopy will provide additional information
about the nature of the surface region (~1-3 nm) and outermost atomic layer (~0.3 nm)
of the photocatalysts where the catalytic active sites are located, respectively. This is
the first study to report on the surface nature of tantalum mixed oxide photocatalysts.
The objectives of this study are to establish (i) fundamental structure-photoactivity
relationship for the highly active tantalum-based photocatalysts and (ii) to examine the
roles of the surface and bulk characteristics of mixed oxide photocatalysts for water
splitting.
2.2. Experimental
2.2.1 Photocatalyst Synthesis
The NaTaO3 and the doped NaTaO3:La photocatalysts were synthesized by
solid-state reactions [11].
Ta2O5 (HC Starck, ceramic grade), Na2CO3 (Aldrich,
99.5%), and La2O3 (Alfa Aesar, 99.99%) were mixed together and calcined at 1170 K
for 1 hour in air followed by intermediate grinding at ambient temperatures and then
calcined in air at 1420 K for 10 hours. The molar ratio of Na:La:Ta was = 1-X:X:1,
with an excess amount of sodium (5% mol) used to compensate for Na volatilization.
The optimized doping of 2 mol% of lanthanum was used in this study. The NiO (0.2
50
wt. %) was subsequently added to the photocatalyst by impregnation of an aqueous

solution of Ni(NO3)2*6H2O (Aldrich, 99.999%). The powdered photocatalyst was
placed into a porcelain crucible and heated over a water bath and the suspension was
stirred using a glass rod until the solution was completely evaporated. The dried
powder was then mildly calcined at 540 K for in air for 1 hour.
The Raman spectra of the photocatalysts were obtained on a Lab Ram-HR
Raman spectrometer (Horiba-Jobin Yvon) equipped with visible (532 nm) laser
excitation utilizing a confocal microscope (Olympus BX-30) for focusing the laser on
the catalyst sample. The visible laser excitation was generated by Nd:YAG laser (10
mW) with the scattered photons directed into a single monochromator and focused
onto a UV-sensitive liquid-N2 cooled CCD detector (Horiba-Jobin Yvon CCD-3000V)
having a spectral resolution of 2 cm-1 for the given parameters. About 5-10 mg of the
catalyst was placed into a high temperature in situ cell (Linkam TS-1500) with a
quartz window and cooled with flowing water. The catalyst samples were treated at
673 K for 1 h in flowing 10% O2/He (Airgas, 30 mL/min) to desorb the adsorbed
moisture and the spectra of the dehydrated samples were collected after cooling the
catalysts back to 373 K in the flowing 10% O2/He gas to ensure that the catalyst
surface was void of moisture. The spectral acquisition time employed was 5 scans of 5
seconds/scan for each spectrum.
System alignment was verified using a silica
reference standard line at 520.7 cm-1.

2.2.3 In Situ UV-Vis Diffuse Reflectance Spectroscopy
51
The UV-Vis Diffuse Reflectance spectra were obtained using a Varian Cary 5E
UV-vis spectrophotometer with a diffuse reflectance attachment (Harrick Praying
Mantis Attachment, DRA-2). The finely ground powder catalyst samples (~20 mg)
were loaded into an in situ cell (Harrick, HVC-DR2) and measured in the 200-800 nm
spectral region with a magnesium oxide reflectance standard used as the baseline. The
UV-vis spectra of the photocatalysts were obtained after the samples were treated at
673 K for 1 h in flowing 10% O2/He (Airgas, 30 mL/min) to desorb the adsorbed
moisture. Below 300 nm, the absorbance signal was unacceptably noisy and a filter
(Varian, 1.5 ABS) was employed to minimize the background noise. Determination of
the Kubelka-Munk function, F ( R ) , was obtained from the UV-vis DRS absorbance
and processed with Microsoft Excel software. The UV-vis edge energy ( E g ) was
determined by finding the intercept of the straight line in the low-energy rise of a plot
1
of [ F ( R )hv] n , where n 0.5 for the direct allowed transition versus hv , where hv is
the energy of the incident photon [16-18].
2.2.4 Photoluminescence Spectroscopy and Lifetime Emissions Decay
Photoluminescence spectra and lifetime emissions decay were obtained using a
tunable Ti:sapphire laser (Mira 900, Coherent), generating 5 ps pulses with 76 MHz
repetition rate and pumped with a frequency-doubled Nd:YVO4 laser (Coherent Verdi
V-18), set at 267 nm and was directed into a tunable Raman/photoluminescence
system (Jobin Yvon Horiba, T64000) with UV objective lens to focus the laser onto
52
the sample. About 5-10 mg of the catalyst sample was placed into a high temperature
in situ cell (Linkam TS-1500) with a quartz window and cooled with flowing water.
The cell was pretreated at 673 K by flowing 10% O2/N2 for 30 minutes to desorb
adsorbed moisture and then cooled back down to 298 K in flowing N2 where the
photoluminescence spectrum was obtained. The emission spectrum was collected in
the 366-700 nm range.
The peak of the photoluminescence spectrum was then
subsequently used as the emissions decay window for photoluminescence lifetime

measurements.
For lifetime decay experiments, the luminescence light was
subsequently backscattered through the objective lenses and focused onto a slit of a
triple-monochromator equipped with a fast gated intensified charge coupled device
(ICCD) camera (Picostar HR12, LaVision). The gate width was set to 500 ps and the
maximum delay was determined by the repetition rate of the Ti:sapphire laser, ~13200
ps. The laser energy at the sample was maintained at approximately 20 mW to
prevent laser-induced sample damage. The experimental decay curves were first fitted
to a simple first order exponential decay model:
t
y A1 exp y 0
t1
(1)
A double first order exponential decay biexponential model was also used to
account for the case where the photoluminescence decay can be described to the decay
of two different excited species back to their ground states independent of one another
[19, 20]:
53
t
t
y A1 exp A2 exp y0
t1
t2
(2)
2.2.5 High Resolution X-ray Photoelectron Spectroscopy (HR-XPS)
The HR-XPS
spectra of the catalysts were obtained on a Scienta ESCA 300
spectrometer equipped with a 300 mm hemispherical electrostatic analyzer and a

monochromatic Al K X-ray source with energy of 1486.6 eV generated from a
rotating anode.
This allows for improved chemical selectivity by narrowing the
spectral peaks of elements and greatly reducing the spectral background signal
compared to conventional XPS spectrometers. Each spectrum was calibrated using a
binding energy (BE) value of 285.0 eV for carbon in the C1s region. The atomic
concentration ratios were calculated by correcting the measured peak area ratios with
relative sensitivity factors employed in the Casa XPS software version 2.3.15
2.2.6 High Sensitivity Low Energy Ion Scattering (HS-LEIS) Spectroscopy
Analysis of the outermost surface layer of the photocatalysts was obtained
with the Qtac100 HS-LEIS Spectrometer (ION-TOF) equipped with a highly sensitive
double torodal analyzer, 3000 times higher sensitivity than conventional LEIS
spectrometers, which allows for static depth profiling. The photocatalyst samples
were first gently cleaned with atomic oxygen to remove surface hydrocarbon
contamination from the atmosphere prior to being transferred inside the analysis
chamber. The HS-LEIS spectra were collected using both 3000 eV He+ with a 8600
pA current and 4000 eV Ne+ with a 2830 pA current as ion sources. For depth
54
profiling, the surface was sputtered by Ar+ gas at 1000 eV at a sputter yield of 1x1015
ions/cm2.
2.3 Results
The in situ Raman spectra of the tantalum-based photocatalysts are shown in
Figure 1. The Raman bands of the bulk Ta2O5 are indicative of the crystalline Ta2O5
(L) phase [21]. The largest band in the spectrum for the Ta2O5 (L) phase is at 100 cm-1
originating from a lattice photon mode along with the band at 199 cm-1. The Raman
bands at 256 and 338 cm-1 have been assigned to Ta-O-Ta and TaO6 bending modes,
respectively. The bands at 489, 631, 708 and 848 cm-1 are assigned to Ta-O-Ta
symmetric stretching, Ta-O symmetric stretching, Ta-O-Ta antisymmetric stretching
and higher order Ta-O symmetric stretching modes, respectively.
The solid state synthesis between Na2CO3 and Ta2O5 greatly changes the bulk
crystal structure of the photocatalyst and the Raman spectrum for the undoped
NaTaO3 photocatalyst is shown in Figure 1. The unmodified NaTaO3 contains Raman
bands at 133, 152, 195 and 213 cm-1 that can be assigned to Na translational vibration
modes [22]. The bands at 261 and 313 cm-1 are assignable to bending modes for TaO6
and the bands at 451, 498, and 629 cm-1 are assignable to Ta-O stretching modes. The
doping of La into the NaTaO3 does not result in any apparent changes to the NaTaO3
Raman spectrum, which suggests that La doping did not significantly perturb the bulk
NaTaO3 structure. Crystalline La2O3 bands expected at 104, 191 and 411 cm-1 are not
55
present [23] indicating that La2O3 is present as an amorphous phase in the NaTaO3
photocatalyst. The crystalline NiO broad overlapping bands at 460 and 500 cm-1 are
not detected either because of their relatively weak Raman bands [24] or the presence
of dispersed NiO species. Thus, the bulk molecular structure of NaTaO3 is not
affected by the addition of the lanthanum and nickel oxide dopants.
2.3.2 UV-Vis Diffuse Reflectance Spectroscopy
The UV-Vis DRS Eg values for the tantalum-based photocatalysts are given in
Table 1. The bulk band gap energies are comparable for all the tantalum-based
photocatalysts and are in agreement with previously reported Eg values in the
literature [7, 11]. The addition of La2O3 and NiO does not have much of an effect on
the band gap energy for the NaTaO3 reflecting the dominant contribution of the
NaTaO3 component.
2.3.3 Photoluminescence Spectroscopy and PL lifetime decay
The photoluminescence emission spectra at 267 nm excitation for the
tantalum-based photocatalysts are presented in Figure 2.
The bulk Ta2O5
photocatalyst exhibits a very intense and broad peak with a peak maximum at 525 nm.
The intensity of PL emissions spectra can give information on the crystallinity of the
Ta2O5 and NaTaO3 [25]. The strong PL emission spectrum indicates the presence of
electron/hole recombination sites in Ta2O5 such as defect sites.
In contrast, the
undoped NaTaO3 PL spectrum is much broader and the peak intensity is significantly
suppressed reflecting a more ordered crystalline structure with less defect sites. The
addition of La2O3 and NiO to the NaTaO3 results in further suppression of the
56
photoluminescence emission intensity suggesting that these dopants further decrease

the concentration of defect sites.
The photoluminescence decay curves for the tantalum photocatalysts are
plotted in Figure 3. Equation 1 and Equation 2 were used to model emissions decay
for the tantalum-based photocatalysts and the fit parameters are given in Table 2. The
bulk Ta2O5 photocatalyst was found to fit the simple first order exponential decay
model while the NaTaO3 photocatalysts fit the biexponential model. The observed
simple first-order exponential decay model for the bulk Ta2O5 catalyst can be
attributed to the homogeneity of this bulk catalysts, where the emission decay is
dominated by only one type of excited tantalum species decaying back to the ground
state [19]. This single species decay is not seen in the NaTaO3 photocatalysts where
multiple excited species exist and decay back to their ground states independently of
each other. Two different regimes in the decay curves can be observed for the NaTaO3
photocatalysts that are identified as the fast component of decay, which dominates at
early decay times, and the slow component of the decay, which dominate at later
decay times. The parameters t1 and t2 refer to the decay constants for the fast and
slow components and A1 and A2 refer to the amplitudes of the fast and slow
components, respectively.
The PL emissions decay is related to the lifetimes of the photo-generated
electron/hole pairs with slower decay rates indicative of a longer lifetime [26, 27]. The
decay constant for Ta2O5 is on the same timescale as the NaTaO3 photocatalysts for
the slow component of decay.
The larger decay constant, t2, for NaTaO3 is

57
indicative of longer decay lifetimes compared to the Ta2O5. For modified NaTaO3
photocatalysts, the addition of La2O3 and NiO leads to a subsequent decrease of the
decay constant of the fast component (t1) with a concurrent increase in the
amplitude (A1) of the fast component. Also the addition of NiO shows a decrease
for the decay constant of the slow component (t2) while the addition of La2O3 leads
to an increase in the decay constant of the slow component (t2).
HR-XPS was employed to determine the elemental composition of the surface region
(~1-3 nm) for the NiO/NaTaO3:La photocatalyst. The XPS survey spectrum for the
NiO/NaTaO3:La photocatalyst is presented in Figure 4. The surface region consists
primarily of Na, La, O and Ta. The main binding energy peak for nickel (Ni 2p) is not
detected due to overlap with the strong La 3d binding energy peak. The appearance of
the Ni LMM Auger peak confirms that Ni is indeed present in the surface region, but
the amount cannot be quantified because of the overlap of the Ni 2p peak with the La
3d peak. The atomic concentrations of the elements in the surface region are tabulated
in Table 3. The bulk atomic concentration for lanthanum is 0.4% and its 3.7%
concentration in the surface region reveals that La is significantly surface enriched in
the NiO/NaTaO3:La photocatalyst system.
2.3.5 High Sensitivity-Low Energy Ion Scattering Spectroscopy
The atomic composition of the outermost surface layer (~0.3 nm) of the 0.2%
NiO/NaTaO3:2%La photocatalyst was determined by HS-LEIS. The HS-LEIS spectra
for the outermost layer of the 0.2% NiO/NaTaO3:2%La photocatalyst, using both He+
58
and Ne+ ion gases, are shown in Figure 5. For the He+ HS-LEIS spectrum, scattering
from Na, O and Ni atoms with low atomic mass are detected in the topmost surface
layer, but He+ ions cannot readily distinguish between the higher atomic mass
elements of La (139 amu) and Ta (181 amu). With Ne+ ions, however, the La and Ta
HS-LEIS peaks can readily be resolved and both elements are also found to be present
on the topmost surface layer of the NiO/NaTaO3:La photocatalyst.
HS-LEIS depth profiling analysis of the NiO/NaTaO3:La photocatalyst was
undertaken to determine its elemental composition with distance from the outermost
surface and are presented in Figure 6. For the bulk components O and Na the signal is
very strong for the outermost layer and increase in intensity with the depth profile
while the La/Ta signal is broader in intensity but also increases with the depth profile
as shown in Figure 6a.
The HS-LEIS Ni peak is only present in the first few
sputtering cycles of Figure 6a revealing that Ni is only present on the outermost layers
of this photocatalyst. The HS-LEIS signal for the La peak in Figure 6b is initially very
strong in the first few layers and decreases with sputtering reflecting its surface
segregation in the NiO/NaTaO3:La photocatalyst system The appearance of a small
La peak at the end of the sputtering indicates that a small concentration of La is also
present in the bulk NiO/NaTaO3:La photocatalyst. In contrast, the HS-LEIS signal for
Ta is extremely small in the outermost surface layer and monotonically increases with
sputtering reflecting its diminished concentration on the outermost surface and in the
surface region. Unlike Ta, the HS-LEIS spectra reveal that Na is present on the
59
outermost surface and that its concentration increases with sputtering suggesting some
surface segregation of Na in the NiO/NaTaO3:La photocatalyst.
2.4 Discussion
2.4.1 Bulk Molecular and Electronic Structures of NaTaO3 Photocatalysts
The bulk molecular structure of the NaTaO3 photocatalysts is not perturbed by
the addition of the La2O3 and NiO promoters since the pure NaTaO3 and promoted
NaTaO3 photocatalysts exhibit the same Raman spectra of the bulk structure. The
bulk electronic structure of the NaTaO3 photocatalysts is also not perturbed by the
addition of the La2O3 and NiO promoters since the pure NaTaO3 and promoted
NaTaO3 photocatalysts possess essentially the same optical band gap value of ~4.1-4.2
eV. The similar bulk molecular and electronic structures of the NaTaO3 photocatalysts
suggest that the La2O3 and NiO promoters are minimally incorporated into the bulk
NaTaO3 lattice.
2.4.2 Surface Composition of Promoted NaTaO3 Photocatalyst
The HS-LEIS analysis of the outermost surface layer (~0.3 nm) of the
promoted NiO/NaTaO3:La photocatalyst shows the presence of O, Na, Ta, Ni and La.
HS-LEIS depth profile analysis demonstrates that both La and Ni are surface
segregated in the promoted photocatalyst since their concentrations decrease during
depth profiling. HR-XPS analysis indicates that the La is enriched by a factor of ~10
in the surface region (~1-3 nm) and probably much greater in the topmost surface
layer. Although Na and a minor amount of Ta are present in the outermost surface
60
layer of the promoted NiO/NaTaO3:La photocatalyst, HS-LEIS depth profiling

indicates that their concentrations are not surface enriched because the intensity of
their HS-LEIS signals increase with depth profiling, especially that of Ta. The only
element whose concentration remains relatively constant is O and reflects the oxide
nature of the promoted NiO/NaTaO3:La photocatalyst. The surface segregation of
NiO and La2O3 in the promoted NiO/NaTaO3:La photocatalyst is consistent with the
synthesis methods employed.
The promoted NaTaO3:La photocatalyst was synthesized by the solid-state
method from physically mixed Ta2O5, NaCO3 and La2O3 at elevated temperatures
(1170-1420 K). Under these conditions, Ta2O5 reacts with NaCO3 to form bulk
NaTaO3 because the molten state of basic Na at such extreme temperatures readily
reacts with the acidic Ta2O5 [28, 29]. The molten basic Na does not have an affinity
for reacting with the basic La2O3 and the low mobility of La2O3 [melting point 2588
K] at these temperatures limits the reaction between acidic Ta2O5 and basic La2O3.
Consequently, La2O3 is not extensively incorporated into the bulk NaTaO3 structure
and remains in the surface region and the topmost surface layer of the promoted
NaTaO3:La photocatalyst. Lanthanum oxide is also well known to be a good additive
for inhibiting particle sintering at high temperatures results in increased BET surface
area and stabilization of small particles [30, 31]. The promotion of NaTaO3 with
La2O3 resulted in an increased BET surface area by a factor of ~8 (see Table 4). The
NiO promoter was added by impregnation of aqueous Ni(NO3)2*6H2O, drying and
calcination at 540 K for only 1 hour. This mild calcination treatment is not sufficient
61
to create a solid-state reaction between NiO and NaTaO3:La and, consequently, Ni is

also surface segregated in the promoted NaTaO3:La photocatalyst. In summary, basic
Na reacted with acidic Ta2O5 at the extreme calcination temperatures to form the bulk
NaTaO3 mixed oxide structure and La2O3 and NiO were not incorporated into the bulk
NaTaO3 and remained segregated in the surface region and topmost surface layer.
2.4.3 Bulk and Surface Structures of the Promoted NaTaO3:La
Photocatalyst
A schematic of the bulk and surface structures of the promoted NaTaO3:La
photocatalyst is depicted in Figure 7. The NaTaO3 bulk phase has a perovskite
structure with ABO3 stoichiometry [22]. Perovskite ABO3 structures have been shown
to be preferentially surface enriched with the A cation (Na) and surface depleted with
the B cation (Ta) [29, 32, 33], which is consistent with the HS-LEIS depth profiling
measurements in this study. The surface segregation of NiO and La2O3 is revealed by
the depth profiling HS-LEIS measurements and the HR-XPS surface analyses, and
indicated in the schematic of the promoted NaTaO3:La photocatalyst.
2.4.4 Generation of Excited Electron/Hole Pairs and Their Lifetimes
The main function of the bulk NaTaO3 mixed oxide support is to control the
materials optical band gap that generates excited electron/hole pairs for the
photocatalytic reactions with water taking place at the surface of the photocatalyst.
The same bulk NaTaO3 structure and optical band gap for the unpromoted and
promoted NaTaO3 photocatalysts implies that the generation of electron/hole pairs is
the same for all the NaTaO3 photocatalysts.
62
The recombination of excited electrons and holes is significantly affected by

promotion of NaTaO3 by the surface NiO and La2O3 additives as reflected by their PL
spectra (see Figure 2). PL spectroscopy monitors the recombination of electrons and
holes and the decrease in the intensity of the PL signal reflects the ability of surface
NiO and La2O3 promoters to create efficient electron traps that help prevent
electron/hole recombination and, thus, become available for photocatalysis [34-36].
The increase in the number of electron traps for La promoted NaTaO3, however, is
directly related to the factor of ~10 increase in surface area of this photocatalytic
material. The increase in the number of electron traps with the NiO promoter reflects
the photoproperties of NiO since the unpromoted and Ni-promoted photocatalyst
possess the same surface area.
The decay part of the PL spectra contains information about the lifetime of the
excited electrons/holes, usually reflected by the slow t2 component, and the ratio
is indicative of the relative population of these long lived electrons with slow
emissions decay [26, 27]. The addition of the efficient NiO electrons traps and the
higher surface area of the La-promoted NaTaO3 also dramatically diminishes the
relative contribution of the slow component of emissions decay with a greater
population of electrons with fast decay lifetimes. Although it is desirable for a
photocatalyst to have a greater population of long lived excited electrons with slow
decay lifetimes, the trapping of the excited electrons by the surface NiO and the high
63
surface area allows more electrons and holes to perform photocatalysis at the oxidewater interface.
2.4.5 Structure-Activity Relationships for Splitting of H2O by NaTaO3
Photocatalysts
The photocatalysis community typically normalizes their activity results per
gram of photocatalyst.
The current findings for the promoted NaTaO3 catalysts,
however, demonstrate that the promoters are only altering the surface characteristics
of the photocatalyst and suggest that the photoactivity should instead be normalized
per unit surface area.
The La2O3 promoter has been proposed to be an electronic promoter for the
NaTaO3 photocatalyst, but when the photoactivity for NaTaO3 and NaTaO3:La are
normalized per unit surface area the photoactivity for both of these catalysts is
essentially the same (see Table 4). This indicates that La2O3 is not an electronic
promoter, but acts as a textural promoter that enhances the BET surface area by a
factor of ~10. As mentioned above, lanthanum oxide is known to be a good additive
for inhibiting particle sintering at high temperatures and stabilizing high BET surface
area [30, 31]. The NiO, however, is an electronic promoter since it dramatically
increases the photoactivity per unit surface area by a factor of ~101-102 both in the
presence and absence of the La2O3 promoter and does not affect the overall BET
surface area of the photocatalyst (see Table 4).
The accepted model for the promoted NiO/NaTaO3:La photocatalyst is that
there is a synergistic interaction of NiO preferentially self-assembling at nanostep
64
structures created by the doped La2O3 [11]. This conclusion was only reached because
the simultaneous addition of NiO and La2O3 to NaTaO3 resulted in extremely
enhanced overall photoactivity for water splitting. The current study employs the
specific photoactivity values for water splitting and demonstrates that La2O3 is a
textural promoter and only NiO is an electronic promoter with no synergistic
interactions between the NiO and La2O3 promoters.
The current study also
demonstrates that it is necessary to compare specific photoactivity rates in order to

fully understand the fundamental structure-activity relationships for photocatalysts.
2.5 Conclusions
The bulk and surface properties of promoted NiO/NaTaO3:La photocatalysts
were investigated with bulk (Raman, UV-vis and PL) and surface (HS-LEIS and HRXPS) spectroscopy. The bulk NaTaO3 perovskite molecular and electronic structures
are not affected by the La2O3 and NiO promoters, which means that the
photogenerated excited electron/hole pairs are the same for all the NaTaO3-based
photocatalysts. Both promoters are surface segregated on the NaTaO3 particles. The
La2O3 additive is a structural promoter that stabilizes small NaTaO3 particles and
increases the surface area by a factor of ~10, but does not affect the specific
photoactivity for water splitting. The higher surface area of the La2O3 promoted
NaTaO3 photocatalyst also increases the total number of defect sites that trap excited
electrons. The NiO additive is an electronic promoter that increases the specific
photoactivity for water splitting by a factor of 101-102 in the presence or absence of
65
the La2O3 promoter.

photocatalysis
The new findings provide fundamental insights about the
mechanism
of
promoted
NiO/NaTaO3:La
photocatalysts
by
emphasizing the role of the surface catalytic active sites and the need to normalize the
photoactivity per unit surface area. The accepted practice of normalizing
photocatalytic performance per unit mass is not fundamentally meaningful since the
splitting of water occurs at surface catalytic active sites and leads to incorrect
photocatalytic models in the literature.
Acknowledgements
This research was financially supported by the Department of Energy grant: DOEFG02-93ER14350. Special thanks goes out to Alexander Puretsky at the Center for
Nanophase Materials Sciences at Oak Ridge National Laboratory for help in obtaining
PL data in conjunction with User Project CNMS2008-075. Thanks to Masashi Tabata
and Dr. Kazunari Domen for help in the synthesis of the NaTaO3 catalysts. The
assistance of Dr. A. Miller at Lehigh University in obtaining and interpreting the HRXPS and HS-LEIS data is also gratefully acknowledged.
66
References
[1] A. Fujishima, K. Honda, "Electrochemical photolysis of water at a semiconductor
electrode." Nature (London), 238, 37, (1972).


Water." Chem. Mater., 20, 35, (2008).

Chem. Soc. Rev., 38, 253, (2009).

[6] K. Sayama, H. Arakawa, "Effect of Na2CO3 addition on photocatalytic

decomposition of liquid water over various semiconductor catalysts." J. Photochem.
Photobiol. , A, 77, 243, (1994).
[7] H. Kato, A. Kudo, "New tantalate photocatalysts for water decomposition into H2
and O2." Chem. Phys. Lett., 295, 487, (1998).
[8] H. Kato, A. Kudo, "Water Splitting into H2 and O2 on Alkali Tantalate

Photocatalysts ATaO3 (A = Li, Na, and K)." J. Phys. Chem. B, 105, 4285, (2001).
67
[9] A. Kudo, H. Kato, "Photocatalytic decomposition of water into H2 and O2 over

novel photocatalyst K3Ta3Si2O13 with pillared structure consisting of three TaO6
chains." Chem. Lett., 867, (1997).
[10] H. Kato, A. Kudo, "Photocatalytic Decomposition of Pure Water into H2 and O2

over SrTa2O6 Prepared by a Flux Method. " Chem. Lett., 1207, (1999).
[11] H. Kato, K. Asakura, and A. Kudo, "Highly Efficient Water Splitting into H2 and
O2 over Lanthanum-Doped NaTaO3 Photocatalysts with High Crystallinity and
Surface Nanostructure." J. Am. Chem. Soc., 125, 3082, (2003).
[12] A. Kudo, H. Kato, "Effect of lanthanide-doping into NaTaO3 photocatalysts for

efficient water splitting." Chem. Phys. Lett., 331, 373, (2000).
[13] A. Kudo, H. Kato, and I. Tsuji, "Strategies for the Development of Visible-lightdriven Photocatalysts for Water Splitting," Chem. Lett., 33, 1534, (2004).
Compounds " Angew. Chem. Int. Ed Engl., 49, 8037-8041, (2010).
[15] I.E. Wachs, K. Routray, "Catalysis Science of Bulk Mixed Oxides " ACS
Catalysis, 2, 1235 <last_page> 1246, (2012).

68

104, 1261, (2000).

Rises and Decays," Kinetics and Catalysis, 44, 5-15, (2003) .
[21] Y. Chen, J.L.G. Fierro, T. Tanaka, and I.E. Wachs, "Supported Tantalum Oxide
Catalysts: Synthesis, Physical Characterization, and Methanol Oxidation Chemical
Probe Reaction," The Journal of Physical Chemistry B, 107, 5243, (2003).
[22] N.G. Teixeira, A. Dias, and R.L. Moreira, "Raman scattering study of the high
temperature phase transitions of NaTaO3," Journal of the European Ceramic Society,
27, 3683, (2007).
[23] G. Burns, F. Dacol, "Anomalous Raman spectra from La2CuO4 " Physical
Review B, 41, 4747, (1990).
69
[24] S.S. Chan, I.E. Wachs, "In situ laser Raman spectroscopy of nickel oxide
supported on gamma-Al2O3 " Journal of Catalysis, 103, 224, (1987).
[25] G.D. Gilliland, "Photoluminescence spectroscopy of crystalline semiconductors "

(2008).
Nanodomains," In preparation for publication, .

(2007).
[29] J.A. Kilner, S.J. Skinner, and H.H. Brongersma, "The isotope exchange depth
profiling (IEDP) technique using SIMS and LEIS " Journal of Solid State
Electrochemistry, 15, 86, (2011).
70
[30] J.S. Church, N.W. Cant, and D.L. Trimm, "Stabilisation of aluminas by rare earth
and alkaline earth ions " Applied Catalysis A: General, 101, 105, (1993).
[31] M. Ferrandon, "Hydrothermal Stabilization by Lanthanum of Mixed Metal

Oxides and Noble Metal Catalysts for Volatile Organic Compound Removal " Journal
of Catalysis, 200, 148, (2001).
[32] I.C. Fullarton, J.-. Jacobs, H.E. Benthem, J.A. Kilner, H.H. Brongersma, P.J.
Scanlon, and B.C.H. Steele, "Study of oxygen ion transport in acceptor doped
samarium cobalt oxide " Ionics, 1, 51, (1995).
[33] M. Viitanen, "Silica poisoning of oxygen membranes " Solid State Ionics, 150,
223, (2002).

[35] C. Hu, C. Tsai, and H. Teng, "Structure Characterization and Tuning of

Perovskite-Like NaTaO3for Applications in Photoluminescence and Photocatalysis " J
Am Ceram Soc, 92, 460, (2009).
[36] C. Hu, Y. Lee, and H. Teng, "Efficient water splitting over Na1-xKxTaO3
photocatalysts with cubic perovskite structure " J.Mater.Chem., (2011).
71
Catalyst
Ta2O5
NaTaO3
0.2% NiO/NaTaO3
NaTaO3: 2%La
0.2% NiO /NaTaO3: 2%La
Eg (eV)
4.1
4.2
4.2
4.2
4.2
Table 2.1: Edge Energy Values from UV-Vis DRS.
72
Excitation 267 nm
Catalyst
Ta2O5
y=A1*exp(-x/t1)+A2*exp(-x/t2)+y0
t1
t2
(ns)
A1 fast
(ns) A2 slow
fast
slow
0
0
5
3.5
A1/
(A1+A2)
A2/
(A1+A2)
1.0000
NaTaO3
0.8
186
17
3.6
0.9810
0.0190
0.2%NiO/NaTaO3
0.5
473
10
2.8
0.9941
0.0059
NaTaO3:2%La
0.2%NiO/NaTaO3:
2%La
0.3
27641
42
2.8
0.9999
0.0001
0.2
1.36E+10
1.0000
0.0000
Table 2.2: Photoluminescence decay fit parameters for catalysts at 267 nm excitation.
73
Elemental Core
Electron
Na 1s
Ta 4d 5/2
O 1s
La 3d 5/2
Ni 2p 3/2
Table
2.3:
XPS
% Atomic
Concentration
10.4
13.5
72.4
3.7
0
surface
region
0.2%NiO/NaTaO3:2%La photocatalyst.
74
atomic
concentration
of
promoted
Photocatalyst
Ta2O5
Photoactivity
(mol H2/h/g) Surface Area (m2/g)
6
4.0
Specific
Photoactivity
(mol
H2/h/m2)
1.5 x 100
1.0% NiO/Ta2O5
1154
4.0
2.9 x 102
NaTaO3
170
0.4
4.3 x 102
0.5%NiO/NaTaO3
2180
0.4
5.5 x 103
NaTaO3:2%La
450
3.2
1.4 x 102
0.2%NiO/NaTaO3:2%La
19800
3.2
6.2 x 103
Table 2.4: Photoactivity of Tantalum Oxide Photocatalysts for Water Splitting with
UV Excitation (>270 nm) obtained from [7, 8, 11].
75
133
195
213
152
261
313
451
498
629
Intensity (a.u.)
NaTaO3
NaTaO3: 2%La
256
338
200
0.2%Ni/NaTaO3: 2%La
631
199
400
708
489
848
600
800
Ta2O5 x10
1000
-1
Raman Shift (cm )
Figure 2.1: Raman Spectra of Tantalum-based Photocatalysts.
76
1200
Intensity (counts)
Ta2O5
NaTaO3:2%La
NaTaO3
0.2% NiO/NaTaO3:2%La
0.2% NiO/NaTaO3
500
600
700
Wavelength (nm)
Figure 2.2: Photoluminescence Spectra at 267 nm Excitation for Tantalum-based

Photocatalysts.
77
NaTaO3
PL Intensity (a.u.)
Ta2O5
2000
4000
6000
8000
10000
12000
14000
2000
4000
6000
0.2%Ni/NaTaO
8000
10000
12000
14000
8000
10000
12000
14000
PL Intensity (a.u.)
NaTaO :2%La
3
2000
4000
6000
8000
10000
12000
14000
2000
4000
6000
Time (ps)
PL Intensity (a.u.)
0.2%Ni/NaTaO :2% La
3
2000
4000
6000
8000
10000
12000
14000
Time (ps)
Figure 2.3:
Photoluminescence Emissions Decay Curves for Tantalum-based
Photocatalysts.
78
O 1s
Intensity (counts)
Na 1s
La 3d5/2
Ni LMM
1000
800
Ta 4d5/2
600
400
200
Binding Energy (eV)
Figure 2.4: XPS survey spectrum of surface region for 0.2%NiO/NaTaO3:2%La

promoted photocatalyst.
79
25
20
Intensity (cts/nC)
Na
15
Ni
La/Ta
10
0
500
1000
1500
2000
2500
3000
Energy (eV)
Intensity (cts/nC)
Ni
La
Ta
0
500
1000
1500
2000
2500
3000
Energy (eV)
Figure 2.5: HS-LEIS spectra for 0.2%NiO/NaTaO3:2%La promoted photocatalyst

80
A
Na
30
25
La/Ta
20
Ni
15
10
Intensity (cts/nC)
35
Sp
ut
te
r
yc
le
s
1000
1500
2000
2500
Energy (eV)
B
0.35
0.30
0.25
La
0.20
0.15
0.10
Intensity (cts/nC)
0.40
Ta
0.05
Sp
ut
te
r
yc
le
0.00
1600
1800
2000
2200
2400
2600
2800
Energy (eV)
Figure 2.6: HS-LEIS Depth Profile for 0.2%NiO/NaTaO3:2%La promoted

81
Figure 2.7: Schematic Diagram of the Bulk and Surface of the Promoted
0.2%NiO/NaTaO3:2%La Photocatalyst.
82
CHAPTER 3
Fundamental Bulk/Surface Structure Photoactivity
Relationships of Supported (Rh2-yCryO3)/GaN Photocatalysts
Abstract
Bulk GaN and supported (Rh2-yCryO3)/GaN photocatalysts were characterized and
investigated for UV activated water splitting. The bulk and surface properties of these
photocatalysts were characterized with Raman, UV-vis, Photoluminescence (PL) High
Resolution-XPS and High Sensitivity-Low Energy Ion Scattering (HS-LEIS)
spectroscopy to assist in the development of fundamental structure photoactivity
relationships.
Raman and UV-vis spectroscopy showed that the molecular and
electronic structures, respectively, of the GaN support were not perturbed by the
deposition of the (Rh2-yCryO3) mixed oxide NPs. High Resolution-XPS and High
Sensitivity-LEIS reveal that the surface regions of GaN and supported (Rh2yCryO3)/GaN
photocatalysts consist of Ga oxynitride and GaO surface layers,
respectively. The supported (Rh2-yCryO3) NPs exclusively consist of Cr+3 and Rh+3
cations.
Photoluminescence (PL) spectroscopy was able to reveal that the (Rh283
yCryO3)
NPs helps to decrease the recombination of electron/holes in the bulk GaN by
acting as efficient electron traps promoting charge transfer to the surface. These new
insights into the surface nature of the (Rh2-yCryO3) NPs show that Rh3+ species on the
outermost surface layer are responsible for enhanced H2 while other surface sites
(Cr3+, GaOx and their contact points) are possibly responsible for O2 production.
3.1 Introduction
Early research on photocatalytic water splitting primarily focused on the use of
semiconductor metal oxide materials with UV irradiation [1-3]. More recently, nonoxide materials have been discovered that are also able to perform photocatalytic
water splitting with visible excitation [4]. Unlike the extensive list (>130) of metal
oxide semiconductors that can perform photocatalytic water splitting, there are only a
handful of these bulk oxynitride materials with d10 electronic configuration that are
active for overall photocatalytic water splitting: Ge3N4 [5-8], (Ga1-xZnx)(N1-xOx) [913], (Zn1+xGe)(N2Ox) [14] and GaN [15-17].
There have also been other bulk
oxynitride materials discovered with d0 electronic configurations (Ti, Ta and Nb) that
are able to produce H2 and O2 with sacrificial reagents, but none are capable of
photocatalytic water splitting in pure water [2, 4].
The limited number of bulk
oxynitride photocatalysts discovered to be capable of water splitting so far suggests

that water splitting is a greater challenge for oxynitrides compared to metal oxide
photocatalysts. Although other mixed metal oxynitride materials are more active than
GaN and can utilize visible light for photocatalytic water splitting, it is important to
84
develop fundamental structure/photoactivity relationships for this basic nitride

photocatalyst because it is a component of (Ga1-xZnx)(N1-xOx) oxynitride
photocatalysts and understanding how the GaN photocatalyst functions is important
for the design of advanced oxynitride photocatalysts.
The present study focuses on the bulk GaN semiconductor photocatalyst.
Although GaN has been found to be unable to photocatalytically split water, it
becomes an active photocatalyst with UV excitation ( > 300 nm) when (Rh2-yCryO3)
mixed oxide NPs are deposited on its surface (quantum efficiency (Q.E.) of 0.7%)
[17].
The present investigation will apply high-resolution X-ray photoelectron
spectroscopy (HR-XPS) and high-sensitivity low energy ion scattering (HS-LEIS)

spectroscopy to provide new insights about the surface region (~1-3nm) and outermost
atomic layer (~0.3 nm) of the supported (Rh2-yCryO3)/GaN photocatalyst system.
Further insight into the molecular and electronic structures of the bulk GaN phase and
influence of the supported (Rh2-yCryO3) NPs upon the bulk GaN phase properties will
be obtained with in situ optical spectroscopic characterization (Raman, UV-vis,
photoluminescence (PL) and time-resolved picosecond PL-Raman).
3.2 Experimental
The bulk GaN was prepared from elemental gallium obtained from Mitsubishi
Chemicals. The GaN was mixed in an evaporating dish with the aqueous precursors
of Cr(NO3)39H2O (Wako Pure Chemicals, 99.9%) and Na3RhCl62H2O (Kanto
85
Chemicals, 97% as Rh) to yield a final composition of 1 wt.% Rh and 1.5wt.% Cr.
This suspension was then placed over a water bath and continuously stirred with a
glass rod until complete evaporation. The powder was then collected and mildly
calcined in air at 623 K for one hour. The supported (Rh2-yCryO3)/GaN photocatalyst
was then washed with distilled water and dried overnight in an oven at 343 K.
The Raman spectra for the photocatalyst were performed on a Lab Ram-HR
excitation and utilizing a confocal microscope (Olympus BX-30) for focusing the laser
on the catalyst sample. The 532nm visible laser excitation was generated by Nd:YAG
laser (10 mW) with the scattered photons directed into a single monochromator and
focused onto a UV-sensitive liquid-N2 cooled CCD detector (Horiba-Jobin Yvon
CCD-3000V) having a spectral resolution of 2 cm-1 for the given parameters. About
5-10 mg of the catalyst was placed into a high temperature in situ cell (Linkam TS1500) with a quartz window and the spectrums were obtained under ambient
conditions. The spectral acquisition time employed was 5 scans of 5 seconds/scan for
each spectrum. System calibration was verified using a silica reference standard line
at 520.7 cm-1.
Ultra
Violet-visible-Near
Infrared
(UV-vis-NIR)
diffuse
reflectance
spectroscopy was utilized to obtain the optical edge energy, Eg, values for the
photocatalysts.
Spectra were obtained using a Varian Cary 5E UV-vis

86
spectrophotometer with a diffuse reflectance attachment (Harrick Praying Mantis

Attachment, DRA-2). The finely ground powder catalyst samples (~20 mg) were
loaded into an in situ cell (Harrick, HVC-DR2) and measured in the 200-800 nm
spectral region with a magnesium oxide reflectance standard used as the baseline. A
filter (Varian, 1.5ABS) was employed to minimize the background noise.
magnesium oxide white reflectance standard baseline was collected under ambient
conditions. Determination of the Kubelka-Munk function, F ( R ) , was obtained from
the UV-vis DRS absorbance and processed with Microsoft Excel software. The edge
energy was determined by finding the intercept of the straight line in the low-energy
1
rise of a plot of [ F ( R )hv] n , where n 0.5 for the direct allowed transition versus hv
, where hv is the energy of the incident photon [18-20].
3.2.4 Photoluminescence Spectroscopy and Photoluminescent Decay
PL spectra and transient PL lifetime measurements of the photocatalysts were
conducted using a Ti:sapphire laser (Coherent Mira 900), tunable in the 685-1000 nm
spectra range, generating 5 ps pulses with a 76 MHz repetition rate, pumped with a
frequency-doubled Nd:YVO4 laser (Coherent Verdi V-18). The output of the laser was
frequency doubled using an ultrafast harmonic generator (Coherent 5-050).
To
perform the luminescence measurements, the excitation light at 400 nm was directed
toward a microscope of a tunable micro-, macro-Raman/photoluminescence system
(Jobin Yvon Horiba, T6400) and was focused using a long distance objective (50x,
N/A=0.5) onto a sample to a spot size of ~ 2 m. The photocatalyst sample was
87
placed into a high temperature in situ microscopy stage (Linkam, TS-1500) and
pretreated as follows: The samples were heated at 10 oC/min to 673 K in flowing 10%
O2/N2 (30 sccm) to remove water, since moisture causes quenching of the PL signal
[21], and to fully oxidize the samples. Upon cooling to room temperature in flowing
inert gas (N2, 30 sccm), the photoluminescence decay measurements were made. To
obtain complete PL spectra, the spectrometer was set to nanometer mode and the
grating was moved several times in order to obtain the intensity over a 400 to 800 nm
range.
The luminescence light was collected through the same objective in
backscattering geometry and focused onto a slit of the triple-monochromator equipped

with a fast gated intensified charge coupled device (ICCD) camera collecting in the
350-900 nm range (LaVision, Picostar HR12). The ICCD camera was gated using a
sequence of 76 MHz pulses propagating with a variable delay relative to the original
train of trigger pulses (76 MHz) from a photodiode in a Ti:sapphire laser. The
minimum gate width was 300 ps and the maximum delay was defined by the laser
repetition rate (~13200 ps). The laser energy at the sample was maintained at
approximately 1.6 mW to prevent photo-degradation of the photocatalyst sample. The
grating was set to monitor decay centered at emission wavelengths from 500 to 700
nm.
3.2.5 High Resolution X-ray Photoelectron (HR-XPS) Spectroscopy
The HR-XPS spectra of the photocatalysts were obtained on a Scienta ESCA
300 spectrometer equipped with a 300 mm hemispherical electrostatic analyzer and a
88
rotating anode.
binding energy (BE) value of 285.0 eV for carbon in the C1s region. The atomic
concentration ratios were calculated by correcting the measured peak area ratios with
relative sensitivity factors employed in the Casa XPS software version 2.3.15.
3.2.6 High Sensitivity Low Energy Ion Scattering (HS- LEISS)
Spectroscopy
Analysis of the outermost surface layer of the photocatalysts was obtained on
the Qtac100 HS-LEIS Spectrometer (ION-TOF) equipped with a highly sensitive
chamber. The HS-LEIS spectra were taken using both 4000 eV 4He+ with 11300 pA
current and 4000 eV 20Ne+ with 478 pA current as ion sources. TOF mass filters were
also utilized for spectra obtained with Ne+ as an ion source for reduced flux
background at low kinetic energies. For depth profiling, the surface was sputtered by
Ar+ gas at 500 eV at a sputter yield of 1x1015 ions/cm2. Metallic Rh and Cr standards
were also analyzed for quantifying the elemental composition of the photocatalysts.
89
3.3.1 Bulk Characteristics of GaN Photocatalysts

The Raman spectra for the ambient GaN photocatalysts are presented in Figure
1 and exhibit bands that are associated with the phonon modes of the hexagonal
wurtzite GaN crystal structure [22, 23]. The Raman band at 140 cm-1 has been
assigned to the E2 phonon mode, the 413 cm-1 band is a combination of optical and
acoustic modes, the 530 and 563 cm-1 bands are the transverse optical (TO) modes of
A1 and E1, respectively, and the 732 cm-1 band is from the combination of the
longitudinal optical modes of A1 and E1 [22, 23].
The Raman spectrum of the
supported (Rh2-yCryO3)/GaN photocatalyst is dominated by the GaN vibrations and is

essentially the same as that of the bulk GaN. The characteristic sharp Raman bands
of crystalline Cr2O3 NPs at 542 and 603 cm-1 [24, 25] are not observed suggesting that
crystalline Cr2O3 NPs are not present. Crystalline Rh2O3 possesses a characteristic
Raman band at 550 cm-1 [26] that cannot be detected because it is overshadowed by
the strong GaN band in this region. Thus, the bulk molecular structure of the GaN
phase is unchanged by the deposition of the (Rh2-yCryO3) mixed oxide NPs.
The bulk optical band gap (Eg) values determined from UV-vis DRS
measurements and are given in Table 1. The optical band gap for bulk GaN is 3.3 eV
and is not perturbed by the deposition of the (Rh2-yCryO3) mixed oxide NPS, which
indicates that the bulk GaN component dominates the UV-vis absorbance spectrum of
the supported (Rh2-yCryO3)/GaN photocatalyst system.
90
3.3.2 Surface Composition of GaN photocatalysts

3.3.2.1 Atomic Composition and Chemical State of the Surface Region
(~1-3 nm)
The XPS survey spectra for the surface region (~1-3 nm) of the GaN
photocatalysts are presented in Figure 2 and the atomic composition is reported in
Table 2. The surface region for the bulk GaN consists only of Ga, O and N and no
contaminants were detected. Note the O/Ga~1 atomic ratio indicating that Ga is
extensively oxidized in the surface region under ambient exposure. The compositions
of the surface region for the supported (Rh2-yCryO3)/GaN and bulk GaN photocatalysts
are compared in Table 2. As expected for deposition of an oxide on GaN, the addition
of the (Rh2-yCryO3) mixed oxide NPs to the GaN support doubles the O concentration,
while decreasing the concentration of the GaN support elements (~50% Ga and ~30%
N).
HR-XPS surface analysis of Cr and Rh for the supported (Rh2-yCryO3)/GaN
photocatalyst was undertaken to determine the chemical states of Cr and Rh in the
surface region and presented in Figure 3. The HR-XPS spectra for the Cr 2p and Rh
3d regions of the supported (Rh2-yCryO3)/GaN photocatalyst reveal the presence of
only Cr3+ and Rh3+ and the absence of Cr6+ or metallic Rh(0) in the surface region.
3.3.2.2 Outermost Atomic Layer Composition (~0.3 nm)
The outermost surface layer (~0.3 nm) and layers below the surface of bulk
GaN were analyzed with dynamic HS-LEIS employing a He+ ion gas source and the
findings are shown in Figure 4. The HS-LEIS signal for N is almost absent from the
91
topmost surface layer and significantly increases in intensity into a definable peak
with further sputtering into the bulk. Although Ga is present on the topmost surface
layer, its HS-LEIS signal also significantly increases with depth profiling. In contrast
to that of Ga and N, the HS-LEIS signal for O is strongest on the outermost layer and
decreases with sputtering into the bulk demonstrating surface enrichment of O. Depth
profiling using Ne+ ion gas source was also performed (not shown for brevity)
showing only a single peak for Ga. The HS-LEIS sputtering findings reveal that for
bulk GaN the outermost surface layer consists of GaOx and that the surface region is
present as a Ga oxynitride (GaOxNy) layer.
The HS-LEIS spectra comparing the untreated and atomic oxygen treated
supported (Rh2-yCryO3)/GaN photocatalyst is presented in Figure 5. This pretreatment
method was used to clean off hydrocarbon deposits on the surface, increasing signal
intensity and does not affect the surface composition of the photocatalyst. The HSLEIS depth profiles for the supported (Rh2-yCryO3)/GaN photocatalyst are presented in
Figure 6. From Figure 6a, the outermost surface layer contains O, Cr, Ga and Rh and
does not contain any detectable N. It appears that the oxidation treatment to form the
supported (Rh2-yCryO3) mixed oxide NPs on the GaN support also oxidized the outer
surface layers of GaN and is consistent with the 100% increase of the XPS O
concentration in the surface region (see Table 2) [27].
More quantitative depth profiling information is provided with Ne+ depth
profiling analysis as shown in Figure 6b and, surprisingly, small concentrations of Sn
and Ba contaminants are also found to be present in the outermost surface layers. The
92
Sn and Ba contaminants were not evident for the bulk GaN and the source of these
impurities is not known, but could be due to handling of the material during and after
impregnation of the Cr(NO3)39H2O and Na3RhCl62H2O precursors. Metallic Cr and
Rh standards were used to quantify the amounts of these two elements with respect to
each other and their quantitative depth profile concentrations are shown in Figure 7.
Both Rh and Cr are surface enriched since their bulk concentrations are 1.0% and
1.5% wt., respectively, while their concentrations in the surface region are 3.2% and
6.7%. The outermost surface layer contains more Cr than Rh, the concentrations of Cr
and Rh are about the same for the first few layers below the outermost surface, and the
concentration of Cr is larger than Rh for the deeper layers below the surface. The
increase in the concentration of Cr at deeper layers is indicative of greater Cr
concentration in the deeper layers of the (Rh2-yCryO3) NPs and the presence of
dissolved Cr beneath the surface. At all depths, the concentrations of Rh and Cr in the
surface region are enriched compared to their bulk concentrations.
3.3.3 Electron/Hole Generation and Recombination
The steady-state PL emissions spectra at 400 nm excitation for the GaN and
supported (Rh2-yCryO3)/GaN photocatalysts are presented in Figure 8.
Both
photocatalysts exhibit broad emission bands with maxima at 710 and 725 nm as
indicated in Table 1. The intensity of the PL emissions spectra is indicative of the
population of electron/hole recombination centers in a photocatalyst [21, 28]. Even
with excitation at 400 nm that is greater than the optical absorption edge energy of the
GaN (3.3 eV or 375 nm); there is still a large population of photo-excited electrons
93
generated in the bulk of the GaN. Deposition of the (Rh2-yCryO3) mixed oxide NPs on
GaN diminishes the PL emissions reflecting that photo-excited electrons are being
trapped by the supported (Rh2-yCryO3) NPs at the mixed oxide/GaN interface and,
therefore, are unavailable for recombining with holes in the GaN bulk. In other words,
the supported (Rh2-yCryO3) NPs on the GaN support act as efficient electron traps for
the photoexcited electrons generated from the GaN bulk and, consequently, minimize
electron/hole recombination and promotes charge transfer to the surface [21, 29, 30].
The transient PL decay curves for the GaN photocatalysts using 400 nm
excitation at different emission wavelengths are plotted in Figure 9. At all the
examined emission wavelengths, both catalysts exhibit a constant emission intensity
signifying no decay emissions. The PL properties of the solid-state GaN have been
well studied due to its importance in optical and electronic applications [31-33].
These studies utilize excitation in the deep UV range (~325 nm) that is able to
generate excited photoelectrons in the GaN so the electron decay can be monitored on
the picosecond timescale. Although a higher light excitation is used (400 nm), the
GaN photocatalysts are able to generate steady-state PL emissions so there are excited
electrons that are decaying and recombining at this wavelength. On the picosecond
timescale, the PL emission measured is generated by fluorescence from the material
[31].
This lack of measurable fluorescence decay emissions on the picosecond
timescale could be due to the powdered GaN exhibiting PL generated from

phosphorescence emissions. Phosphorescence materials tend to emit PL over a longer
time period (ms - hrs) so there will not be detectable emissions decay on the
94
picosecond timescale. The constant PL emissions intensity after the pulsed 400 nm
excitation can, thus, be attributed to the GaN photocatalysts exhibiting
phosphorescence longer than the picosecond timescale.
3.3.4 Bulk/Surface Photoactivity Relationships of GaN Photocatalysts
The supported (Rh2-yCryO3)/GaN photocatalyst functions as a two-component
system. The role of bulk GaN is to generate electron/hole pairs using UV excitation.
Surface analysis was able to show that the GaN bulk phase dynamically changes: GaN
GaOxNy GaOx in the surface region with some dissolved Cr3+ beneath the
outermost surface layer. The (Rh2-yCryO3) NPs are responsible for transferring the
photoexcited electrons to the surface and act as the catalytic active sites. In the
previous work, GaN was unable to photocatalytically split water by itself only (Rh2yCryO3)/GaN
was able to produce stoichiometric amount of H2/O2 [17].
It was
proposed that (Rh2-yCryO3) were the sites for H2 production from using MeOH as a
sacrificial reagent. The GaN was proposed as the sites for O2 production from using
AgNO3 as a sacrificial reagent. For another oxynitride photocatalyst, (Ga1-xZnx)(N1xOx),
the loading of only Rh2O3 resulted in the nonstoichiometric production rate 50
H2 : 1 O2 [10, 34]. However with the addition of Cr2O3 the production of H2 was
increased by ~ 77x with stoichiometric O2 production. These observations show that
Rh3+ is primarily responsible for H2 evolution but it is still not clear which site is
active for O2 evolution. From the surface analysis, the possible sites for O2 evolution
are the Cr3+ in the (Rh2-yCryO3) NPs, the GaOx surface layer or at the contact point
95
between the two.

yCryO3)/GaN
A schematic modeling the photocatalytic process on (Rh2-
is shown in Figure 10.
3.4 Conclusions
The bulk and surface properties of (Rh2-yCryO3)/GaN photocatalysts were
investigated using bulk (Raman, UV-vis, and PL) and surface (HS-LEIS and HR-XPS)
spectroscopic techniques. The bulk molecular and electronic structure of the GaN
phase was not affected by the addition of the (Rh2-yCryO3). The (Rh2-yCryO3) was
found to decrease the recombination of excited electron/holes in the bulk GaN via
charge transfer of the excited electrons to the surface promoting the surface
photocatalytic reaction. Surface analysis was able to reveal that the composition of
the bulk GaN dynamically changes from GaN GaOxNy GaOx near the surface
region. The surface composition of the (Rh2-yCryO3)/GaN was revealed for the first
time showing that Rh3+ present on the outermost surface layer was responsible for
enhanced H2 while the O2 production can possibly be attributed to other surface sites
(Cr3+, GaOx, or at their contact points).
Acknowledgements
This research was financially supported by the Department of Energy grant:
DOE-FG02-93ER14350. Special thanks goes out to Alexander Puretsky at the Center
for Nanophase Materials Sciences at Oak Ridge National Laboratory and along with
Charles A. Roberts for assistance with the PL experiments in conjunction with User
96
Project CNMS2008-075.
Thanks to K. Maeda and Professor K. Domen at the
University of Tokyo for their assistance with the synthesis of the GaN photocatalysts.
The assistance of Dr. A. Miller and R. Pafchek at Lehigh University in obtaining and
interpreting the HR-XPS and HS-LEIS data is also gratefully acknowledged.
97
References
Water." Chem. Mater., 20, 35, (2008).

Chem. Soc. Rev., 38, 253, (2009).

Hydrogen Generation. Chem. Rev., 110, 6503, (2010).

[5] J. Sato, N. Saito, Y. Yamada, K. Maeda, T. Takata, J.N. Kondo, M. Hara, H.

Kobayashi, K. Domen, and Y. Inoue, "RuO2-Loaded beta -Ge3N4 as a Non-Oxide
Photocatalyst for Overall Water Splitting." J. Am. Chem. Soc., 127, 4150, (2005).
[6] Y. Lee, T. Watanabe, T. Takata, M. Hara, M. Yoshimura, and K. Domen, "Effect

of High-Pressure Ammonia Treatment on the Activity of Ge3N4 Photocatalyst for
Overall Water Splitting." J. Phys. Chem. B, 110, 17563, (2006).
[7] K. Maeda, N. Saito, D. Lu, Y. Inoue, and K. Domen, "Photocatalytic Properties of

RuO2-Loaded beta -Ge3N4 for Overall Water Splitting." J. Phys. Chem. C, 111, 4749,
(2007).
98
[8] K. Maeda, N. Saito, Y. Inoue, and K. Domen, "Dependence of Activity and

Stability of Germanium Nitride Powder for Photocatalytic Overall Water Splitting on
Structural Properties." Chem. Mater., 19, 4092, (2007).

Phys. Chem. C, 111, 7554, (2007).

Chem. B, 110, 13753, (2006).

99
[13] K. Maeda, K. Teramura, N. Saito, Y. Inoue, and K. Domen, "Improvement of

splitting by co-loading Cr and another transition metal." J. Catal., 243, 303, (2006).

[15] N. Arai, N. Saito, H. Nishiyama, Y. Inoue, K. Domen, and K. Sato, "Overall

water splitting by RuO2-dispersed divalent-ion-doped GaN photocatalysts with d10
electronic configuration." Chem. Lett., 35, 796, (2006).
[16] N. Arai, N. Saito, H. Nishiyama, K. Domen, H. Kobayashi, K. Sato, and Y.

Inoue, "Effects of divalent metal ion (Mg2+, Zn2+ and Be2+) doping on
photocatalytic activity of ruthenium oxide-loaded gallium nitride for water splitting."
Catal. Today, 129, 407, (2007).
[17] K. Maeda, K. Teramura, N. Saito, Y. Inoue, and K. Domen, "Photocatalytic

overall water splitting on gallium nitride powder." Bull. Chem. Soc. Jpn., 80, 1004,
(2007).
100

104, 1261, (2000).


[22] W. Han, Z. Liu, and H. Yu, "Synthesis and optical properties of GaN/ZnO solid
solution nanocrystals " Appl. Phys. Lett., 96, 183112, (2010).
[23] V. Davydov, Y. Kitaev, I. Goncharuk, A. Smirnov, J. Graul, O. Semchinova, D.

Uffmann, M. Smirnov, A. Mirgorodsky, and R. Evarestov, "Phonon dispersion and
Raman scattering in hexagonal GaN and AlN " Physical Review B, 58, 12899, (1998).
[24] E.L. Lee, I.E. Wachs, "In Situ Spectroscopic Investigation of the Molecular and
Electronic Structures of SiO2 Supported Surface Metal Oxides." J. Phys. Chem. C,
111, 14410, (2007).
[25] D.S. Kim, J.M. Tatibouet, and I.E. Wachs, "Surface structure and reactivity of
chromium trioxide/silica catalysts," J. Catal., 136, 209, (1992).
101
[26] C.T. Williams, C.G. Takoudis, and M.J. Weaver, "Methanol Oxidation on
Rhodium As Probed by Surface-Enhanced Raman and Mass Spectroscopies:
Adsorbate Stability, Reactivity, and Catalytic Relevance " The Journal of Physical
Chemistry B, 102, 406, (1998).

(2007).
[28] G.D. Gilliland, "Photoluminescence spectroscopy of crystalline semiconductors "


Am Ceram Soc, 92, 460, (2009).
photocatalysts with cubic perovskite structure " J.Mater.Chem., (2011).
[31] M.A. Reshchikov, H. Morkoc, "Luminescence properties of defects in GaN," J.

Appl. Phys., 97, 061301, (2005).
[32] S. Nakamura, "III-V nitride based light-emitting devices." Solid State Commun.,
102, 237, (1997).
102
[33] E. Calleja, M. Snchez-Garca, F. Snchez, F. Calle, F. Naranjo, E. Muoz, U.

Jahn, and K. Ploog, "Luminescence properties and defects in GaN nanocolumns
grown by molecular beam epitaxy " Physical Review B, 62, 16826, (2000).
[34] K. MAEDA, K. TERAMURA, N. SAITO, Y. INOUE, and K. DOMEN,

"Improvement of photocatalytic activity of (Ga1-xZnx)(N1-xOx) solid solution for
overall water splitting by co-loading Cr and another transition metal " Journal of
Catalysis, 243, 303, (2006).
103
Excitation 400 nm
Catalyst
Eg (eV)
Peak Maximum (nm)
GaN
(Rh2-yCryO3) /GaN
3.3
3.3
725
710
Table 3.1: Eg values from UV-vis DRS, and peak emission wavelengths from PL
spectroscopy of the GaN photocatalysts at 400 nm excitation.
104
Element
O 1s
N 1s
Ga 2p 3/2
Cr 2p 3/2
Rh 3d
GaN
19.0%
62.0%
19.0%
0.0%
0.0%
Table 3.2:
(Rh2-yCryO3)/GaN
39.5%
41.1%
9.5%
6.7%
3.2%
HR-XPS surface region atomic composition (~1-3nm) of GaN
photocatalysts.
105
Intensity (a.u.)
563
530
732
140
GaN
413
(Rh2-yCryO3)/GaN x10
200
400
600
800
-1
Raman Shift (cm )
Figure 3.1: Raman spectra for GaN photocatalysts (532 nm).
106
1000
1200
Intensity (a.u.)
Ga 2p3/2
O 1s
N 1s
1200
1000
800
600
400
200
200
Binding Energy (eV)
Intensity (a.u.)
Ga 2p3/2
O 1s
Cr 2p3/2
N 1s
Rh 3d
1200
1000
800
600
400
Binding Energy (eV)
Figure 3.2: XPS survey spectra of (a) GaN and (b) supported (Rh2-yCryO3)/GaN
photocatalysts.
107
3+
Cr 2p3/2
B.E.= 576.0 eV
6+
Intensity (a.u.)
Cr 2p3/2
B.E.= 580.0 eV
590
585
580
575
570
565
Binding Energy (eV)
B
3+
Rh 3d3/2
Intensity (a.u.)
B.E.= 309.7 eV
Rh(0) 3d3/2
B.E.= 307.4 eV
320
315
310
305
300
Binding Energy (eV)
Figure 3.3: HR-XPS spectra of (a) Cr 2p and (b) Rh 3d regions of the supported
(Rh/Cr2O3)/GaN photocatalyst.
108
Ga
6
Outermost Layer
2
Ion Dose 1.9 E16 /cm
2
Ion Dose 6.7 E16 / cm
2
Ion Dose 1.2 E17 / cm
Intensity (cts/nC)
0
1000
2000
3000
4000
Energy (eV)
Figure 3.4: HS-LEIS Depth Profile for GaN photocatalyst using He+ ion gas.
109
10
Intensity (cts/nC)
Before O Treatment
Cr
Zn/Ga
Rh
After O Treatment
0
1000
2000
3000
4000
Energy (eV)
Figure 3.5: HS-LEIS spectra of untreated and atomic O treated supported (Rh2yCryO3)/GaN
photocatalyst.
110
4
3
Cr
2
Ga
Intensity (cts/nC)
Rh
Sp
ut
te
rC
yc
le
1000
1500
2000
2500
3000
Energy (eV)
3500
4000
B
Ga
Cr
25
20
15
Rh
10
5
Ba
Sp
ut
te
rC
yc
le
Sn
Intensity (cts/nC)
30
500
1000
1500
2000
2500
3000
Energy (eV)
Figure 3.6: HS-LEIS depth profile for the supported (Rh2-yCryO3)/GaN photocatalyst
111
Cr
Rh
10
9
% of Pure Metal Monolayer
8
7
6
5
4
3
2
1
0
0
10
12
14
16
Sputter Cycle
Figure 3.7: HS-LEIS Depth Profile of Cr and Rh for supported (Rh2-yCryO3) /GaN
photocatalyst using Ne+ ion gas.
112
Intensity (Counts)
GaN
(Rh2-yCryO3)/GaN
450
500
550
600
650
700
750
800
Wavelength (nm)
Figure 3.8: PL emissions spectra for GaN photocatalysts (400 nm excitation).
113
A
550 nm
600 nm
Intensity (Counts)
500 nm
700 nm
650 nm
2000
4000
6000
8000
10000
12000
14000
Delay Time (ps)
550 nm
Intensity (Counts)
600 nm
650 nm
500 nm
700 nm
0
2000
4000
6000
8000
10000
12000
Delay Time (ps)
Figure 3.9: PL lifetimes spectra for (a) GaN and (b) (Rh2-yCryO3)/GaN photocatalysts
at different emissions wavelengths (400 nm excitation).
114
Figure 3.10: Schematic of (Rh2-yCryO3)/GaN photocatalysts.
115
CHAPTER 4
Nature of Catalytic Active Surface Sites Present on Bulk
(Ga1-xZnx)(N1-xOx) Photocatalysts
Abstract
The
supported
mixed
oxide
(Rh2-yCryO3)/(Ga1-xZnx)(N1-xOx)
and
core/shell
(Rh0/Cr2O3)/(Ga1-xZnx)(N1-xOx) photocatalyst, active for splitting of H2O, were

synthesized, extensively characterized for their bulk and surface properties, and
examined for water splitting with the objective of developing fundamental structurephotoactivity relationships.
Raman and UV-vis spectroscopy revealed that the
molecular and electronic structures, respectively, of the oxynitride (Ga1-xZnx)(N1-xOx)

support are not perturbed by the deposition of the Rh-Cr NPs. Photoluminescence
(PL) spectroscopy, however, showed that the oxynitride (Ga1-xZnx)(N1-xOx) support is
the source of excited electrons/holes and the Rh-Cr NPs greatly reduce the undesirable
recombination of photoexcited electron/holes by acting as efficient electron traps as
well as increase the lifetimes of the excitons.
High Resolution-XPS and High
Sensitivity-LEIS surface analyses reveal that the surfaces of the Rh-Cr NPs consist of
Rh+3 and Cr+3 mixed oxide NPs in spite of the desire to exclusively synthesize metallic
116
Rh0 NPs that are encapsulated by Cr2O3 films for the core/shell catalysts.
The
function of the Rh-Cr-O mixed oxide NPs is to trap the excited electrons and holes to
harness them for the photocatalytic splitting of water. The Rh+3 are the H2 evolution
sites and Cr+3, GaZnOx or their contact points are the possible O2 evolution sites. The
presence of some exposed metallic Rh0 in the core/shell photocatalyst, however,
decreases the photocatalytic efficiency relative to the mixed oxide photocatalyst by
catalyzing the undesirable back reaction between H2 and O2 to form water. The
current investigation establishes the fundamental structure-photoactivity relationships
of these visible light activated photocatalysts. This is also the first study to employ
High Sensitivity-Low Energy Ion Scattering to determine the composition of the
outermost surface layers of photocatalysts that is critical for a complete understanding
of complex, multicomponent photocatalysts.
4.1 Introduction
Throughout much of the 40 years of research on photocatalytic water splitting,
efforts have been dedicated towards finding suitable materials that are able to produce
H2 and O2 at significant quantum efficiency (Q.E.).
The focus has been on
discovering semiconductor mixed metal oxide materials that are able to efficiently
produce H2 and O2 with NiO/NaTaO3:La being the most active semiconductor metal
oxide photocatalyst system under UV irradiation with a Q.E. of 56% [1-4]. This
achievement demonstrated that it is possible for a photocatalyst system to be more
efficient than the targeted goal of 30% Q.E. for the start of commercialization of
117
photocatalytic hydrogen production on the industrial scale [3]. The goal, however, is
for 30% Q.E. under visible light irradiation and not just UV light irradiation for
semiconductor mixed metal oxide photocatalyst systems.
The research focus has now shifted towards the design of novel photocatalytic
materials that are able to photocatalytic split water under visible light irradiation. The
design of advanced, robust solar powered photocatalyst systems is the ultimate
objective in photocatalysis research since the visible region covers a much larger
portion of the suns radiation compared to the narrow UV region [5]. The large band
gap energies of semiconductor mixed metal oxides hinder their ability to fully utilize
visible light energy.
Only a few of these semiconductor mixed metal oxides
(NiO/InTaO4 (Q.E. = 0.7%) [6], RuO2/InTaO4 (Q.E. = 0.4%) [7] and RuO2/YBiWO6
(Q.E. = 0.2%) [8]) have been found to be able to perform the water splitting reaction
under visible light excitation with very low quantum efficiencies.
In order to overcome the low Q.E. of mixed metal oxides, the Domen research
group investigated the calcination of physically mixed metal oxides under NH3 flow
that form novel, advanced bulk mixed metal oxynitride materials, (Ga1-xZnx)(N1-xOx)
and (Zn1+xGe)(N2Ox), that are able to generate electron/hole pairs under visible light
irradiation for photocatalytic water splitting [1, 9, 10]. Whereas GaN absorbs in the
UV region (3.3 eV), the mixed (Ga1-xZnx)(N1-xOx) oxynitride solid solution absorbs in
the visible region (2.4-2.8 eV with increasing ZnO lowering the band gap) that allows
for absorption of visible light for electron/hole generation. The narrowing of the band
gap is attributed to the presence of p-d repulsion between Zn3d and N2p electrons that
118
increases the valence band maximum allowing for visible light irradiation [11-13].
Although (Ga1-xZnx)(N1-xOx) generates electron/holes pairs in its bulk phase, it is
unable to photocatalytically split water at its surface.
xZnx)(N1-xOx)
Modification of the (Ga1-
with various transition metal oxides (Ni, Ru, Rh, Ir, and Pt), however, is
able to activate the photocatalyst system for hydrogen and oxygen production [14].
The photoactivity was found to be further enhanced (between 2.5 860x) when the
transition metal oxides were co-loaded with Cr2O3. The co-loaded (Rh-Cr2O3)/(Ga1xZnx)(N1-xOx)
catalyst is the most active of all the oxynitride photocatalysts under
visible light water splitting ( > 400nm) with an optimized apparent Q.E of 5.9% [1518].
Two different synthesis methods have been developed in the photocatalysis
literature for the loading of Rh and Cr onto the (Ga1-xZnx)(N1-xOx) substrate. The first
synthesis method involves a one-step co-impregnation of Na3RhCl62H2O and
Cr(NO3)39H2O onto the (Ga1-xZnx)(N1-xOx) followed by calcination in air at 623K
that forms supported Rh2-yCryO3 mixed oxide nanoparticles (NPs) on the bulk (Ga1xZnx)(N1-xOx)
support.
The supported Rh2-yCryO3 NPs were proposed to be the
catalytic active site for H2 production and the bulk (Ga1-xZnx)(N1-xOx) support
providing the active sites for O2 production for this preparation technique [15]. The
second synthesis method is a two-step process, aqueous Na3RhCl62H2O is initially
photodeposited onto the bulk (Ga1-xZnx)(N1-xOx) support, dried at 343 K, followed by
Cr photodeposition from an aqueous K2CrO4 solution and again dried at 343 K. The
motivation for this photodeposition synthesis method is to cover up the metallic Rh0
119
NPs with a thin Cr2O3 layer creating a core/shell arrangement to prevent or minimize
the O2 and H2 back reaction to form H2O that readily takes place with exposed
metallic Rh0 [15]. It was proposed that H2 evolves from the metallic Rh0 component
while the Cr2O3 shell prevents the O2/H2 back reaction to form water and that the bulk
(Ga1-xZnx)(N1-xOx) provides the active sites for O2 production [17].
These
photocatalysis models were proposed from application of bulk characterization

techniques and to fully understand these complex systems requires complementary
surface characterization information since the catalytic splitting of water takes place at
the surface of photocatalysts [19, 20]. The lack of surface information about the (RhCr)/(Ga1-xZnx)(N1-xOx)
oxynitride
photocatalysts
prevents
fundamental
understanding of the functioning of these novel materials and limits our ability to have
guiding principles to design advanced visible light activated mixed oxynitride
photocatalysts.
The present study utilizes cutting edge surface characterization methods (highresolution X-ray photoelectron spectroscopy (HR-XPS) that analyzes ~1-3nm of the
surface region and high-sensitivity low energy ion scattering (HS-LEIS) that analyzes
~0.3nm of outermost or topmost surface layer) for the (Rh-Cr)/(Ga1-xZnx)(N1-xOx)
photocatalyst systems prepared by the different synthesis methods. The fresh and used
photocatalysts will be also compared to determine how their surfaces are affected by
the photocatalytic water splitting environment.
In situ optical spectroscopic
characterization techniques (Raman, UV-vis, photoluminescence (PL) and timeresolved picosecond PL-Raman) will also be applied to provide further insights into
120
the bulk molecular and electronic structures to determine (i) how (Ga1-xZnx)(N1-xOx)
differs from the bulk Ga2O3 and ZnO precursors, and (ii) how Rh2-yCryO3 NPs affect
the bulk properties of the (Ga1-xZnx)(N1-xOx) support. The photocatalytic activity of
the two synthesis methods will also be determined to help develop fundamental
structure-photoactivity
relationships
for
these
novel,
advanced
oxynitride
photocatalysts for water splitting.

4.2 Experimental
The (Ga1xZnx)(N1xOx) with x=0.12 as measured from energy-dispersive
(EDX) analysis was synthesized by a nitridation method [13]. A mixture of Ga2O3
(High Purity Chemicals, 99.9%) and ZnO (Kanto Chemicals, 99%) in a 1:2 molar ratio
was first well mixed in an agate mortar. The physical mixture was then calcined at
1123 K under NH3 flow (250 ml/min) for 15 hours in a custom quartz nitridation
reactor. For the co-impregnation synthesis method, the (Ga1xZnx)(N1xOx) was mixed
in an evaporating dish with the aqueous precursors, Cr(NO3)39H2O (Wako Pure
Chemicals, 99.9%) and Na3RhCl62H2O (Kanto Chemicals, 97% as Rh), yielding 1
wt.% Rh and 1.5wt.% Cr [15]. This suspension was then placed over a water bath and
continuously stirred with a glass rod until complete evaporation. The powder was
then collected and calcined in air at 623 K for one hour. The catalyst obtained from
the co-impregnation method will be referred to as (Rh2-yCryO3)/(Ga1-xZnx)(N1-xOx), so
named in the literature.
For the photodeposition synthesis method, the

121
(Ga1xZnx)(N1xOx) powder is placed in a Pyrex inner irradiation-type reaction vessel

connected to a glass closed gas circulation system in an aqueous solution (~370 ml)
containing Na3RhCl62H2O (Kanto Chemicals, 97% as Rh) [17] yielding 1 wt% Rh.
After the system was evacuated, the reactor was irradiated with a 450 W high-pressure
Hg lamp with a sodium nitrite aqueous solution as a filter to block ultraviolet light for
4 hours. The temperature of the reactant solution was maintained at room temperature
by flowing cooling water. The powder was then separated from the solution via
filtration, washed thoroughly with distilled water and dried overnight in an oven at
343 K. This procedure is then repeated where the Rh-deposited (Ga1xZnx)(N1xOx) is
placed in the Pyrex inner irradiation-type reaction vessel in an aqueous K2CrO4
solution yielding 2.5% Cr. After irradiation for 4 hours with a 450 W high-pressure
Hg lamp with a sodium nitrite aqueous solution as a filter, the filtrated powder is
washed thoroughly with distilled water and dried overnight in an oven at 343 K. The
catalyst obtained from the photodeposition method will be referred to as
(Rh/Cr2O3)/(Ga1-xZnx)(N1-xOx) named so in the literature.
Raman spectroscopy was utilized to obtain the molecular structure of the fresh
photocatalysts and was performed on a Lab Ram-HR Raman spectrometer (HoribaJobin Yvon) equipped with visible (442 and 532 nm) laser excitation and utilizing a
confocal microscope (Olympus BX-30) for focusing the laser on the catalyst sample.
The 532nm visible laser excitation was generated by Nd:YAG laser (10 mW) and the
442 visible laser excitation was generated by a He-Cd laser ( ~7 mW) with the
122
scattered photons directed into a single monochromator and focused onto a UVsensitive liquid-N2 cooled CCD detector (Horiba-Jobin Yvon CCD-3000V) having a
spectral resolution of 2 cm-1 for the given parameters. About 5-10 mg of the catalyst
was placed into a high temperature in situ cell (Linkam TS-1500) with a quartz
window and the spectrums were obtained under ambient conditions. The spectral
acquisition time employed was 5 scans of 5 seconds/scan for each spectrum. System
alignment was verified using a silica reference standard line at 520.7 cm-1.
Ultra
Violet-visible-Near
Infrared
(UV-vis-NIR)
diffuse
reflectance
spectroscopy (DRS) was utilized to obtain the optical edge energy, Eg, values for the
fresh photocatalysts.

123
The HR-XPS spectra of the fresh and used photocatalysts were obtained on a
Scienta ESCA 300 spectrometer equipped with a 300 mm hemispherical electrostatic
analyzer and a monochromatic Al K X-ray source with energy of 1486.6 eV
generated from a rotating anode. This allows for improved chemical selectivity by
narrowing the spectral peaks of elements and greatly reducing the spectral background
signal compared to conventional XPS spectrometers. Each spectrum was calibrated
using a binding energy (BE) value of 285.0 eV for carbon in the C1s region. The
atomic concentration ratios were calculated by correcting the measured peak area
ratios with relative sensitivity factors employed in the Casa XPS software version
2.3.15.
4.2.5 High Sensitivity Low Energy Ion Scattering (HS-LEISS)
Spectroscopy
Analysis of the outermost surface layer of the fresh and used photocatalysts
was obtained on the Qtac100 HS-LEIS Spectrometer (ION-TOF) equipped with a
highly sensitive double torodal analyzer, 3000 times higher sensitivity than
conventional LEIS spectrometers, which allows for static depth profiling.
The
photocatalyst samples were first gently cleaned with atomic oxygen to remove surface
hydrocarbon contamination from the atmosphere prior to being transferred inside the
124
analysis chamber. The HS-LEIS spectra were taken using both 4000 eV 4He+ with
7245 pA current and 3000 eV 20Ne+ with 2959 pA current as ion sources. TOF mass
filters were also utilized for spectra obtained with Ne+ as an ion source for reduced
flux background at low kinetic energies. For depth profiling, the surface was sputtered
by Ar+ gas at 500 eV at a sputter yield of 1x1015 ions/cm2. Metallic Rh and Cr
standards were also analyzed for quantifying the elemental composition of the
photocatalysts.
4.2.6 Photoluminescence Spectroscopy and Photoluminescent Decay
Spectrally resolved PL spectra and transient PL lifetime measurements of the
fresh photocatalysts were conducted using a Ti:sapphire laser (Coherent Mira 900),
tunable in the 685-1000 nm spectra range, generating 5 ps pulses with a 76 MHz
repetition rate, pumped with a frequency-doubled Nd:YVO4 laser (Coherent Verdi V18). The output of the laser was frequency doubled using an ultrafast harmonic
generator (Coherent 5-050).
To perform the luminescence measurements, the
excitation light at 400 nm was directed toward a microscope of a tunable micro-,

macro-Raman/photoluminescence system (Jobin Yvon Horiba, T6400) and was
focused using a long distance objective (50x, N/A=0.5) onto a sample to a spot size of
~ 2 m.
The photocatalyst sample was placed into a high temperature in situ
microscopy stage (Linkam, TS-1500) and pretreated as follows: The samples were
heated at 10 oC/min to 673 K in flowing 10% O2/N2 (30 sccm) to remove water, since
moisture causes quenching of the PL signal [24], and to fully oxidize the samples.
Upon cooling to room temperature in flowing inert gas (N2, 30 sccm), the
125
photoluminescence decay measurements were made. To obtain complete PL spectra,

the spectrometer was set to nanometer mode and the grating was moved several times
in order to obtain the intensity over a 400 to 800 nm range. The luminescence light
was collected through the same objective in backscattering geometry and focused onto
a slit of the triple-monochromator equipped with a fast gated intensified charge
coupled device (ICCD) camera collecting in the 350-900 nm range (LaVision, Picostar
HR12). The ICCD camera was gated using a sequence of 76 MHz pulses propagating
with a variable delay relative to the original train of trigger pulses (76 MHz) from a
photodiode in a Ti:sapphire laser. The minimum gate width was 300 ps and the
maximum delay was defined by the laser repetition rate (~13.2 ns). The laser energy at
the sample was maintained at approximately 1.6 mW to prevent photo-degradation of
the photocatalyst sample. The grating was set to monitor decay centered at emission
wavelengths from 500 to 700 nm. The grating setting used to monitor PL lifetimes
decay was based on the PL peak maximum from the PL spectra. Experimental decay
curves were then fit to a double first-order exponential decay model to account for an
observed fast (t1) and slow (t2) components [25, 26]:
t
t
y A1 exp A2 exp y0
t1
t2
(1)

About 0.1 g of the photocatalyst powder was placed in a top down Pyrex
reactor cell filled with 100 ml of distilled water. A magnetic stirrer was also placed in
the reactor to allow for continuous stirring of the suspension. The reactor was then
126
attached to a closed-gas circulation system connected to a vacuum pump. Air in the

system was removed prior to the reaction so no atmospheric oxygen can be detected in
the system. A Xe arc lamp was used to irradiation the sample for the reaction and cutoff filters were utilized to only allow > 400 nm. A Liebig condenser connected to
the closed gas circulation system was used to prevent gas phase water from escaping
the reactor so only H2 and O2 products can be found in the rest of the closed gas
circulation system. A cooling water system was also utilized to keep the reactor at
room temperature. A gas chromatograph was used to measure H2, O2, and N2 products
in the system.
After the reaction, the used catalysts were filtered out from the
suspension to allow for further surface analysis (HR-XPS and HS-LEISS).

4.3 Results
4.3.1 Bulk Molecular Structures
The Raman spectra for the bulk Ga2O3 and ZnO precursors used for the
synthesis of the (Ga1xZnx)(N1xOx) are shown in Figure 1. The bulk Ga2O3 contains
Raman bands characteristic of -Ga2O3 [27, 28]: bands below 200 cm-1 are assignable
to the translation and libration of GaO4 and Ga2O6 chains, bands between 300500
cm-1 are assigned to Ga2O6 symmetric stretching and bending modes, and higher
frequency bands between 500700 cm-1 are from GaO4 stretching and bending modes.
The Raman spectrum of ZnO contains bands from phonon modes indicative of the
ZnO wurtzite structure [29]: the bands between 100-540 cm-1 are dominated by
acoustic phonon modes, the bands between 540-820 cm-1 originate from optical and
127
acoustic phonons modes, and bands between 820-1200 cm-1 are from optical phonon
modes and their overtones.
The Raman spectra of the synthesized (Ga1xZnx)(N1xOx) and (Rh2yCryO3)/(Ga1xZnx)(N1xOx)
are presented in Figure 2.
The (Ga1xZnx)(N1xOx)
contains strong bands below 200 cm-1 that can be attributed to the translation and
libration modes of GaO4 and Ga2O6 chains similar to the for bulk Ga2O3. The band at
332 cm-1 has been assigned to multiple photon modes of ZnO and can be assigned to a
phonon mode in the (Ga1xZnx)(N1xOx) solid solution while the 637 cm-1 is from the
O-Zn-O local structure in the (Ga1xZnx)(N1xOx) solid solution [30]. The Raman
bands at 412, 556, and 717 cm-1 are indicative of GaN optical phonon modes in the
(Ga1xZnx)(N1xOx) solid solution [30, 31].
The Raman spectra suggest that the
(Ga1xZnx)(N1xOx) oxynitride also contains impure phases of Ga2O3, GaN and ZnO.
The Raman spectrum of (Rh2-yCryO3)/(Ga1xZnx)(N1xOx ) does not differ noticeably
from the Raman spectrum of the bulk (Ga1xZnx)(N1xOx) phase. Crystalline Cr2O3
characteristic Raman bands at 542 and 603 cm-1 [32, 33] are not detected, which may
be due to overlap with the bulk (Ga1xZnx)(N1xOx) vibrations. The weak and broad
Rh2O3 Raman band at 550 cm-1 [34] is also not detected because of the strong bulk
(Ga1xZnx)(N1xOx) vibrations.
Although Raman spectroscopy does not provide
information about the supported Rh2-yCryO3 NPs, it does reveal that the bulk molecular
structure of the (Ga1xZnx)(N1xOx) support is not modified by the addition of the Rh2yCryO3
NPs.
4.3.2 Bulk Electronic Structures
128
The UV-vis DRS optical bulk Eg values of the photocatalysts are tabulated in
Table 1. The bulk band gap energy values for -Ga2O3 (4.7 eV or 264 nm absorption
edge) and ZnO (3.2 eV or 388 nm) are in agreement with previously reported Eg
values [13, 35]. The bulk band gap energy for the (Ga1xZnx)(N1xOx) is 2.6 eV (477
nm) and falls in the reported range of 2.4-2.8 eV (442 - 517 nm) for visible light active
(Ga1xZnx)(N1xOx) materials [13].
Deposition of Rh2-yCryO3 NPs on the
(Ga1xZnx)(N1xOx) support essential does not perturb the overall band gap energy of
the composite photocatalyst.
4.3.3 Atomic Composition of Surface Region (~1-3nm)
The atomic composition obtained from the XPS survey spectrum for the bulk
(Ga1xZnx)(N1xOx) is shown in Table 2. No contaminants were detected in the sample
and the O concentration was found to be slightly higher (15%) compared to the bulk
concentration (12%). The XPS survey spectra of the fresh and used supported (Rh2yCryO3)/(Ga1xZnx)(N1xOx)
photocatalysts are compared in Figure 3. The surface
region of the fresh (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) consists primarily of Ga, Zn, Cr,

O, N and Rh with some Na and Cl residual contaminants from the Na3RhCl62H2O
precursor. Compared to the bulk (Ga1xZnx)(N1xOx), the O concentration was found
to increase while the N and Ga concentration decreased showing the surface region
became oxidized after the calcination procedure.
After being employed as a
photocatalyst for the splitting of water, the surface Na contaminant is gone and the
surface Cl concentration decreased by more than 50% most probably because of their
129
aqueous solubility.
The elemental composition of the surface regions for both
photocatalysts are compared in Table 3 and indicate an increase of O, but almost the
same concentrations of Ga, Zn, Cr, O, N and Rh.
This trend suggests that the Na and
Cl contaminants were occupying some of the O sites in the surface region of the
photocatalyst.
The XPS survey spectra (not shown for brevity) for the fresh and used
supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) photocatalysts did not exhibit any
contaminants from the K2CrO4 precursor. The atomic compositions obtained from the
XPS survey spectra for the supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) photocatalysts
are listed in Table 4. The used photocatalyst was found to possess slightly less O (5.3
%), and slightly more N (3.3%) and Ga (2.0%) while the concentration of the other
elements staying relatively constant in the surface region. These changes are within
the margin of error ( 5%) for the XPS spectrometer and, thus, surface composition of
the supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) photocatalyst remained relatively
constant during photocatalytic splitting. In addition, all the photocatalysts contained
~1% Rh and ~5% Cr for the supported (Rh2-yCryO3) /(Ga1xZnx)(N1xOx) photocatalyst
while the supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) photocatalysts contained ~3.5%
Cr in the surface region.
The HR-XPS spectra for Cr 2p and Rh 3d peaks in the surface region were also
collected in order to determine their oxidation states. The Cr 2p and Rh 3d transitions
for the supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) photocatalysts are presented in
Figure 4 and reveals the presence of only Cr3+ and Rh3+ species in the surface region.
130
Neither Cr6+ or metallic Rh(0) species were found for these photocatalysts in the
surface
region.
The
Cr
2p
and
Rh
3d
peaks
for
the
supported
(Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalysts are presented in Figure 5.

The Cr 2p region reveals the existence of only Cr3+ without any Cr6+ species. The
peak in the Rh 3d region shows that Rh3+ is the dominant Rh phase on the surface with
only a small shoulder corresponding to the presence of a trace of metallic Rh(0).
4.3.4 Atomic Composition of Outermost Surface Layer (~0.3nm)
The atomic composition of the outermost surface layer (~0.3 nm) and the
layers beneath the topmost surface layer of the photocatalysts were analyzed by HSLEIS spectroscopy depth profiling. The HS-LEIS spectra for the (Ga1xZnx)(N1xOx)
support using both 4He+ and 20Ne+ as ion gas sources are presented in Figures 6A and
6B. Resolvable peaks for the N, O and Zn/Ga in the outermost surface layer with the
4
He+ gas ion source are shown in Figure 6A. The atomic oxygen pretreatment cleaned
off hydrocarbons deposited on the surface which did not affect the surface
composition of the other elements and increased signal intensity for the sample. The
other photocatalyst samples were also pretreated with atomic oxygen and the surface
composition of non-carbon elements were likewise not affected. The close atomic
masses of Zn and Ga did not allow for resolution between these two elements using
He+ and Ne+ gas ions as shown in Figure 6B. Further attempts were made to resolve
these two elements with the heavier
40
Ar+ gas ions, but resolving the Ga and Zn
signals was also unsuccessful.
131
The HS-LEIS depth profiles for fresh (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) using

both 4He+ and 20Ne+ as ion gas sources are presented in Figures 7A and 7B. The HSLEIS resolvable peaks are for the O, Na, Cl, Cr, Zn/Ga, and Rh in the outermost
surface layer with the 4He+ gas ion source are shown in Figure 7A. The absence of a
peak for N is possibly due to the slow velocity of the 4He+ gas ions which makes it
difficult to obtain good elemental sensitivity for low mass elements like N without the
use of 3He+ as gas ion source [36]. The evolution of the HS-LEIS signals for Cl, Na
and Zn/Ga, normalized against the relatively constant O signal, are shown in Figure 8
during the dynamic depth profiling. The Cl and Na contaminants slightly decrease
with depth profiling reflecting some surface enrichment of Na and Cl. The Zn/Ga
signal significantly increases during the depth profiling as expected for the bulk nature
of the Zn and Ga oxides. The HS-LEIS signals for Cr and Rh are more easily resolved
with Ne+ gas ions as seen in Figure 7B. The Rh and Cr signals were calibrated against
metallic Rh and Cr standards in order to quantify these two elements as shown in
Figure 9. The outermost surface layer contains slightly more Rh than Cr, and the Cr
concentration markedly increase with depth profiling while that of Rh markedly
decreases. These concentration profiles reveal that Rh is surface segregated while Cr
becomes more prevalent with increasing depth. The ~5 times greater concentration of
Cr at the end of the depth profile is much higher than the expected ~1.5 times greater
Cr expected from the catalysts loading.
The
HS-LEIS
yCryO3)/(Ga1xZnx)(N1xOx)
depth
profiles
for
the
used
supported
(Rh2-
photocatalyst are presented in Figure 10. A very small

132
shoulder for N is visible in the depth profile of Figure 10A indicating that N is being
exposed
on
the
surface
of
the
supported
(Rh2-yCryO3)/(Ga1xZnx)(N1xOx)
photocatalysts during the water splitting reaction.
Traces of the Na and Cl
contaminants are not detected during HS-LEIS depth profiling, unlike XPS analysis of
the surface region, which suggests that the trace Cl contaminant is not present on the
outermost surface layers of this used photocatalyst. The Rh and Cr depth profiles for
the used (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) photocatalyst in Figure 11 reveal
comparable amounts of Rh and Cr on the outermost surface and the Cr concentration
increases while the Rh concentration decreases with depth profiling as found for the
fresh (Rh2-yCryO3) /(Ga1xZnx)(N1xOx) photocatalyst. The similar depth profile trends
for the Rh and Cr for the fresh and used supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx)
photocatalysts suggests a static situation exists during photocatalysis for the supported
(Rh2-yCryO3) NPs.
The
HS-LEIS
depth
profiles
for
the
fresh
supported
(Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst are presented in Figure 12 and

show that the outer surface region is void of any contaminants and N. The Cr and Rh
depth profiles for the fresh supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell
photocatalyst are shown in Figure 13 and reveal that the outermost layers contain
about twice as much Rh as Cr. With increasing depth from the outer surface, the
concentration of Cr decreases only slightly while that of Rh falls rapidly and becomes
lower than that of Cr at the end of the sputtering. The corresponding HS-LEIS depth
profiles
for
the
used
supported
(Rh/Cr2O3)/(Ga1xZnx)(N1xOx)
133
core/shell
photocatalysts are presented in Figure 14 and the main spectral difference with the
fresh core/shell photocatalyst is the existence of some N (see Figure 14A). The
appearance of N in the outermost layers was also found for the used supported (Rh2yCryO3)/(Ga1xZnx)(N1xOx)
mixed oxide photocatalyst suggests possible
surface
enrichment during the photocatalytic water splitting reaction. The HS-LEIS Cr and
Rh depth profiles for the used supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell
photocatalyst are given in Figure 15 and the trend is similar to that of the fresh
supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst: Rh concentration is
about twice as great in the outermost layers and rapidly decreases while the Cr
concentration only slightly decreases with increasing depth.
Also found for the
supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) mixed oxide photocatalysts, the Rh and Cr

depth profile concentrations for the supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx)
core/shell photocatalysts are not altered by the photocatalytic reaction environment.
4.3.5 Dynamics of Photoexcited Electrons and Holes
The photoluminescence (PL) emission spectrum monitors the recombination
dynamics of excited electrons and holes.
The PL intensity is indicative of the
population of electron/hole recombination centers (self-trapped electrons, oxygen

vacancies, defect sites, impurities, reduced metal ions, etc.) in the bulk phase of
excited photocatalysts with high intensity reflecting a greater number of
recombination centers [24, 37]. The PL emission spectra of bulk ZnO, Ga2O3 and
(Ga1xZnx)(N1xOx) photocatalysts in response to laser excitation at 400 nm are
presented in Figure 16 and their peak maxima are listed in Table 1. The much lower
134
PL emissions intensities for bulk ZnO and Ga2O3 might at first suggest that these
oxides appear to be more suitable photocatalysts than the (Ga1xZnx)(N1xOx)
oxynitride, but the low PL emissions intensities are actually a consequence of the 400
nm excitation energy being lower than their optical band gap energy values (388 and
264 nm). The optical absorption edge values for bulk ZnO and Ga2O3 are in the UVrange, thus, the lack of PL emissions is due to the use of visible light irradiation
causing a decreased population of excited electrons being able to recombine. The
addition of the (Rh2-yCryO3) NPs greatly diminishes the PL emissions of the
(Ga1xZnx)(N1xOx) showing that electron/hole recombination is decreased in the
presence of the supported NPs because of electrons being trapped and unable to
recombine with holes in the bulk phase to produce emission.
The photoluminescence emission decay curves for the bulk ZnO, Ga2O3 and
oxynitride photocatalysts are plotted in Figure 17. The PL emission decays for all the
photocatalysts were modeled with Equation 1 that is based on two different species of
electrons decaying at different rates and the fit parameters are given in Table 1. The
parameters t1 and A1 refer to decay constant and amplitude of the fast component of
electron decay while t2 and A2 refer to the decay constant and amplitude of the slow
component of electron decay. Comparison of the PL emission decays from the bulk
ZnO, Ga2O3 and (Ga1xZnx)(N1xOx) indicates that both t1 and t2 increase with
decreasing optical band gap energy, which is indicative of longer lifetimes of photogenerated electron/hole pairs [38, 39]. Deposition of the (Rh2-yCryO3) NPs on the
135
(Ga1xZnx)(N1xOx) support prolongs the lifetime of excited electrons/holes in this

multicomponent photocatalyst system.
The evolution of the H2 and O2 products from steady-state photocatalytic water
splitting
was
monitored
over
hours
for
the
supported
(Rh2-yCryO3)/
(Ga1xZnx)(N1xOx) mixed oxide and supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx)

core/shell photocatalysts and are presented in Figure 18. Both photocatalysts were able
to produce H2 and O2 at the proper 2:1 stoichiometric ratio. The production rate (units
of mols/ g catalyst/h) for the supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) mixed oxide
photocatalyst was 161 for H2 and 79.4 for O2 while the production rate for the
supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst was 71.1 for H2 and
33.4 for O2. The BET surface areas of both photocatalyst systems were dominated by
that of the (Ga1xZnx)(N1xOx) oxynitride support at 8 m2/g [13], and the surface area
normalized
activities
(units
yCryO3)/(Ga1xZnx)(N1xOx)
of
mols/
m2/h)
for
the
supported
(Rh2-
mixed oxide photocatalyst are 20.1 H2/9.9 O2 and 8.9
H2/4.2 O2 for the supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst.

Thus, the supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) mixed oxide photocatalyst is
slightly more than 2 times as active as the supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx)
core/shell photocatalyst. No formation of N2 was detected and indicates that N3oxidation to N2 did not occur at the surface of the oxynitride photocatalysts [13].
4.4 Discussion
136
4.4.1 Bulk and Electronic Structures of the (Ga1xZnx)(N1xOx) Oxynitride

Support
The bulk molecular structure of the oxyntride support was confirmed to be the
(Ga1xZnx)(N1xOx) solid solution with Raman spectroscopy (see Figure 2).
The
corresponding UV-vis spectrum confirms that the (Ga1xZnx)(N1xOx) oxynitride phase

is a solid solution and is not just a physical mixture of Ga2O3, ZnO and GaN because
of its much lower optical band gap than the starting Ga2O3, ZnO and GaN materials
for its synthesis. The presence of small amounts of beta-Ga2O3, ZnO and GaN NPs,
which presumably reside on the surface of the oxynitride solid solution support, were
also detected with Raman spectroscopy. The bulk molecular and electronic structures
of the (Ga1xZnx)(N1xOx) oxynitride support phase were not perturbed by the
deposition of the (Rh2-yCryO3) mixed oxide NPs as reflected in the resulting Raman
and UV-vis spectra (see Figures 2 and Table 1). Thus, the deposition of the (Rh2yCryO3)
mixed oxide NPs onto the surface (Ga1xZnx)(N1xOx) support had almost no
measurable effect on the bulk oxynitride phase.

4.4.2 Atomic Distribution of the Surface Region of Supported
4.4.2.1 Model of the Supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) Mixed
Oxide Photocatalyst
Although the surface of the fresh supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx)
photocatalyst initially was contaminated with Na and Cl from the rhodium precursor,
137
Na and Cl were easily dissolved in the aqueous environment from the outermost
surface layer (~0.3nm) during photocatalysis. The Na is also completely absent from
the surface region (~1-3nm) of the used photocatalyst and the Cl concentration
diminishes more than 50% from the surface region (Table 2). LEIS was able to show
a small peak for N on the (Ga1xZnx)(N1xOx) which is not present for the (Rh2yCryO3)/(Ga1xZnx)(N1xOx).
The
surface
region
for
the
(Rh2-yCryO3)/
(Ga1xZnx)(N1xOx) was also found to be enriched with O relative to N (surface ~0.5

vs. bulk ~ 0.14), which is not too surprising since the deposited (Rh2-yCryO3) NPs do
not possess any N. The lack of N on the (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) combined
with the surface enriched O gives evidence for the existence of a GaZnO x thin film on
the surface of the (Ga1xZnx)(N1xOx) which arises due to the calcination treatment in
air. The deposited (Rh2-yCryO3) NPs on the oxynitride support exclusively contain
Cr3+ and Rh3+ species both in the fresh and used photocatalysts (see HR-XPS spectra
in Figure 4). The Rh+3 species is surface enriched and its concentration significantly
decreases with depth while the Cr+3 concentration increases with depth from the
surface for the fresh and used photocatalysts (see HS-LEIS Figures 9 and 11). The
increase in Cr3+ concentration with depth profile can be explained by the formation of
some C2O3 NPs on the surface, dissolved Cr3+ in the GaZnOx layer, and increased
Cr3+ concentration in the (Rh2-yCryO3) NPs with depth from the surface.
The
increased Ga/Zn concentration with depth profiling indicates that Ga and Zn are not
surface enriched (see HS-LEIS Figures 7 and 10). The lower XPS ratio of Ga to Zn in
the surface region, ~5, compared to the bulk, ~7, suggests that Ga is depleted relative
138
to Zn in the surface region of the oxynitride support (see Table 2). A schematic of the
structure of the supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) photocatalyst is depicted in
Figure 19.
4.4.2.2 Model of the Supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx)
Core/Shell Photocatalyst
The photodeposition synthesis method yielded a relatively contaminant-free
photocatalyst whose surface did not contain the Na and Cl from the Rh precursor as
well as the K from the Cr precursor. The lack of N on the outermost surface with
surface enriched O shows that a GaZnOx film covers the surface of the
(Ga1xZnx)(N1xOx). Surface Cr3+ and Rh3+ are the dominant chemical states in the
Rh/Cr2O3 core/shell NPs with only a small amount of metallic Rh0 (see HR-XPS
Figure 5). The Rh concentration in the outermost surface layer is almost ~2 is much
as the Cr and much higher than the ~ 1:1 ratio for the supported (Rh2yCryO3)/(Ga1xZnx)(N1xOx)
photocatalyst and its concentration significantly decreases
with depth profiling (see HS-LEIS Figures 13 and 15). The most surprising new
insights about the surface region is that Rh is fully oxidized and surface enriched
given that the photodeposition synthesis was supposed to encapsulate the metallic Rh0
NPs with a thin Cr2O3 film.
There are two possibilities to explain why the Rh
concentration is much higher on the outermost surface layer. The 2:1 Rh:Cr ratio on
the surface can be attributed to Rh0 NPs that where not fully encapsulated by the
Cr2O3 shell and becomes oxidized forming a Rh3+ shell over the Rh0 NP. The second
possibility is that during the photodeposition synthesis the Rh was able to become
139
oxidized and diffuse through the Cr2O3 layer, presumably related to the Rh2O3 having
similar structure as Cr2O3 and being the driving force to surface segregate. These
surprising findings are in contrast to the supported core/shell model (metallic Rh core
encapsulated by a Cr2O3 film) previously proposed [17]. The previous investigation
employed bulk EXAFS/XANES to analyze the Rh and the bulk nature of this
spectroscopic method indicates that the majority Rh species in the core/shell
photocatalyst is indeed metallic Rh0, but provides no information about the surface
region [17]. The current surface measurements, however, demonstrate that it is Rh+3
that is surface enriched in the topmost surface layer and surface region of the
core/shell photocatalyst where the photocatalytic splitting of water takes place. The
LEIS shows that the Cr concentration stays constant with depth indicating uniform Cr
concentration in the Rh-Cr NPs along with the formation of some Cr2O3 NPs. A
schematic of the supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst is
depicted in Figure 20.
4.4.3 Generation of Excited Electron/Hole Pairs and Their Lifetimes
Although the (Ga1xZnx)(N1xOx) phase is able to generate excited
electrons/holes with visible light excitation, it is unable to photocatalytically split
water at the surface/water interface [13]. The function of the bulk (Ga1xZnx)(N1xOx)
phase, or support, of the multicomponent photocatalyst system is to generate excited
electron/holes upon excitation by visible light. PL spectroscopy is able to measure the
recombination of electrons/holes in the photocatalyst and the presence of the
supported
(Rh2-yCryO3)
particles
suppresses
140
the
PL
signal
of
the
bulk
(Ga1xZnx)(N1xOx), which indicates that the supported (Rh2-yCryO3) particles prevent

recombination of electrons and holes.
This suggests that the supported (Rh2-yCryO3)
mixed oxide particles are efficient electron traps that minimize electron and hole
recombination and, thus, allow for their consumption at the surface for photocatalytic
water splitting [24, 40, 41]. Transient PL spectroscopy is able to measure the lifetime
of the excited electrons/holes from the decay emissions parameters based on the
slow t2 component and where the ratio
is indicative of the relative population
of these long lived electrons with slow emissions decay [38, 39]. The electron traps
from the surface (Rh2-yCryO3) were able to prolong the lifetimes of excited
electrons/holes as shown in Table 1 from the increased lifetimes not only from the
slow t2 component of decay but also for fast t1 component.
The relative
population of long lived excited electrons was also found to be enhanced by the
surface (Rh2-yCryO3). PL and transient PL decay spectroscopy demonstrated that the
surface modified (Rh2-yCryO3) possessed all of the measurable desired properties for
an efficient photocatalyst system.
The surface (Rh2-yCryO3) decreased the
recombination of electron/holes in the bulk by transferring then to the surface which

increased the lifetimes and population of the excited electrons and holes. This greatly
increases the probability for the photocatalytic splitting of H2O into H2 and O2 at the
surface.
HS-LEISS depth profiling revealed that the Cr2O3 species is more
concentrated than the Rh2O3 near the bulk is seen in Figure 9 so the Cr2O3 may be
more responsible for these charge transfer effects.
141
4.4.4 Structure-Photoactivity Relationships for Splitting of H2O by

4.4.4.1 Supported (Rh2-yCryO3)/(Ga1x Znx)(N1xOx) Mixed Oxide
Photocatalyst
The supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) photocatalyst is a two
component photosystem. The bulk (Ga1xZnx)(N1xOx) support phase has a band gap
of 2.6 eV that allows homogeneous generation of electrons and holes inside this phase
with visible light activation. The bulk generated electrons and holes must avoid
recombination and diffuse to the surface to be able to perform the photocatalytic
reaction. The supported (Rh2-yCryO3) mixed oxide NPs are the catalytic active sites
that trap and harness the electrons and holes to perform photocatalytic splitting of
water to H2 [15]. It was previously shown that although supported Rh2O3/(Ga1x
Znx)(N1xOx) photocatalyst is able to split H2O into H2, it is unable to produce any
significant O2 [14].
Only when Cr2O3 is added to the supported Rh2O3/(Ga1x
Znx)(N1xOx) photocatalyst is the stoichiometric amount of O2 simultaneously evolved

[14] . These observations suggest that the function of the Rh+3 sites is to generate H2
while the O2 evolution site can exist at several points: on the Cr3+ sites, GaZnOx or at
their contact points (see Figure 19). There is not enough evidence to determine which
the active site is but the surface analysis reveals those are the likely sites for O 2
evolution.
The current HS-LEISS surface analysis reveals that the ideal surface
composition for the (Rh2-yCryO3) NPs consists of a ~ 1:1 ratio of Rh:Cr on the
142
outermost surface layer for the
supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx)
photocatalysts.
4.4.4.2 Supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) Core/Shell
Photocatalyst
The supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst is a
three-component system.
The bulk oxynitride (Ga1xZnx)(N1xOx) support phase
generates the bulk electrons and holes with visible light excitation. The supported
metallic Rh NPs trap the electrons arriving at the Rh/(Ga1xZnx)(N1xOx) interface.
The metallic Rh NPs are encapsulated by a film of (Rh2-yCryO3) that requires the
electrons to diffuse from the metallic Rh NPs to the surface of the (Rh2-yCryO3) mixed
oxides NPs to participate in the photocatalytic water splitting reaction.
The supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalyst was
designed to optimize evolution of H2 [17]. Although metallic Rh is known as a good
H2 evolution photocatalyst, it also suffers from catalyzing the back reaction between
H2 and O2 to water [3]. To minimize the back reaction by metallic Rh NPs it was
proposed to encapsulate the metallic Rh NPS with a Cr2O3 film [17]. The current HSLEIS analysis, however, reveals that some of the metallic Rh was able to become
oxidized during Cr2O3 photodeposition and, surprisingly, to diffuse to the surface of
the encapsulating chromia film (see Figure 12). This suggests that the supported
core/shell photocatalyst may actually function similarly to the supported (Rh2-yCryO3)
mixed oxide NPs where the Rh+3 evolves H2 with Cr3+, GaZnOx or their contact points
as the possible active sites for O2 evolution. Although the metallic Rh component may
143
enhance the trapping of electrons and supply them to the (Rh2-yCryO3) mixed oxide
film, the presence of some metallic Rh0 is also detected in the surface region (see HRXPS Figure 5) that catalyzes the undesirable back reaction between H2 and O2 to form
water.
Consequently, the supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell
photocatalyst is actually only ~50% as efficient as the supported (Rh2yCryO3)/(Ga1xZnx)(N1xOx)
mixed oxide photocatalyst because the desired catalyst
was not achieved in the photodeposition process [17]. The high 2:1 ratio for Rh:Cr
found on the outermost surface layer may also be a factor in the decreased activity
giving evidence for the possibility of exposed metallic Rh0 NPs with a thin Rh3+ shell
which were not encapsulated by the (Rh2-yCryO3) mixed oxide film. The model for the
photocatalytic process is given in Figure 20.
4.5 Conclusions
The bulk and surface properties of supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx)
mixed oxide and (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell visible light activated
photocatalysts were investigated with bulk (Raman, UV-vis, and PL) and surface (HSLEIS and HR-XPS) spectroscopic techniques. The bulk molecular and electronic
structures of the oxynitride (Ga1xZnx)(N1xOx) support phase were not affected by the
addition of the Rh-Cr NPs.
The supported Rh-Cr NPs, however, affected the
recombination of excited electrons and holes revealing their ability to trap electrons
and holes and harness them for photocatalytic splitting of water. The Rh2-yCryO3 NPs
were responsible for evolution of the H2 (Rh+3) and several sites are proposed (Cr+3,
144
GaZnOx or their contact points) for O2 photocatalytic reaction products. The presence
of some metallic Rh0 in the supported (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell
photocatalysts is responsible for its lower water splitting efficiency because the
metallic Rh0 catalyzes the undesirable backward reaction of H2 and O2 to water. This
study establishes for the first time the fundamental structure-photoactivity
relationships for the novel visible light active supported (Rh-Cr)/(Ga1xZnx)(N1xOx)
photocatalysts.
Acknowledgements
DOE-FG02-93ER14350. Special thanks goes out to Alexander Puretsky at the Center
for Nanophase Materials Sciences at Oak Ridge National Laboratory and Charles
Roberts for assistance with the PL experiments in conjunction with User Project
CNMS2008-075. Thanks to A. Xiong, Dr. J. Kubota and Dr. K. Domen at the
University of Tokyo for their assistance with the synthesis of the (Ga1xZnx)(N1xOx)
photocatalysts as well as generating the photocatalytic activity data. The assistance of
Dr. A. Miller at Lehigh University in obtaining and interpreting the HR-XPS and HSLEIS data is also gratefully acknowledged.
145
References

Water." Chem. Mater., 20, 35, (2008).

Chem. Soc. Rev., 38, 253, (2009).

[5] K. Maeda, K. Domen, "Photocatalytic Water Splitting: Recent Progress and Future
Challenges " The Journal of Physical Chemistry Letters, 1, 2655, (2010).
[6] Z. Zou, J. Ye, K. Sayama, and H. Arakawa, "Direct splitting of water under visible
light irradiation with an oxide semiconductor photocatalyst." Nature (London, U. K. ),
414, 625, (2001).
[7] Z. Zou, H. Arakawa, "Direct water splitting into H2 and O2 under visible light
irradiation with a new series of mixed oxide semiconductor photocatalysts." J.
Photochem. Photobiol. , A, 158, 145, (2003).
146
[8] H. Liu, J. Yuan, W. Shangguan, and Y. Teraoka, "Visible-light-responding

BiYWO6 solid solution for stoichiometric photocatalytic water splitting." J. Phys.
Chem. C, 112, 8521, (2008).
[9] K. Maeda, K. Teramura, N. Saito, Y. Inoue, H. Kobayashi, and K. Domen,

"Overall water splitting using (oxy)nitride photocatalysts." Pure Appl. Chem., 78,
2267, (2006).

[11] T. Hirai, K. Maeda, M. Yoshida, J. Kubota, S. Ikeda, M. Matsumura, and K.

Domen, "Origin of Visible Light Absorption in GaN-Rich (Ga1-xZnx)(N1-xOx)
Photocatalysts." J. Phys. Chem. C, 111, 18853, (2007).
[12] W. Wei, Y. Dai, K. Yang, M. Guo, and B. Huang, "Origin of the Visible Light
Absorption of GaN-Rich Ga1-x ZnxN1-xOx (x = 0.125) Solid Solution " Journal of
Physical Chemistry C, 112, 15915, (2008).
[13] K. Maeda, K. Teramura, T. Takata, M. Hara, N. Saito, K. Toda, Y. Inoue, H.

Kobayashi, and K. Domen, "Overall Water Splitting on (Ga1-xZnx)(N1-xOx) Solid
Solution Photocatalyst: Relationship between Physical Properties and Photocatalytic
Activity." J. Phys. Chem. B, 109, 20504, (2005).
147


Chem. B, 110, 13753, (2006).

Phys. Chem. C, 111, 7554, (2007).

148
Compounds " Angew. Chem. Int. Ed Engl., 49, 8037, (2010).
[20] I.E. Wachs, K. Routray, "Catalysis Science of Bulk Mixed Oxides " ACS
Catalysis, 2, 1235, (2012).

104, 1261, (2000).


Rises and Decays," Kinetics and Catalysis, 44, 5, (2003) .
149
[27] Y. Zhao, R.L. Frost, "Raman spectroscopy and characterisation of -gallium

oxyhydroxide and -gallium oxide nanorods " J. Raman Spectrosc., 39, 1494, (2008).
[28] R. Rao, A.M. Rao, B. Xu, J. Dong, S. Sharma, and M.K. Sunkara, "Blueshifted
Raman scattering and its correlation with the [110] growth direction in gallium oxide
nanowires " J. Appl. Phys., 98, 094312, (2005).
[29] R. Cusc, E. Alarcn-Llad, J. Ibez, L. Arts, J. Jimnez, B. Wang, and M.

Callahan, "Temperature dependence of Raman scattering in ZnO " Physical Review B,
75, (2007).
[30] W. Han, Z. Liu, and H. Yu, "Synthesis and optical properties of GaN/ZnO solid
solution nanocrystals " Appl. Phys. Lett., 96, 183112, (2010).
[31] V. Davydov, Y. Kitaev, I. Goncharuk, A. Smirnov, J. Graul, O. Semchinova, D.

Uffmann, M. Smirnov, A. Mirgorodsky, and R. Evarestov, "Phonon dispersion and
Raman scattering in hexagonal GaN and AlN " Physical Review B, 58, 12899, (1998).
[32] E.L. Lee, I.E. Wachs, "In Situ Spectroscopic Investigation of the Molecular and
Electronic Structures of SiO2 Supported Surface Metal Oxides." J. Phys. Chem. C,
111, 14410, (2007).
[33] D.S. Kim, J.M. Tatibouet, and I.E. Wachs, "Surface structure and reactivity of
chromium trioxide/silica catalysts," J. Catal., 136, 209, (1992).
150
[34] C.T. Williams, C.G. Takoudis, and M.J. Weaver, "Methanol Oxidation on
Rhodium As Probed by Surface-Enhanced Raman and Mass Spectroscopies:
Adsorbate Stability, Reactivity, and Catalytic Relevance " The Journal of Physical
Chemistry B, 102, 406, (1998).
[35] T. Yanagida, Y. Sakata, and H. Imamura, "Photocatalytic decomposition of H2O

into H2 and O2 over Ga2O3 loaded with NiO." Chem. Lett., 33, 726, (2004).

(2007).
[37] G.D. Gilliland, "Photoluminescence spectroscopy of crystalline semiconductors"

(2008).
Nanodomains," In preparation for publication, .
151

Am Ceram Soc, 92, 460, (2009).
photocatalysts with cubic perovskite structure" J.Mater.Chem., (2011).
152
Excitation 400 nm
Catalyst
Eg
Peak
(eV) Maximu
m (nm)
t1
(ps)
fast
y=A1*exp(-t/t1)+A2*exp(-t/t2)+y0
A1 fast
t2
A2
A1/
A2/
(ps) slow (A1+ (A1+A2)
slow
A2)
Ga2O3
4.7
713
131417
2404
0.99
9.40E-06
ZnO
(Ga1xZnx)(
N1xOx )
(Rh2-yCryO3)
/(Ga1xZnx)(
N1xOx ).
3.2
2.6
637
657
789
941
29
179
6818
7421
8
57
0.79
0.76
0.21
0.24
2.5
725
1003
26
8002
15
0.64
0.36
Table 4.1: Eg values, peak emission wavelengths, and decay fit parameters of the
catalysts at 400 nm excitation. The decay parameters displayed are those found in the
PL spectra taken in the emission range of the peak value.
153
Element
O 1s
15.9%
N 1s
60.2%
Ga 2p 3/2 22.2%
Zn 2p 3/2 1.7%
Table 4.2: XPS surface region atomic composition (~1-3nm) of (Ga1xZnx)(N1xOx).
154
Element
O 1s
N 1s
Ga 2p 3/2
Zn 2p 3/2
Cr 2p 3/2
Rh 3d
Cl 2p
Na 1s
Fresh
23.5%
50.2%
10.9%
2.4%
5.2%
0.8%
3.5%
3.4%
Used
31.1%
48.6%
10.6%
1.9%
5.1%
1.1%
1.5%
0.0%
Table 4.3: XPS surface region atomic composition (~1-3nm) of fresh and used (Rh2yCryO3)
/(Ga1xZnx)(N1xOx) mixed oxide photocatalysts.
155
Element
O 1s
N 1s
Ga 2p 3/2
Zn 2p 3/2
Cr 2p 3/2
Rh 3d
Cl 2p
Na 1s
K 2s
Fresh
31.7%
50.8%
10.9%
2.1%
3.4%
1.1%
0.0%
0.0%
0.0%
Used
26.4%
54.1%
12.9%
1.7%
3.7%
1.2%
0.0%
0.0%
0.0%
Table 4.4: HR-XPS surface region atomic composition (~1-3nm) of fresh and used
(Rh/Cr2O3)/(Ga1xZnx)(N1xOx) core/shell photocatalysts.
156
Intensity (a.u.)
437
138
332 383
202
480 586
1158
662
ZnO
196
347
318
167
200
415
478
400
629
767
655
600
800
Ga2O3
1000
1200
-1
Raman Shift (cm )
Figure 4.1: Raman Spectra of -Ga2O3 and ZnO precursors (532 nm) under ambient
conditions.
157
Ga2O3
GaN
Intensity (a.u.)
195
160
717
ZnO GaN
332
140
GaN ZnO
556
412
637
(Ga1-xZnx)(N1-xOx)
(Rh2-yCryO3)/(Ga1-xZnx)(N1-xOx) x2
200
400
600
800
1000
1200
-1
Raman Shift (cm )
Figure
4.2:
Raman
Spectra
of
(Ga1-xZnx)(N1-xOx)
/(Ga1xZnx)(N1xOx) (432 nm) under ambient conditions.
158
and
(Rh2-yCryO3)
Ga 2p3/2
Intensity (a.u.)
Cr 2p3/2
O 1s
Zn 2p3/2
Na
N 1s
Rh 3d
1200
1000
800
600
400
Cl
200
Binding Energy (eV)
Intensity (a.u.)
Ga 2p3/2
O 1s
Cr 2p3/2
Zn 2p3/2
N 1s
Rh 3d
1200
1000
800
600
400
Cl
200
Binding Energy (eV)
Figure 4.3: XPS Survey spectra of (a) fresh and (b) used (Rh2-yCryO3)
/(Ga1xZnx)(N1xOx) photocatalysts.
159
3+
Cr 2p3/2
3+
Cr 2p3/2
B.E. 577.1 eV
B.E. 577.1 eV
6+
B.E. 580.5 eV
590
588
586
584
582
6+
Cr 2p3/2
Intensity (a.u.)
Intensity (a.u.)
Cr 2p3/2
580
578
576
574
572
570
B.E. 580.5 eV
590
588
586
584
Binding Energy (eV)

A
582
580
578
576
574
572
570
Binding Energy (eV)

B
3+
Rh 3d5/2
3+
Rh 3d5/2
B.E. 309.9 eV
Intensity (a.u.)
Intensity (a.u.)
B.E. 309.9 eV
Rh(0) 3d5/2
Rh(0) 3d5/2
320
318
316
314
312
310
308
306
320
Binding Energy (eV)
318
316
314
312
310
308
306
Binding Energy (eV)
Figure 4.4: HR-XPS spectra of Cr 2p and Rh 3d regions of (a) fresh and (b) used
supported (Rh2-yCryO3)/(Ga1xZnx)(N1xOx) mixed oxide photocatalysts.
160
B
3+
Cr 2p3/2
Intensity (a.u.)
B.E. 576.8 eV
3+
6+
B.E. 580.0 eV
590
585
580
Cr 2p3/2
Intensity (a.u.)
Cr 2p3/2
575
570
565
6+
Cr 2p3/2
B.E. 580.0 eV
590
Binding Energy (eV)
B.E. 576.7 eV
585
580
575
570
565
Binding Energy (eV)
3+
Rh 3d5/2
3+
Rh 3d5/2
B.E. 309.4 eV
Intensity (a.u.)
Intensity (a.u.)
B.E. 309.7 eV
Rh(0) 3d5/2
B.E. 307.4 eV
Rh(0) 3d5/2
B.E. 307.7 eV
320
318
316
314
312
310
308
306
304
302
320
318
Binding Energy (eV)
316
314
312
310
308
306
304
302
Binding Energy (eV)
Figure 4.5: HR-XPS spectra of Cr 2p and Rh 3d regions of (a) fresh and (b) used
(Rh/Cr2O3)/(Ga1xZnx)(N1xOx) photocatalysts.
161
A
O
Intensity (cts/nC)
After atomic O Pretreatment

4
Zn/Ga
Untreated
1
0
1000
2000
3000
4000
Energy (eV)
35
B
Zn/Ga
30
Intensity (cts/nC)
25
20
15
10
5
0
500
1000
1500
2000
2500
Energy (eV)
Figure 4.6: HS-LEIS spectra for (Ga1xZnx)(N1xOx) photocatalyst using (a) He+ ion
gas and (b) Ne+ ion gas.
162
10
Cl
Na
Cr
Zn/Ga
2
Rh
Sp
ut
te
r
yc
le
1000
1500
2000
2500
3000
3500
4000
Energy (eV)
Cl
15
Zn/Ga
Cr
10
Intensity (cts/nC)
20
Rh
Sp
ut
te
rC
yc
le
500
1000
1500
2000
2500
Energy (eV)
Figure 4.7: HS-LEIS Depth Profile for the fresh (Rh2-yCryO3) /(Ga1xZnx)(N1xOx)
163
Na/O
Cl/O
(Zn/Ga)/O
1.0
0.9
0.8
Intensity Ratio
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0
10
12
14
16
Sputter Cycle
Figure 4.8: HS-LEIS depth profile intensity ratios of elements/oxygen for the fresh
(Rh2-yCryO3) /(Ga1xZnx)(N1xOx) photocatalyst using He+ ion gas.
164
Cr
Rh
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0
10
12
14
Sputter Cycle
Figure 4.9: HS-LEIS Depth Profile of Cr and Rh for the fresh (Rh2-yCryO3)
165
16
10
N
Cr
Zn/Ga
Intensity (cts/nC)
Rh
Sp
ut
te
rC
yc
le
1000
1500
2000
2500
3000
3500
4000
Energy (eV)
25
Zn/Ga
20
15
Cr
10
Rh
Intensity (cts/nC)
30
Sp
ut
te
rC
yc
le
500
1000
1500
2000
2500
Energy (eV)
Figure 4.10: HS-LEIS Depth Profile for the used supported (Rh2-yCryO3)
/(Ga1xZnx)(N1xOx) photocatalyst using (a) He+ ion gas and (b) Ne+ ion gas.
166
Cr
Rh
0
0
10
12
14
16
Sputter Cycle
Figure 4.11: HS-LEIS Depth Profile of Cr and Rh for used supported (Rh2-yCryO3)
167
Cr
4
Zn/Ga
Intensity (cts/nC)
Rh
Sp
ut
te
rC
yc
le
1000
1500
2000
2500
3000
3500
4000
Energy (eV)
Zn/Ga
30
25
20
15
10
Cr
Intensity (cts/nC)
Rh
Sp
ut
te
rC
yc
le
500
1000
1500
2000
2500
Energy (eV)
Figure 4.12: HS-LEIS Depth Profile for the fresh supported (Rh/Cr2O3)
/(Ga1xZnx)(N1xOx) core/shell photocatalyst using (a) He+ ion gas and (b) Ne+ ion gas.
168
Cr
Rh
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
10
12
14
Sputter Cycle
Figure 4.13: HS-LEIS Depth Profiles of Cr and Rh for the fresh supported
169
16
10
N
6
Cr
Zn/Ga
Intensity (cts/nC)
Rh
Sp
ut
te
rC
yc
le
1000
1500
2000
2500
3000
3500
4000
Energy (eV)
Zn/Ga
40
30
25
20
15
Intensity (cts/nC)
35
10
Cr
Rh
Sp
ut
te
rC
yc
le
500
1000
1500
2000
2500
Energy (eV)
Figure 4.14: HS-LEIS Depth Profile for the used supported (Rh/Cr2O3)
/(Ga1xZnx)(N1xOx) core/shell photocatalyst using (a) He+ ion gas and (b) Ne+ ion gas.
170
3.0
Cr
Rh
2.5
2.0
1.5
1.0
0.5
0.0
0
10
12
14
Sputter Cycle
Figure 4.15: HS-LEIS Depth Profiles for Cr and Rh for the used supported
171
16
Intensity (Counts)
(Ga1-xZnx)(N1-xOx)
(Rh2-yCryO3)/(Ga1-xZnx)(N1-xOx)
-Ga2O3 x2
ZnO x2
500
550
600
650
700
750
Wavelength (nm)
Figure 4.16: PL spectra of bulk -Ga2O3, ZnO and oxynitride catalysts at 400 nm
excitation.
172
Ga2O3
PL Intensity (a.u.)
PL Intensity (a.u.)
ZnO
2000
4000
6000
8000
10000
12000
14000
2000
4000
Delay Time (ps)
6000
8000
10000
12000
14000
10000
12000
14000
Delay Time (ps)

(Rhy-2CryO3)/(Ga1-xZnx)(N1-xOx)
PL Intensity (a.u.)
PL Intensity (a.u.)
(Ga1-xZnx)(N1-xOx)
2000
4000
6000
8000
10000
12000
14000
2000
Delay Time (ps)
4000
6000
8000
Delay Time (ps)
Figure 4.17: PL decay curves for the photocatalysts.
173
H2
140
140
Co-Impregnation
Synthesis
Products formed (mol)
120
120
100
100
80
80
60
60
40
40
20
20
Photodeposition
Synthesis
Time (hrs)
Time (hrs)
Figure 4.18: Comparison of Photocatalytic Water Splitting Activity ( > 400 nm).
174
O2
Figure 4.19: Schematic Model of Rh-Cr NPs for Co-Impregnation Synthesis Method.
175
Figure 4.20: Schematic Model of Rh-Cr NPs for Photodeposition Synthesis Method.
176
CHAPTER 5
Investigating the Surface Nature of TaON Photocatalysts
Abstract
Bulk TaON and supported RuO2/TaON photocatalysts used in Z-scheme
photocatalytic water splitting were synthesized and characterized for their bulk and
surface characteristics. The bulk properties of these photocatalysts were determined
with Raman and UV-vis spectroscopy and show that the molecular and electronic
structures, respectively, of the TaON support are not perturbed by the deposition of the
RuO2.
For the first time, the surface properties of these photocatalysts were
determined using High Resolution-XPS (HR-XPS) and High Sensitivity-Low Energy

Ion Scattering (HS-LEIS) spectroscopy. The High Sensitivity-LEIS and HR-XPS
surface measurements reveal that the outermost surface layers of TaON are present as
a TaOx thin film and that the deposited RuO2 is present as dissolved Ru+4 cations in
the tantalum oxide layers. These new structural insights of the surface region of
supported RuO2/TaON photocatalysts demonstrate that the photoactive catalytic sites
for O2 evolution are dissolved Ru+4 cations in the tantalum oxide thin film and not
RuO2 nanoparticles as previously proposed for this photocatalytic system.
177
5.1 Introduction
In recent years, the use of oxynitride materials for photocatalytic water
splitting has garnered much attention due to their response to visible light irradiation
[1].
Bulk oxynitride materials with d0 electronic configuration: LaTiO2N [2],
CaNbO2N [3], CaTaO2N [4], SrTaO2N [4], BaTaO2N [4], TaON [5-7], Y2Ta2O5N2 [8]
and TiNxOyFz [9] have been found to be able to evolve H2 and O2 in the presence of
sacrificial reagents under visible light irradiation. Unlike bulk oxynitride materials
with d10 electronic configuration, (Ga1-xZnx)(N1-xOx) [10-14] and (Zn1+xGe)(N2Ox)
[15], the bulk oxynitrides with d0 electronic configuration cannot photocatalytically
split pure water into H2 and O2 without sacrificial reagents. The main interest in
studying the d0 bulk oxynitrides is in their use as part of a two component Z-Scheme
photocatalyst system [16]. Pt/TaON [17], Pt/CaTaO2N [18] and Pt/BaTaO2N [18]
have been found to work as the H2 evolution component with Pt/WO3 as the O2
evolution component using IO3-/I- as the shuttle redox mediator in a Z-scheme system.
The most active Z-scheme photocatalyst system (Pt/ZrO2/TaON with Pt/WO3) has an
optimized quantum yield of 6.3% at > 420 nm [19] that is higher than any single
component photocatalyst system. There is still much research that can be done to
improve the photoactivity for Z-scheme systems since this is a more recently
developed technology. Thus, the study of advanced, novel d0 bulk oxynitride materials
is essential for the future design of more efficient Z-scheme systems.
178
One of the more recently discovered Z-scheme systems involves combining

supported Pt/ZrO2/TaON with supported RuO2/TaON using IO3-/I- as the shuttle redox
couple [20, 21].
It was proposed that supported Pt/ZrO2/TaON acts as the H2
evolution photocatalysts while supported RuO2/TaON acts as the O2 evolution

photocatalyst. Scanning electron microscopy (SEM) images of the supported
RuO2/TaON photocatalyst showed featureless RuO2 nanoparticles on the TaON
support and have been attributed to the greatly increased O2 evolution for the system.
From these observations it was hypothesized that the RuO2 NPs are the catalytic active
sites for O2 evolution, although surface characterization was not performed to
substantiate the proposed photocatalysis model.
The present study will focus on one half of a Z-scheme photocatalyst system to
determine the surface nature of the supported RuO2/TaON photocatalyst. Information
about the surface region (~1-3nm) and outermost atomic layer (~0.3 nm) of the
supported RuO2/TaON photocatalyst system will be acquired by characterization with
high-resolution X-ray photoelectron spectroscopy (HR-XPS) and high-sensitivity low
energy ion scattering (HS-LEIS) spectroscopy, respectively. Complementary bulk
molecular and electronic structural information about the TaON support phase and
influence of the addition of the deposited RuO2 NPs upon the TaON phase properties
will be acquired with in situ optical spectroscopic characterization (Raman and UVvis, respectively).
5.2 Experimental
179

The bulk TaON phase was prepared by heating Ta2O5 powder (Rare Metallic,
99.9%) in a quartz reactor tube under flowing NH3 (100 ml/min) and N2 (1 ml/min)
gases that were bubbled through water at 1123 K for 10 hrs [7]. The TaON was mixed
in an evaporating dish with an aqueous solution of the (NH4)2RuCl6 (Aldrich)
precursor yielding 0.5 wt.% Ru [20]. This suspension was then placed over a water
bath and continuously stirred with a glass rod until complete evaporation of water.
The powder was then collected and calcined in air at 623 K for one hour.
The Raman spectra for the photocatalyst were performed on a Lab Ram-HR
excitation and utilizing a confocal microscope (Olympus BX-30) for focusing the laser
on the catalyst sample. The 532nm visible laser excitation was generated by Nd:YAG
laser (10 mW) with the scattered photons directed into a single monochromator and
focused onto a UV-sensitive liquid-N2 cooled CCD detector (Horiba-Jobin Yvon
CCD-3000V) having a spectral resolution of 2 cm-1 for the given parameters. About
5-10 mg of the catalyst was placed into a high temperature in situ cell (Linkam TS1500) with a quartz window and the spectrums were obtained under ambient
conditions. The spectral acquisition time employed was 5 scans of 5 seconds/scan for
each spectrum. System calibration was verified using a silica reference standard line
at 520.7 cm-1.
5.2.3 UV-vis NIR Diffuse Reflectance Spectroscopy (DRS)
180
Ultra
Violet-visible-Near
Infrared
(UV-vis-NIR)
diffuse
reflectance
spectroscopy was utilized to obtain the optical edge energy, Eg, values for the
photocatalysts.

1
5.2.4 High Sensitivity-Low Energy Ion Scattering (HS-LEISS)
Spectroscopy
Analysis of the outermost surface layer of the photocatalysts was obtained on
the Qtac100 HS-LEIS Spectrometer (ION-TOF) equipped with a highly sensitive
181
chamber. The HS-LEIS spectra were taken using 4000 eV 4He+ with 10994 pA
current as an ion source. For depth profiling, the surface was sputtered by Ar+ gas at
1000 eV at a sputter yield of 1x1015 ions/cm2.
The HR-XPS spectra of the photocatalysts were obtained on a Scienta ESCA
300 spectrometer equipped with a 300 mm hemispherical electrostatic analyzer and a
rotating anode.
binding energy (BE) value of 285.0 eV for carbon in the C1s region. For depth
profiling, the surface was sputtered ~30 for each cycle with Ar+ gas at 1500 eV.
The atomic concentration ratios were calculated by correcting the measured peak area
ratios with relative sensitivity factors employed in the Casa XPS software version
2.3.15.
5.3.1 Bulk Characteristics of TaON Photocatalysts
The Raman spectra for Ta2O5 and the TaON photocatalysts are presented in
Figure 1. The Raman spectrum of the bulk Ta2O5 starting material is characteristic of
182
the crystalline Ta2O5(L) phase [25]. The strongest Raman band in the spectrum of the
Ta2O5(L) phase occurs at 100 cm-1 and originates from the bulk lattice photon mode
along with a second bulk lattice mode at 199 cm-1. The Raman bands at 256 and 338
cm-1 have been assigned to bridging Ta-O-Ta and TaO6 bending modes, respectively.
The bands at 489, 631, 708 and 848 cm-1 are assigned to bridging Ta-O-Ta symmetric
stretching, Ta-O symmetric stretching, bridging Ta-O-Ta antisymmetric stretching and
higher order Ta-O symmetric stretching modes, respectively.
The detailed vibrational assignments for bulk TaON are currently not available
in the literature. The Raman band at 259 cm-1, however, can be assigned to the
bridging Ta-O-Ta bending mode. The Raman bands at 125 and 175 cm-1 can also be
assigned to transverse acoustic and longitudinal acoustic phonon modes of TaNx (0.94
x 1.37) [26]. The TaNx also contains a broad optical phonon mode at 550 cm-1
which was not able to be detected for the TaON. The vibrational spectrum of TaON
demonstrates that the synthesized TaON phase is not just a physical mixture of the
Ta2O5 and TaN phases and is an oxynitride material. The Raman spectrum of bulk
TaON was not perturbed by the deposition of the Ru precursor and its oxidation to
RuO2. The vibrations of crystalline RuO2 (characteristic Raman bands at 528, 646,
and 716 cm-1[27]) are not present in the Raman spectrum of the supported
RuO2/TaON photocatalyst against the strong bands from the TaON support phase.
Thus, the bulk molecular structure of the TaON support phase is essentially
unperturbed by the deposition of RuO2 component.
183
The optical band gap (Eg) values of bulk Ta2O5, TaON and 0.5% Ru/TaON
were determined with UV-vis DRS are listed in Table 1. The obtained Eg values are
in agreement with the values reported in the literature for these bulk phases [28]. The
deposition of Ru and its oxidation to RuO2 do not have a significant effect on the band
gap of the bulk TaON photocatalyst.
5.3.2 Surface Atomic Composition of TaON photocatalysts
5.3.2.1 Outermost Surface Layer (~0.3 nm)
The outermost surface layer (~0.3 nm) and layers below the surface of the bulk
TaON photocatalyst were analyzed with dynamic HS-LEIS employing a He+ ion gas
source after atomic O pretreatment and the results are presented in Figure 2. The HSLEIS depth profile only reveals the presence of O and Ta on the surface with no signal
for the N. Although it is difficult to obtain good elemental sensitivity for low mass
elements like N without the use of 3He+ as the gas ion source [29], the complete
absence of N from the outermost surface layers suggests that this region became
oxidized by exposure to the ambient environment. The HS-LEIS depth profile reveals
that the signal for O is strongest on the outermost surface layer and decreases with
depth while the signal for Ta increases with depth. The spectra for the TaON before
atomic O treatment and after are compared in Figure 3. The treatment was found to
increase signal intensity but did not affect the surface composition of the TaON
photocatalyst. Thus, the outermost surface layers of bulk TaON do not contain any
detectable N, are enriched in O and depleted in Ta.
184
The HS-LEIS depth profiles for the supported 0.5%RuO2/TaON photocatalysts

are presented in Figure 4. Surprisingly, no Ru is detected on the outermost surface
layers and only the Ta and O signals are detected. The amount of RuO2 in the
photocatalyst is approximately an order of magnitude above the HS-LEIS detection
limit (500 ppm). Furthermore, deposition of the Ru on the TaON support should lead
to its greater concentration on the outermost surface. The lack of a HS-LEIS Ru
signal suggests that most of the Ru species are diffusing into the bulk lattice of the
TaON support during the calcination step at 623K.
5.3.2.2 Surface Region (~1-3 nm)
The XPS survey spectra of the surface region (~1-3 nm) for the TaON
photocatalysts are presented in Figure 5 and the atomic compositions (calculated from
the HR-XPS region) are reported in Table 2. The surface region of the bulk TaON
photocatalyst consists only of Ta, O, and N with no contaminants detected (e.g., Cl
from the (NH4)2RuCl6 precursor). The O/N~2, O/Ta > 1 and N/Ta~0.6 atomic ratios
demonstrate that the surface region is enriched in O and depleted in N, which is in
agreement with the HS-LEIS findings. Unlike HS-LEIS, Ru species are able to be
detected in the surface region with HR-XPS. The deposition of 0.5% RuO2 on the
TaON support did not greatly affect the surface region Ta, O and N atomic
compositions.
The composition of 0.5%RuO2/TaON with depth profiling is also
compared in Table 2 and indicates increased concentrations of Ta with decreased

concentrations of O and N. The concentration of Ru was found to increase with the
first 2 sputter cycles and the concentration was found to decrease afterwards. The
185
increasing surface concentrations of Ta and Ru and the decreasing surface

concentrations on O and N are consistent with the HS-LEIS analysis revealing that O
is surface enriched and that Ta and Ru are surface depleted and become more
concentrated below the outermost surface layers. The initial increase in concentration
of Ru with depth is consistent with the scenario that Ru has diffused into the
subsurface lattice of the bulk TaON support. The decrease in concentration with
further depth profiling indicates that it not uniformly distributed in the bulk phase and
Ru is localized beneath the surface layer.
The HR-XPS spectra of the Ru 3d region for the unsputtered and sputtered
supported RuO2/TaON samples are compared in Figure 6. The Ru 3d5/2 binding
energy indicates that Ru4+ cations predominate in the supported RuO2/TaON
photocatalysts. The HR-XPS spectra of the Ta 4f region for the unsputtered and
sputtered supported RuO2/TaON samples are also compared in Figure 7. The position
of Ta 4f binding energies on the surface indicate that only Ta5+ oxidized species are
present. However with depth profiling, the Ta5+ 4f bands decrease in intensity and a
shoulder appears at a lower binding energy. This shift is indicative of reduced Ta and
is consistent with the presence of greater Ta-N interaction beneath the surface layer
[30]. The broad nature of the Ta 4f doublet bands suggests that both Ta-O and Ta-N
interactions are present in the bulk.
5.3.2.3 Surface Nature of TaON Photocatalysts
The HS-LEIS analysis showed that Ta and O were the only elements present
on the outermost surface layer. The position of the HR-XPS Ta 4f binding energies
186
show that Ta is in an oxidized state on the surface layer and there exist a TaOx outer
layer for the photocatalyst. Depth profiling revealed that the chemical state of the Ta
changes from exclusively TaOx to a combination of TaOx and TaNx which confirms
that TaOx is surface enriched and that TaON is the bulk state. The deposited Ru was
proposed to be concentrated on the surface, however the lack of a Ru signal in the HSLEIS spectra shows that is not the case. HR-XPS confirmed that R1u is not surface
enriched and it is mostly concentrated below the surface region. The schematic for the
surface of the photocatalyst is shown in Figure 8.
5.3.3 Structure-Photoactivity Relationships
The new insights indicate the supported RuO2/TaON photocatalyst system is a
much more complex material than was previously proposed with RuO2 NPs deposited
on the external surface of the TaON support with the RuO2 NPs the catalytic
photoactive sites proposed to be responsible for O2 evolution [20]. Rather than RuO2
NPs being the catalytic photoactive sites for O2 evolution, it appears that the Ru+4 sites
dissolved in the TaON support surface region, which is enriched with O and depleted
in N, are responsible for O2 evolution. Higher RuO2 loadings on the TaON support
were found to significantly decrease the photoactivity and may indeed be related to
formation of RuO2 NPs that are not photoactive. The function of the bulk TaON
support is to supply visible light excited photoelectrons that can be utilized by the
Ru+4 catalytic active sites, in the oxygen rich oxynitride TaON surface region, to
evolve O2 during photodecomposition of H2O. The new surface insights are changing
187
our models about how complex semiconductor photocatalysts function and will guide
the development of advanced photocatalytic materials.
5.4 Conclusions
The bulk molecular and electronic structures of the TaON phase are not
affected by the deposition of RuO2 on the TaON support.
For the first time, the
natures of the surface region of TaON and supported RuO2/TaON photocatalysts have
been determined. The outermost surface layers primarily consist of TaOx phase and is
depleted in N. The RuO2 component is not present as NPs decorating the TaON
support as previously concluded, but the Ru4+ cations are dissolved in the TaOx outer
layers and absent from the outermost surface layer. The new surface insights are
changing our models about how complex semiconductor photocatalysts function and
will guide the development of advanced photocatalytic materials.
Acknowledgements
DOE-FG02-93ER14350. Thanks to A. Xiong, Dr. J. Kubota and Professor K. Domen
at the University of Tokyo for their assistance with the synthesis of the TaON
photocatalysts. The assistance of Dr. A. Miller and R. Pafchek at Lehigh University in
obtaining and interpreting the HR-XPS and HS-LEIS data is also gratefully
acknowledged
188
References
[2] A. Kasahara, K. Nukumizu, G. Hitoki, T. Takata, J.N. Kondo, M. Hara, H.

Kobayashi, and K. Domen, "Photoreactions on LaTiO2N under Visible Light
Irradiation." J. Phys. Chem. A, 106, 6750, (2002).
[3] G. Hitoki, T. Takata, U.N. Kondo, M. Hara, H. Kobayashi, and K. Domen,

"(Oxy)nitrides as new photocatalysts for water splitting under visible light irradiation."
Electrochemistry (Tokyo, Jpn. ), 70, 463, (2002).
[4] D. Yamasita, T. Takata, M. Hara, J.N. Kondo, and K. Domen, "Recent progress of
visible-light-driven heterogeneous photocatalysts for overall water splitting." Solid
State Ionics, 172, 591, (2004).
[5] G. Hitoki, T. Takata, J.N. Kondo, M. Hara, H. Kobayashi, and K. Domen, "An
oxynitride, TaON, as an efficient water oxidation photocatalyst under visible light
irradiation (l ? 500 nm)." Chem. Commun. (Cambridge, U. K. ), 1698, (2002).
[6] M. Hara, J. Nunoshige, T. Takata, J.N. Kondo, and K. Domen, "Unusual

enhancement of H2 evolution by Ru on TaON photocatalyst under visible light
irradiation." Chem. Commun. (Cambridge, U. K. ), 3000, (2003).
189
[7] M. Hara, T. Takata, J.N. Kondo, and K. Domen, "Photocatalytic reduction of water
by TaON under visible light irradiation." Catal. Today, 90, 313, (2004).
[8] M. Liu, W. You, Z. Lei, G. Zhou, J. Yang, G. Wu, G. Ma, G. Luan, T. Takata, M.
Hara, K. Domen, and C. Li, "Water reduction and oxidation on Pt-Ru/Y2Ta2O5N2
catalyst under visible light irradiation." Chem. Commun. (Cambridge, U. K. ), 2192,
(2004).
[9] K. Maeda, Y. Shimodaira, B. Lee, K. Teramura, D. Lu, H. Kobayashi, and K.

Domen, "Studies on TiNxOyFz as a Visible-Light-Responsive Photocatalyst." J. Phys.
Chem. C, 111, 18264, (2007).

Phys. Chem. C, 111, 7554, (2007).

Chem. B, 110, 13753, (2006).
190



[16] R. Abe, "Development of a New System for Photocatalytic Water Splitting into
H2 and O2 under Visible Light Irradiation," Bull. Chem. Soc. Jpn., 84, 1000, (2011).
[17] R. Abe, T. Takata, H. Sugihara, and K. Domen, "Photocatalytic overall water

splitting under visible light by TaON and WO3 with an IO3-/I- shuttle redox
mediator." Chem. Commun. (Cambridge, U. K. ), 3829, (2005).
191
[18] M. Higashi, R. Abe, K. Teramura, T. Takata, B. Ohtani, and K. Domen, "Two

step water splitting into H2 and O2 under visible light by ATaO2N (A = Ca, Sr, Ba)
and WO3 with IO-3/I- shuttle redox mediator." Chem. Phys. Lett., 452, 120, (2008).
[19] K. Maeda, M. Higashi, D. Lu, R. Abe, and K. Domen, "Efficient Nonsacrificial

Water Splitting through Two-Step Photoexcitation by Visible Light using a Modified
Oxynitride as a Hydrogen Evolution Photocatalyst." Zhonghua Jianyan Yixue Zazhi,
32, 5858, (2009).
[20] K. Maeda, R. Abe, and K. Domen, "Role and Function of Ruthenium Species as
Promoters with TaON-Based Photocatalysts for Oxygen Evolution in Two-Step Water
Splitting under Visible Light" The Journal of Physical Chemistry C, 115, 3057,
(2011).
[21] M. Higashi, R. Abe, A. Ishikawa, T. Takata, B. Ohtani, and K. Domen, "Zscheme overall water splitting on modified-TaON photocatalysts under visible light (l
< 500 nm)." Chem. Lett., 37, 138, (2008).

104, 1261, (2000).
192

[25] Y. Chen, J.L.G. Fierro, T. Tanaka, and I.E. Wachs, "Supported Tantalum Oxide
Catalysts: Synthesis, Physical Characterization, and Methanol Oxidation Chemical
Probe Reaction," The Journal of Physical Chemistry B, 107, 5243, (2003).
[26] M. Stoehr, C.-. Shin, I. Petrov, and J.E. Greene, "Raman scattering from epitaxial
TaN[sub x] (0.94=x=1.37) layers grown on MgO(001) " J. Appl. Phys., 101, 123509,
(2007).
[27] A.V. Korotcov, Y. Huang, K. Tiong, and D. Tsai, "Raman scattering

characterization of well-aligned RuO2 and IrO2 nanocrystals " J. Raman Spectrosc.,
38, 737, (2007).
[28] T. Takata, G. Hitoki, J.N. Kondo, M. Hara, H. Kobayashi, and K. Domen,

"Visible-light-driven photocatalytic behavior of tantalum-oxynitride and nitride." Res.
Chem. Intermed., 33, 13, (2007).

composition analysis by low-energy ion scattering" Surface Science Reports, 62, 63,
(2007).
193
[30] W. Chun, A. Ishikawa, H. Fujisawa, T. Takata, J.N. Kondo, M. Hara, M. Kawai,

Y. Matsumoto, and K. Domen, "Conduction and Valence Band Positions of Ta2O5,
TaON, and Ta3N5by UPS and Electrochemical Methods " The Journal of Physical
Chemistry B, 107, 1798, (2003).
194
Catalyst
Ta2O5
TaON
0.5%Ru/TaON
Eg (eV)
4.1
2.8
2.8
Table 5.1: Uv-vis DRS Edge Energy Values for Tantalum-based Photocatalysts.
195
Element
Ta 4f
O 1s
N 1s
Ru 3d 5/2
TaO
N
Atom
ic %
37.26
40.40
22.33
0.00
Ru/TaON
Surface
Atomic
%
35.62
41.89
21.65
0.84
Ru/TaON
30
Atomic
%
43.77
35.57
19.73
0.93
Ru/TaON
60
Atomic
%
45.14
35.68
18.18
1.00
Ru/TaON
90
Atomic
%
45.93
36.12
17.07
0.88
Ru/TaON
120
Atomic
%
46.92
36.36
16.07
0.65
Table 5.2: HR-XPS Surface Region Atomic Composition of TaON Photocatalysts.
196
125
256
631
199
Intensity (a.u.)
708
489
338
848
175
TaON
0.5%RuO2/TaON
Ta2O5
200
400
600
800
1000
-1
Raman Shift (cm )
Figure 5.1: Raman Spectra of Tantalum-based Photocatalysts (532 nm).
197
1200
Ta
12
10
8
6
4
2
Sp
ut
te
rC
yc
le
1000
1500
2000
2500
3000
3500
4000
Energy (eV)
Figure 5.2: HS-LEIS Depth Profile of the Bulk TaON Photocatalyst.
198
Intensity (cts/nC)
14
10
After atomic O Treatment
Ta
Intensity (cts/nC)
Untreated
2
0
1000
2000
3000
4000
Energy (eV)
Figure 5.3: HS-LEIS spectra of Bulk TaON Photocatalyst Before and After Atomic
O Pretreatment.
199
Ta
14
Ru
12
10
8
6
4
2
Sp
ut
te
r
yc
le
1000
1500
2000
2500
3000
3500
Energy (eV)
Figure 5.4: HS-LEIS Depth Profile of the Supported RuO2/TaON Photocatalyst.
200
Intensity (cts/nC)
16
Ta 4f
Intensity (counts)
O 1s
N 1s
1200
1000
800
600
Binding Energy
Figure 5.5: XPS Survey Spectra of Bulk TaON.
201
400
200
Intensity (Counts)
C 1s
Ru
4+
3d5/2
Ru
0
120 Angstrom
60 Angstrom
Surface
290
288
286
284
282
280
278
Binding Energy (eV)
Figure 5.6:
HR-XPS surface analysis of Rh 3d region for supported
202
5+
Ta 4f7/2
Intensity (Counts)
5+
Ta 4f5/2
120 Angstrom
Surface
30
28
26
24
60 Angstrom
22
20
Binding Energy (eV)
Figure 5.7: HR-XPS surface analysis of Ta 4f region for supported 0.5%RuO2/TaON

photocatalysts.
203
Figure 5.8:
Schematic Model for the Surface Region of the Supported 0.5%
RuO2/TaON photocatalyst.
204
CHAPTER 6
Conclusions and Future Studies
6.1 Conclusions
Much progress has been achieved over the past 40 years in discovering
advanced materials that are capable of photocatalytic water splitting. Unfortunately,
the progress in the fundamental scientific understanding of these advanced
photocatalytic materials has significantly lagged.
One of the major issues with
previous studies of the semiconductor photocatalysts has been the focus on utilizing
bulk characterization techniques and attempting to correlate photocatalyst bulk
structural properties with the corresponding photoactivity. Photocatalysis is a process
that involves both bulk (electron/hole generation), diffusion of the electrons/holes to
the surface and reaction of H2O with excited electrons and holes at the surface to produce H2
and O2. Although bulk characterization techniques effectively elucidate bulk molecular and
electronic structures of complex semiconductor photocatalytic systems, the catalytic
reactions between the electrons/holes and water take place on the outermost surface
layer and further progress in photocatalysis will come from surface characterization
studies that will provide new insights. Only by correlating surface properties with
photoactivity can realistic photocatalytic models be developed.
205
The fundamental
bulk/surface structurephotoactivity relationships for a few of the more interesting

photocatalyst systems were investigated in this dissertation and the main conclusions
for each chapter are summarized below.
Chapter 2
The most active photocatalyst under UV irradiation (0.2%NiO/NaTaO3:2%La)
was studied in this chapter to determine the effect of the La2O3 and NiO species on
photoactivity. The addition of both La2O3 and NiO did not cause any change in the
bulk molecular and electronic structure of the NaTaO3.
HR-XPS and HS-LEIS
spectroscopy revealed that the NiO and La2O3 promoters are surface segregated on the
bulk NaTaO3 phase. La2O3 was found to be a surface structural promoter, stabilizing
the NaTaO3 particles and increases the surface area by a factor ~10. The specific
photoactivity was found not to be enhanced by the La2O3. NiO was found to act as
efficient electron traps preventing electron/hole recombination in the bulk and helping
to facilitate surface migration for photocatalysis. This leads to the enhancement of
photoactivity by a factor of 101-102 in the presence or absence of the La2O3 promoter.
These new findings brings into question the validity of the previously accepted model
for the NiO/NaTaO3:La photocatalyst where a synergistic interaction of NiO
preferentially self-assembling at nanostep structures created by the doped La2O3 was
responsible for the enhanced photoactivity.
The importance of normalizing
photocatalytic activity by unit surface area was emphasized as well since the accepted
practice of normalizing by unit mass is not fundamentally meaningful.
Chapter 3
206
The fundamental bulk/surface structure photoactivity relationship was

investigated on bulk GaN and supported (Rh2-yCryO3)/GaN photocatalysts. Raman
and UV-vis spectroscopy was able to show that the bulk structure of the GaN phase
was not perturbed by the (Rh2-yCryO3) NPs. The surface of the GaN phase was found
to be altered in the surface region: GaN GaOxNy GaOx. The (Rh2-yCryO3) NPs
were found to consist of Rh3+ and Cr3+ species both found on the outermost surface
layer. The addition of (Rh2-yCryO3) assisted in the transfer of photoexcited electrons
to the surface promoting the photocatalytic surface reaction. Thus, it was determined
that Rh3+ is responsible for enhanced H2 and Cr3+, GaOx or at their contact points are
the possible sites for O2 production.
Chapter 4
Two different methods for the preparation of Rh-Cr NPs on (Ga1-xZnx)(N1-xOx)
were compared to determine surface/photoactivity relationships. (Rh2-yCryO3) NPs on
(Ga1-xZnx)(N1-xOx) were found to consist of Rh+3 surface active sites for the generation
of H2 with Cr+3, GaZnOx or at their contact points as the possible surface active sites
for the generation of O2 similar to (Rh2-yCryO3) NPs on GaN. Rh+3 and Cr+3 were also
found to be the dominant surface species for the (Rh/Cr2O3)/(Ga1xZnx)(N1xOx) with
trace amounts of Rh0 on the surface. The presence of the exposed metallic Rh0, which
catalyzes the back reaction, was found to be responsible for the ~50% less H2/O2
production for the (Rh/Cr2O3)/(Ga1xZnx)(N1xOx). PL spectroscopy was also able to
show that the Rh-Cr NPs affected the recombination of electron/holes in the bulk
207
(Ga1xZnx)(N1xOx) by increasing the lifetime of excited electrons and the population

of these long lived electrons so they can be collected at the surface and harnessed for
photocatalytic water splitting. These new insights were used to propose new modified
models for the photocatalyst systems.
Chapter 5
The surface nature of bulk TaON and RuO2/TaON photocatalysts was the
focus of investigation for this chapter. The bulk molecular and electronic structure
was found to not be perturbed by the addition of RuO2. Analysis of the outermost
surface layer revealed only the presence of TaOx. Depth profiling with HR-XPS was
able to show that concentration of Ru4+ increased with sputtering showing that most of
the Ru4+ was diffusing into the bulk and were not surface enriched. The presence of
the dissolved Ru4+ in the bulk was proposed to be responsible for enhanced activity for
the RuO2/TaON photocatalysts.
A new model was then developed for the
RuO2/TaON photocatalyst from the surface analysis.

6.2 Future Studies
In the past, the lack of utilizing surface characterization techniques has led to
proposed photocatalytic models based on bulk structure that has hindered the progress
in development of fundamental photocatalytic models that can guide the design of
advanced photocatalytic materials. The hypothesis of this research proposal is that
only when bulk and surface characterization techniques are combined can fundamental
structure-photoactivity relationships be established for complex, multicomponent
208
photocatalysts. The methodology for determining fundamental structure-photoactivity

relationships in this dissertation can thus be applied to other photocatalyst systems to
help in the scientific understanding of how photocatalyst function.
The insights
gained can then be applied in designing more efficient photocatalyst systems. For
example,
the
conclusions
from
this
study
show
that
the
synthesis
of
(Rh/Cr2O3)/(Ga1xZnx)(N1xOx) can be improved by the modification of the GaZnOx

surface by adding metal oxide dopants/other known metal oxide active sites and by
exchange Cr2O3 with other transition metal promoters.
Critical surface information can be obtained with HS-LEIS and HR-XPS
characterization that are able to determine the atomic composition and chemical state
of the surface of the photocatalysts, however the signal from these characterization
techniques are averaged over a large raster size. The elements present on the surface
are able to be quantified but these techniques do not give information on how the
elements are distributed on the surface. Additional information about the distribution
of the various elements in the complex photocatalysts can come from electron
microscopy with the use of STEM-EELS and STEM-HAADF that is able to provide
atomic resolution.
The main drawback to HS-LEIS and HR-XPS, as well as electron microscopy,
is that these characterization techniques require the measurements to be performed in
ultrahigh vacuum and, thus, are unable to operate during photocatalysis with
condensed water.
The application of in situ ATR-IR and transient ATR-IR
spectroscopy allows analysis of photocatalysts at the solid/liquid interphase during

209
photocatalysis with condensed water as long as the photocatalyst possesses a high

surface area (number of surface sites). ATR-IR spectroscopy can provide information
about aqueously dissolved and adsorbed reactants and products, adsorbed
intermediates, byproducts, and spectators.
Such fundamental studies can help to
determine the most abundant reactive intermediate on the surface and identify the ratedetermining-step for the photocatalytic process. This will provide the much needed
surface kinetics of the different bond making and bond breaking chemical steps and
their sequence under the photocatalytic process.
The low surface area for most
promising photocatalyst systems, however, will make it quite challenging to obtain

measurable signals. If such measurements become successful in the future, then the
knowledge gained will significantly assist in advancing the scientific foundation for
the photocatalytic water splitting surface reactions. Such fundamental information
about the mechanism for water splitting is severely lacking in the literature and only
such experiments will help to clarify the photocatalysis process by semiconductor
materials.
Photocatalytic water splitting is still a developing technology and the science
behind the photocatalytic process is still not fully understood. A little over 10 years
ago, photocatalytic water splitting with the use of visible light was still considered to
be a dream reaction. The advances made in the past decade have demonstrated that
there are several photocatalyst systems that are capable for photocatalytic water
splitting using visible light. The next challenge is to increase the efficiency of these
visibly activated photocatalyst systems so that the process can become commercially
210
viable. An efficient photocatalysis process has the potential to help address global
energy and environmental problems, and be the ultimate source of sustainable green
energy. Establishing this fundamental foundation is critical for the rational design of
future photocatalyst systems.
211
VITAE
Education
Ph.D., Chemical Engineering, Lehigh University
2006-2013
Dissertation Title: Fundamental Characterization Studies of Advanced Photocatalytic

Materials
Advisor: Israel E. Wachs
B.S., Chemical Engineering, Lehigh University
2002-2006
Research Experience
Research Assistant, Lehigh University
Synthesized and characterized supported Pt/ZrO2/SiO2 and WO3/ZrO2/SiO2 by

controlling ZrO2 nano-domain size for methanol oxidation reactions.
Synthesized and characterized supported TiO2/SiO2, MOx/TiO2/SiO2, and
Ta2O5/SiO2 for photocatalytic water splitting reactions.
Surface characterized UV-light activated bulk mixed Tantalum photocatalysts
to determine the nature of the catalytic active sites.
Collaborated with a research group at the University of Tokyo for surface
characterization and improved catalyst design of visible light activated bulk
mixed oxide photocatalysts.
Teaching Assistant for Chemical Engineering Lab I, Chemical Equilibria in
Aqueous Systems and Transport Processes (graduate-level).
Undergraduate Student Researcher, Lehigh University
2005-2006
Opportunity for Student Innovation Program on examining small-scale fluids

systems.
Designed mm-scale fluid systems using PDMS polymer utilizing electroosmosis for fluid flow.
Velocity profiles and pressure drop properties were examined to compare
electro-osmosis and pressure driven flow for running small-scale chemical
reactions.
Research Intern, Mallinckrodt Baker, Phillipsburg, NJ
2006-2013
Summer 2005
Worked in clean room environment to develop sample preparation methods for

reduced analysis time of contaminants in acids and organic solvents.
212
Developed experimental protocol for microwave evaporation for trace metal

analysis.
Performed trace metal analysis of acids and organic solvents for contamination
using ICP-OES and ICP-MS.
Student Researcher, South Dakota School of Mines and Technology,

Rapid City, SD
Summer 2003
Research Experience for Undergraduates on pretreatment of biomass

for enhanced ethanol production.
Processed the breakdown of lignocellulosic materials into fermentable
sugars for ethanol production.
Ran simultaneous saccharification and fermentation reactions in flask
batch reactor analyzing product composition with HPLC.
Professional Affiliations
Tau Beta Pi Engineering Honors Society, American Institute of Chemical Engineers,
American Chemical Society, Catalysis Club of Philadelphia, and New York Catalysis
Society
Publications
Determination of a Structure-Photocatalytic Relationship for Water Splitting on
Well-Defined TiO2 Nanodomains, C.A. Roberts, A.A. Puretzky , S.P. Phivilay and
I.E. Wachs (in preparation).
Influence of Oxide Support Domain Size on the Characteristics of Supported Metal
and Metal Oxide Nanophases, S.P. Phivilay, W. Zhou, E.I. Ross-Medgaarden, C.J.
Kiely, and I.E. Wachs (in preparation).
Selected Presentations
Design and Characterization of Photocatalysts for Solar Powered Hydrogen
Generation, S.P. Phivilay, C.A. Roberts, A. Puretzky, A. Xiong, J. Kubota, K. Maeda,
K. Domen and I.E. Wachs, Lehigh University Chemical Engineering Student Seminar,
Bethelehem, PA (March 30, 2012).
Fundamental Surface Structure-Photoactivity Relationships of Tantalum-based
Photocatalysts, S.P. Phivilay, X. Yang, G.F. Liu, B. Koel, and I.E. Wachs, The 22nd
NAM Meeting, Detroit, MI (June 6, 2011).
213
Fundamental Surface Structure-Photoactivity Relationships of Advanced

Photocatalysts, S.P. Phivilay, C.A. Roberts, X. Yang, G.F. Liu, B. Koel, and I.E.
Wachs, The 241th ACS National Meeting, Anaheim, CA (March 30, 2011).
Photocatalysts, S.P. Phivilay, C.A. Roberts, and I.E. Wachs, Gordon Research
Conference student poster competition, New London, NH (June 27 - July 2, 2010).
Photocatalysts, S.P. Phivilay, C.A. Roberts, and I.E. Wachs, The 239th ACS Annual
meeting, San Francisco, CA (March 21-24, 2010).
The Structure-Photocatalytic Activity Relationships of TiO2/SiO2 Nanocatalysts,
S.P. Phivilay and I.E. Wachs, Catalysis Club of Philadelphia student poster
competition, Claymont, DE (March 19, 2009).
The Structure-Photocatalytic Activity Relationships of TiO2/SiO2 Nanocatalysts,
S.P. Phivilay and I.E. Wachs, Spring Symposium of the Catalysis Society of
Metropolitan New York student poster competition, Princeton, NJ (March 18, 2009).
Influence of ZrO2 Nanoligands on the Catalytic Performance of Supported
Pt/ZrO2/SiO2 Catalysts, S.P. Phivilay, W. Zhou, C. Kiely, and I.E. Wachs, AIChE
Annual Meeting, Philadelphia, PA (Nov 16-21, 2008).
214

Fundamental Characterization Studies of Advanced Photocatalytic M PDF

Uploaded by

Copyright:

Available Formats

Fundamental Characterization Studies of Advanced Photocatalytic M PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fundamental Characterization Studies of Advanced Photocatalytic M PDF

Uploaded by

Copyright:

Available Formats

Lehigh University

Fundamental Characterization Studies of

Follow this and additional works at: http://preserve.lehigh.edu/etd

Fundamental Characterization Studies

Copyright 2013 by Somphonh Peter Phivilay

Overview of Heterogeneous Photocatalyst Literature for Water Splitting

Nature of Catalytic Active Sites Present on the Surface of Advanced Bulk

Fundamental Bulk/Surface Structure Photoactivity Relationships of

3.2.1 Catalyst Synthesis

Nature of Catalytic Active Surface Sites Present on Bulk (Ga1-xZnx)(N1-xOx)

4.2.7 Photocatalytic Water Splitting

Investigating the Surface Nature of TaON Photocatalysts

5.2.5 High Resolution X-ray Photoelectron (HR-XPS)

Conclusions and Future Studies

Table 2.2: Photoluminescence decay fit parameters for catalysts at 267 nm

Table 2.3: XPS surface region atomic concentration of promoted

Table 2.4: Photoactivity of Tantalum Oxide Photocatalysts for Water Splitting

Table 3.2: HR-XPS surface region atomic composition (~1-3nm) of GaN

Table 5.2: HR-XPS Surface Region Atomic Composition of TaON

Figure 1.2: Schematic Energy Diagram of photocatalytic water-splitting systems:

Figure 2.2: Photoluminescence Spectra at 267 nm Excitation for Tantalumbased Photocatalysts.

Figure 2.3: Photoluminescence Emissions Decay Curves for Tantalum-based

Figure 2.4: XPS survey spectrum of surface region for 0.2%NiO/NaTaO3:2%La

Figure 3.5: HS-LEIS spectra of untreated and atomic O treated supported

Figure 3.6: HS-LEIS depth profile for the supported (Rh2-yCryO3)/GaN

Figure 3.10: Schematic of (Rh2-yCryO3)/GaN photocatalysts.

Figure 4.2: Raman Spectra of (Ga1-xZnx)(N1-xOx) and (Rh2-yCryO3)

Figure 4.4: HR-XPS spectra of Cr 2p and Rh 3d regions of (a) fresh and

Figure 4.6: HS-LEIS spectra for (Ga1xZnx)(N1xOx) photocatalyst using

Figure 4.7: HS-LEIS Depth Profile for the fresh (Rh2-yCryO3)/(Ga1xZnx)(N1xOx)

Figure 4.11: HS-LEIS Depth Profile of Cr and Rh for used supported

Figure 4.12: HS-LEIS Depth Profile for the fresh supported

Figure 4.14: HS-LEIS Depth Profile for the used supported

Figure 4.18: Comparison of Photocatalytic Water Splitting Activity

Figure 4.19: Schematic Model of Rh-Cr NPs for Co-Impregnation Synthesis

Figure 4.20: Schematic Model of Rh-Cr NPs for Photodeposition Synthesis

Figure 5.2: HS-LEIS Depth Profile of the Bulk TaON Photocatalyst.

Figure 5.6: HR-XPS surface analysis of Rh 3d region for supported

Figure 5.7: HR-XPS surface analysis of Ta 4f region for supported

and TaON) were investigated to establish fundamental bulk/surface

structure photoactivity relationships. The bulk molecular and electronic structures of

the photocatalysts were determined with Raman and UV-vis spectroscopy.

photocatalysts were determined with high resolution X-ray photoelectron spectroscopy

multicomponent photocatalyst systems.

Mimicking this process for production of hydrogen fuel will help to

The designs of these PEC

Unlike PEC cells, the use of heterogeneous

semiconductor materials allows for greater simplicity in design for photocatalytic

It is not always clear which

component in a multi-component photocatalyst system is the catalytic active site so it

production and the undetectable amounts of O2 evolution attributed to adsorbed O2

From all of the discovered materials for

photocatalytic water splitting, only certain semiconductor materials possess suitable

Tungstates (WO3, 2.8 eV) are less active than

Some of the effective photocatalysts consist of the bulk

have shown great

photocatalytic water splitting. Kudos group has demonstrated that tantalum-based

discovered to date with an apparent quantum efficiency (number of reacted electrons/

Metal oxynitride catalysts ((Ga1-xZnx)(N1-xOx) and (Zn1+xGe)(N2Ox))