Photothermal and Photochemical Nanostructured Catalyst Engineering Towards Efficient Solar Fuel Production

by

Jia Jia

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Department of Materials Science and Engineering University of Toronto

Copyright by Jia Jia 2017

ii

Photothermal and Photochemical Nanostructured Catalyst

Engineering Towards Efficient Solar Fuel Production

Jia Jia

Doctor of Philosophy

Department of Materials Science and Engineering

University of Toronto

2017

Abstract

With global energy demand rising alongside the advancement of climate change, the conversion

of greenhouse gas carbon dioxide into value-added chemicals and fuels is attracting increasing

attention. The crux for the successful development of this promising technology is the

exploration and discovery of highly active, selective, and stable catalyst materials. Herein we

demonstrate that the reverse water gas shift (RWGS) reaction can be driven by Nb2O5 nanorod-

supported Pd nanocrystals, without external heating, using visible and near infrared (NIR) light.

By measuring the dependence of the RWGS reaction rates on the intensity and spectral power

distribution of filtered light incident on the nanostructured Pd@Nb2O5 catalyst, we determine the

RWGS reaction to be initiated by heat generated from thermalized charge carriers in the Pd

nanocrystals that are excited by inter-band and intra-band absorption of visible and NIR light.

We also demonstrate that the catalytic activity and selectivity of CO2 reduction to CO and CH4

products can be systematically tailored, by varying the size of the Pd nanocrystals, to acheive

champion turnover frequencies (0.61 s1) and efficient conversion of solar energy to stored

chemical energy. The remarkable control over the catalytic performance of Pd@Nb2O5 stems

from a combination of photothermal, electronic, and size effects. The insight gleaned from this

detailed experimental-theoretical study provides a blueprint for how to tailor the performance

iii

metrics of earth-abundant, low-cost metal-metal oxide (M@M'Ox) analogues. Finally, we report

a lattice strain and defect controlled strategy that enables the high-performance of low-cost

photocatalystic materials. Lattice compressed ultrafine nonstoichiometric indium oxide dots,

In2O3-x(OH)y, grown on the surface of niobium pentoxide nanorods were fabricated; the

optimized hybrid structure exhibits 44-fold increase in efficiency compared with pristine In2O3-

x(OH)y, along with extremely long-term operational stability, potentially originating from the

increased number of active oxygen vacancies, prolonged excited-state lifetimes, and enhanced

photo-generated carrier energies.

iv

Acknowledgments

I thank my supervisors, Prof. Geoffrey Ozin, Prof. Doug Perovic and Prof. Nazir Kherani for

their strong support and insightful guidance and for the great degree of scientific freedom that

they has afforded me throughout my PhD studies. I would also like to thank my committee

members, Profs. Chandra Veer Singh, Robert Morris, Yujie Xiong, Zheng-Hong Lu, and Dwight

Seferos, for their advice and encouragement over the course of this project.

I am very grateful to have been able to work with so many excellent collaborators. In particular, I

would like to thank all the engineers in the Solar Fuel ClusterDr. Thomas Wood, Dr. Paul

OBrien, and Amit Sandhel, without whom this project would not have been possible, Dr. Paul

OBrien for all of many great discussions and for all of his hard work, Dr. Hong Wang and Dr.

Miaomiao Ye for the synthetic insights and great collaboration, and Prof. Amr Helmy for the

help and guidance with Raman, and Joller Lu, Mireille Ghoussou, Dr. Kulbir Kaur Ghuman for

all of the computational insights, and Dr. Qiao Qiao for her hard work on the high resolution

imaging, and Dr. Lu Wang and Abdinoor Jelle for the help with XPS analysis presented herein,

and Dr. Paul Duchesne for the help with XAS analysis, and Peicheng Li for UPS analysis, and

Yuchan, Mireille Ghoussou as well as Dr. Laura Reyes for BET, TGA, and DRIFT

measurements, and Teng Fei, Ziqi Zheng as well as Yue Shao for their excellent undergraduate

research.

I am also extremely grateful to my colleague in the Solar Fuel Cluster both past and present for

all of their camaraderie and support, over the past 5 years. In particular, I would like to thank Sue

Mamiche for always keeping the group running so smoothly, Dr. Mohammed Hmadeh, Dr. Paul

OBrien, Dr. Le He, Dr. Miaomiao Ye, Dr. Hong Wang, and Dr. Lu Wang, Dr. Benoit Mahler

for their constant supply of helpful and entertaining discussions as senior colleague, Dr. Wei Sun,

Dr. Chenxi Qian, Dr. Laura Hock, Dr. Jon Moir, Mireille Ghoussou, Yuchan Dong, Meikun Xia,

Annabelle Wong, Kenny Chen, Dr. Laura Reyes, Dr. Melanie Matronardi, Dr. Navid Soheilnia,

Dr. Daniel Faulkner, Dr. Liwei Wang, Prof. Changlong Chen, and Prof. Dongzhi Chen, for

encouragement, advice, and friendship.

v

I also appreciate all the help from the administrative staff in the department, Maria Fryman,

Fanny Strumas-Manousos, Krista Haldenby, and Luke Y. H. Ng, for their help with my life and

study in Canada.

Financial support from the Doctoral Completion Award, the Atsumi Ohno Scholarship, the

Lachlan Gilchrist Fellowship Fund, the Nanotechnology Network Graduate Student

Enhancement Award, and the Department of Materials Science and Engineering, and University

of Toronto is gratefully acknowledged.

Finally, I would like to thank my family, my boyfriend Wei Zhao, and my friends for their

continued love and support in everything I choose to do. I would certainly not be where I am

today without them and knowing that no matter where they are they will always be proud of me.

vi

Table of Contents

Acknowledgments .......................................................................................................................... iv

Table of Contents ........................................................................................................................... vi

List of Tables ................................................................................................................................. ix

List of Figures ................................................................................................................................ xi

Copyright Acknowledgements ..................................................................................................... xxi

Chapter 1 Introduction and Background ......................................................................................... 1

1 Introduction and Background ..................................................................................................... 1

1.1 Introduction ......................................................................................................................... 1

1.2 Methanation ........................................................................................................................ 2

1.3 CO Production .................................................................................................................... 4

1.4 Methanol Synthesis ............................................................................................................. 9

1.5 C2 C11 ............................................................................................................................. 11

1.6 The Goals of this Doctoral Thesis Work .......................................................................... 13

Chapter 2 Photothermal Catalysed Hydrogenation of Gaseous CO2 over Nanostructured

Pd@Nb2O5 ............................................................................................................................... 15

2 Photothermal Catalysed Hydrogenation of Gaseous CO2 over Nanostructured Pd@Nb2O5... 15

2.1 Abstract ............................................................................................................................. 15

2.2 Introduction ....................................................................................................................... 16

2.3 Experimental ..................................................................................................................... 17

2.3.1 Synthesis of Nb3O7(OH) Nanorods ...................................................................... 17

2.3.2 Synthesis of Nanostructured Pd@Nb2O5 .............................................................. 18

2.3.3 Physical Characterization ...................................................................................... 18

2.3.4 Gas Phase Catalytic Measurements ...................................................................... 20

vii

2.4 Results and Discussion ..................................................................................................... 22

2.4.1 Characterization of Nanostructured Pd@Nb2O5 ................................................... 22

2.4.2 Investigation of Photothermal RWGS Catalysis on Nanostructured Pd@Nb2O5. 31

2.4.3 Investigating the Effect of Spectral Distribution and Light Intensity on the

Photothermal Catalytic Activity ........................................................................... 32

2.4.4 Investigating the Effect of the Temperature Dependence of RWGS Reaction

Rates over Pd@Nb2O5 in the Dark ....................................................................... 35

2.4.5 Raman Investigation of the Visible Light Induced Photothermal Effect .............. 37

2.4.6 Investigating In-situ Generated Oxygen Vacancies and Reduced Oxidation

State Surface Niobium Sites in the Photothermal Catalytic CO2 Reduction of

Nanostructured Pd@Nb2O5................................................................................... 44

2.5 Conclusions ....................................................................................................................... 47

Chapter 3 Photothermal Catalyst Engineering with High Activity and Tailored Selectivity ....... 49

3 Photothermal Catalyst Engineering with High Activity and Tailored Selectivity ................... 49

3.1 Abstract ............................................................................................................................. 49

3.2 Introduction ....................................................................................................................... 50

3.3 Experimental ..................................................................................................................... 51

3.3.1 Preparation of Pd@Nb2O5 Heteronanostructure with Different Loadings of Pd.. 51

3.3.2 Characterization .................................................................................................... 51

3.3.3 Gas Phase Catalytic Measurements ...................................................................... 53

3.4 Results and Discussion ..................................................................................................... 54

3.4.1 Synthesis and Characterization of Pd@Nb2O5 Heteronanostructures .................. 54

3.4.2 Photothermal Materials Engineering of the Gas-Phase CO2 Catalytic

Reduction .............................................................................................................. 63

3.5 Conclusion ........................................................................................................................ 68

Chapter 4 Mechanistic Insights into the Photocatalytic Reduction of CO2 on Nanostructured

Pd@Nb2O5 ............................................................................................................................... 69

4 Mechanistic Insights into the Photocatalytic Reduction of CO2 on Nanostructured

Pd@Nb2O5 ............................................................................................................................... 69

viii

4.1 Experimental ..................................................................................................................... 69

4.1.1 Physical Characterization ...................................................................................... 69

4.1.2 Theoretical Methods ............................................................................................. 70

4.2 Investigating the RWGS Reaction Pathway over Pd@Nb2O5 .......................................... 72

4.3 Investigating the Size Effect of Pd Nanocrystals on the CO2 to CO versus CH4

Selectivity and the Reaction Mechanisms. ....................................................................... 78

Chapter 5 Lattice Strain Engineering of Photocatalyst for Highly-Efficient Gas-Phase CO2

conversion ................................................................................................................................ 89

5 Lattice Strain Engineering of Photocatalyst for Highly-Efficient Gas-Phase CO2

Conversion ............................................................................................................................... 89

5.1 Abstract ............................................................................................................................. 89

5.2 Introduction ....................................................................................................................... 90

5.3 Synthesis and Characterization of Lattice Strained In2O3-x(OH)y grown on Nb2O5

nanorods. ........................................................................................................................... 91

5.4 Photocatalytic Performance of Lattice Strained In2O3-x(OH)y grown on Nb2O5

nanorods. ......................................................................................................................... 101

5.5 Proposed Mechanism ...................................................................................................... 102

5.6 Conclusion ...................................................................................................................... 103

Chapter 6 Conclusion and Future Work ..................................................................................... 104

6 Conclusion and Future Work ................................................................................................. 104

6.1 Conclusion ...................................................................................................................... 104

6.2 Future Work .................................................................................................................... 105

References ................................................................................................................................... 106

ix

List of Tables

Chapter 2

Table 2.1. Estimation of reaction temperatures over Pd@Nb2O5 under irradiation from the 300 W

Xe lamp using different cut-off filters for batch reactions A through F and without any filter at

different light intensities for batch reactions I through V. ............................................................ 36

Table 2.2. Summary of the Raman band frequency for the Nb=O stretching vibrations of

Pd@Nb2O5 and Nb2O5 at different power levels using a 633 nm laser. Temperature estimation

was performed using the slope of the dependence of the Raman band shift on temperature in

prior work.92 .................................................................................................................................. 39

Table 2.3. Estimation of the temperature of Pd@Nb2O5 and Nb2O5 under the power level of 24

mW using a laser pump wavelength of 785 nm. The signals of the Raman Stokes and anti-Stokes

scattering at around 640 cm-1 are obtained from the deconvolution of the spectra by LabSpec

software, as shown in Figure 17. The estimation of temperatures is calculated using the ratio of

Stokes signals to anti-Stokes signals and the band position of the Raman modes. ...................... 43

Chapter 3

Table 3.1. Elemental analysis of different Pd loading samples .................................................... 56

Table 3.2. Parameter values obtained from Pd K-edge EXAFS fitting. Values marked with an

asterisk indicate that these parameters were correlated during EXAFS fitting. ........................... 61

Chapter 4

Table 4.1. The surface areas for the different Pd loading samples ............................................... 75

Table 4.2. CO chemisorption, the amount of exposed Pd surface atoms, apparent metal

dispersion, CO production rates, CH4 production rates, Nb2O5 active sites, and turnover

frequency (TOF) for 3% Pd@Nb2O5 and 10% Pd@Nb2O5 under 4.2 W cm-2 in light-batch

system and for 10% Pd@Nb2O5 under 2.1 W cm-2 in light-flow system ..................................... 79

x

Table 4.3. The energy conversion efficiency in light-batch system over 0.1% Pd@Nb2O5 under

4.2 W cm-2 and in light-flow system over 10% Pd@Nb2O5 under 2.1 W cm-2 ............................ 80

Table 4.4. Locations of surface sites used in DFT calculations .................................................... 85

Table 4.5. Adsorption energies of relevant intermediates on different reactive sites.

Corresponding site locations are shown in in Table 4. Values marked in bold denote the

adsorption energy of the most stable intermediate. Note that for the *CO, *CHO, and *CH

intermediates on Pd(111), only the fcc site has been calculated since previous studies have shown

fcc to be the most stable adsorption site in all three of these cases. 128, 129 ................................... 86

Chapter 5

Table 5.1. PXRD and ICP-AES results of In2O3-x(OH)y and In2O3-x(OH)y/Nb2O5 hybrids ......... 93

Table 5.2. BET surface area of different samples ....................................................................... 100

xi

List of Figures

Chapter 1

Figure 1.1 Methanation of Carbon Dioxide. a) Visualizing strong metal support interaction,

SMSI, states by in situ scanning transmission electron microscope (STEM) images of sintered

6% Rh/TiO2 after treatment in 5% H2 and 95% N2 at 550 C for 10 min. b) Visualizing

adsorbate-mediated SMSI (A-SMSI) states by in situ STEM images of sintered 6% Rh/TiO2 after

treatment in 20CO2:2H2 at 250C for 3 h. c) Infrared analysis of selectivity switch for thermally

reduced (red), 20CO2:2H2-treated (blue) and re-reduced (black) catalysts. Reprinted with

permission.3 Copyright 2016, Nature Publishing Group. d) Illustration of charge transfer cross

the surface facet homojunction of CeO2. Reprinted with permission.5 Copyright 2015, American

Chemical Society.e) Isolation of Cu atoms in Pd nanoparticle lattice for highly selective

conversion of CO2 to CH4. 6 f) Average production rates of CH4 and CO in photocatalytic CO2

reduction with H2O by bare TiO2, Pd1Cu1TiO2, Pd3Cu1TiO2, Pd5Cu1TiO2, Pd7Cu1TiO2,

Pd9Cu1TiO2,Pd11Cu1TiO2, and PdTiO2 hybrid structures under UV-light (

xii

Academy of Sciences. f) Nanostructured indium oxide coated silicon nanowire arrays as a hybrid

photothermal-photochemical approach to solar fuels. 26 g) The absorption spectrum of

In2O3x(OH)y nanoparticles and the photon utilization of the solar irradiance for

In2O3x(OH)y/SiNW hybrid materials. h) 13CO production rates of evenly coated

In2O3x(OH)y/SiNW, bilayer In2O3x(OH)y/SiNW, and In2O3x(OH)y/glass in the dark and under

illumination, with or without external heating. Reprinted with permission.26 Copyright 2016,

American Chemical Society. .......................................................................................................... 6

Figure 1.3. Reactor design. a) Schematic of the solar reactor configuration for splitting CO2 into

separate streams of CO2 and O2 via a 2-step thermochemical redox cycle. b) Photographs of the

solar reactor, showing the front face of the solar reactor with the windowed aperture and its

interior containing the octagonal reticulated porous ceramics (RPC) structure lined with alumina

thermal insulation.28 Copyright 2017, by The Royal Society of Chemistry, under the Creative

Commons Attribution-NonCommercial 3.0 License.c) Photograph of a photoreactor in operation

Reprinted with permission.9 Copyright 2014, American Chemical Society.d-e) Setup for

photothermal conversion of CO2 in d) batch system and e) flow system. Reprinted with

permission. 7, 29 Copyright 2014, Wiley. Copyright 2016, Wiley. ................................................. 8

Figure 1.4. Methanol synthesis from CO2. a) Theoretical activity volcano plot for CO2

hydrogenation to methanol. b) Zn 2p3/2 XPS binding energies measured after performing the

hydrogenation of CO2 on the Zn/Cu(111) catalyst. Reprinted with permission.31 Copyright 2014,

Nature Publishing Group. c) Potential energy diagram for the hydrogenation of CO2(g) to

CH3OH(g) on ZnO/Cu(111) via the RWGS + CO-hydroxycarbonyl and formate pathways. (Inset)

Structures of *HCOO on ZnO/Cu(111). Cu, brown; Zn, blue; O, red; H, white; C, gray. d)

Coverage of surface reaction intermediates on ZnCu(211) and ZnO/Cu(111) under reaction

conditions. Reprinted with permission.33 Copyright 2017, American Association for the

Advancement of Science. e) Ex situ MAS-NMR 1H-13C HETCOR spectra of Cu/ZrO2 reacted

with H2/13CO2 (3:1) at 230C for 12 h at 5 bars. f) Reaction scheme derived from the

spectroscopic measurements. Reprinted with permission.34 Copyright 2017, Wiley. .................. 10

Figure 1.5. FT synthesis. a) CO conversion and product distribution at different H2/CO ratios in

syngas over a catalyst with a mass ratio of ZnCrOx/mesoporous SAPO zeolite (MSAPO) = 1.4 at

a space velocity of 4800ml/hgcat b) A stability test of a composite catalyst with

xiii

ZnCrOx/MSAPO ratio = 0.9 at 6828 ml/hgcat and H2/CO of 2.5. Reprinted with permission.35

Copyright 2016, American Association for the Advancement of Science. c) The

CeO2Pt@mSiO2Co tandem catalyst. d) CO2 conversion and hydrocarbon distribution at

different H2/CO2 ratios over the tandem catalyst at 250 C. Reprinted with permission.36

Copyright 2017, American Chemical Society. e) Reaction scheme for CO2 hydrogenation to

gasoline-range hydrocarbons.37 Copyright 2017, Nature Publishing Group, under the Creative

Commons Attribution 4.0 International License .......................................................................... 12

Chapter 2

Figure 2.1. a) Schematic illustration of the photothermal-reactor system used for gas-phase

catalysis testing. b) Film fabrication process. c) The spectral output of the 300W xenon lamp. . 21

Figure 2.2. Synthesis, morphology and structure of Nb3O7(OH) nanorods and Pd@Nb2O5

nanocrystal-nanorod samples. a) Scheme of the synthesis of Nb3O7(OH) and Pd@Nb2O5 nanorod

samples. b) TEM images of Nb3O7(OH) nanorod sample. c) TEM images of Pd@Nb2O5

nanocrystal-nanorod sample. ........................................................................................................ 22

Figure 2.3. a) HAADF image of the Nb3O7(OH) nanorods at low magnification. b) HAADF

image taken from a nanorod in its [001] direction. The preferred growth direction (longitudinal)

of the nanorod is [010]. c) FFT of (b) showing (200), (110) and (020) spots. d) Atomic model

indicated by the red rectangle in Figure 3b, where the dotted blue rectangle indicates a

Nb3O7(OH) unit cell. ..................................................................................................................... 23

Figure 2.4. Powder X-ray diffraction patterns of Nb2O5, Nb3O7(OH), 0.5% Pd@Nb2O5 (after pre-

treatment) and 0.5% Pd@Nb3O7(OH) (before pre-treatment). ..................................................... 24

Figure 2.5. TEM images of a) Nb3O7(OH) nanorods and b) Nb2O5 nanorods. ............................ 25

Figure 2.6. a) HAADF image of a Pd nanocrystal decorated Nb2O5 nanorod. The cyan arrow

indicates the [010] growth direction of the nanorod, and the dashed yellow rectangle marks the

area where EELS elemental mapping was taken. Scale bar, 5 nm. b) Enlarged image showing

nanocrystalline Pd viewed from its [112] direction. The d-spacing of the (111) and (022) planes

were measured. c) Inverse FFT by masking spots only contributed from Pd, as indicated in the

xiv

inserted figure by the red circles. The inserted figure is the FFT of a), showing an overlap of

contributions from Pd and Nb2O5. d) EEL spectra from the Pd nanocrystal (red), Nb2O5 nanorod

(cyan), and interface between the Pd and Nb2O5 (blue). The Pd M4,5-edge and Pd M3-edge are

visible in the red spectrum; the Nb M3-edge, Nb M2-edge and O K-edge are visible in the cyan

spectrum; a superposition of these edges are exhibited in the blue spectrum. e) HAADF image

taken while simultaneously collecting the EELS map. f) Fitting coefficient of Pd showing the

spatial distribution of the Pd signal. g) Fitting coefficient of Nb2O5 showing the spatial

distribution of the Nb2O5 signal. ................................................................................................... 27

Figure 2.7. High resolution XPS spectra of the a) Nb3d, b) O1s. c) and d) Pd 3d regions taken on

Nb2O5, Pd@Nb2O5, 0.5% Pd@Nb2O5 and 1% Pd@Nb2O5, respectively. ................................... 28

Figure 2.8. a) Diffuse reflectance spectra of Nb3O7(OH), Nb2O5 and Pd@Nb2O5 films dispersed

on a borosilicate filter. b) The band energy diagram of Nb2O5 in comparison with the work

function of Pd. ............................................................................................................................... 29

Figure 2.9. Estimation of the electronic band gap of Nb2O5. By using a modified Kubelka-Munk

function, (F(R)*h)n is plotted as a function of photon energy for Nb2O5 where F(R) = (1-R)2/2R.

Here R is the diffuse reflectance of the films loaded onto the borosilicate sample supports and n

was set to 2 for Nb2O5. The linear portion of the plot was extrapolated and its intercept with the

abscissa is considered to be the band-gap. .................................................................................... 29

Figure 2.10. XPS Spectra of Nb2O5 a) secondary electron cut-off region. b) Valence band region

(3.3 eV is relative to the work function.) ...................................................................................... 30

Figure 2.11. Mass spectrometry of photo-thermally generated 13CO from 13CO2. The 28 AMU

mass fragment peak at approximately 1.32 min corresponds to N2 and the 29 AMU mass

fragment peak at approximately 1.345 min corresponds to 13CO. The fact that there is no peak in

the vicinity of 1.345 min retention time for the 28 AMU curve shows that there is no 12CO in the

products that could have been generated from sources of adventitious 12C. ................................ 32

Figure 2.12. Photothermal catalytic performance of the nanostructured Pd@Nb2O5 samples. a)

CO production rates over Pd@Nb2O5 in the dark at room temperature (RT), under irradiation

from a 300 W Xe lamp, and in the dark at a reaction temperature of 160 C. b) Spectral

irradiance incident onto the Pd@Nb2O5 catalyst with different cut-off filters for batch reaction

xv

tests A through F. c) Spectral irradiance incident onto the Pd@Nb2O5 catalyst for batch reactions

I through V. d) The RWGS reaction rates plotted as a function of absorbed power for the series

of batch reactions A - F (red line) and I V (black line). ............................................................. 33

Figure 2.13. TEM image of Pd@Nb2O5 after gas-phase catalytic testing. ................................... 35

Figure 2.14. a) RWGS reaction rate measured in the dark as a function of reaction

temperature b) The effective reaction temperature as a function of radiant power absorbed

by the Pd@Nb2O5 catalyst. The equation used to estimate Te is provided in Figure 5b. .. 37

Figure 2.15. Dependence of the Raman frequency from 100 cm-1 to 1500 cm-1 for Nb=O

stretching vibrations of a) Pd@Nb2O5 and b) Nb2O5 nanorods at different power levels. ........... 40

Figure 2.16. Dependence of the Raman frequency for Nb=O stretching vibrations of a)

Pd@Nb2O5 and b) Nb2O5 nanorods at different power levels. c) Estimated temperatures for

Pd@Nb2O5 and Nb2O5 at different power levels. ......................................................................... 41

Figure 2.17. Selected wavelength Stokes and anti-Stokes Raman spectra between 500 cm 1 and

900 cm1 with representative deconvolution of the spectra analyzed by LabSpec software for a)

Pd@Nb2O5 Stokes spectra; b) Pd@Nb2O5 anti-Stokes spectra; c) Nb2O5 Stokes spectra and d)

Nb2O5 anti-Stokes spectra. ............................................................................................................ 42

Figure 2.18. Photothermal catalysis by nanostructured Pd@Nb2O5 (reduced) sample. a) Without

exposure to air, the production rates of CO over Pd@Nb2O5 kept monotonically increasing from

the 1st run to the 4th run. When the sample was exposed to air for 24 h, the production rate of the

5th run dropped to that of the 1st. b) Diffuse reflectance spectra and appearance of (i) pristine

Pd@Nb2O5 and (ii) reduced Pd@Nb2O5 nanostructured films. c) Schematic illustration of in-situ

generation of reduced surface niobium sites and/or oxygen vacancies during photothermal

reaction. ......................................................................................................................................... 45

Figure 2.19. EPR spectra of Nb2O5 before (black) and after (red) being illuminated under a 300

W Xe lamp for 2 hours, recorded at 20 K. .................................................................................... 46

Chapter 3

xvi

Figure 3.1. Schematic illustration of the growth of Pd nanocrystals on Nb2O5 nanorods a, low

loading of Pd precursor and b, high loading of Pd precursor. ...................................................... 55

Figure 3.2. TEM bright field images of a) 0.1%Pd@Nb2O5, b) 0.5% Pd@ Nb2O5, c) 1% Pd@

Nb2O5, d) 2% Pd@ Nb2O5, e) 3% Pd@ Nb2O5, f) 5% Pd@ Nb2O55, g) 10% Pd@ Nb2O5 and h)

15% Pd@ Nb2O5. Scale bar: 20 nm. ............................................................................................. 56

Figure 3.3. Morphology and structural analysis of Pd@Nb2O5 samples. Bright field TEM

images of a, 0.1% Pd@Nb2O5 sample; b, 3% Pd@Nb2O5 sample; c, 10% Pd@Nb2O5 sample;

Scale bar: 20 nm. HRTEM image of d, 0.1% Pd@Nb2O5 sample; e, 3% Pd@Nb2O5 sample; f,

10% Pd@Nb2O5 sample; Scale bar: 2 nm. g, Pd K-edge XANES spectra from the supported Pd

nanocrystals. h, Pd K-edge FT-EXAFS spectra from the supported Pd nanocrystals. ................. 57

Figure 3.4. Pd particle size distribution determined for a) 10% Pd@Nb2O5, b) 3% Pd@Nb2O5,

and c) 0.1% Pd@Nb2O5 from TEM images. ................................................................................ 58

Figure 3.5. a) HRTEM images of Nb2O5 nanorod viewed from its [[ 0 4] direction, showing

(010) and (401) planes. b) HRTEM image of 10% Pd-Nb2O5 interface, where the (111) plane of

Pd and (401) plane of Nb2O5 are visible. c-d) HRTEM images of 3% Pd-Nb2O5 interface. For the

3% sample, Pd particles are round with no obvious facets. .......................................................... 59

Figure 3.6. Powder X-ray diffraction patterns of Nb2O5, 0.1% Pd@Nb2O5, 0.5% Pd@Nb2O5, 1%

Pd@Nb2O5, 2% Pd@Nb2O5, 3% Pd@Nb2O5, 5% Pd@Nb2O5, 10% Pd@Nb2O5 and 15%

Pd@Nb2O5..................................................................................................................................... 60

Figure 3.7. Optical and Raman characterization of Pd@Nb2O5 samples. a, Digital photographs

of pristine Nb2O5 and Pd@Nb2O5 samples with different loadings of Pd. b, Diffuse reflectance

spectra of pristine Nb2O5 and different Pd@Nb2O5 samples dispersed on borosilicate filter films.

Nb=O stretching vibrations of c, 0.1% Pd@Nb2O5, d,

0.5% Pd@Nb2O5 and e, 5% Pd@Nb2O5 at different incident light power levels. f, The estimated

temperatures of 0.1% Pd, 0.5% Pd, and 5% Pd loaded onto Nb2O5 nanorods at different incient

light power levels .......................................................................................................................... 62

xvii

Figure 3.8. Fragmentation patterns measured by GC-MS for different products. On the top is CH4

and on the bottom is CO. (For CO,the mass of 13CO is 29; the mass of 12CO is 28.). ................. 64

Figure 3.9. Photothermal catalytic performance of Pd@Nb2O5 samples. a, CO production rates

over Pd@Nb2O5 film samples with different Pd loadings under irradiation from a 300 W Xe

lamp. b, CH4 selectivities over Pd@Nb2O5 film samples with different Pd loadings. c, CH4

selectivities over 10% Pd@Nb2O5 with different sample weights. d, Comparison of TOFPd methane

and TOFPd CO for 3% Pd@Nb2O5 and 10% Pd@Nb2O5 in a batch photoreactor under 4.2 W cm-2,

and for 10% Pd@Nb2O5 in flow photoreactor under 2.1 W cm-2. ................................................ 65

Figure 3.10. CH4 production rates over Pd@Nb2O5 film samples with different Pd loadings under

irradiation from a 300 W Xe lamp. ............................................................................................... 66

Figure 3.11. CO and CH4 production rates over 10% Pd@Nb2O5 under different light intensities.

....................................................................................................................................................... 66

Figure 3.12. CO and CH4 production stability on the presence of light for 3% Pd@Nb2O5 sample,

where each test lasted for 30 minutes. .......................................................................................... 67

Chapter 4

Figure 4.1. Atomic models used to simulate the Pd(111) and Pd(211) surfaces and Pd

nanoparticle in DFT calculations. ................................................................................................. 72

Figure 4.2. Possible reaction pathway for the reverse water gas shift reaction (RWGS) on the

Pd@Nb2O5 hybrid catalyst............................................................................................................ 73

Figure 4.3. CO production rate from CO2 in the presence and absence of H2 over Pd@Nb2O5. . 74

Figure 4.4. The dependence of the CO2 adsorption capacity on the Pd loading in hybrids. The

CO2 adsorption capacity is normalized to the surface area of each sample. ................................. 76

Figure 4.5. Thermogravimetric analysis (TGA) and differential thermogravimetric analysis of the

fresh catalyst and used catalyst after catalytic reaction tests of 40 h. ........................................... 77

Figure 4.6. High resolution Pd 3d XPS spectra ............................................................................ 81

xviii

Figure 4.7. XPS, UPS and in situ DRIFT analysis. a, Pd 3d5/2 binding energy position as a

function of Pd loading. b, The Fermi-level position of pristine Nb2O5 in comparison with that of

0.1% Pd@Nb2O5 and 10% Pd@Nb2O5 measured by UPS (Fermi-levels are referenced to the

vacuum level, i.e., the materials work functions). c,d, In situ DRIFT spectra of 0.1% Pd@Nb2O5

under a gas mixture of H2 and CO2 (1:1 ratio) at 240 C. e,f, In situ DRIFT spectra of 10%

Pd@Nb2O5 under a gas mixture of H2 and CO2 (1:1 ratio) at 240 C. .......................................... 82

Figure 4.8. Density functional theory analysis and reaction mechanism on low Pd loading and

high Pd loading samples. a, Calculated free energy diagram for *CO intermediate reduction to

CH4 on Pd(111), Pd(211) and Pd55. b, Pd(111), stepped Pd(211), and Pd55 computational

models for representing surface terrace sites, edge and corner sites. c, Calculated activation

barrier for the rate determine step on Pd(111), Pd(211) and Pd55. d, Schematic illustration

depicts the weak CO bonding of small Pd nanocrystals in low Pd loading sample. .................... 84

Figure 4.9. The weight fraction of adsorption sites (step, corner, terrace and bulk sites) on ideal

octahedral Pd NPs as a function of the cluster diameter. .............................................................. 87

Figure 4.10. TEM images of 1%Pd@Nb2O5 a) before gas-phase catalytic testing, b) after 10

consecutive gas-phase catalytic tests, where each test lasted for 30 minutes. Scale bar: 100 nm. 88

Chapter 5

Figure 5.1. TEM image of Nb2O5 nanorods. ................................................................................ 91

Figure 5.2. Schematic illustration of the preparation of In2O3-x(OH)y via slowly decomposition of

urea in the H2O/ethanol mixture solvent and how the lattice strain induced to In2O3-x(OH)y. (A)

The growth of lattice strained In2O3-x(OH)y on the surface of Nb2O5 and synthesis of pristine

In2O3-x(OH)y as reference. (B) PXRD patterns of pristine In2O3-x(OH)y, Nb2O5 and their hybrids.

....................................................................................................................................................... 92

Figure 5.3. Thermal analysis of In(OH)3, S3 and Nb2O5. (A) TGA and (B) DTA curves ........... 93

Figure 5.4.. PXRD patterns of pristine In(OH)3, Nb2O5 and their hybrids. These samples are

denoted as S1-S4, labeled with gradually increasing In(OH)3 contents. It can be seen clearly

that there is no XRD peaks shifts of In(OH)3 from S1 to S4 samples. ....................................... 94

xix

Figure 5.5. Representative enlarged PXRD patterns of lattice compressed In2O3-x(OH)y along (A)

(211), (B) (222) and (C) (411) facets, respectively. ..................................................................... 95

Figure 5.6. XPS spectrum of pristine In2O3-x(OH)y, physical mixture and S3 sample. (A) Full

spectrum of pristine In2O3-x(OH)y, S3 sample and physical mixture. (B) C1S (284.6 eV)

spectrum which serve as reference for calibration. (C) Enlarged In3d5/2 and In3d3/2 peaks of

pristine In2O3- x(OH)y, physical mixture and S3 sample. (D) Fitting curves of the In3d5/2 and

In3d3/2 peaks of pristine In2O3-x(OH)y, physical mixture and S3 sample. ..................................... 95

Figure 5.7. Structural and component characterizations of In2O3-x(OH)y/Nb2O5 hybrid (S3

sample). a-c, TEM and HRTEM images of S3. D-E, HRTEM image of pristine and compressed

In2O3-x(OH)y with (222) lattice fringe are aligned for the direct evidence of lattice constraint,

scale bar 0.5 nm. (i) TEM image and corresponding elemental (Nb, In, O) mappings. ............... 97

Figure 5.8. EDX Spectrum. The energy dispersive X-ray (EDX) spectrum of the ultrafine In2O3-

x(OH)y nanodots with a diameter of ~5 nm uniformly grown on the surface of Nb2O5 nanorods.

....................................................................................................................................................... 98

Figure 5.9. TEM image of Nb2O5 nanorods and In2O3-x(OH)y composite prepared with ammonia

method. Clearly, the Nb2O5 nanorods and In2O3-x(OH)y species are separated existing in the

composite. ..................................................................................................................................... 98

Figure 5.10. PXRD pattern of In2O3-x(OH)y/Nb2O5 composite prepared with ammonia method.

The 2 angle of the peaks of (211), (222) located at 25.0o and 31.6o, indicating this method could

not lead to lattice compressed In2O3-x(OH)y in In2O3-x(OH)y/Nb2O5 composite. ......................... 99

Figure 5.11. Pore size distribution of In2O3-x(OH)y, Nb2O5 and their hybrids. ............................. 99

Figure 5.12. N2 absorptiondesorption isotherms of In2O3-x(OH)y, Nb2O5 and their hybrids. ... 100

Figure 5.13. Photocatalytic performance of as-synthesized samples. (A) S3 deposited on the

borosilicate filter film. (B) Photocatalytic performance of various photocatalysts. (C) Long-term

stability measurements of S3 sample. ......................................................................................... 101

Figure 5.14. Mass spectrometry of S3 sample generated 13CO from 13CO2. The 28 AMU mass

fragment peak at approximately 1.32 min corresponds to N2 and the 29 AMU mass fragment

xx

peak at approximately 1.345 min corresponds to 13CO. The fact that there is no peak in the

vicinity of 1.345 min retention time for the 28 AMU curve shows that there is no 12CO in the

products that could have been generated from sources of adventitious 12C. .............................. 102

xxi

Copyright Acknowledgements

The thesis has been compiled using excerpts with permissions from the following published and

unpublished work:

(a) J. Jia, N. P. Kherani, D. D. Perovic, G. A. Ozin, Recent developments in gaseous

carbon dioxide conversion powered by light and heat. In preparation .

(b) J. Jia, H. Wang, Z. Lu, P. G. O'Brien, M. Ghoussoub, P. Duchesne, Z. Zheng, P. Li, Q.

Qiao, L. Wang, A. Gu, A. A. Jelle, Y. Dong, Q. Wang, K. K. Ghuman, T. Wood, C. Qian,

Y. Shao, M. Ye, Y. Zhu, Z. H. Lu, P. Zhang, A. S. Helmy, C. V. Singh, N. P. Kherani, D.

D. Perovic, and G. A. Ozin. Photothermal Catalyst Engineering: Hydrogenation of

Gaseous CO2 with High Activity and Tailored Selectivity Adv. Sci. 2017, 1700252.

Copyright 2017, published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

(c) J. Jia, P. G. O'Brien, L. He, Q. Qiao, T. Fei, L. M. Reyes, T. E. Burrow, Y. Dong, K.

Liao, M. Varela, S. J. Pennycook, M. Hmadeh, A. S. Helmy, N. P. Kherani, D. D.

Perovic, G. A. Ozin, Visible and Near Infrared PhotothermalCatalysed Hydrogenation

of Gaseous Carbon Dioxide over Nanostructured Pd@Nb2O5 Adv. Sci. 2016, 3,

1600189-n/a. Copyright 2016, published by WILEY-VCH Verlag GmbH & Co. KGaA,

Weinheim

(d) M. Ye, J. Jia, Z. Wu, C. Qian, R. Chen, P. G. O'Brien, W. Sun, Y. Dong, G. A. Ozin,

Synthesis of Black TiOx Nanoparticles by Mg Reduction of TiO2 Nanocrystals and their

Application for Solar Water Evaporation Adv. Energy Mater. 2017, 7, 1601811-n/a.

Copyright 2017, published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

(e) H., Wang*, J. Jia*(equal contribution), G. Casillas, M.Wei, M. Liu, Y. Dong, L. Wang,

M. Ghoussoub, S. Ye, Z. Zheng, Q. Wang, N. P. Kherani, D. D. Perovic, A. Walsh, G. A

Ozin. Lattice Strain Engineering of Photocatalyst for Highly-efficient CO2 Conversion.

Submitted. .

(f) H. Wang, J. Jia, P. Song, Q. Wang, D. Li, S. Min, C. Qian, L. Wang, Y. F. Li, C. Ma, T.

Wu, J. Yuan, M. Antonietti, G. A. Ozin. Efficient Electrocatalytic Reduction of CO2 by

xxii

Nitrogen-Doped Nanoporous Carbon/Carbon Nanotube Membranes: A Step Towards the

Electrochemical CO2 Refinery. Angew. Chem. Int. Ed. 2017. Copyright 2017,

published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

1

Chapter 1 Introduction and Background

This chapter has been compiled using excerpts from previously published work and unpublised

work:

(a) Recent developments in gaseous carbon dioxide conversion powered by light and heat.

In preparation.

1 Introduction and Background

1.1 Introduction

Global warming, caused by greenhouse gases, has a significant effect on climate and

environment. For decades, the concentration of CO2, mainly originating from the combustion of

fossil fuels, continuously grows in the atmosphere. Another serious issue associated with fossil

fuels is their unsustainability.

Because of climate change and the energy crisis, researchers have proposed using CO2 as a

feedstock to store the energy from renewable energy sources in the form of chemical energy. By

converting CO2 into chemicals and fuels, the energy from renewable sources can be stored in

chemical bonds. Upon combustion, the energy of these renewable fuels can power our industries,

transportation systems, buildings and homes around the world. CO2 will be recycled in this

sustainable carbon-neutral carbon-cycle. If the existing infrastructure in the chemical and

petrochemical industry can be developed into CO2 refineries, together with other approaches,

they will play a vital role to facilitate greenhouse gas reduction and ameliorate climate change.

It has been reported that CO2 can be converted into a variety of value-added chemicals and fuels,

including carbon monoxide, methane, ethylene, methanol, ethanol, urea, dimethyl ether and other

higher hydrocarbons. Among different CO2 conversion approaches, gas-phase heterogeneous

catalytic reduction of CO2, driven by renewable forms of energy, is considered as a promising

strategy, benefiting from its potential to be easily integrated into existing chemical and

petrochemical industry infrastructure.1, 2 However, there are still remaining challenges to be

overcome. Currently, there is an urgent need to discover and design highly active, selective and

2

stable heterogeneous catalysts for the production of fuels powered by light and/or heat from the

largest renewable energy resource, the sun.

In this chapter, we present a summary of advances specifically for the heterogeneous catalytic

conversion of gaseous carbon dioxide into different classes of chemicals and fuels.

1.2 Methanation

Research on the thermal reduction of CO2-to-CH4 has yielded experimental and theoretical

mechanistic insights into the activity and selectivity of the reaction, with a focus on different

reaction pathways and the formation, binding and chemistry of intermediate species. In the case

of oxide-supported Rh catalysts, it has been demonstrated that the thermal treatment of TiO2

supported Rh nanoparticles in a H2 atmosphere at 500 C creates a strong metal support

interaction (SMSI) encapsulated state, as shown in Figure 1.1a, which facilitates the selective

production of CH4. 3 An adsorbate-mediated SMSI (A-SMSI) overlayer can be formed in CO2

H2 environments at temperatures of 150300 C, as shown in Figure 1.1b, which causes a

selectivity switch from CH4 to CO production. Probed by in situ spectroscopy, the high coverage

of adsorbates (HCOx) on the A-SMSI overlayer, illustrated in Figure 1.1c, enables oxygen

vacancy formation and the selectivity switch. Theoretically, it has been demonstrated by Density

Functional Theory (DFT) calculations that the difference in selectivity of metal metal oxide

catalysts is associated with different reaction pathways and the binding strength of reaction

intermediates. In the case of PtCo bimetallics supported on reducible metal oxides for the CO2

hydrogenation, by changing the metal oxide support, selectivity could be modified as a result of

different modes of binding of C and O-bound and O-bound species at the metal metal oxide

interface.4

For photochemical reduction of CO2 to CH4, several distinct materials chemistry strategies have

been devised for increasing the yield and lifetime of charge carriers or improving adsorption and

activation of CO2, to tailor the reactivity of heterogeneous catalysts. The effect of creating

homojunctions between different facets of CeO2 has been explored.5 Intentionally formed

homojunctions are found to enable the efficient separation and fast transfer of photoinduced

charge carriers, as a means for optimizing the photocatalytic performance, as illustrated in

Figure 1.1d. In the case of Cu atoms in a Pd catalyst, the observed correlation between adjacent

Cu-Pd sites and reduction of CO2 to CH4, as illustrated in Figure 1.1e-f, implies they are

3

responsible for enhanced CO2 adsorption.6 Based on synchrotron radiation characterizations and

theoretical simulations, the Cu atoms in the Pd lattice can play dual roles: (1) providing the

paired CuPd sites for enhanced CO2 adsorption and suppressed H2 evolution; and (2) elevation

of the Cu d-band in Cu-Pd, considered to be the reason for the improved activation of CO2.

Figure 1.1 Methanation of Carbon Dioxide. a) Visualizing strong metal support interaction,

SMSI, states by in situ scanning transmission electron microscope (STEM) images of sintered

6% Rh/TiO2 after treatment in 5% H2 and 95% N2 at 550 C for 10 min. b) Visualizing

adsorbate-mediated SMSI (A-SMSI) states by in situ STEM images of sintered 6% Rh/TiO2 after

treatment in 20CO2:2H2 at 250C for 3 h. c) Infrared analysis of selectivity switch for thermally

reduced (red), 20CO2:2H2-treated (blue) and re-reduced (black) catalysts. Reprinted with

permission.3 Copyright 2016, Nature Publishing Group. d) Illustration of charge transfer cross

4

the surface facet homojunction of CeO2. Reprinted with permission.5 Copyright 2015, American

Chemical Society.e) Isolation of Cu atoms in Pd nanoparticle lattice for highly selective

conversion of CO2 to CH4. 6 f) Average production rates of CH4 and CO in photocatalytic CO2

reduction with H2O by bare TiO2, Pd1Cu1TiO2, Pd3Cu1TiO2, Pd5Cu1TiO2, Pd7Cu1TiO2,

Pd9Cu1TiO2,Pd11Cu1TiO2, and PdTiO2 hybrid structures under UV-light (

5

thermal CO2-to-CO conversion, as illustrated in Figure 1.2a-c.4, 10-15 In situ diffuse reflectance

infrared Fourier transform (DRIFT) spectroscopy has been widely used to investigate details of

the reaction mechanism. By changing the oxide support, the metal-metal oxide heterogeneous

catalysts can be tuned to selectively produce CO, which is related to the binding strength of

intermediate species.4 Alternative catalysts to precious metals, such as molybdenum carbide

(Mo2C), have been demonstrated to be highly active for selective reduction of CO2 to CO.16

Photochemical catalytic conversion of CO2 involves light-induced electron-hole pair generation,

charge carrier seperation and transfer, activation of H2 and CO2 by charge carriers, and CO2-to-

CO conversion. A class of materials capable of photocatalytically reducing CO2 are oxygen

deficient metal oxides. For example, defect laden In2O3-xOHy[O]z was discovered to be an

efficient heterogeneous photocatalyst for the hydrogenation of CO2.17 Detailed experimental and

theoretical studies on In2O3-xOHy[O]z have revealed the function of oxygen vacancies and

hydroxyls, the Lewis acidity and Lewis basicity of surface active sites, and kinetics and

mechanistic insights into the photochemical CO2-to-CO conversion. Following band gap

excitation, oxygen vacancies and hydroxyls, the former located below the conduction band and

the latter above the valence band, serve as traps for photo-generated electrons and holes

respectively, as illustrated in Figure 1.2e.18. This has both the effect of slowing down carrier

relaxation times and enhancing the Lewis acidity and Lewis basicity of the defect centres in the

excited state. These two effects work synergistically and result in the reduction of the activation

energy for the rate determining carbon dioxide hydrogenation step by 20 kJ/mole, in the light

compared to the dark,19 thereby enhancing the rates for the reverse water gas shift reaction.

In2O3-xOHy[O]z also served as an archetype for tailoring the effect of defects on its photocatalytic

properties.18-24 Recent work uncovered the role of defects in affecting electronhole separation at

the atomic level to achieve highly efficient solar CO2 reduction, as illustrated in Figure 1.2d.25

Verified by DFT and ultrafast transient absorption spectroscopy, the presence of zinc vacancies

increased charge density, carrier transport and separation rates. Together with increased

photoabsorption, higher CO2 adsorption capacity, and more pronounced surface hydrophilicity,

the zinc vacancy rich compared to vacancy poor, single-unit-cell thick ZnIn2S4 layers boost the

carbon monoxide formation rate by roughly 3.6 times.

Powered by both the heat and light from the sun, the photothermal strategy is extremely

appealing for the development of highly efficient and stable gas-phase CO2 reduction catalysts.26

6

An example is a hybrid catalyst consisting of a vertically aligned silicon nanowire (SiNW)

support evenly coated by In2O3x(OH)y nanoparticles that utilizes the vast majority of the solar

irradiance to simultaneously produce both the photogenerated charge carriers and heat required

to reduce CO2 to CO at a rate of 22.0 molgcat1h1, illustrated in Figure 1.2f-g. Attributed to

reduced reflection losses and improved light harvesting efficiency within the SiNW array, a

nearly 6-fold enhancement over identical In2O3x(OH)y films on roughened glass substrates has

been observed. The even distribution of the In2O3x(OH)y nanoparticles on the SiNW array also

facilitate heat transfer from the SiNW support to the In2O3x(OH)y nanoparticles. This strategy of

utilizing nanostructured photothermal supports to provide a direct source of heating and light

absorption of photocatalytic materials is demonstrated to be an effective way for improving the

conversion rates for gas-phase reactions that require both light and heat energy.

Figure 1.2 Reverse Water Gas Shift (RWGS) Reaction CO2 + H2 CO + H2O. a)

Heterogeneous catalysis of CO2 reduction on supported metals. b) CO2 conversion on Pd/Al2O3

7

at different loadings and temperatures. Reprinted with permission.11 Copyright 2013, American

Chemical Society. c) DRIFT spectrum obtained from a saturated layer of CO adsorbed at 300 K

on 4% Rh/TiO2. Reprinted with permission.13 Copyright 2015, American Chemical Society.d)

Scheme for the photoreduction CO2 into CO on the VZn-rich one-unit-cell ZIS layers. Reprinted

with permission.25 Copyright 2017, American Chemical Society.e) Schematic illustrating the

different charge carrier relaxation processes in In2O3x(OH)y[O]z.18 Copyright 2016, National

Academy of Sciences. f) Nanostructured indium oxide coated silicon nanowire arrays as a hybrid

photothermal-photochemical approach to solar fuels. 26 g) The absorption spectrum of

In2O3x(OH)y nanoparticles and the photon utilization of the solar irradiance for

In2O3x(OH)y/SiNW hybrid materials. h) 13CO production rates of evenly coated

In2O3x(OH)y/SiNW, bilayer In2O3x(OH)y/SiNW, and In2O3x(OH)y/glass in the dark and under

illumination, with or without external heating. Reprinted with permission.26 Copyright 2016,

American Chemical Society.

The goal to create CO2 refineries will also require the establishment of laboratories that

specialize in the design and development of catalyst reactors and operando analytical

instrumentation, specifically for the purpose of evaluating the efficacy, durability and

reproducibility of catalytic materials able to convert gaseous CO2 into chemicals and fuels. In

this context, a solar thermal reactor has been designed for converting CO2 to CO.27 As shown in

Figure 1.3a, the reactor consists of ceria catalysts, gas-collection gap, thermal insulation layer,

quartz window, gas inlet, and gas outlet. Concentrated solar radiation passes through the quartz

window and is focused on the chemical reaction chamber. Powered by thermal energy from the

concentrated light, the temperature of the reactor can be as high as 1500 C. Porous ceria

catalysts convert CO2 to CO efficiently and selectively at such elevated temperatures. The solar

reactor consists of a 100 mm-inner diameter and a 75 mm-depth cylindrical cavity-receiver,

demonstrating the potential for the development of large scale CO2 conversion demonstrators.

There have been recent advances in the design of photothermal reactors for CO2 conversion as

well. Instead of generating high uniform temperatures throughout the entire catalyst sample,

photothermal catalysts can generate high local temperatures at the nanoscale of the catalyst,

without causing significant temperature increase in the environment. A photothermal reactor

system has been designed with a quartz window using a solar simulator coupled with an AM1.5

8

Figure 1.3. Reactor design. a) Schematic of the solar reactor configuration for splitting CO2 into

separate streams of CO2 and O2 via a 2-step thermochemical redox cycle. b) Photographs of the

solar reactor, showing the front face of the solar reactor with the windowed aperture and its

interior containing the octagonal reticulated porous ceramics (RPC) structure lined with alumina

thermal insulation.28 Copyright 2017, by The Royal Society of Chemistry, under the Creative

Commons Attribution-NonCommercial 3.0 License.c) Photograph of a photoreactor in operation

Reprinted with permission.9 Copyright 2014, American Chemical Society.d-e) Setup for

photothermal conversion of CO2 in d) batch system and e) flow system. Reprinted with

permission. 7, 29 Copyright 2014, Wiley. Copyright 2016, Wiley.

filter as the light source (Figure 3c).9 Two other kinds of photothermal reactors based on batch

and a flow system (Figure 3d,e), have also been developed.7, 30 When the illumination from the

lamp is focused on the metal-metal oxide catalysts, it will absorb the incident radiation and

convert it into heat locally thereby elevating the temperature of these catalysts by a few hundred

degrees celsius. It has been demonstrated that the mechanism of photothermal reactions may be

9

identical with thermal analogues. The difference is the distinct ways of heat generation and

transfer to the catalytically active site. Photothermally induced local heating to drive CO2

heterogeneous catalytic hydrogenations opens new doors in the design of catalysts and reactors.

1.4 Methanol Synthesis

If methanol could be efficiently obtained from captured carbon dioxide, solar-generated

hydrogen, renewable forms of energy and produced economically competitive with methanol

made from non-renewable fossil fuels, then a transition from an unsustainable carbon-positive to

a sustainable carbon-neutral future will be brought closer to reality. Recent advances in synthetic,

characterisation and computational techniques provide new approaches to the rational design of

heterogeneous catalysts for methanol synthesis.

A Ni-Ga catalyst which reduces CO2 to methanol at ambient pressure has been discovered.

Identified through a combination of a descriptor-based analysis of the process and computational

methods, the Ni-Ga catalyst demonstrates the same or better methanol synthesis activity

compared to conventional Cu/ZnO/Al2O3 catalysts, as illustrated in Figure 1.4a. 31 The power of

materials theory and simulation has been demonstrated by insightful mechanistic studies,

identification of active sites, the synergy effect of Cu and ZnO at the interface and catalyst

optimization. 32, 33 By combining X-ray photoemission spectroscopy, DFT calculations, and

kinetic Monte Carlo simulations, ZnCu and ZnO/Cu model catalysts have been identified and

characterized for methanol synthesis, as illustrated in Figure 1.4b-d. 33 Both experimental and

theoretical results indicate that ZnCu undergoes surface oxidation under the reaction conditions

and surface Zn transforms into ZnO with activity comparable to ZnO/Cu with the same Zn

coverage. Additonally, an in-depth mechanistic study of the hydrogenation of CO2 on Cu/ZrO2

has been performed using a combination of kinetics, in situ IR and NMR spectroscopies, isotopic

labeling strategies, and DFT calculations, as illustrated in Figure 1.4e-f. 34 It is confirmed by

NMR spectroscopic investigations that the initially formed formate intermediate species

adsorbed on zirconia converts successively to acetal, methoxy and finally methanol.

10

Figure 1.4. Methanol synthesis from CO2. a) Theoretical activity volcano plot for CO2

hydrogenation to methanol. b) Zn 2p3/2 XPS binding energies measured after performing the

hydrogenation of CO2 on the Zn/Cu(111) catalyst. Reprinted with permission.31 Copyright 2014,

Nature Publishing Group. c) Potential energy diagram for the hydrogenation of CO2(g) to

CH3OH(g) on ZnO/Cu(111) via the RWGS + CO-hydroxycarbonyl and formate pathways. (Inset)

Structures of *HCOO on ZnO/Cu(111). Cu, brown; Zn, blue; O, red; H, white; C, gray. d)

Coverage of surface reaction intermediates on ZnCu(211) and ZnO/Cu(111) under reaction

conditions. Reprinted with permission.33 Copyright 2017, American Association for the

Advancement of Science. e) Ex situ MAS-NMR 1H-13C HETCOR spectra of Cu/ZrO2 reacted

11

with H2/13CO2 (3:1) at 230C for 12 h at 5 bars. f) Reaction scheme derived from the

spectroscopic measurements. Reprinted with permission.34 Copyright 2017, Wiley.

1.5 C2 C11

Recent progress has been made in the hydrogenation of gaseous CO2 to C2 C11 hydrocarbons

via Fischer-Tropsch (FT) synthesis by either an indirect or a direct route. The indirect route is

often performed using different catalysts, reactors and conditions with CO and H2. The CO can

be produced from hydrogenation of CO2 through the reverse water-gas shift (RWGS) reaction.

Since it was developed almost a century ago, Fischer-Tropsch (FT) synthesis has played a vital

role as it is the only effective technology to date for direct conversion of synthesis gas (syngas)

to C2 C11 hydrocarbons, producing synthetic lubricants and synthetic fuels from coal, natural

gas, or biomass. However, the bottomneck of this technology is low hydrocarbon selectivity and

high methane selectivity, as well as carbon coking. The key challenges of selective formation of

hydrocarbon from FT synthesis are the precise control of CC coupling, the degree of

hydrogenation, and the suppression of methane formation. Rising to this challenge, a

bifunctional catalyst has been designed, which separates CO activation and CC coupling onto

two different types of active sites with complementary properties. This ZnCrOx/ mesoporous

SAPO zeolite (MSAPO) catalyst enabled activation of CO2 to CO on the surface of the reduced

oxide (ZnCrOx) and C-C coupling within the acidic pores of zeolites. As a result, C2-C4

selectivity as high as 80% with long-term stability was achieved (Figure 1.5a-b).35

The direct CO2 hydrogenation is usually a two-step process, including the reduction of CO2 to

CO via reverse water-gas shift (RWGS) reaction and subsequent hydrogenation of CO to

hydrocarbons via FT synthesis. A model catalyst that enables the direct route is a well-defined

nanostructured catalyst CeO2Pt@meso SiO2Co, which functions as a tandem catalysis for the

conversion of CO2 to C2C4 hydrocarbons (Figure 1.5c-d). The reaction mechanism involves the

conversion of CO2 and H2 to CO at the Pt/CeO2 interface, and a subsequent production of C2C4

hydrocarbons through a FischerTropsch process at the neighboring Co/meso SiO2 interface. 36

Another interesting recent development concerns studies that are designed to achieve direct

conversion of CO2 to gasoline-range (C5C11) hydrocarbons with selectivity up to 78%, driven

thermochemically under industrial relevant conditions, as illustrated in Figure 1.5e.37 In this

context, detailed experimental studies have revealed that the high selectivity is achieved by a

12

multifunctional catalyst consisting of three types of active sites (Fe3O4, Fe5C2 and Brnsted acid

sites), which cooperatively catalyse a tandem reaction. The appropriate proximity of three types

of active sites is essential to enable the successive and synergetic catalytic conversion of CO2 to

C5C11 hydrocarbons. Remarkably, the multifunctional catalyst exhibits a stability for 1,000 h

on stream.37

Figure 1.5. FT synthesis. a) CO conversion and product distribution at different H2/CO ratios in

syngas over a catalyst with a mass ratio of ZnCrOx/mesoporous SAPO zeolite (MSAPO) = 1.4 at

13

a space velocity of 4800ml/hgcat b) A stability test of a composite catalyst with

ZnCrOx/MSAPO ratio = 0.9 at 6828 ml/hgcat and H2/CO of 2.5. Reprinted with permission.35

Copyright 2016, American Association for the Advancement of Science. c) The

CeO2Pt@mSiO2Co tandem catalyst. d) CO2 conversion and hydrocarbon distribution at

different H2/CO2 ratios over the tandem catalyst at 250 C. Reprinted with permission.36

Copyright 2017, American Chemical Society. e) Reaction scheme for CO2 hydrogenation to

gasoline-range hydrocarbons.37 Copyright 2017, Nature Publishing Group, under the Creative

Commons Attribution 4.0 International License

1.6 The Goals of this Doctoral Thesis Work

Among different approaches that can convert CO2 into value-added products, gas-phase light-

assisted reduction of CO2 is a practical option, since gas phase processes can be easily scaled and

integrated with existing chemical and petrochemical industry infrastructure.1 The activity of

numerous semiconductor photocatalyts towards the light-assisted reduction of CO2 to useful

fuels has been studied.38-49 Photothermal and photochemical approaches in which gas-phase

light-assisted hydrogenation of CO2 to useful fuels over supported metal or metal oxide

nanostructured catalysts have been investigated.

The photothermal approach to the catalytic hydrogenation of CO2 has now achieved conversion

rates as high as mmol to mol gcat1 h1.7-9 This tactic exploits the transformation of light into heat,

generating high local temperatures in nanostructured catalysts that drive CO2 reduction reactions.

The mechanistic details of photothermal catalysis remain to be clarified for different classes of

materials and supports.

For the case of metal nanocrystals, the photothermal effect can arise in different ways that

include non-radiative relaxation of photoexcited surface plasmons, and intra-band or inter-band

non-radiative relaxation of photoexcited charge carriers. The plasmonic effect occurs when

absorbed radiant energy is stored in the collective resonant oscillations of conduction electrons.[6,

10-12] Inter-band and intra-band absorption commonly occur in transition metals when electrons

are photoexcited to higher energetic states within the same electronic band or between different

bands, respectively.[13, 14] By identifying the key factors which underpin the photothermal effect

that promotes catalytic CO2 reduction reactions, it will prove possible to design and make

nanoscale catalysts with compositions and structures, chemical and physical properties that serve

14

to maximize the rate of reduction of gaseous CO2 to chemicals and fuels, using both the heat and

light from sunlight.

The catalyst support may also influence the photothermal effect. Recent efforts towards the

development of photothermal catalysts for gas-phase CO2 reduction have focused on metal

nanocrystals dispersed on insulating metal oxide supports.[9] Photothermal catalysts based on

metal nanocrystals deposited on metal oxide semiconductor supports, especially reducible metal

oxide supports, have received much less attention. It is therefore essential to identify the roles

played by the catalyst supports, since redox activity in semiconductor supports may influence the

catalytic reactions through different mechanisms.

Another important factor is the size and shape of the nanocrystals.[6] However, this has not been

explored in the photothermal catalytic hydrogenation of CO2. We aim at investigating the effect

of size and shape of nanostructured catalysts in the photothermal hydrogenation of CO2. A goal

is to pinpoint the mechanistic details that underpin the size and shape dependence of the activity

and selectivity of the photothermal hydrogenation process, experimentally and theoretically,

using a powerful combination of in situ diffuse reflectance infrared Fourier transform (DRIFT)

spectroscopy, Density Functional Theory (DFT) calculations, X-ray photoemission spectroscopy

(XPS) and ultra-violet photoemission spectroscopy (UPS), thermogravimetric analysis (TGA),

and chemisorption analysis.

Semiconducting metal oxides have also been widely investigated as photocatalysts for CO2

reduction.50-54 We have shown that intentionally engineered defects in non-stoichiometric indium

oxide, In2O3-x(OH)y, are active for the photocatalytic hydrogenation of CO2-to-CO.17, 20 Despite a

significant amount of effort, developing CO2 photocatalysts with high activity, selectivity and

stability and at low cost, remains a great challenge. We aim to explore new strategies to achieve

high-performance yet low-cost photocatalysts for reduction of gaseous CO2 to CO.

15

Chapter 2 Photothermal Catalysed Hydrogenation of Gaseous CO2 over

Nanostructured Pd@Nb2O5

This chapter has been compiled using excerpts from previously published work:

(a) J. Jia, P. G. O'Brien, L. He, Q. Qiao, T. Fei, L. M. Reyes, T. E. Burrow, Y. Dong, K.

Liao, M. Varela, S. J. Pennycook, M. Hmadeh, A. S. Helmy, N. P. Kherani, D. D.

Perovic, G. A. Ozin, Ambient Temperature Visible and Near Infrared

PhotothermalCatalysed Hydrogenation of Gaseous Carbon Dioxide over Nanostructured

Pd@Nb2O5 Adv. Sci., 3, 1600189

(b) M. Hmadeh, V. Hoepfner, E. Larios, K. Liao, J. Jia, M. Jose-Yacaman, and G. A. Ozin,

ChemSusChem, 2014, 7, 2104-2109.

2 Photothermal Catalysed Hydrogenation of Gaseous CO2 over Nanostructured Pd@Nb2O5

2.1 Abstract

The reverse water gas shift (RWGS) reaction driven by Nb2O5 nanorod-supported Pd

nanocrystals without external heating using visible and near infrared (NIR) light is demonstrated.

By measuring the dependence of the RWGS reaction rates on the intensity and spectral power

distribution of filtered light incident onto the nanostructured Pd@Nb2O5 catalyst, it is determined

that the RWGS reaction is activated photothermally. That is the RWGS reaction is initiated by

heat generated from thermalization of charge carriers in the Pd nanocrystals that are excited by

inter-band and intra-band absorption of visible and NIR light. Taking advantage of this

photothermal effect, a visible and NIR responsive Pd@Nb2O5 hybrid catalyst that efficiently

hydrogenates CO2 to CO at an impressive rate as high as 1.8 mmol gcat-1 h1 is developed. The

mechanism of this photothermal reaction involves H2 dissociation on Pd nanocrystals and

subsequent spillover of H to the Nb2O5 nanorods whereupon adsorbed CO2 is hydrogenated to

CO. This work represents a significant enhancement in our understanding of the underlying

mechanism of photothermally driven CO2 reduction and will help guide the way towards the

16

development of highly efficient catalysts that exploit the full solar spectrum to convert gas-phase

CO2 to valuable chemicals and fuels.

2.2 Introduction

Solar fuels are attracting increasing attention owing to their potential of being a viable alternative

to fossil fuels. One appealing route to the production of solar fuels is to use renewable sources of

hydrogen to reduce the greenhouse gas CO2 to value-added chemicals and fuels.55 If realized in

practice at a technologically significant efficiency, economically competitive cost and

industrially relevant scale, this could simultaneously address several significant global

challenges: climate change, renewable energy, and protection of the environment.38, 39, 45, 47, 56-60

For the case of metal nanocrystals, the photothermal effect can arise in different ways that

include Surface Plasmon Resonance (SPR) non-radiative relaxation, and intra-band or inter-band

non-radiative relaxation of photoexcited charge carriers. The SPR effect occurs when absorbed

radiant energy is stored in the collective resonant oscillations of conduction electrons.61, 62 On the

other hand inter-band and intra-band absorption commonly occur in transition metals when

electrons are photoexcited to higher energetic states within the same electronic band or between

different bands, respectively.63, 64 For the case of Pd nanocrystals below 10 nm in size, the SPR is

around 200-250 nm and these nanocrystals exhibit strong ultraviolet (UV) light absorption.65-68

Furthermore, visible and near infrared (NIR) light absorption in these Pd nanocrystals is

dominated by inter-band electron transitions (between the d band and s-p conduction band) and

intra-band transitions (between filled and empty states in the d and s-p bands).61, 62, 64, 69

The catalyst support may also influence the photothermal effect. Recent efforts towards the

development of photothermal catalysts for gas-phase CO2 reduction have focused on metal

nanocrystals dispersed on insulating metal oxide supports.7 Photothermal catalysts based on

metal nanocrystals deposited on metal oxide semiconductor supports, especially reducible metal

oxide supports, have received much less attention. Intuitively, such photoactive reducible

semiconductor supports may further accelerate the catalytic reactions through different

mechanisms. For example, defects in metal oxide semiconductors, such as oxygen vacancies

hydroxyl groups and reduced or co-ordinately unsaturated metal sites, can promote the affinity of

CO2 to the surface of the catalysts, prolong photo-generated charge-carrier lifetimes and enhance

the reactivity of CO2.45, 70-74

17

Herein we report the reverse water gas shift (RWGS) reaction driven with visible and near-

infrared photons at ambient temperature over hybrid catalysts comprised of palladium

nanocrystals dispersed on Nb2O5 nanorods denoted Pd@Nb2O5. Relevant prior work on Pd

nanocrystals includes electrocatalytic,75, 76 photocatalytic,77 and thermocatalytic hydrogenation of

CO2,11, 78-82 as well as CO2 photothermal methanation.

7 The thermally driven systems promote

CO2 hydrogenation at high rates but require high operating temperatures (523 K 900 K) and

high pressures.11, 78-82

In this study we use nanostructured Pd@Nb2O5 as a model hybrid catalyst system to demonstrate

the potential of driving the RWGS reaction using light, without providing external heating to the

catalyst. Specifically, we observe CO2-to-CO hydrogenation rates as high as 5 mmol gcat-1 h1

over these hybrid Pd@Nb2O5 catalysts, normalized to the weight of the Pd nanocrystals, when

subjected to light emitted from a Xe lamp at 25 Suns intensity. This is only about two orders of

magnitude less than a rate of technological significance. Notably, the CO2 hydrogenation

reaction proceeded at a rate of 2 mmol gcat-1h-1 using light in the visible and NIR spectral region,

normalized to the weight of the Pd nanocrystals. We then attempt to pinpoint the key factors

responsible for the photothermal hydrogenation of CO2-to-CO over Pd@Nb2O5 by investigating

the reaction rate dependence on the spectral power distribution and intensity of the incident light.

We also utilize variable temperature Raman spectroscopy measurements to investigate the

photothermal effect on the local temperature of individual and assemblies of the nano-

components that comprise our Pd@Nb2O5 hybrid catalysts. Additionally, we have observed an

enhancement of the CO2 conversion rate that is connected to the in situ generation of Nb4+ and/or

O vacancy defects. This effect more than doubles the RWGS rate.

2.3 Experimental

2.3.1 Synthesis of Nb3O7(OH) Nanorods

Niobium powder (680 mg of Nb, 325 Mesh, Aldrich) was dissolved in a hydrochloric acid

solution (9 mL HCl, SigmaAldrich; 10 mL deionized water) in a pyrex beaker. The aqueous

solution was ultra-sonicated for half an hour and then stirred for 15 min. The solution was

subsequently placed in a Teflon-lined stainless steel autoclave with 100 mL capacity. The

hydrothermal reaction was performed at T=200 C for 24 h.83 After cooling to room temperature,

the white product was collected through a centrifugation process and washed three times with

18

deionized water to remove non-reacted residues. Finally, the sample was dried in a vacuum oven

at T=70 C for 12 h.

2.3.2 Synthesis of Nanostructured Pd@Nb2O5

Nanostructured Pd@Nb2O5 was synthesized via a microwave-assisted reaction. Typically, 50 mg

of Nb3O7(OH) nanorods were suspended in anhydrous ethanol (20 mL) in a pyrex vessel with 40

mL capacity. A stock solution of Na2PdCl4 (39.3 mg, Na2PdCl43H2O, Alfa Aesar) was prepared

in anhydrous ethanol (20 mL), the concentration of which is 1 mg/mL. 0.25 mL of Pd precursor

solution, equivalent to 0.5 wt%, was added to the dispersion of 50 mg Nb3O7(OH) under

sonication. After 30 min of sonication, the vessel was capped and transferred to the microwave

reactor (CEM Discover, 220 W, 220 psi, T=150 C, 20 min). After centrifugation and being

washed with deionized water, the sample was placed into a vacuum oven at T=70 C for 24 h.

The dried Pd@Nb3O7(OH) sample was then treated at T=120 C in air for 24 h to obtain the final

Pd@Nb2O5 nanostructured samples ready for characterization and catalytic testing.

2.3.3 Physical Characterization

Sample morphology was determined using a Hitachi H-7000 transmission electron microscopy

(TEM) at 100 kV. The scanning transmission electron microscopy (STEM), high angle annular

dark field (HAADF) imaging and electron energy-loss spectroscopy (EELS) analysis were

performed using an aberration-corrected Nion UltraSTEM-200, equipped with a cold-field

emission gun. The Nion UltraSTEM-200 was operated at 200 kV when performing HAADF

imaging and EELS mapping. EEL spectra were collected using a Gatan Enfinium Dual EELS

spectrometer. Principal component analysis (PCA) was performed on the EELS data to remove

random noise.84

Powder X-ray diffraction (PXRD) was performed on a Bruker D2-Phaser X-ray diffractometer,

using Cu K radiation at 30 kV.

The surface area of each sample was determined through volumetric nitrogen adsorption at 77 K

using a Quantachrome Autosorb-1-C and calculated using Brunauer-Emmet-Teller (BET) theory.

Samples were outgassed overnight at 80C prior to being analysed.

19

Diffuse reflectance spectra of the samples were measured using a Lambda 1050 UV/VIS/NIR

spectrometer from Perkin Elmer equipped with an integrating sphere with a diameter of 150 mm.

The film samples were prepared by drop-casting aqueous dispersions of Pd@Nb2O5 and Nb2O5

samples onto borosilicate glass microfiber filters.

X-ray photoelectron spectroscopy (XPS) was performed using a Perkin Elmer Phi 5500 ESCA

spectrometer. The spectrometer uses an Al K X-ray source to generate X-rays with an energy of

1486.7 eV. The samples used in XPS analyses were prepared by drop-casting aqueous

dispersions of Pd@Nb2O5 and Nb2O5 samples onto p-doped Si(100) wafers. All measurements

were conducted in an ultrahigh vacuum chamber with a base pressure of 1x10-9 Torr. Data

analyses were performed using the Multipak program and all binding energies were referenced to

the NIST-XPS database and the Handbook of X-ray Photoelectron Spectroscopy.85, 86

CO2 adsorption capacity was determined using thermogravimetric analysis (Discovery TGA, TA

Instruments) with the following procedure. Each sample (approximately 5-10 mg) was first

heated at a rate of 10C/min to a final temperature of 120C, under a nitrogen flow of 100

mL/min. This temperature and nitrogen flow was maintained for 1 hour to remove any adsorbed

moisture. The gas flow was then switched to carbon dioxide, maintaining the same flow rate and

temperature for 2 hours. The observed weight gain between the N2 and CO2 gas streams was

directly used to calculate CO2 capacity, and normalized against the surface area of each sample.

TGA was performed to determine whether or not carbon deposits reside on the used catalyst. 35,

87, 88 These experiments were carried out using a Discovery TGA (TA Instruments). A catalyst

film sample was prepared on frosted glass substrates (Erie Scientific) and tested for a duration of

40 h under testing conditions. After the catalytic tests, the used sample was scraped off the

frosted glass substrate for TGA measurements. Two catalyst samples, one fresh sample and one

used sample, were each heated under flowing CO2, at a rate of 100 mL/min. The samples were

heated at a rate of 1C/min to 700C. The same TGA experiments were also performed using air

since it is known to efficiently remove carbon deposits.

The Raman spectra were obtained using the Horiba Jobin Yvon confocal system; namely a

LabRAM HR 800 spectrometer equipped with a continuous wave HeNe laser (=633 nm, with

r=400 nm) and a CW diode laser (=785 nm, with r=500 nm). The maximum laser power

available using the HeNe laser was 12 mW. The samples were illuminated using a 100x

20

objective. For the Raman spectra, which were obtained as a function of laser power, the exposure

time varied from 1s to 40s, with averaging over 4 spectra depending on the power used. The

exposure time was increased as the incident power was reduced to maintain a signal to

background ratio conducive to accurate peak fitting. Prior to fitting the Raman peaks, the spectra

were identified and isolated from the measured data by subtracting a polynomial curve fit to the

background signal. The stokes/anti-stokes spectra were obtained using the 785 nm laser with a

power of 24 mW. The samples for Raman measurements were prepared by drop-casting aqueous

dispersions of Pd@Nb2O5 and Nb2O5 samples onto borosilicate glass microfiber filters.

Electron paramagnetic resonance (EPR) measurements were performed at 20K using a Bruker

ECS-EMX X-band EPR spectrometer equipped with an ER4119HS cavity. An Oxford ESR 900

helium cryostat controlled by an Oxford ITC503 temperature controller was utilized. Typical

operating parameters were as follows: microwave frequency 9.381799030 Ghz, microwave

power 21.63 mW, modulation amplitude 4 G, sweep width 512 G centred at 3455.45 G, time

constant 0.01 ms, total sweep time 512 s, 4096 points were acquired. The EPR analysis was

applied to nanostructured Nb2O5 samples, which were sealed in the EPR tubes in the glove-box

under a N2 gas atmosphere.

2.3.4 Gas Phase Catalytic Measurements

For gas-phase photothermal catalytic testing, samples were prepared by drop casting

nanostuctured Nb2O5 and Pd@Nb2O5 from an aqueous dispersion onto 1 inch by 1 inch binder

free borosilicate glass microfiber filters (Whatman, GF/F, 0.7 m).

The gas phase photothermal catalytic measurements were conducted in a custom-built 1.8-mL

stainless steel batch reactor with a fused silica view port sealed with a Viton O-ring (see Figure

2.1). The reactor was evacuated using an Alcatel dry pump prior to being purged with H2

(99.9995%) at a flow rate of 20 mL/min. After purging, the reactor was infiltrated with H2 and

CO2 gas at a 1:1 ratio to a total pressure of 27 psi prior to being sealed. The pressure inside the

reactor was monitored during the reaction using an Omega PX309 pressure transducer. Reactors

were irradiated with a 300W Xe lamp for a duration of 3 hours. The spectral output from the

300W Xe lamp was measured using a StellarNet Inc spectrophotometer and the power of the

incident irradiation was measured using a Spectra-Physics Power meter (model 407A) (see

Figure 2.1). The radiant power absorbed by the sample for each run was determined using the

21

equation ; where P and t, are the power emitted by the Xe lamp and the fraction

of light transmitted through the quartz window as well as any filters used in the measurement,

respectively. The light absorbed by the sample was calculated using the equation a = 1-R,

where R is the diffuse reflectance measured at wavelength . The reactors were not connected

to a heating source for the photothermal tests, although the reactor temperature increased to 50C

over the 3 hour duration of the test due to the heat generated from the incident irradiation. For

thermal tests in the dark, the reactor temperature was controlled by an OMEGA temperature

controller combined with a thermocouple placed in the reactor. Product gases were analysed with

a flame ionization detector (FID) insta