SYNTHESIS, CHARACTERIZATION AND PHOTOCATALYTIC STUDIES
OF CdS INCORPORATED TITANOSILICATE FOR HYDROGEN
GENERATION
NG YEW CHOO
UNIVERSITI TEKNOLOGI MALAYSIA
“I hereby declare that I have read this thesis and in my opinion this
thesis is sufficient in terms of scope and quality for the award of
the degree of Master of Science (Chemistry)
Signature
: …………………………………………..
Name of Supervisor : Prof. Dr. Mustaffa Shamsuddin
Date
: 27/08/2009
BAHAGIAN A – Pengesahan Kerjasama*
Adalah disahkan bahawa projek penyelidikan tesis ini telah dilaksanakan melalui
kerjasama antara _______________________ dengan _______________________
Disahkan oleh:
Tandatangan
:
Nama
:
Jawatan
(Cop rasmi)
:
Tarikh :
* Jika penyediaan tesis/projek melibatkan kerjasama.
BAHAGIAN B – Untuk Kegunaan Pejabat Sekolah Pengajian Siswazah
Tesis ini telah diperiksa dan diakui oleh:
Nama dan Alamat Pemeriksa Luar
: Prof. Madya Dr. Mohammad Kassim
Pusat Pengajian Sains Kimia
& Teknologi Makanan
Falkuti Sains and Teknologi
Universiti Kebangsaaan Malaysia
43600 UKM Bangi, Selangor
Nama dan Alamat Pemeriksa Dalam
:
Prof. Dr. Wan Azlee Abu Bakar
Jabatan Kimia, Fakulti Sains
Universiti Teknologi Malaysia
81310 UTM Skudai, Johor
Nama Penyelia lain (jika ada)
:
-
Disahkan oleh Timbalan Pendaftar SPS:
Tandatangan :
Nama
:
Tarikh :
SYNTHESIS, CHARACTERIZATION AND PHOTOCATALYTIC STUDIES OF
CdS INCORPORATED TITANOSILICATE FOR HYDROGEN GENERATION
NG YEW CHOO
A thesis submitted in fulfilment of the
requirements for the award of the degree of
Master of Science (Chemistry)
Faculty of Science
Universiti Teknologi Malaysia
AUGUST 2009
ii
I declare that this thesis entitled “Synthesis, Characterization and
Photocatalytic Studies on CdS Incorporated Titanosilicate for Hydrogen
Generation” is the result of my own research except as cited in the
references. The thesis has not been accepted for any degree and is not
concurrently submitted in candidature of any other degree.
Signature
: .............................
Name
: Ng Yew Choo
Date
: 27/08/2009
iii
Specially dedicated to my family members, my supervisor, co-workers,
friends and anybody always by my side……
iv
ACKNOWLEDGEMENT
First and foremost, I would like to send my gratitude to my project supervisor
Prof. Dr. Mustaffa Shamsuddin for his continuous guidance, encouragement and
supports through out this research. It was a great pleasure for me to conduct this
research under his supervision.
I would like to express thousands of thank you to all the lecturers, laboratory
officers and research officers in the Department of Chemistry, the Solid State
Laboratory, and the Institute Ibnu Sina for all the research facilities and
instrumentation expertise. Grateful acknowledge to the financial support from the
Ministry of Science, Technology and Innovation Malaysia (MOSTI) through Science
Fund (03-01-06-SF0273) and National Science Fellowship.
A special gratitude should goes to all the co-workers of Inorganic Research
Laboratory, faculty of science, especially to Mr. Wong Hon Loong, Ms. Jei Ching
Yih, Ms. Wan Nazihah Wan Ibrahim, Mr. Ridzuan Omar and Ms. Najmah. Not
forgotten other research team members, Ms. Lau Su Chien, Ms. Quek Hsiao Pei, Ms.
Ching Kuan Yong and Mr. Chin Tian Kae for their valuable friendship and also
wonderful suggestions along my work.
Their unselfishness and continuously
sharing of the ideas have inspired me.
Last but not least, I would like to express my deepest appreciation to my
beloved family especially my parent and brothers for their mental support during my
studies.
v
PREFACE
This thesis is the result of my work carried out in the Department of Chemistry;
Universiti Teknologi Malaysia between July 2006 to December 2008 under
supervision of Prof. Dr. Mustaffa Shamsuddin. Part of my work described in this
thesis has been submitted in the following exhibition, publication and presentations:
1.
Yew-Choo Ng, Ching-Yih Jei and Mustaffa Shamsuddin. Titanosilicate ETS10 Derived from Rice Husk Ash. Microporous. Mesoporous Materials 122
(2009) 195-200.
2.
Bronze Medal Award. Nanostructured CdS on ETS-10 for In-Situ Hydrogen
Generation. 10th Industrial Art and Technology Exhibition (INATEX). UTM
skudai. 2008.
3.
Gold Medal Award. Nanostructured CdS on ETS-10 for In-Situ Hydrogen
Generation. 8th Malaysian Technology Expo (MTE). PWTC Kuala Lumpur.
2009.
4.
Ng Yew Choo and Mustaffa Shamsuddin. Physicochemical Studied of CdS
Nanoparticles-Titanosilicate Hybrid. Regional Annual Fundamental Science
Seminar (RAFSS). Poster Presentation. UTM Skudai. 2007.
5.
Ng Yew Choo and Mustaffa Shamsuddin. Analysis of Titanosilicate
Supported CdS Photocatalyst for Water Splitting Reaction. 21th Simposium
Kimia Analysis Malaysia (SKAM-21). Oral Presentation. UMS Sabah. 2008.
6.
Yew-Choo Ng and Mustaffa Shamsuddin. Solid State Morphology and Band
Gap Studied of ETS-10 Supported CdS Nanoparticles. Journal of Iranian
Chemical Society. Manuscript No.: PS-08-240-08.
vi
ABSTRACT
This study relates to a development of heterogeneous solid catalyst,
Engelhard titanosilicate (ETS-10) supported cadmium sulfide (CdS) for water
splitting reaction to generate hydrogen under visible light irradiation.
Highly
crystalline truncated bipyramid shape of ETS-10 was successfully synthesized with
the molar composition of TiO2:3.75SiO2:1.5NaOH:0.54KF:21.25H2O at 220oC for
52 hours. The as-synthesized CdS crystallized at the size of approximately 8 nm in
the cubic structure with the lattice constant a=0.5818 nm and reflection peaks of
(111), (220) and (331) lattice planes. The effect of the synthesis route of CdS, the
effect of percentage loading of CdS and reusability of the catalysts towards water
splitting were also reported. The conduction edge of the photocatalyst was found to
be more negative than H+/H2 redox potential. The highly crystalline phases of
photocatalyst were able to prevent a charge recombination leading to enhancement in
the hydrogen production yield.
The existence of co-catalyst (ETS-10) in the
catalytic system induces reduction of water and increases the efficiency of charge
separation.
The hybrid photocatalyst was found to be more stable and do not
undergo photo-corrosion. CdS derived from in-situ sulphur reduction method (CdSIS) performed better than CdS derived from reverse micelle method (CdS-RM). For
0.1 g of catalyst, the average rate of reaction for the first 5 hours was found to be
68.69 μmol/hr and 49.05 μmol/hr for CdS-IS and CdS-RM respectively. Besides,
the results showed that the higher the percentage of CdS loaded on ETS-10, the
higher amount of hydrogen gas liberated. The reusability of the photocatalysts was
demonstrated in three cycles and the hydrogen gas evolved slightly decreased with
the number of the reusability.
vii
ABSTRAK
Kajian ini berkaitan dengan pembangunan pepejal mangkin heterogen
titanosilikat, kadmium sulfida (CdS) berpenyokong Engelhard titanosilikat (ETS-10)
dalam tindak balas pemecahan air bagi penghasilan gas hidrogen di bawah pancaran
cahaya nampak.
Hablur ETS-10 yang berbentuk dwipiramid terpotong telah
disintesis dengan komposisi molar TiO2:3.75SiO2:1.5NaOH:0.54KF:21.25H2O pada
suhu 220oC selama 52 jam. CdS yang disintesis menghablur pada saiz 8 nm dalam
struktur kiub dengan pemalar kekisi a= 0.5818 nm dan jalur pemantulan pada satah
kekisi (111), (220) and (331).
Kesan cara sintesis CdS, kesan peratusan
pertambahan CdS dan penggunaan semula mangkin terhadap tindak balas
pemecahan air juga dilaporkan. Jalur konduksi foto-mangkin ini didapati bersifat
lebih negatif daripada keupayaan redoks H+/H2.
Fasa berhablur foto-mangkin
mampu mengelakkan gabungan semula cas justeru meningkatkan penghasilan gas
hidrogen.
Kewujudan ko-mangkin (ETS-10) di dalam sistem pemangkinan
menggalakkan penurunan air dan meningkatkan keberkesanan pemisahan cas. Fotomangkin hibrid ini didapati lebih stabil dan tidak mengalami pengaratan foto. CdS
yang disediakan daripada cara penurunan in-situ sulfur (CdS-IS), berfungsi lebih
baik berbanding dengan CdS yang disediakan daripada cara misel berbalik (CdSRM). Dengan 0.1 g mangkin, purata kadar tindak balas bagi 5 jam pertama adalah
68.69 μmol/jam and 49.05 μmol/jam bagi CdS-IS and CdS-RM masing-masing.
Selain itu, permerhatian menunjukkan semakin tinggi peratusan CdS yang disokong
di atas ETS-10, semakin tinggi jumlah gas hidrogen yang dibebaskan. Penggunaan
semula foto-mangkin juga dikaji sebanyak tiga kitaran, di mana pembebasan gas
hidrogen semakin berkurang dengan bilangan penggunaan semula.
viii
TABLE OF CONTENTS
CHAPTER
SUBJECT
PAGE
TITLE PAGE
1
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENT
iv
PREFACE
v
ABSTRACT
vi
ABSTRAK
vii
TABLE OF CONTENTS
viii
LIST OF TABLES
xii
LIST OF FIGURES
xiv
LIST OF ABBREVIATIONS
xx
INTRODUCTION
1.1
Research Background
1
1.1.1 Renewable Resources
3
1.1.2 Hydrogen Economy
4
1.1.3 Hydrogen Production
5
1.1.4 Photocatalysis
7
1.2
Problem Statement
10
1.3
Objectives of Research
11
1.4
Scope of Research
12
1.5
Thesis Outline
12
ix
2
LITERATURE REVIEW
2.1
Solar Energy Distribution
14
2.2
Semiconductor
15
2.3
Thermodynamic Limitation of Water
Photo-splitting
19
2.4
Hole Scavenger Agents
21
2.5
CdS as Photocatalyst
23
2.6
Supports in Photocatalysis
25
2.6.1 Engelhard Titanosilicates (ETS-10)
28
CdS Composites
30
2.7
3
EXPERIMENTAL
3.1 Apparatus and Special Equipments
33
3.2 Synthesis of CdS Nanoparticles by
Reverse Micelle Method
34
3.3 Synthesis of CdS Nanoparticles by In-situ
Sulphur Reduction Method
35
3.4 Synthesis of ETS-10 by Hydrothermal
Method
35
3.5 Modification of ETS-10
36
3.6 CdS Nanoparticles Impregnated on ETS10
37
3.7 Characterization Techniques
3.7.1
X-ray Diffraction (XRD)
3.7.2
Fourier Transform Infrared (FTIR)
Spectroscopy
3.7.3
38
38
Energy Dispersive Spectroscopy
(EDAX)
3.7.6
38
Field Emission Scanning Electron
Microscopy (FESEM)
3.7.5
37
Diffuse Reflectance UV-Vis (DRUV) Spectroscopy
3.7.4
37
Transmission Electron Microscopy
39
x
(TEM)
39
3.8 Experimental Set-up for Photocatalytic
Testing
40
3.9 Catalytic Testing
3.9.1
Hydrogen Gas Calibration GCTCD
3.9.2
44
Photocatalytic Testing by Using
Microreactor
4
43
44
RESULTS AND DISCUSSION
4.1
Characterization of the Photocatalysts
4.1.1
Preparation of CdS
Nanoparticles
4.1.2
52
Physico-chemical Studies of
CdS/ETS-10
4.1.5
48
Physico-chemical Studies of
ETS-10
4.1.4
46
Physico-chemical Studies of
CdS
4.1.3
46
56
Photo-absorption Properties of
CdS
60
4.1.6 Photo-absorption Properties of
ETS-10
61
4.1.7 Photo-absorption Properties of
4.1.8
4.2
CdS/ETS-10
62
Band Gap Studies
63
Photocatalytic Activity
4.2.1
Hydrogen Detection by GCTCD
4.2.2
66
The Mechanism Study of
Water Photo-splitting
4.2.3
66
The Effect of Synthesis Route
68
xi
of CdS
4.2.4
The Effect of CdS Loading on
ETS-10
4.2.5
4.2.6
5
REFERENCES
70
72
The Effect of CdS Loading on
METS-10
77
Reusability Test
80
CONCLUSION AND RECOMMENDATIONS
5.1
Conclusion
82
5.2
Recommendations
83
85
xii
LIST OF TABLES
TABLE NO.
TITLE
PAGE
1.1
Heating value for the selected fuels
5
1.2
Hydrogen production based on the types of fuel
6
2.1
The classification of solid according to their band gap,
carrier density and typical conductivity at room
temperature
2.2
16
The classification of semiconductors according to their
crystal structure
2.3
Several
potential
18
mechanisms
involved
in
a
photocatalytic system and its typical suitable support
material
27
4.1
The elemental analysis for CdS
48
4.2
The elemental analysis for ETS-10
53
4.3
The details of the wavenumber and the type of
vibration present for ETS-10 synthesized from Ludox30 and RHA.
56
xiii
4.4
Hydrogen generation of CdS/ETS-10 for the first 24
hours
4.5
4.6
74
Hydrogen generation of CdS/METS-10 for the first 24
hours
78
Data of hydrogen generation of the catalysts in 3 cycles
81
xiv
LIST OF FIGURES
FIGURE NO.
TITLE
1.1
The world energy consumption from year 1982 to
2007
1.2
PAGE
2
The regional energy consumption pattern in year
2007
2
1.3
Principle of the PEM fuel cell
4
1.4
Photocatalytic process
8
2.1
Standard solar spectra for the usage of (a) space
(b) terrestrial
2.2
Schematic diagram of the electronic feature of a
metal and semiconductor
2.3
15
17
Crystal structure of the common semiconductors
(a) diamond cubic (b) zincblende (c) wurtzite (d)
rock salt
2.4
18
Energy level diagram of various semiconductors
in aqueous medium (a) OR type (b) R type (c) O
type
21
xv
2.5
Schematic diagram of electron donor (D)
2.6
Schematic diagram of several functions assigned
22
to a support in photocatalysis. Symbols are as
follows: S or Si are substrates to be transformed;
P or Pi are final or intermediate product; D or A
represents electron donor or electron acceptor
respectively.
Dote lines represents the light
irradiation. (a) adsorption of the substrate; (b)
adsorption of the substrate and intermediate
products in a restricted geometry; (c) molecular
assembly for energy transfer toward a reaction
center; (d) separating redox intermediates by
double layer effects; (e) bifunctional catalytic
system
2.7
26
The ETS-10. (a) structure arrangement of ETS-10
where the red colour represents the [SiO4]4tetrahedra and the blue colour represents the
[TiO6]8- octahedra (b) molecular structure.
2.8
29
Schematic diagram illustrated the holes transfer
from valance band of CdS to valance band of
LaMnO3.
3.1
32
Front view of microreactor frame. Grey dots
indicate hole size 1.5 cm whereas black dot
indicate hole size 2.3 cm. (scale in cm)
41
3.2
Rear view of microreactor frame. (scale in cm)
41
3.3
Diagram of experimental set-up for catalytic
testing
42
xvi
3.4
dimension of (a) pyrex reaction vessel (b) pyrex
glass tube
4.1
43
The EDAX spectrum for as-synthesized CdS
sample
4.2
XRD
49
pattern
of
the
CdS
nanoparticles
synthesized from cadmium acetate by in-situ
reduction method with different solvents (a) DMF
(b) DMSO. (c) CdS synthesized from cadmium
nitrate by reverse micelle method.
4.3
50
XRD patterns of CdS nanoparticles synthesized
with different cadmium salts. (a) cadmium acetate
(b) cadmium chloride.
4.4
50
FESEM micrographs of as-synthesized CdS (a)
CdS-IS (b) CdS-RM
51
4.5
TEM micrographs of as-synthesized CdS
52
4.6
XRD patterns of (a) P25 TiO2 (b) ETS-10 from
RHA (c) ETS-10 from Ludox-30. Peaks assigned
to P25 TiO2 are indicated by asterisks.
4.7
FESEM micrographs of ETS-10 synthesized by
different silica sources (a) RHA (b) Ludox-30
4.8
4.9
53
54
IR spectra for ETS-10 synthesized by different
silica sources (a) Ludox-30 (b) RHA.
55
The EDAX spectrum for 10CdS/ETS-10 sample.
57
xvii
4.10
XRD patterns of samples (a) ETS-10 (b) CdS (c)
CdS/ETS-10
4.11
FESEM
micrographs
58
of
the
samples
(a)
10CdS/ETS-10 (b) 5CdS/ETS-10
59
4.12
TEM micrographs of CdS/ETS-10
59
4.13
DR-UVspectra (a) CdS-IS (b) CdS-RM (c) bulk
CdS
4.14
DR-UV spectra of (a) ETS-10 from RHA (b)
ETS-10 from Ludox-30 (c) METS-10
4.15
65
The band gap studied of (a) ETS-10 (b)
10CdS/ETS-10 (c) 5CdS/ETS-10 (d) CdS-IS
4.19
64
The band gap studied of (a) ETS-10 (b)
10CdS/ETS-10 (c) METS-10
4.18
63
The band gap studied of CdS samples (a) CdS-IS
(b) CdS-RM (c) bulk CdS
4.17
62
Diffuse reflectance UV-vis spectra of the
samples.
4.16
60
66
The view of the micro-reactor coupled online
GC-TCD
67
4.20
Chromatogram of water photo-splitting.
68
4.21
Schematic diagram of the energy level and
charges separation mechanism of CdS/ETS-10.
70
xviii
4.22
Amount of hydrogen generated by the catalysts
(a) CdS-IS (b) CdS-RM
71
4.23
Schematic energy diagram of (a) CdS-IS (b) CdSRM
4.24
72
Schematic energy diagram of (a) 10CdS-IS/ETS10
(b)
10CdS-RM/ETS-10
(straight
lines
represent the original band edge and dotted lines
represent new band edge)
4.25
73
The amount of hydrogen generated by the
catalysts:
(a)
5CdS-IS/ETS-10
(b)
15CdS-
IS/ETS-10 (c) 10CdS-IS/ETS-10 (d) 5CdSIS/ETS-10
4.26
75
The amount of hydrogen generated by the
catalysts: (a) 5CdS-RM/ETS-10 (b) 10CdSRM/ETS-10 (c) 15CdS-RM/ETS-10 (d) 20CdSRM/ETS-10.
4.27
76
Schematic energy diagram of (a) 10CdSIS/METS-10 (b) 10CdS-RM/METS-10 (straight
lines represent the original band edge and dotted
lines represent new band edge).
4.28
77
The amount of hydrogen generated by the
catalysts: (a) 5CdS-IS/METS-10 (b) 10CdSIS/METS-10 (c) 15CdS-IS/METS-10 (d) 20CdSIS/METS-10
4.29
The amount of hydrogen generated by the
catalysts: (a) 5CdS-RM/METS-10 (b) 10CdS-
79
xix
RM/METS-10
(c)
15CdS-RM/METS-10
(d)
20CdS-RM/METS-10
4.30
79
The amount of hydrogen generated in 3 cycles by
the catalysts (a) 20CdS-IS/ETS-10 (b) 20CdSIS/METS-10 (c) CdS-IS (d) CdS-RM.
80
xx
LIST OF ABBREVIATIONS
HHV
-
Higher heating value
CdS
-
Cadmium sulfide
ETS-4
-
Engelhard titanosilicate-4
ETS-10
-
Engelhard titanosilicate-10
UV
-
Ultra violet
Vis
-
Visible
RHA
-
Rice husk ash
KF
-
Potassium fluoride
NaOH
-
Sodium hydroxide
H2O2
-
Hydrogen peroxide
GC-TCD
-
Gas chromatography thermal conductivity detector
Å
-
Dimension unit, Amstrongs (x10-10m)
TiO2
-
Titanium dioxide
ASTM
-
American Society for Testing and Materials
CB
-
Conduction band
VB
-
Valence band
ccm/g
-
Cubic centimeter per gram
nm
-
Nanometer (x10-9m)
ZnO
-
Zinc(II) oxide
Fe2O3
-
Iron(III) oxide
ZnS
-
Zinc(II) sulfide
H2
-
Hydrogen gas
EG
-
Energy gap
fcc
-
Face-centered-cubic
hcp
-
Hexagonal close-packed
CdS/ETS-10
-
Cadmium Sulfide supported on ETS-10
Si/Ti
-
Silicon to titanium ratio
xxi
OR type
-
Strong oxidation and reduction power
O type
-
Strong oxidation power
R type
-
Strong reduction power
X type
-
Weak oxidation and reduction
KBr
-
Potassium bromide
H+/H2
-
Reduction of hydrogen ion to hydrogen gas
D
-
Electron donor / hole scavenger
A
-
Electron acceptor
EDTA
-
Ethylenediaminetetraacetic acid
TEOA
-
Triethanolamine
SO32-
-
Sulfite ion
H2PO2-
-
Hypophosphite ion
XRD
-
X-ray diffraction
FTIR
-
Fourier Transform Infrared spectroscopy
DR-UV
-
Diffuse reflectance UV-Vis Spectroscopy
FESEM
-
Field Emission Scanning Electron Microscopy
TEM
-
Transmission Electron Microscopy
EDAX
-
Energy Dispersive Spectroscopy
W
-
Watt
CHAPTER 1
INTRODUCTION
1.1
Research Background
Recently, one of the biggest issues that capture most of our concern is the
fuel‟s market value. It has been the all-time front page news in the newspaper for
the last several months. Everyone is talking about the tremendous increase in prices
of the petrol and diesel around the world. It draws a huge attention because the
fuel‟s market value is directly affecting the prices of goods, or even the economy of
countries. In addition to transportation and power generation, mass quantities of
petroleum and petroleum derived chemicals are required as raw materials and fuels
in many industries such as manufacturing, food processing and pharmaceutical
industries.
The world energy consumption obtained from BP Statistical Review is
shown in Figure 1.1 (BP, 2008). Fossil fuels such as petroleum, coal and natural gas
are still remain as the most important energy resources. Petroleum is still the
world‟s leading fuel, but has lost its global share of the market for six consecutive
years while coal remained as the fastest growing fuel and has gained the share of the
market for six years. Petroleum remains as the dominant fuel in all regions except
Europe and Eurasia and Asia Pacific. Coal dominates in the Asia Pacific region
primarily because it meets 70% of China‟s energy needs. The regional energy
consumption pattern in year 2007 is shown in Figure 1.2 (BP, 2008).
2
Million tonnes oil equivalent
Year
Figure 1.1: The world energy consumption from year 1982 to 2007 (BP, 2008)
Percentage (%)
Region
Figure 1.2: The regional energy consumption pattern in year 2007 (BP, 2008)
3
1.1.1
Renewable Resources
Due to the depletion of petroleum and the rising demand for alternative
resources, several approaches must be carried out in order to overcome this obstacle
especially in the development of renewable resources to replace the unstable and
diminishing natural fuel. A number of renewable resources technologies are now
commercially available, the most notable being wind power, photovoltaics, solar
thermal systems, biomass and the various forms of water power (Thring, 2004).
Biomass is gaining high attention as it is one of the most available renewable energy
resources that can be used to reduce the dependency on fossil resources (Williams
and Nugranad, 2000). Biomass refers to living and recently dead biological material
that can be used as fuel or for industrial production. Some of the agricultural wastes
that consist of carbon could be utilized as raw materials to generate the heat and
electricity for milling processes. Since enormous numbers of agriculture activities
are conducted locally, therefore this method could be applied in energy recovery
scheme.
However, the utilization of biomass for energy conversion through
combustion is still limited due to its poor fuel properties such as high moisture and
ash contents, low bulk density, low energy content (William and Nugranad, 2000).
One of the alternatives for energy generation is the usage of hydrogen fuel
cell. Several researches have been developed in order to obtain a highly efficient and
effective hydrogen fuel cell (James and Michael, 2005). A hydrogen fuel cell
converts chemical energy directly into electricity by combining oxygen from the air
with hydrogen gas. However, unlike a battery, a fuel cell does not run down or
require recharging.
It will produce electricity as long as fuel, in the form of
hydrogen, is supplied. No pollution is produced and the only byproducts are water
and heat.
Figure 1.3 shows the functional principle of the Proton Exchange
Membrane (PEM) fuel cell.
4
Figure 1.3: Principle of the PEM fuel cell (Strasser and Siemens, 1995)
1.1.2
Hydrogen Economy
Hydrogen, the first element on the periodic table, is the least complex and the
most abundant element in the universe (Mohammad and Chen, 2005). However, not
much is available in pure form on earth and is available either as water (when
combined with oxygen) or as a hydrocarbon (when combined with carbon). For this
reason, hydrogen is only an energy carrier and not a primary energy source.
Extraction of hydrogen from its compounds was studied with the presence
(Lindström et.al., 2003; Chang et.al., 2005; Hu and Lu, 2007; Wang et.al., 2007) or
absence of catalysts (Marty and Grouset, 2003).
A hydrogen economy has been proposed as a way to reduce global
greenhouse gas emission due to its properties as a very clean fuel and burning it
results in no greenhouse emissions or undesirable carbon compounds (James and
Michael 2005). Hydrogen contains the highest energy density on a mass basis of all
chemical fuels which can be quantified by its higher heating value (HHV) of 141.9
MJ/kg.
The higher heating value is the amount of energy released during the
oxidation reaction of a fuel with air at a starting and finishing temperature of 25oC.
5
Heating values of hydrogen and other fossil fuels are tabulated in Table 1.1, which
clearly indicates that hydrogen has the highest energy density (Lo et.al, 2006).
Table 1.1: Heating value for selected fuels (Lo et.al., 2006)
Fuel
Coal
Methane
Natural gas
Propane
Gasoline
Diesel
Hydrogen
1.1.3
HHV (MJ/kg)
34.1
55.5
42.5
48.9
46.7
45.9
141.9
Hydrogen Production
The production of hydrogen is more costly than any other fuel. Hydrogen
can be produced by reforming hydrocarbons, in which the steam reforming of
methane being the most efficient method (Christofoletti et.al., 2005; Lo et.al., 2006).
Steam reforming is a very established technology and is used to generate large
quantities of hydrogen in industrial processes. This reaction is an endothermic
process; slow to start-up and relatively inflexible with regard to non-steady state
operation. The mechanism of the steam reforming is shown in equations below
(Thring, 2004).
CnHm + nH2O  nCO + (n+m/2)H2
(1.1)
CH4 + 2H2O  4H2 + CO2
(1.2)
In contrast, partial oxidation reforming is an exothermic process and work
faster and more responsive if compared to steam reforming. However, it produces
relatively higher levels of carbon monoxide and lower concentration of hydrogen.
The third system which involves a combination of these two processes is known as
autothermal reforming. The mechanism of the partial oxidation reforming is shown
in equations below (Thring, 2004).
6
CnHm + (n/2)O2  nCO + (m/2)H2
(1.3)
CH4 + (1/2)O2  2H2 + CO
(1.4)
Hydrogen also can be produced via coal gasification. However, formation of
carbon monoxide and carbon dioxide from hydrocarbon reformation and coal
gasification is inevitable. Moreover, the yield collected is not 100 % pure (Zhu
et.al., 2005; Praveen and Alan 2005; Praveen and William 2006). Hydrogen gas
derived from biomass and biogas was found to contain impurities (Osamu, 2006;
Hadi and Tomohiro, 2006).
Another method of producing hydrogen is by water splitting using electricity
generated from renewable resources such as solar, wind, hydro and biomass. In
addition, water splitting can be conducted in thermochemical cycles when operating
at high temperatures (Panini and Srinivas 2006). Hydrogen production based on the
type of fuel is shown in Table 1.2.
Table 1.2: Hydrogen production based on the types of fuel (Panini and Srinivas
2006)
Fuel
Amount
Percentage
Method of production
(billions of
Nm3/year)
Natural gas
Oil
Coal
water
240
150
90
20
48%
30%
18%
4%
Steam reforming
Partial oxidation reforming
Coal gasification
Electrolysis
Methane or natural gas is the fuel of choice and almost 50% of industrial
hydrogen production uses methane as a fuel. For higher hydrocarbons like gasoline
or diesel partial oxidation reforming is generally used.
During recent years,
autothermal reforming is increasingly being employed (Praveen and Alan, 2005).
7
Hydrogen obtained from gasification of coal currently contributes 18% of the
world hydrogen consumption.
Among those commercial techniques, hydrogen
production from electrolysis of water is the only method that produces hydrogen gas
in high purity.
However, only 4% of the total world hydrogen production is
produced via this method. In general, hydrogen production from renewable sources
is economically impractical due to the current technical constraint. The storage of
the hydrogen for consumer applications has proved to be a difficult challenge as well
(Lo et.al., 2006).
1.1.4
Photocatalysis
Photocatalysis is a division of chemistry studying catalytic reactions
proceeding under the action of light. In photocatalytic process, the surface reaction
involved is redox reaction (Serpone, 1989). The basic of the photocatalysis process
is shown in Figure 2.2. The initial step of photocatalysis is the adsorption of photons
by a molecule to produce highly reactive electronically excited states. The photon
needs to have energy of hυ equal to or more than the band gap energy of the
semiconductor. The energy absorbed will cause an electron to be excited from the
valence band to the conduction band, leaving a positive hole in the valence band.
This movement of electrons forms e-/h+ or negatively charged electron/positively
charged hole pairs (Serpone, 1989). The positively charged holes in valence band
are powerful oxidants, whereas the negatively charged electrons in conduction band
are good reductants. The void region which extends from the top of the filled
valence band to the bottom of the vacant conduction band is called the bandgap
(Serpone, 1989).
8
Figure 1.4: Photocatalytic process.
Photocatalysis can be defined as acceleration of a photoreaction by the
presence of a catalyst. The catalyst may accelerate the photoreaction by interaction
with the substrate in its ground or excited state and/ or with a primary photoproduct,
depending on the mechanism of the photoreaction (Bard, 1979). According to
Serpone (1989), when the light is absorbed by the catalyst (Equation 2.1), the system
represents a sensitized photoreaction which may occur either via energy transfer
(Equations 2.2 and 2.3) or via electron transfer (Equations 2.4 to 2.6).
catalyst + hυ  *catalyst
*catalyst + substrate  *substrate + catalyst
*substrate  product
*catalyst + substrate  substrate- + catalyst+
(2.1)
(2.2)
(2.3)
(2.4)
substrate-  product-
(2.5)
catalyst+ + product-  product + catalyst
(2.6)
The very first photoinduced redox reaction was discovered by Fujishima and
Honda in 1972 (Fujishima and Honda, 1972). They used an n-type titanium dioxide
(TiO2) semiconductor electrode, which was connected through an electrical load to a
platinum black counter electrode and exposed to near UV light. The extensive
9
research was conducted on producing hydrogen from water as a means of solar
energy conversion. Later, this redox reaction was utilized in the decomposition of
organic and inorganic compounds for environment protection.
Most of the organic and inorganic substances decomposition involves the
formation of intermediate in a form of radical. Irradiation of semiconductor with
light of energy higher than the band gap results in creation of holes in the
semiconductor valence band (VB) and electrons in the conduction band (CB)
followed by trapping of separated charges shallow traps (tr) at the solid-solution
interface (Izumi et. al., 1981). These traps carriers can be recombine or react with
the substrates to form various groups of radicals such as hydroxyl radicals and
hydrogen radicals as shown in the following equations.
(TiO2) + hυ  ecb- + hvb+
(2.7)
ecb-  etr-
(2.8)
hvb+  htr+
(2.9)
OH- + htr+  .OH
(2.10)
H+ + etr-  H
(2.11)
Semiconductors such as TiO2, ZnO, Fe2O3, CdS and ZnS can act as
sensitizers for light-induced redox processes due to their electronic structure, which
is characterized by a filled valence band and empty conduction band.
characteristics
enable
them
to
generate
e-/h+
or
negatively
These
charged
electron/positively charged hole pairs when they received a photon with sufficient
energy.
The excited electrons and created holes can recombine and scatter the input
energy as heat or get trapped in meta-stable surface states. The excited electrons and
created holes can also react with electron donors and acceptors adsorbed on the
semiconductor surface or within the surrounding electrical double layer of the
charged particles. The reactants, reaction intermediates and products are transported
between the semiconductor surface and the bulk solution all the time (Fox and
10
Dulay, 1993).
The rate of photocatalysis is proportional to the adsorption and
desorption rate (Maldotti et. al, 2002).
1.2
Problem Statement
From the literature review, cadmium sulfide (CdS) nanoparticles prepared by
reverse micelle method showed photocatalytic activity for hydrogen generation from
water under visible light irradiation (Guan et. al., 2005). However, the way of
preparing CdS nanoparticles by these methods are complicated, wasting a lot of
solvents and surfactants (create disposal problem) and time consuming. At present,
there is still no report of the CdS nanoparticles being prepared by in-situ sulphur
reduction method used in hydrogen generation. Since CdS nanoparticles prepared
with this method is more feasible, nearly zero waste and economically practical, it
becomes very suitable for low cost hydrogen generation.
Many researchers have reported preparation of Engelhard titanosilicates
(ETS-10) using Ludox and water glass as silica sources showed photocatalytic
activity for hydrogen generation from water under ultra violet (UV) irradiation
(Guan et. al., 2005; Sitharamam et. al., 2004). In this research, rice husk ash was
used as silica source due to its high percentage of silica and locally available in
cheaper price. In addition, the ETS-10 that derived from rice husk ash (RHA) was
found to be more crystalline and posses higher surface area up to 35.5% (Jei et. al.,
2008). The high crystallinity character of ETS-10 is very desirable in water splitting
reaction. Therefore, ETS-10 is suggested as the support material to enhance the
function of CdS in water splitting reaction. As-synthesized ETS-10 was further
treated by hydrogen peroxide (H2O2) to narrow down its band gap to enable it
function under visible light irradiation.
The well known photocatalyst such as titanium dioxide only function well
under UV light irradiation due to its wide band gap energy properties.
Unfortunately, the usage of titanium dioxide is not desirable because only 3% of sun
11
energy reaches the earth in that UV region. Therefore, the development of visible
light driven photocatalyst was gaining much more attention. CdS solely easily
undergoes photocorrosion. In order to solve this problem, several approaches have
been carried out to promote an efficient charge separation. As reported earlier, CdS
nanoparticles prepared by precipitation in the zeolite matrix showed a significant
result in water splitting reaction (Sathish et. al., 2006). In this research, an in-situ
hydrogen generation from water by CdS nanoparticles supported on pure ETS-10 is
suggested as a photocatalyst. CdS nanoparticles supported on modified ETS-10 was
also tested for comparison purposes.
1.3
Objectives of Research
The main aim of this research is to investigate the influence of synthesis and
modification techniques on CdS/ETS-10 photocatalyst towards the hydrogen gas
yield under visible light irradiation. This thesis extensively discusses the design of
the continuous flow micro-reactor coupled with gas chromatography to thermal
conductivity detector (GC-TCD) for the online detection of hydrogen gas. The
details of the research objectives are as shown below:
(i)
To synthesize and modify nano-sized CdS and microporous titanosilicate
ETS-10.
(ii)
To impregnate the CdS on ETS-10 by incipient wet technique.
(iii)
To characterize the physic
o-chemical properties of the as-synthesized
samples.
(iv)
To set-up and calibrate the microreactor coupled with online GC-TCD for
hydrogen detection.
(v)
To evaluate the performance of the samples in hydrogen generation from
water under visible light irradiation.
12
(vi)
To optimize condition of the reaction including the catalysts preparation
method, CdS to ETS-10 ratios, light sources used and the function of
sacrificed agents in water.
1.4
Scope of Research
This research focuses on the synthesis and modification pathway of the
CdS/ETS-10 hybrid photocatalyst and examines its physicochemical properties.
Two methods were applied in the CdS preparation, including reverse micelle method
and in-situ sulphur reduction method.
ETS-10 used is generally prepared by
hydrothermal synthesis route with Degussa P25 titanium dioxide as titanium source
and RHA as silica source. ETS-10 derived from Ludox-30 as silica source was also
prepared for comparison purposes.
The effect H2O2 treatment on ETS-10 was
performed to evaluate the effectiveness of the charge separation effect compared
with the original ETS-10. The photocatalyst prepared was tested in water splitting
process under visible light irradiation. The volume of hydrogen gas generated was
recorded by water displacement method whereas the purity of hydrogen gas being
confirmed by online GC-TCD. The optimization condition of the reaction includes
catalysts preparation method, CdS to ETS-10 ratios, light sources used and the
presence and absence of sacrificed agents in water.
1.5
Thesis Outline
This thesis consists of 5 chapters and completed with a list of references.
Chapter 1 has presented a brief introduction to the research background and some
advantages or application of the output. It has also discussed the problem statement
and summarizes the objectives of the research.
Chapter 2 presents some
fundamental concepts of photocatalyst, water photo-splitting, semiconductor and
role of hole scavengers as electron donors. Additionally, chapter 2 includes some
13
previously relevant literature reviews on the hydrogen generation from other type of
photocatalysts. Chapter 3 discusses the experimental steps including: synthesis,
characterization, reactor set-up and catalytic testing of the samples. Chapter 4 shows
the results and discussion of the finding, whereas the conclusion and suggestions of
research was reported in chapter 5.
CHAPTER 2
LITERATURE REVIEW
2.1
Solar Energy Distribution
The solar spectrum changes throughout the day and the location. In 1982, the
American Society for Testing and Materials (ASTM) adopted consensus standard
solar terrestrial spectra to provide standard spectra for photovoltaic (PV)
performance applications, solar energy systems and materials degradation.
Gueymard and co-workers (2002) have summarized the definition of atmospheric
parameters, spectral range, accuracy and resolution, and documentation of the
standards. The new reference spectrum proposed was claimed to be more realistic
which might maximize the use of concentrators that utilize only the direct
component. The effect of various pollutants on solar spectra irradiation was also
studied (Jacovides et.al., 2000).
The high urban polluted air has significantly
reduced the atmospheric spectral transmittance.
The standard spectrum for space and terrestrial applications is presented in
Figure 2.1. The magnitude of spectral irradiation at top of atmosphere was found to
be greater than radiation at sea level. The existence of atmospheric particles such as
ozone, aerosol and water vapor act as the radiation absorbent and leads to the
decreasing of radiation number (British Standard, 1991).
At sea level, UV
irradiation appearing in the region of 100 - 400 nm was found only consists only 3%
of the total solar energy, whereas visible light in the region of 400 - 780 nm consists
of 44%. Near Infrared region (780 - 2500 nm) consists the largest proportion up to
15
53%.
However, the consumption of energy below the UV light region is
unbreakable, particularly in photocatalysis field due to the limitation of
semiconductor properties. The well known photocatalyst such as titanium oxide
only function well under UV light irradiation. In order to fully utilize the spectrum
of sun light, the development of visible light driven photocatalyst was gaining much
more attention.
UV
Visible
Near Infrared
Spectral Irradiance (W m-2 nm -1)
2.00
1.75
1.50
1.25
1.00
(a)
0.75
0.50
(b)
0.25
0.00
250
500
750
1000
1250
1500
1750
2000
2250
2500
Wavelength (nm)
Figure 2.1: Standard solar spectra for the usage of (a) space (b) terrestrial.
2.2
Semiconductor
Since the first discovery of electrons by J.J. Thomson in year 1897, many
development of the theoretical descriptions of the conduction of metal was
demonstrated. In year 1931, A.H. Wilson introduced the concept of holes and the
relationship of the energy gap toward the reverse breakdown of semiconductors and
insulators in large electric field. There are several ways to define a semiconductor.
In the early stage, semiconductor has been used to denote materials with a much
16
higher conductivity than insulator but a much lower conductivity than metals at
room temperature (Turner, 1961). Recently, the definition is much more related to
the energy gap through the free carrier concentration at room temperature. As
shown in Table 2.1, metals and semimetals have a rather largest carrier density;
semiconductors exhibit a moderate carrier density while insulators have a negligible
carrier density at room temperature.
However, the real semiconductors always
contain some impurities, which can act as dopants leading to larger values of the
carrier density (Grahn, 1999).
Table 2.1: The classification of solids according to their energy gap, carrier density
and typical conductivity at room temperature (Grahn, 1999).
Type of solid
Energy gap, EG
Carrier
Conductivity,
(eV)
density, n
σ (cm-1)
Example
(cm-3)
Metal
No energy gap
Semimetal
Semiconductor
Insulator
EG ≤ 0
1022
17
105- 1010
21
10 -10
2
5
Au, Cu, Pb
10 -10
Graphite, HgTe
0 < EG < 4
< 1017
10-9-102
Si, Ge, GaAs
EG ≥ 4
<< 1
< 10-9
Quartz, CaF2
The characteristic electronic feature of a metal is extended band cut by the
Fermi level as shown in Figure 2.2 (a). In a metal, a continuous set of former vacant
level above the Fermi energy is occupied during illumination and an equivalent set
of vacant level below the Fermi energy is generated. However, the lifetime of
excited electrons and excited hole is extremely short due to the continuous density of
energy states (Serpone, 1989). Semiconductors are characterized by two separated
energy bands: a filled low-energy valence band (with the Fermi level at the top) and
an empty high-energy conduction band (Figure 2.2 (b)). Unlike the metal, the
existence of the band gap in semiconductors prevents rapid recombination of the
excited electron-hole pairs.
17
h‫ט‬
h‫ט‬
Fermi energy
Fermi energy
+
+
(a) metal
E
(b) semiconductor
Figure 2.2: Schematic diagram of the electronic feature of a metal and a
semiconductor.
In general, semiconductor structures are characterized by covalent bonding,
low coordination number with open lattice structures.
Most of the common
semiconductors structures are diamond cubic, zincblende, wurtzite and rock salt
(Figure 2.3). Diamond cubic exhibit tetrahedral bonding form by sp3-hybridization
with each atom is surrounded by four nearest neighbors located at the four corners of
a regular tetrahedron. It can be presented as two face-centered-cubic (fcc) lattices
with one fcc lattice displaced from the other. In zincblende structure, one of the
interpenetrating fcc lattices consists of one element, and the other fcc lattice is
composed by the other element (Grahn, 1999). The wurtzite structure consists of
two interpenetrating hexagonal close-packed (hcp) lattice constructed of two
different atoms. A number of semiconductors may interconvert between the wurtzite
and zincblende structures in response to slight changes in temperature and/ or
pressure. Some binary compounds composed of elements from the Group IV family
and the Group VI family display in rock salt structure and typically have small band
gaps (Serpone, 1989). The classification of the semiconductors according to their
crystal structure is shown in Table 2.2 (Grahn, 1999).
18
Table 2.2: The classification of semiconductors according to their crystal structure.
Structure
Group
Materials
Diamond
VI
Diamond (c), Ge, Si, α-Sn
Zincblende
III-V
AlAs, AlP, AlSb, BN, GaAs, InAs
II-VI
CdS, CdSe, HgSe, ZnS, ZnTe
I-VII
γ-CuBr, γ-CuCl, γ-CuI
III-V
AlN, BN, GaN, InN
II-VI
CdS, CdSe, ZnO, ZnS, ZnSe
I-VII
β-AgI
IV-VI
PbS, PbSe, SnTe
Wurtzite
Rock salt
(a)
(c)
(b)
(d)
Figure 2.3: Crystal structure of common semiconductors (a) diamond cubic (b)
zincblende (c) wurtzite and (d) rock salt.
19
2.3
Thermodynamic Limitation of Water Photo-splitting
Photosynthesis represents a natural photochemical system for the conversion
of radiant solar energy to chemical energy. In fact, a number of researches has been
attempted to mimic natural photosynthesis by using colloidal semiconductor
particles. Splitting of water into gaseous hydrogen and oxygen by illumination of
the semiconductor powder suspensions is regarded as a promising method for
harnessing sunlight into storable energy (Kutty and Avudaithai, 1988). In water
splitting, it involves the oxidation and reduction of water to produce O2 and H2. The
photo-driven conversion of liquid water to gaseous hydrogen and oxygen at room
pressure is showed in equation 2.12, where n is the number of photons used per
electron transferred in the reaction (Schiavello, 1985; Bard and Fox, 1995).
2H2O (l)
4nhν
2H2 (g) + O2 (g)
∆Go= 474 kJ/mol
(2.12)
The nature of the semiconductor and the fluid-irradiated semiconductor
interface affects the photocatalytic activity. The nature of the interface depends on
various parameters such as electronic and chemical properties of the semiconductor,
presence of additives in fluid, morphological features, donor-acceptor and acid-base
properties of fluid, temperature, pressure and others (Schiavello, 1985; Serpone,
1989; Peral and Mills, 1993).
From the electronic and chemical properties
perspective, the energy of irradiation must equal or greater than the band gap or
threshold energy (Eg) in order to promote a charge separation. The excess energy
will be absorbed as vibrational energy or other forms of energy; and being lost to the
surrounding as heat. The Mie theory can be used to predict the relationship between
threshold wavelength (λg) and the band gap energy via equation 2.13 (Serpone,
1989; Raymond, 1998).
λg (nm) =
1240
Eg (eV)
(2.13)
20
The knowledge of the band edge position is particularly useful in water
photo-splitting.
Figure 2.4 represents the standard potentials for several redox
couple, which indicates the thermodynamic limitations for the photoreaction that can
be carried out with the charge carriers (Schiavello, 1985). The reduction of H+ to H2
only occurs when the conduction band of the semiconductor located more negatively
edge than the relevant redox potential level. In contrast, the oxidation process of
H2O to form O2 only executed when the valance band of the semiconductor located
more positively edge than redox potential of O2 (+1.23 eV).
The potential of these semiconductors can be classified into four groups from
the water splitting reaction point of view.
There are OR, O, R and X type
semiconductors. The OR type semiconductor posses both strong oxidation and
reduction power thermodynamically.
The O type semiconductor shows strong
oxidation power but weak reduction power thermodynamically; vice versa for R type
semiconductor. Lastly, the X type semiconductor with the conduction and valence
bands located in the between of relevant redox levels showed the weak power in
both oxidation and reduction process (Serpone, 1989).
From the thermodynamic point of view, the OR type and R type
semiconductors are predicted to be suitable in hydrogen generation form water due
to the conduction band edge position which is more negative than H+/H2. However,
there are many obstacles when finding a suitable photocatalyst for water splitting
reaction. The semiconductors with too narrow band gap energy (below 1.7 eV) were
found to be inactive in water photo-splitting due to the fast recombination of
charges. Besides, a few numbers of semiconductors were found unstable or easily
undergo certain reaction when in contact with the substrate or fluid.
The
semiconductors with the band gap energy equal or more than 3.1 eV were also
inactive in visible light irradiation. Cadmium sulfide (CdS) with the band gap
energy of 2.4 eV was found to be a potential visible light driven photocatalyst due to
its conduction edge that is more negative that H+/H2. Furthermore, it is thermally
stable under mild temperature and pressure.
21
Negative
Potential (V/NHE)
-2
-1
ZnS
CdS
CdT
CdS Si
e
e
TiO2
1.1
MoS2
3
H+/H2
eV
Fe
O
2
3
WO3
3.6
2.4 1.4
3.2
1.75
eV
eV eV 1.7
eV
eV
eV O2/H2O
3.0
2.3
2.8
eV
eV
eV
SrTiO
0
+1
+2
Positive
(a)
(b)
(c)
Figure 2.4: Energy-level diagram of various semiconductors in aqueous media. (a)
OR type (b) R type (c) O type.
2.4
Hole Scavenger Agents
The photocatalytic activity in photocatalytic system is highly dependence on
the additives in fluid or solution. Generally, water splitting reaction was conducted
under alkaline condition rather than acidic condition. This is due to the maximum
hydrogen evolution rate being located in strong alkaline region (pH 13-14). In
addition, most of the photocatalyst is more structurally stable in alkaline solution.
The additives used in photocatalysis are commonly defined by its function as
electron donor (D) or electron acceptor (A). In hydrogen generation reaction, a
strong reduction power has to be generated at the conduction band of the
photocatalyst.
This condition can be achieved by preventing the charge
recombination between the electron and hole pair. The presence of electron donor in
the irradiated solution was able to scavenge the photogenerated hole at the valence
band irreversibly, hence these materials are known as scarifying agents or hole
22
scavenger agents (Schiavello, 1985). Therefore, the backward electron transfer from
photocatalyst to the oxidized scavenger agents is unlikely and consequently, the
reductive side is attained in high quantum yield (Kitamura et.al, 1983).
The
mechanism of hole scavenge by the electron donor is illustrated in Figure 2.5.
H+
e-
hυ
H2
D+
h+
D
Figure 2.5: Schematic diagram of electron donor (D).
Kutty and Avudaithai had reported the use of ethylenediaminetetraacetic acid
(EDTA), triethanolamine (TEOA), sulfite (SO32-) or hypophosphite (H2PO2-) in the
photolysis of water by several photocatalyst composite such as: TiO2 fine powder
and Pt-mounted TiO2 (1988); SrTiO3 and Pt- or Rh- mounted SrTiO3 (1990). When
TiO2 applied as photocatalyst, inorganic ions such as SO32- ions and H2PO2- ions are
less effective as hole scavengers if compared to organic molecules.
However,
H2PO2- ions was found to be the best among the hole scavengers in SrTiO3 catalytic
system. Oxygen gas was found absence in the all photoreaction except for Pt- or Rhmounted SrTiO3 particles suspended in water without any hole scavengers.
The use of hole scavenger agent might cause a post-irradiation effect when
the irradiation was terminated.
This phenomenon was ascribed to an electron
transfer from radical species formed by hole scavenging to the TiO2 particles
(Serpone et.al., 2000). This mechanism further produces electron and hole pairs and
cause the reaction growth at dark. However, this observation only happened in
oxygen-free system.
23
2.5
CdS as Photocatalyst
Recently, Meng Ni and co-workers (2007) published a review on the current
development topics in photocatalytic water splitting by using TiO2 as the
photocatalyst. Various issues have been discussed in order to obtain high yield of
hydrogen production especially on the topics of photocatalyst modification and role
playing of additive in water. However, the photocatalytic conversion efficiency is
too low for the technology to be economically practicing. The main barriers are the
low solar energy conversion efficiency, rapid charge recombination between the
photogenerated electron and hole pairs or known as photo-corrosion, fast catalyst
deactivation as well as the backward reaction. These are the main factors that affect
the low conversion efficiency for hydrogen generation from water.
In order to resolve the above listed problems and make solar photocatalytic
hydrogen production feasible, continuous efforts have been made to promote the
photocatalytic activity and enhance the visible light response. Various factors such
as addition of electron donors or hole scavengers (Jang et. al., 2007; Kida et. al,
2004)), addition of carbonate salts (Sayama et.al., 1998), noble metal loading (Jin
and Shiraishi, 2004), dye sensitization (Dhanalakshmi et. al., 2001; Abe et. al, 2002)
and composite semiconductors (Hirai and Bando, 2005) have been investigated and
are proved to be useful to enhance hydrogen production.
Various types of photocatalysts such as TiO2, CdS, ZnO, NiO/Sr3Ti2O7 and
Fe2O3 have been developed for hydrogen generation from water under light
irradiation (Karsten et.al., 2006; Kanade et.al., 2006; Jeong et.al., 2006; Khan et.al.
2008). Among the photocatalysts developed so far, TiO2 was the most important
photocatalyst due to economy feasibility, high activity and excellent chemical
stability. However, TiO2 is a wide-band-gap semiconductor and only function well
under UV light irradiation.
In order to utilize photocatalyst in solar energy
conversion, it is indispensable that the photocatalyst shall also be sensitive to visible
light.
24
Nano-sized CdS photocatalytic water splitting technology has great potential
for low cost and environmental friendly solar hydrogen energy production to support
the future hydrogen economy. CdS is one of the few examples of a visible-lightdriven photocatalyst.
This material has excellent properties for the hydrogen
generation with the band gap of 2.4 eV that corresponds well with the spectrum of
sunlight. The conduction band edge is more negative than the H+/H2 redox potential
and this allows the H2 evolution from water over CdS under sunlight (Darwent and
Porter, 1981). Matsumura and co-worker (1985) have studied the effect of crystal
structure and preparation of CdS towards the hydrogen production efficiency. Ptloaded CdS powder with a hexagonal crystal structure has been found to be much
more efficient as a photocatalyst in hydrogen production from aqueous solutions of
sulfite than that by using Pt-loaded CdS powder with a cubic crystal structure.
Generally, it has also been found that the activity of the photocatalyst is
reduced by the mechanical damage caused by grinding. The defect on the catalyst
surface will lead to an increase of the number of electron-hole recombination centers
and reduce the conversion efficiency.
The enhancement of the photocatalytic
hydrogen evolution can be further achieved by the addition of EDTA or other
chelating agents in the solution. It is explained as being caused by the upward shift
of the conduction band energy of CdS due to the negative charge of the chelating
agents (Uchihara et.al., 1990).
Recent approach on water photo-splitting research direction of CdS focuses
on the minimization of photocorrosion effect by enhancing the charge separation.
This can be achieved by addition of dye sensitizer on CdS or by a composite
semiconductor system. With the addition of dye sensitizer, the excited dyes can
inject electrons to the CB of CdS to initiate the catalytic reaction under light
irradiation.
The fast electron injection and slow backward reaction make dye-
sensitized semiconductors feasible for energy conversion.
In the composite
semiconductors system, the electrons located at the CB can be injected from the
small band gap semiconductor to the larger band gap semiconductor. The reduction
of water is then occurred at the CB of the semiconductor that posses larger band gap
energy.
25
2.6
Supports in Photocatalysis
In heterogeneous catalysis, the usage of supports is inevitable due to several
advantages other than providing high surface area and recovery possibility.
In
general, the supports commonly exist in porous microtexture and appear in a unique
interfacial interaction in order to maintain in the dispersed state. Due to these
physical requirements, the supports often interfere either directly or indirectly in the
reaction mechanism via its active site.
The porosity properties also affect the
adsorption and desorption rate of the substrates, intermediate and the final products.
Figure 2.6 presents the schematic diagram of several functions assigned to a support
in a photocatalytic system.
Water photo-splitting by composite semiconductors
often occur according to the mechanism illustrated in Figure 2.6 (e) (Casal et.al.,
1985). The separation of charges can be achieved when the electrons and holes
diffuse in opposite direction. The potential mechanism involved in a photocatalytic
system and its favourable support materials are shown in Table 2.3.
26
P1
*s
s
*s
P1
P2
(a)
(b)
P3
P2
e*D D D D DT A
D D D D *DT A
eD
-
-
+
+
S1
-
e-
+
A
-
-
-
+
+- +
+
S2
-
A
-
D A
-
*D
-
e
S1 D+
-
-
(c)
-
A
-
D+
A
-
-
-
-
-
+
+
+
+
+
S2
-
P1
-
D A
-
-
P2
(d)
(e)
-
Figure 2.6: Schematic diagram of several functions assigned to a support in a
photocatalysis. Symbols are as follows: S or Si are substrates to be transformed; P or
Pi are final or intermediate product; D or A represents electron donor or electron
acceptor respectively. Dote lines represents the light irradiation. (a) adsorption of
the substrate; (b) adsorption of the substrate and intermediate products in a restricted
geometry; (c) molecular assembly for energy transfer toward a reaction center; (d)
separating redox intermediates by double layer effects; (e) bifunctional catalytic
system. (Serpone, 1989).
Table 2.3: Several potential mechanisms involved in a photocatalytic system and its typical suitable support material (Serpone, 1989)
Support
Function
Adsorbing the substrate to be transformed
„Caging‟ the substrate or the primary products
Silicas,
Aluminas
√
Silicas,
Colloidal
Zeolites semiconductors
√
Supporting a particular catalyst
√
Organizing the system at supramolecular level
Clays, Silicas
colloidal
TiO2
semiconductors
√
√
√
√
√
TiO2,
clays
√
√
Separating redox intermediates
Silicas, clays,
√
Absorbing light
Anchoring a molecular catalyst or a sensitizer
Polymer,
√
27
28
2.6.1
Engelhard Titanosilicates (ETS-10)
Currently, there are many researches carried out on the synthesis of nanoporous
materials due to their wide range of applications.
Among the porous materials,
microporous titanosilicate is an example of interlinked octahedral and tetrahedral
structures, unlike TS-1 (Taramasso et.al., 1983), zeolite A (Leonard, 1981) and
aluminosphosphates that only consist of tetrahedral coordinated framework. ETS-10
with the pore size of 0.8 nm is a microporous titanosilicate molecular sieves family with
octahedral coordinated titanium framework ions that was first discovered by Engelhard
in 1989 (Kuznicki, 1989). The porosity properties of ETS-10 will affect the migration
rate of gases where the lighter gas will move faster and vice versa.
The ETS-10 comprises of corner-sharing [SiO4]4- tetrahedra and [TiO6]8octahedra linked through bridging oxygen atoms, forming a three-dimensional 12membered ring network. The titanium (IV) is found in the centre of the corner-sharing
octahedra while the silicon is in the centre of corner-sharing tetrahedra. This produces
an anionic framework whereby, whenever titanium is present in the structure, there is an
associated two minus charge which is compensated by extra-framework cations (Na+
and K+ in as-synthesized ETS-10) (Kuznicki, 1990; Kuznicki, 1991).
The extra-
framework charging might increase the rate of electrons transfer from CB of CdS to CB
of ETS-10 and thus lead to a more effective charge separation. This attraction forces is
expected to be stronger than common supports such as zeolite, clay, alumina and silica.
The ETS-10 exhibits good thermal stability up to 550oC in air at ambient
pressure.
Anderson and co-workers (1994) has reported that the ETS-10 has a
composition of M2TiSi5O13 . nH2O (M = K or Na). The Si to Ti ratio of as-synthesized
ETS-10 was found approximately 5.0 and it is higher than ETS-4 that is only 2.7 (Kim
et. al., 2000). The Ti-O bonds in the – Ti – O – Ti – O – Ti  chains was found to be
alternately long and short as determined by EXAFS spectroscopy (Sankar et.al.1996).
The framework structure of ETS-10 is shown in Figure 2.7.
29
Si
Si
Si
Si
O
O
O
O
O
Ti
Ti
O
O
O
O
O
O
Si
Si
Si
Si
(a)
(b)
Figure 2.7: The ETS-10. (a) structure arrangement of ETS-10 where the red colour
represents the [SiO4]4- tetrahedra and the blue colour represents the [TiO6]8- octahedra
(b) molecular structure.
The ETS-10 has wide interesting properties, making it a potential material for a
wide range of industrial applications. ETS-10 had shown a great catalytic behaviour
especially on the removal of the heavy metals, such as Cu2+, Co2+, Mn2+, Zn2+, Pb2+ and
Cd2+ (Choi et.al., 2006; Choi et. al., 2006; Lu Lv et.al.,2007). ETS-10 also could be
applied as solid base catalyst in n-hexane reforming reactions (Philippou et.al., 1998),
Knoevenagel condensation (Goa et.al, 2004) and transestrification (Lόpez et.al., 2005;
Suppes et.al.,2004). The adsorption of argon, oxygen, and nitrogen gases were studied
on silver exchanged titanosilicate molecular sieve ETS-10. It was found that silver
exchanged ETS-10 would appear to be a promising potential adsorbent for the
production of high purity oxygen streams (Ansόn et.al., 2008). The ETS-10, transition
metal incorporated ETS-10 and ion-exchanged ETS-10 also shown a significant
photocatalytic activities in the decomposition of acetaldehyde (Uma et.al., 2004) and the
photo-oxidation of ethene (Krisnandi and Howe, 2006). The application of ETS-10 in
photocatalytic system has gained attention nowadays (Nash et.al., 2008).
30
2.7
CdS Composites
Semiconductor composition or known as coupling is one of the methods to
increase the hydrogen conversion efficiency. This process involves coupling two or
more semiconductors with different band gap energy.
When a large band gap
semiconductor is coupled with a small band gap semiconductor with a more negative
CB level, CB electrons can be injected from the small band gap semiconductor to the
large band gap semiconductor.
Thus, the photo-corrosion of small band gap
semiconductor could be prevented by a wide electron hole separation.
Successful
coupling of the two semiconductors for photocatalytic water-splitting hydrogen
production under visible light irradiation can be achieved when the following conditions
are met:
(i)
semiconductors should be photocorrosion free,
(ii)
the small band gap semiconductor should be able to be excited by visible
light,
(iii)
the CB of the small band gap semiconductor should be more negative
than the large band gap semiconductor,
(iv)
the CB of the large band gap semiconductor should be more negative
than redox potential of H+/H2 and
(v)
the electron injection from CB of small band gap semiconductor to large
band gap semiconductor should be fast as well as efficient.
Doong and co-workers (2001) have studied the CdS/TiO2 composite for 2chlorophenol degradation under UV irradiation. The combination of the two
semiconductors showed better photocatalytic activity due to better charge separation.
Similar response was observed when in the photo-degradation of 4-chlorophenol where
the coupling of CdS/TiO2 was more effective than CdS and TiO2 that were used
separately (Kang et.al., 1999).
31
Several reports on the CdS/TiO2 photocatalyst have been published in water
photo-splitting process. Optical absorption spectral analysis showed that CdS/TiO2
could absorbed photons with wavelength up to 520 nm (So et.al., 2004). Under visible
light illumination, CdS/TiO2 composite semiconductors produced hydrogen at a higher
rate than CdS and TiO2 that were used separately. The physicochemical properties of
bulk CdS/TiO2 composite photocatalysts and the optimization of their photocatalytic
activity of hydrogen production from water containing Na2S and Na2SO3 as a sacrificial
reagents under visible light irradiation (λ >420 nm) has been extensively studied
recently (Jang et.al, 2008; Zhang et.al., 2008).
The crystallinity of CdS is more
important than the crystallinity of TiO2 and the formation of CdS(bulk)/TiO2 composite
photocatalyst is a more effective strategy than single CdS photocatalyst (Jang et.al,
2007).
The photocatalytic activity depended significantly on modification techniques,
such as loading, proton exchange, and intercalation.
CdS intercalated composites
showed higher activity and stability. The formation of a „„nest‟‟ on the particle surface
promoted a uniform distribution and strong combination of the nano-sized particles on
the surface of catalysts. Shangguan (2007) has reported the photocatalytic activity of
K4Ce2M10O30 (M =Ta, Nb) in water photo-splitting reaction. It was found that the
activity of hydrogen generation of K4Ce2M10O30 was enhanced by the incorporation of
Pt, RuO2 and NiO as co-catalysts.
Nano-sized CdS particles embedded in ETS-4 zeolite nano-pores showed stable
photocatalytic activity in an aqueous solution containing Na2S and Na2SO3 electron
donors and the energy conversion efficiency (ECE) was improved by combining CdS
with ETS-4. The results suggest that the encapsulation of CdS in ETS-4 zeolite is
effective for separating charge-carriers photogenerated in CdS and for improving the
activity as well as the stability (Guan et.al, 2004; Guan et.al., 2005). LaMnO3/CdS
nanocomposite prepared by a reverse micelle method has been reported to show higher
photocatalytic activity than CdS solely. This result suggests that photogenerated holes
32
in the valence band of CdS could be transferred to the valence band of LaMnO3, which
can reduce the probability of charge-recombination and improve the activity (Kida et.al,
2003).
The mechanism of the holes transfer of LaMnO3/CdS nanocomposite as
illustrated in Figure 2.8.
Figure 2.8: Schematic diagram illustrated the holes transfer from valance band of CdS
to valance band of LaMnO3.
CHAPTER 3
EXPERIMENTAL
3.1
Apparatus and Special Equipments
For solution preparation, the apparatuses used were 50 mL and 100
volumetric flasks. Adjustable volume micropipette Eppendorf 100-1000μL with
plastic tips was used for accurate measurement of sample solution in small quantity.
Glass beakers of several sizes were used as container for non-silica compounds
during weighing. In addition, 50, 100 and 250 mL plastic beakers were used for
silica compounds.
Moreover, 100 mL and 250 mL two necked round bottomed flasks and three
necked round bottomed flasks were used as solution mixture container for
synthesization process that involve inert atmosphere. PTFE coated magnetic stirring
bars were used to ensure the mixture was evenly heated and mixed homogenously.
The 240 mL and 500 mL teflon-lined pressure vessel were used as container for
hydrothermal synthesis of titanosilicates, ETS-10. General centrifuge manufactured
by Kubota Japan was used to separate out the nanoparticles from its mixture. The
as-synthesized catalysts are kept in a vaccum desiccator at ambient temperature. All
glasses and ceramics apparatus are cleaned and dried prior to use.
34
3.2
Synthesis of CdS Nanoparticles by Reverse Micelle Method
A nano-sized CdS was synthesized based on previously reported methods
(Kida et.al., 2003; Guan et.al, 2004).
Experimentally, CdS nanoparticles were
prepared by a reverse micelle method using Triton-X (Scharlau, extra pure 98%) as
surfactants, n-hexanol (Aldrich-Sigma, 98%) as a co-surfactant and cyclohexane
(Merck, 99.%) as the solvent.
Cadmium nitrate hexahydrate, Cd(NO3)2.6H2O
(Scharlau, 99%) and sodium sulfide (Riedel-deHaën, 60-62%) were used as
cadmium and sulfide sources. At the final stage of the synthesis, tetrachloromethane
(Avondale Chemicals, 99.5%), ethanol (HmbG Chemicals, 95%) and double distilled
water was used as a solvent during the washing process of CdS precursor particles.
Microemulsions of A and B were first prepared, where microemulsion A and
B contained an aqueous solution of Na2S (0.4 M) and Cd(NO3)2 (0.4 M)
respectively. Each microemulsion composition consists of aqueous solution (10
wt%), Triton-X (20 wt%), n-hexanol (10 wt%) and cyclohexane (60 wt%).
Microemulsion B was then added to microemulsion A dropwisely and the mixture
was maintained at 5oC under ice bath with vigorously stirring during the reaction.
Bright yellow-orange colour indicates the formation of CdS precursor particles.
After 12 hours of aging period, the CdS precursor particles formed in the
microemulsions were collected and washed by centrifuging at 4000 rpm with the
several solvents. The sequence of the solvents used has to follow the polarity at the
ascending order: tetrachloromethane, ethanol and double distilled water for a total
more than 10 times. It is easily anticipated that if surfactant remains over the CdS
nanoparticles. The presence of the surfactant in CdS suspension will create large
amount of foam when the suspension is being shook. The final CdS nanoparticles
paste was dried in air for 3 days. Due to its fine nature, the dried CdS nanoparticles
were collected without further grinding.
35
3.3
Synthesis of CdS Nanoparticles by In-situ Sulphur Reduction Method
A CdS nanoparticle was synthesized based on the method reported by
Khanna and Subbarao (2004). Commercial cadmium acetate (Aldrich-Sigma, 98%)
and cadmium chloride (Comak, 99%) was used as cadmium source while
commercial sulphur powder (Emory, 99%) as the sulfide source.
N,N‟-
dimethylformamide (Merck, 99.8%) and dimethyl sulfoxide (Fisher Scientific,
99.8%) were distilled and dried separately by molecular sieve 4 Å in an amber
colour bottle prior to use.
Initially, cadmium acetate was dissolved in an appropriate ratio of N,N‟dimethylformamide in a round bottom flask. Sulphur powder was added to this
solution in 1: 1 ratio with respect to cadmium acetate. The reaction mixture was
then kept under nitrogen flow and stirred at 120 oC for 4 hours to obtain a bright
yellow suspension. After 4 hours of heating, the colloidal suspension was cooled
and filtered followed by washing with water and methanol. The final product was
dried in a desiccator overnight.
In order to study the effect of cadmium salts and solvents towards the
formation of CdS nanoparticles, the experimental procedure was repeated by using
cadmium chloride to replace cadmium acetate and dimethyl sulfoxide to replace
N,N‟-dimethylformamide for different batches of syntheses.
3.4
Synthesis of ETS-10 by Hydrothermal Method
Presently, ETS-10 was prepared by a method similar to the experimental
procedure reported by Liu and Thomas (1996) using commercial titanium oxide P25
produced by Degussa having the oxide composition 76 % anatase and 24 % rutile.
In this study, the rice husk ash (RHA) containing 97 % of SiO2 was used as silica
source. Sodium hydroxide (Merck, 99%) and potassium floride (GCE, 99%) were
introduced into the mixure in order to provide suitable extra-framework cations and
36
offered an alkaline medium for the growing process of ETS-10 precursor. For
comparison, the commercial colloidal silica, Ludox AM-30 (Aldrich-Sigma, 30%
SiO2) was used to replace RHA as silica source. The ETS-10 prepared with the
chemical composition of TiO2 : 3.75SiO2 : 1.5 NaOH : 0.54KF : 21.25H2O was
found most crystalline (Jei et.al, 2008).
A 8.00 g of P25 TiO2 was dispersed by stirring into 170.00 g of distilled
water. While stirring, 12.00 g of NaOH pellets and 4.30 g KF were added to the
TiO2 suspension. After 5 minutes of stirring, 30.00 g of RHA was added slowly into
the mixture while stirring vigorously. After 30 minute of stirring, the homogenized
slurry was transferred into a 500 mL teflon-lined pressure vessel which was sealed
tightly and heated in an oven for 52 hours at 220oC. The amount of raw materials
was reduced to half of the above quantity when the 240 mL teflon-line pressure
vessel was employed.
After 52 hours, the pressure vessel was directly cooled overnight to room
temperature.
The crystalline samples were filtered and washed to separate the
filtrate and impurities by using filter paper (Whatman). Due to the very tiny nature
of the product, the ETS-10 could be collected as a filtrate while the impurities
remain at the top of filter paper. The filtrate was then centrifuged at 3000 rpm and
washed twice by dispersing into double distilled water. The final white solid paste
remains in centrifuge tubes was dried in air for about 3 days. The dry sample cake
was ground with mortar and pestle into a fine powder and lastly, the ETS-10 powder
was collected in a plastic container and kept inside a desiccator for at least three days
to yield fully hydrated product.
3.5
Modification of ETS-10
The effect of hydrogen peroxide (Scharlau, 30%) treatment on ETS-10 and
CdS/ETS-10 was studied. The pure ETS-10 and CdS/ETS-10 samples was soaked
separately into an appropriate amount of hydrogen peroxide inside a round bottom
37
flask and the suspension was kept stirring overnight. The colour of ETS-10 was
changed from white to pale yellow during the reaction. The final samples were
collected by centrifuging, washed and dried at room temperature. The modified
ETS-10 and CdS/ETS-10 were labelled as METS-10 and CdS/METS-10
respectively.
3.6
CdS Nanoparticles Impregnated on ETS-10
Preparation of CdS nanoparticles supported on ETS-10 or METS-10 was
accomplished through impregnation by wet incipient technique. In this preparation,
four samples were prepared based on different ratios of CdS nanoparticles, 5wt%,
10wt%, 15wt% and 20wt% by adding 0.0263 g, 0.0555 g, 0.0818 g and 0.1081 g
into 0.5000g of ETS-10 respectively. Deionised water was used as aqueous medium
with the water to ETS-10 ratio of 1 to 3.
The mixture was stirred at room
temperature for 5 hours and dried on a hot-plate. The product was kept overnight in
an oven at 100oC to yield a fully dried samples.
3.7
Characterization Techniques
3.7.1
X-ray Diffraction (XRD)
All dried fine powder samples of CdS nanoparticles, ETS-10 and its
substitutes were spread on the sample holder. The ETS-10 and its substitute were
measured in a 2 range of 5 to 50 degrees at room temperature with the step interval
of 0.05 degrees and step time of 1 second per step.
The as-synthesized CdS
nanoparticles and CdS bulk sample were measured in a 2 range of 20 to 70 degrees
at room temperature with the step interval of 0.05 degrees and step time of 1 second
per step. The X-ray power diffractogram patterns was recorded with a Bruker
38
Advance D8 with Siemens 5000 diffractometer, using Cu Kα radiation (= 1.5418Å,
40 kV, 40 mA). The EVA software was used to calculate the integrated intensity of
the selected peaks.
3.7.2
Fourier Transform Infrared (FTIR) Spectroscopy
KBr pellet technique was used in samples preparation with the ratio of the
sample to KBr at 1 to 100. The mixture was ground homogenously by using pestle
and mortar and pressed by hydraulic press at 5000 psi into a transparent pellet. The
pellet was put in a sample holder and the spectrum was recorded. The FTIR spectra
were recorded by using a Perkin Elmer Spectrum One FTIR spectrometer with a 4
cm-1 resolution and 10 scans in the mid IR region (400-4000 cm-1).
3.7.3
Diffuse Reflectance UV-Vis (DR-UV) Spectroscopy
The fine powder samples were spread over the sample holder quartz window
and sealed nicely and tightly. The quartz window at the female nut has to be kept
dried and clean without any fingerprint during the sample preparation. A barium
sulfate (BaSO4) standard was used as reference for background scanning. The
spectra were represented as the Kubelka-Munk function versus wavelength. The
diffuse reflectance UV-Vis Spectra were recorded by using a Perkin Elmer
Instrument Lambda 900 Ultraviolet-Visible-Near IR Spectrometer.
The scan
parameter was set with slit size 2 nm and scan in the UV region (200-700 nm).
3.7.4
Field Emission Scanning Electron Microscopy (FESEM)
The small amounts of dried samples were dispersed as a thin layer on
conductive carbon tap aluminium stubs and keep in desiccators to prolong the
39
storage of the sample in dried condition. The images of the gold-coated samples
were collected on a Zeiss Supra 35VP FE-SEM with accelerating voltage 15 kV and
6 mm working distance in the secondary electron imaging mode. Samples were
magnified until a clear visual image was obtained. The dimensions and size of the
crystals were estimated by the visual analysis of the FE-SEM images of the products.
3.7.5
Energy Dispersive Spectroscopy (EDAX)
The elemental ratio was measured using energy dispersive X-ray
spectroscopy utilizing an Oxford Instrument INCA X-sight analyzer equipped with a
Sapphire super ultra window detector attached to the Zeiss Supra 35VP FE-SEM.
The accelerating voltage of the electron gun was set to 15 kV. Uncoated samples
were dispersed on aluminium specimen stubs coated with conductive carbon tap.
For each sample, a measurement at different magnifications was performed at least 3
times. The cadmium to sulfur ratio was measured for CdS nanoparticles samples. In
a meanwhile, the silicon to titanium ratio was measured for crystalline ETS-10.
3.7.6
Transmission Electron Microscopy (TEM)
A small quantity of as-synthesized samples was added in an appropriate
amount of acetone and the mixture was dispersed by ultrasonification for about 15
minutes. A small amountt of the ultrasonicated mixture was dropped on a Formvar
film coated 300 mesh copper grid. The TEM micrographs of the samples were
recorded by Oxford Instrument JEM-2100 Electron Microscope JEOL at 160 kV
accelerate voltage.
40
3.8
Experimental Set-up for Photocatalytic Testing
The micro-reactor frame was constructed from painted iron bar of 25 mm x
25 mm x 1.6 mm hollow section. This material was selected for rigidity to the
reactor so that the measurement of gas flow of argon in the cylinder vessel will not
be disturbed during the reaction. The front view of the micro-reactor was fabricated
by a plate of stainless steel of 3 mm thickness with the four holes size of 1.5 cm and
a hole size of 2.3 cm at the related measurement as shown in Figure 3.1. The
different dimension of holes was designed for the installation of flow meter, 2-ways
valves and 6-ways valve. An extra iron bar with similar dimension was linked 10
cm from the top so that it can be used to set up a lamp for the pyrex reaction vessel
from top. The rear view of the micro-reactor frame is shown in Figure 3.2.
Figure 3.3 shows the diagram of experimental set-up for hydrogen detection
in 2-D dimension. The design is then further fixed on the stainless steel plate located
in the micro-reactor frame. The flow meter was vertically installed in the 2 holes
located 8 cm from left while two pieces of the 2-ways valves were installed in the
holes 10 cm from top, respectively. The 6-ways valve was finally fixed on the hole
size of 2.3 cm and the internal gas stream was developed by 1/8 inch stainless steel
tubing. The gas line was tested by Snoop Test and was proved to be free of leakage.
The reaction vessel was made from pyrex glass with the glass tubes 0.9 cm
OD as the connector to the stainless steel gas stream line. The dimension of reaction
vessel and the glass tube is further shown in Figure 3.4.
During the reaction,
molecular sieve 4 Å (Merck) was filled in between the fine quartz wool (TOSH)
inside the glass tube. The water photo-splitting reaction was conducted inside the
reaction vessel under a continuous argon gas flow. Argon gas was chosen as the
carrier gas as well for gas chromatography coupled thermal conductivity detection
(GC-TCD) due to its high sensitivity towards the hydrogen gas detection.
41
10
41 20
10
18
28
50
35
8
40
8
30
50
Figure 3.1: Front view of microreactor frame. Grey dots indicate hole size 1.5 cm
whereas black dot indicate hole size 2.3 cm. (scale in cm)
10 50
15 30
50
Figure 3.2: Rear view of microreactor frame. (scale in cm)
GC-TCD
Check valve
Pyrex glass tube
Argon
6-ways valve
Regulator
2-ways valve
Argon
Pyrex reaction vessel
Figure 3.3: Diagram of experimental set-up for catalytic testing
42
43
0.9 cm
3
cm
3
cm
4
cm
(a)
8
cm
0.9 cm
(b)
20 cm
Figure 3.4: dimension of (a) pyrex reaction vessel (b) pyrex glass tube
3.9
Catalytic Testing
The water photo-splitting reaction was conducted by adding photocatalyst
(0.1 g) in an alkaline solution of S2-/SO32- (50 mL). The sulfide ions were supplied
from sodium sulfide, whereas the sulfite ions were supplied from sodium sulfite
(Sigma-aldrich, 98+%). The alkaline solution contains Na2S (0.1M), Na2SO3 (0.5M)
and NaOH (1.0 M) dissolved properly in double distilled water (50 mL). This
alkaline solution was transferred into the pyrex reaction vessel after which the
catalyst (0.1 g) was added into the reaction vessel before attaching to the gas stream
line.
The effect of the light source on the hydrogen generation was studied by
several commercial visible light sources including: Kwzone submersible lamp (6W),
Sunkyo cool daylight lamp E27 (18W), Phlight halogen lamp (150 W) and Phlight
halogen lamp (500 W). It was found that only Phlight halogen lamp is active as the
irradiation source for the water photo-splitting reaction. This probably due to higher
energy powers of the Phlight halogen lamp that lead to more successful electron
excitation. In this reaction, the photocatalytic activity of the series of catalyst was
tested under illumination of Phlight halogen lamp. The volume of hydrogen gas
evolved was collected by water displacement method and the average rate of
hydrogen generation was calculated.
44
The catalyst regenerateability was tested in three cycles. After the first
usage, the suspension of the catalyst was centrifuged, washed and dried at 80oC for
24 hours. The post-treated sample was re-applied as the catalyst into a fresh load of
the aqueous alkaline S2-/SO32- solution for another two times. The hydrogen gas
liberated was recorded at 2 hours interval for the first 8 hours and last point at 24
hours. The reusability of the catalyst was observed.
The hydrogen gas collected in the unit of centimeter cubic was converted into
the unit of mol via the ideal gas law as shown in equation 3.1; where, P is the
absolute pressure of gas, V is the volume of gas, n is the number of mole of gas, R is
the universal gas constant and T is the absolute temperature.
The conversion
efficiency of hydrogen evolved was demonstrated in the function of hydrogen gas
volume over 0.1 g catalyst (μmol/0.1 g) versus time (hr)
PV = nRT
3.9.1
(3.1)
Hydrogen Gas Calibration by GC-TCD
The standard gases calibration was conducted to determine the retention time
of the particular gases on a chromatogram. The retention peaks of the hydrogen and
oxygen gases were confirmed by using the ultra purified hydrogen gas and industrial
oxygen gas supplied from MOX-Linde Gases Sdn. Bhd.
3.9.2
Photocatalytic Testing by Using Microreactor
The hydrogen generation from water was carried out in the continuous flow
micro-reactor. The powder sample (0.1 g) was added into an alkaline solution-filled
pyrex reaction vessel and attached to the gas stream line. The whole gas line was
pre-treated under argon flow at room conditions for 20 minutes. The water photo-
45
splitting reaction was conducted by light irradiation from the bottom of the reaction
vessel. Since reaction vessel was made from quartz, therefore it is assumed that no
incident photon in the region of visible is being absorbed by the reaction vessel
glass. However, the lost of irradiation to the surrounding is inevitable. Mass flow
rate was set at 50 mL/min for the pre-treatment and 20 mL/min during the reaction.
The products of the water photo-splitting reaction were directly separated and
detected by an online Agilent 7890N gas chromatograph equipped with TCD and
packed column (Supelco, 13 X molecular sieves). The purity of hydrogen gas
produced was determined based on appearance of the hydrogen peak on
chromatograms.
CHAPTER 4
RESULTS AND DISCUSSION
4.1
Characterization of the Photocatalysts
4.1.1
Preparation of the CdS Nanoparticles
CdS nanoparticles were synthesized by both reverse micelle and in-situ
sulphur reduction methods.
It was found that the colour intensity of CdS
nanoparticles prepared by reverse micelle method (CdS-RM) was similar to the
commercial CdS (Sigma-Aldrich) which was yellowish-orange.
Whereas, CdS
nanoparticles synthesized by in-situ sulphur reduction method (CdS-IS) gave a
bright yellow solid. Both as-synthesized CdS nanoparticles appear as a very fine
powder after drying process without any further mechanical grinding.
Solvent affects toward the CdS formation were studied in the in-situ
reduction method.
N,N‟-dimethylformamide (DMF) and dimethyl sulfoxide
(DMSO) were used as solvents appropriately in the different batches of the
syntheses. Both solvents gave a good dissolution effect thus CdS nanoparticles
could be formed effectively. DMF is known as a dipolar aprotic solvent with a high
dielectric property. The hydrogen gas can be generated in the presence or absence of
a suitable catalyst when DMF is heated with water (Yu et.al., 1990). The suggested
mechanism of the reaction is shown in the equations below. The reaction involved is
47
the reduction water that leads to the formation of hydrogen gas wherein, this
hydrogen gas causes the in-situ reduction of sulphur (Khanna and Subbarao, 2004).
HCONMe2 + H2O  Me2NCOOH + H2
(4.1)
H2 + S  H2S
(4.2)
2+
+
+ H2S  CdS + 2H
(4.3)
Me2NCOOH  CO2 + Me2NH
(4.4)
Cd
When DMSO was applied as solvent, two main reaction mechanisms were
proposed by Rivka and co-worker (2001). The first suggestion for the mechanism
was a reduction of Cd2+ to Cd0, followed by a chemical reaction of Cd0 with the
dissolved sulphur to give CdS. In contrast, the second mechanism suggested a
different pathway including a reduction of S to S2-, followed by a chemical reaction
with Cd2+ to form CdS. It is possible that both mechanisms operate to a greater or
lesser extent depending on the reaction conditions.
The effect of cadmium salts toward the reaction has also been studied by
using cadmium acetate and cadmium chloride, respectively. When cadmium acetate
was used as the cadmium source, homogenized bright yellow solid formed after 4
hours of reaction time. This bright yellow solid indicates the formation of cadmium
sulfide. This method gave the highest product yield more than 60% compared to the
other methods. The reverse micelle method using cadmium nitrate as cadmium
source only gave the product yield of 50%.
During the reaction of cadmium chloride with sulphur, it was observed that
the original pale yellow colour solution attributed by the sulphur remain unchanged
after heating process was extended to 4 hours. This observation suggested that there
was no significant sign to indicate the formation of CdS nanoparticles by using
cadmium chloride as the cadmium source. It was probably due to the least solubility
of cadmium chloride in DMF and DMSO solvents.
48
The as-synthesized CdS were used as a reference rather than calcined
samples in characterization and photocatalytic testing. This is due to the CdS fact
that the cubic phase as-synthesized CdS was found to be thermally unstable and
could easily undergo phase transformation from cubic to hexagonal phase (Tetsuya
et.al., 2003). Furthermore, the calcination of CdS might results decrease in the
photocatalytic activity due to the increase in the particle size and the decrease in the
surface area. The heat treatment of the CdS samples might also lead to the structural
defection.
4.1.2
Physico-chemical Studies of CdS
All elemental data were collected in iterations of three by using Energy
Dispersive Spectroscopy (EDAX). There are slightly different of calculated atomic
percentage compared to the EDAX elemental analysis. As can be seen in Table 4.1,
the calculated atomic percent of Cd and S ratio in CdS-RM showed a close ratio to
the EDAX elementals analytical data. The application of microelmusion technique
in the preparation of CdS-RM effectively controls the Cd and S ratio. The EDAX
spectrum of the as-synthesized CdS nanoparticles is shown in Figure 4.1
Table 4.1: The elemental analysis for CdS
CdS
Calculated atomic%
EDAX atomic %
Cd
S
Cd
S
In-situ reduction method
50.00
50.00
56.49
43.51
Reverse micelle method
50.00
50.00
50.54
49.46
49
Figure 4.1: The EDAX spectrum for as-synthesized CdS sample.
The formation of CdS nanoparticles was studied by using X-ray Diffraction
(XRD) technique. It was found that crystallinity of CdS prepared from cadmium
acetate by in-situ reduction method is similar to the XRD pattern of CdS prepared
from cadmium nitrate by reverse micelle method. The XRD patterns indicated that
these nanocrystals were crystallized in the cubic structure with the lattice constants
a= 0.5818 nm. Both samples display reflection peaks (111), (220) and (311) lattice
planes of cubic CdS that corresponded well with previous research (Yang et.al.,
2005; Guan et.al., 2005; Khanna et.al., 2004). The appearance of broad humps
suggested the formation of nano-sized crystallites. The XRD patterns of the CdS
synthesized by in-situ reduction method and reverse micelle method are shown in
Figure 4.2.
Besides that, it was also found that CdS nanoparticles could not be produced
when cadmium chloride was used as cadmium source. This suggestion was further
confirmed by the XRD pattern as illustrated in Figure 4.3 (b) in which no lattice
plane corresponded to CdS was observed. This was probably due to the least
solubility of cadmium chloride in solvents medium and lead to the low degree of
ionization. Unlike cadmium chloride, cadmium nitrate and cadmium acetate were
found more readily soluble. The sharp peaks in the region of 20-30o might be
attributed to other crystalline structure that is still unidentified.
311
220
Intensity (a.u.)
111
50
(c)
(b)
(a)
2-Theta (o)
Figure 4.2: XRD pattern of the CdS nanoparticles synthesized from cadmium
acetate by in-situ reduction method with different solvents (a) DMF (b) DMSO. (c)
Intensity (%)
CdS synthesized from cadmium nitrate by reverse micelle method.
(b)
(a)
2-Theta (o)
Figure 4.3: XRD patterns of CdS nanoparticles synthesized with different cadmium
salts. (a) cadmium acetate (b) cadmium chloride.
51
The surface morphology and particles size of as-synthesized CdS was
evaluated by using Field Emission Scanning Electron microscopy (FESEM). As can
be seen in Figure 4.4, both of the as-synthesized CdS showed similar surface
morphology characteristic in term of shape and particle size. The aggregation and
agglomeration was observed on some portion of the samples due to the heat
treatment at 100oc during drying process. Under a magnification of 80000 times, it
was found that the CdS particle size exists in the particle range of 25-40 nm. This is
in a good agreement with XRD result, which shows the broad humps at 3 different
reflection points. However, the shape of the particle could not be confirmed due to
the low resolution of FESEM.
(a)
(b)
100 nm
100 nm
Figure 4.4: FESEM micrographs of as-synthesized CdS (a) CdS-IS (b) CdS-RM
The morphology of the samples was further observed by using Transmission
Electron Microscopy (TEM). From the TEM micrographs, as-synthesized CdS was
found to be crystalline form in cubic shape with the particles size of 8-10 nm. This
evaluation was found to be totally different with the FESEM analysis. The variation
of the observation is due to the water elimination step of CdS during the preparation
of stub for FESEM analysis. However, the drying process in preparing stub for
FESEM analysis is inevitable. Thus, the TEM result is more reliable since the
sample preparation of Formvar film coated sample grid does not involve any heat
treatment. The ultrasonification step for 15 minutes in acetone further facilitates the
52
separation of CdS particles. The TEM micrographs of the as-synthesized CdS at
different magnifications are shown in Figure 4.5.
Figure 4.5: TEM micrographs of as-synthesized CdS
A pure CdS nanoparticles are not active in the infrared region (Rivka et.al.,
2001). Therefore, no report has been published regarding the infrared vibration data
for CdS nanoparticles.
4.1.3
Physico-chemical of ETS-10
Elemental data of ETS-10 synthesized from rice husk ash (RHA) and Ludox30 was collected by EDAX analysis. It was found that the calculated Si/Ti starting
gel ratio for the RHA synthesized ETS-10 shows the nearest Si/Ti ratio to the EDAX
analysis (Table 4.3). The unreacted silica in RHA could be easily separated from the
ETS-10 suspension during filtration process. It can be collected as the residue
whereas the pure ETS-10 was collected as the filtrate.
53
Table 4.2: The elemental analysis for ETS-10
ETS-10
Calculated Si/Ti ratio
EDAX Si/Ti ratio
Ludox-30
3.75
5.14
RHA
3.75
3.79
The crystallinity and the purity of as-synthesized ETS-10 were tested using
XRD.
It was found that highly crystalline ETS-10 with high purity could be
obtained by hydrothermal synthesis under condition of reaction at 220 oC for 52
hours. The XRD pattern of the product corresponded well with previous report
(Chapman, 1990). The starting gel of ETS-10 consists of molar composition of TiO2
: 3.75SiO2 : 1.5NaOH : 0.54KF : 21.25H2O.
Figure 4.6 shows the XRD
diffractograms for ETS-10 synthesized by using RHA and Ludox-30 as silica
sources. It is noticeable that a small amount of unreacted P25 TiO2 is still remaining
in ETS-10 synthesized by RHA. It might be due to slightly incomplete reaction.
However, the priority of RHA as silica source remains unchanged for low cost
production of ETS-10.
Intensity (a.u)
(c)
(b)
*
(a)
5
10
20
30
40
2-Theta (o)
Figure 4.6: XRD patterns of (a) P25 TiO2 (b) ETS-10 from RHA (c) ETS-10 from
Ludox-30. Peaks assigned to P25 TiO2 are indicated by asterisks.
54
The surface morphology of the as-synthesized ETS-10 was characterized by
FESEM. It can be seen that both of ETS-10 samples from different Si source
displays truncated bipyramid structure. The formation of ETS-10 is believed to be
related to the fine particles of P25 TiO2 and Ludox-30 or RHA, which directly led to
the nucleation of ETS-10. ETS-10 synthesized from RHA showed a small amount
of anatase TiO2. This can be seen in micrograph of Figure 4.7 (b) where the
appearance of impurity presents in the sample. It might be due to the incomplete
reaction or structure distortion since ETS-10 is one of the meta-stable microporous
titanosilicates. Figure 4.7 (a) and (b) presents the FESEM micrographs for ETS-10
under magnification of 25000 times.
Generally, the ETS-10 particles distribute in a wide range of size of 200 to
500 nm. Ten specimens of both of the samples were measured and the average
crystal dimensions along a, b and c directions were determined. The dimension of
ETS-10 synthesized from RHA gave 407±49 nm, 207±62 nm and 142±36 nm in the
direction of a, b and c, respectively. Similar dimension for the sample ETS-10
synthesized from Ludox-30 was observed for a, b and c which were 433±47 nm,
225±69 nm and 139±24 nm.
(a)
200 nm
(b)
200 nm
Figure 4.7: FESEM micrographs of ETS-10 synthesized by different silica sources
(a) RHA (b) Ludox-30
55
Figure 4.8 presents the IR spectra for ETS-10 synthesized from Ludox-30
and RHA, respectively. In general, both of the samples showed a similar vibration
bands. The fundamental vibration of octahedral [TiO6]8- and tetrahedral [SiO4]4occurred in mid-infrared region between 450-1300 cm-1. The position of the major
vibration mode matched well with the literature value (Das et.al., 1995). The high
frequency range (above 800 cm-1) of the spectra is dominated by SiO bond
stretching modes, which generate a stronger and intense band. Whereas, the SiO
bending and TiO stretching modes are likely to appear in the middle frequency
range. The details of IR spectra and the type of vibration present in ETS-10 are
summarized in Table 4.3.
657
T (%)
(b)
746
653
555
449
(a)
1026
747
554
447
1028
1400
1200
1000
800
600
400
-1
Wavenumber (cm )
Figure 4.8: IR spectra for ETS-10 synthesized by different silica sources (a) Ludox30 (b) RHA.
56
Table 4.3: The details of the wavenumber and the type of vibration present for ETS10 synthesized from Ludox-30 and RHA.
Wavenumber for ETS-10 (cm-1)
Type of Vibrations
Ludox-30
RHA
Si-O stretching
1028
1026
Ti-O stretching
747
746
Ti-O stretching
653
657
Si-O rocking, O-Ti-O bending
554
555
O-Si-O bending, O-Ti-O bending, Ti-O stretching 447
449
4.1.4
Physico-chemical Studies of CdS/ETS-10
CdS supported on ETS-10 was prepared by a wet impregnation method. It
was found that the colour of ETS-10 changed from white to pale yellow after loading
with CdS. This colour changes served as an indication to successful attempt of
incorporating CdS on the ETS-10. The colour becomes more intense with the
increase in CdS loading onto ETS-10.
From an elemental analysis, elements of sulphur and cadmium could be
detected in the samples of CdS/ETS-10. This indicates the presence of cadmium and
sulphur in the ETS-10 samples.
The EDAX spectrum of the 10 wt% of CdS
supported on ETS-10 is shown in Figure 4.9.
57
S
Figure 4.9: The EDAX spectrum for 10CdS/ETS-10 sample.
The crystallinity and lattice plane of ETS-10 was studied after loading of
CdS nanoparticles. From the XRD pattern of CdS/ETS-10 (Figure 4.10), a slight
drop in intensity was noticed by a comparison with the pure ETS-10. This was
suggested from the decrease in crystallinity degree after the modification process.
As can be seen here, there is no any formation of new peak observed. This probably
suggested the absence of extensive formation of CdS over the ETS-10 framework.
The addition of CdS in ETS-10 does not affect the lattice structure of ETS-10. This
observation corresponded well with the previous publication by Guan and coworkers (2005). It is suggested that CdS nanoparticles seated adjacently besides the
ETS-10 without causing pore blockage. The surface contact between CdS composite
would definitely promote the electrons transfer from conduction band of CdS to the
conduction band of ETS-10 under light illumination.
58
Intensity (a.u)
(c)
(b)
(a)
2-Theta (o)
Figure 4.10: XRD patterns of samples (a) ETS-10 (b) CdS (c) CdS/ETS-10
Figure 4.11 (a) and (b) shows the FESEM micrographs for samples with the
5 wt% and 10 wt% of CdS nanoparticles supported on microporous ETS-10. It was
found that the CdS nanoparticles disperse well in the external framework of ETS-10.
The pore blockage could be prevented due to the pore of ETS-10 approximately 0.8
nm while the diameter of as-synthesized CdS averagely around 30 nm. A large
crystal in the shape of hexagonal was observed in the sample 10CdS/ETS-10. This
hexagonal crystal attributes to the agglomeration of CdS nanoparticles under heat
treatment during the sample preparation for FESEM scanning. The agglomeration
process does not occur in 5CdS/ETS-10 due to the low ratios number of CdS to ETS10.
59
(a)
(b)
200 nm
200 nm
Figure 4.11: FESEM Micrographs of the samples (a) 10CdS/ETS-10 (b) 5CdS/ETS10
The interaction between CdS coupling materials was also studied by TEM
analysis as illustrated in Figure 4.12.
A large truncated bipyramid shape was
assigned to the ETS-10 and the small spots in its surrounding were CdS
nanoparticles. CdS nanoparticles were found became larger when attached to ETS10. This was probably due to the formation of CdS cluster arrays when in contact
with pore of ETS-10. This CdS clusters were interconnected electronically by a
quantum tunneling effect, forming CdS cluster arrays with the geometric structures
imposed by internal pore structure of zeolite. Similar interaction was observed in
CdS/Zeolite Y system (Yahiro et.al., 2002).
Figure 4.12: TEM micrographs of CdS/ETS-10
60
4.1.5
Photo-absorption Properties of CdS
The diffuse reflectance UV-vis spectra of all samples were recorded in
Kubelka Munk unit.
UV-visible absorption spectra of CdS showed a strong
absorption edge at 488 and 532 nm for CdS-IS and CdS-RM, respectively in Figure
4.13. It was found that the absorption edge of the as-synthesized CdS is apparently
blue-shifted compared to bulk commercial CdS which occurred at 538 nm. This
phenomenon of blue shift of absorption edge has been ascribed to an increase in
band gap energy of semiconductor. In semiconductor, it is well known that band gap
energy increase with the decrease in particles size (Yang et.al.,2005). With the
decreasing radius, the onset of absorption is shifted to higher energy due to the size
quantization effect.
As-synthesized CdS demonstrated a high light absorption
capability in visible region, making it a potentially important material as visible-light
driven photocatalyst. In addition to blue shift, a certain degree of band broadness
was also observed, most probably due to the wide distribution of particle size and
some crystal lattice distortion (Yang et.al., 2005).
10
9
8
7
K-M
6
(c)
5
4
(b)
3
2
1
(a)
0
400
420
440
460
480
500
520
540
560
580
wavelength (nm)
Figure 4.13: DR-UVspectra (a) CdS-IS (b) CdS-RM (c) bulk CdS
600
61
4.1.6
Photo-absorption Properties of ETS-10
The diffuse-reflectance UV absorption spectra of the ETS-10 prepared using
RHA and Ludox-30 as silica sources are shown in Figure 4.14.
Two obvious
absorption bands were observed in the regions of 260 nm and 280 nm.
The
absorption band at 260 nm indicated the presence of Ti atom bonded to four silica
tetrahedra. This band was assigned to the charge transfer from the Si, Ti-linking
oxygen atoms to the Ti(IV) central atom in directions perpendicular to the -Ti – O –
Ti – O- chains. The 280 nm absorption band was attributed to the charge transfer
within the -Ti – O – Ti – O- chains of the ETS-10 structure (Das et.al., 1996;
Philippou et.al., 2000).
There is a weak shoulder band at 370 nm observed in the DR-UV spectra of
the ETS-10 synthesized from RHA, assigned to the existence of a small amount of
anatase TiO2 phase.
This result corresponded well with the XRD pattern and
FESEM analysis of ETS-10.
The photo-absorption properties of the hydrogen peroxide treated ETS-10
(METS-10) was also studied for the comparison purposes. The METS-10 that
underwent chemical treatment showed the colour changed from white to pale yellow
colour. This sample showed wide band broadness in the DR-UV spectra, probably
due to the some distortion of crystal lattice during the chemical treatment.
K-M
62
5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
296
(c)
254
(b)
260
280
278
(a)
370
200
250
300
350
400
450
500
Wavelength (nm)
Figure 4.14: DR-UV spectra of (a) ETS-10 from RHA (b) ETS-10 from Ludox-30
(c) METS-10
4.1.7
Photo-absorption Properties of CdS/ETS-10
The diffuse-reflectance UV absorption spectra of the as-synthesized CdS,
ETS-10 and ETS-10 supported CdS with the 5 % and 10% CdS loaded samples are
shown in Figure 4.15. It is noticeable that pure ETS-10 is not active in the visible
region. However, it gave a broad absorption band up to visible region after the
modification process by attaching small amounts of CdS on the ETS-10 surface.
This result was found to be correspond well with other types of CdS composites such
as CdS/TiO2 (Kang et.al., 1999; Doong, et.al., 2001) and CdS/ETS-4 (Guan et.al.,
2004).
63
8
CdS
ETS-10
10CdS/ETS-10
5CdS/ETS-10
7
6
K-M
5
4
3
2
1
0
200
250
300
350
400
450
500
550
600
Wavelength (nm)
Figure 4.15: Diffuse reflectance UV-vis spectra of the samples.
4.1.8
Band Gap Studies
The band gap of as-synthesized CdS samples were calculated from the linear
correlation of [F(R) hυ]2 and hυ (Guan et.al., 2004; Guan et.al., 2005). The function
of F(R), represents the light reflection intensity of the samples and the details could
be obtained as Kubelka-Munk unit. The function of hυ, represents the energy axis
with the function of h as the Planck‟s constant
The intercept at energy axis gave the band gap energy of CdS-RM to be 2.40
eV which is lower than CdS-IS at 2.58 eV. Whereas, the calculated band gap energy
of commercial CdS was found to be the lowest (2.36 eV), probably due to the size
quantization effect since the commercial CdS was prepared in micro-sized region.
Larger particle size of the CdS ascribed to the narrow band gap energy.
However, factors of quantization effect towards the band gap energy could
not be applied for the as-synthesized CdS. This is due to these samples having been
proved to be present in similar particles dimension by using microscopic techniques.
64
The differences of calculated band gap value of CdS are due to the remaining initial
reagents on the samples. For example, the use of Triton X-100 as surfactant and
hexanol as co-surfactant in the preparation of CdS-RM precursor might not be fully
removed during the washing process. The calculated and extrapolated band gap
energy of CdS samples is shown in Figure 4.16.
700
600
[F(R) hv]^2
500
(c)
400
(b)
300
200
(a)
100
0
2.00 2.05 2.10 2.15 2.20 2.25 2.30 2.35 2.40 2.45 2.50 2.55 2.60 2.65 2.70 2.75 2.80 2.85 2.90 2.95 3.00
hv (eV)
Figure 4.16: The band gap studied of CdS samples (a) CdS-IS (b) CdS-RM (c) bulk
CdS
The band gap of ETS-10, METS-10 and ETS-10 supported CdS were
calculated with the method as described above. ETS-10 supported CdS shows a
wide absorption range from visible range up to ultra violet range. The band gap
energy of pure ETS-10 was found to be 4.03 eV. As can be seen in Figure 4.17,
ETS-10 in the 10CdS/ETS-10 also has the same amount of calculated band gap
energy. This observation signified that the coupling work does not alter the band
gap energy of ETS-10.
However, when ETS-10 underwent hydrogen peroxide
treatment, the band gap energy was found to decrease drastically to 3.41 eV. The
formation of new species on the ETS-10 surface such as TiOH was proposed.
65
400
350
[F(R)hv]^2
300
250
(c)
200
150
(b)
100
(a)
50
0
3.00 3.10 3.20 3.30 3.40 3.50 3.60 3.70 3.80 3.90 4.00 4.10 4.20 4.30 4.40 4.50
hv (eV)
Figure 4.17: The band gap studied of (a) ETS-10 (b) 10CdS/ETS-10 (c) METS-10
As mentioned earlier, the CdS coupling ETS-10 does not alter the band gap
energy of ETS-10. However, the band gap energy of CdS-IS in CdS/ETS-10 was
noticeable slightly reduced when calculated using the same approach as described
above. This is due to the slightly increase in the particle size of CdS where CdS
forms cluster arrays when contact with the pores of ETS-10. This statement is
further supported by TEM micrographs.
As can be seen in Figure 4.18, the
significance of the band gap energy shifted was highly depended on the amount of
CdS loaded on ETS-10. The more CdS portion in the CdS composites, the more it is
likely to be as untainted CdS band gap energy. The band gap energy of pure CdS-IS,
10CdS/ETS-10 and 5CdS/ETS-10 were calculated to be in a descending order of
2.58 eV, 2.56 eV and 2.55 eV, respectively.
66
6
5
[F(R)hv]^2
(c)
4
(d)
(b)
3
2
1
(a)
0
2.20 2.25 2.30 2.35 2.40 2.45 2.50 2.55 2.60 2.65 2.70 2.75 2.80 2.85 2.90 2.95 3.00
hv (eV)
Figure 4.18: The band gap studied of (a) ETS-10 (b) 10CdS/ETS-10 (c) 5CdS/ETS10 (d) CdS-IS
4.2
Photocatalytic Activity
4.2.1
Hydrogen Detection by GC-TCD
Hydrogen generation of water photo-splitting was conducted in a pyrex
reaction vessel with the light irradiation from bottom. The whole system was pretreated and saturated with argon gas before the reaction was start. The hydrogen gas
generated was then separated and analyzed with online GC-TCD in the argon as the
carrier gas. Figure 4.19 presents the outlook view of the micro-reactor coupled
online GC-TCD.
67
The hydrogen gas evolution was tested in the presence of two types of
additive added into water. The groups of the additives are classified as organic hole
scavenger represented by methanol and inorganic hole scavenger represented by a
mixture of sulfide and sulfite (S2-/S2O32) alkaline solution. In was found that when
methanol was added in the ratio of 1 to 1 with water, the volume of hydrogen gas
generated was negligible. However, there were significant hydrogen gas observed
when the S2-/S2O32- was added into the alkaline solution. The evolution of hydrogen
gas was confirmed by GC-TCD, where the hydrogen gas peak occurred at the
retention time of 0.910 minute as illustrated in Figure 4.20. The blank was ascribed
to the reaction condition with the absence of photocatalyst.
Figure 4.19: The view of the micro-reactor coupled online GC-TCD
1.200
H2
0.910
1.435
68
Blank
H2O + catalyst + hv
Figure 4.20: Chromatogram of water photo-splitting.
4.2.2
The Mechanism Study of Water Photo-splitting
The water photo-splitting reaction was tested in both the gas-solid phase and
liquid-solid phase. No reaction occurred when the photocatalytic activity conducted
in a gas-solid phase, probably due to the lesser contact time between the water with
the catalyst. In contrast, significant amount of hydrogen was produced in liquidsolid phase. The liquid-solid phase offer several advantages, such as high extinction
coefficients, fast carrier diffusion to the interface and suitable positioning of valance,
conduction bands to achieve high efficiencies in light energy conversion processes
(Schiavello, 1985).
CdS is not stable in water because of the anodic photocorrosion, which leads
to the formation of sulfate ions and sulfur in the presence and absence of oxygen
respectively (Equations 5.1 and 5.2). A quantitative analysis showed that four holes
are consumed to form one sulfate ion in the presence of oxygen gas (Meissner et.al.,
69
1986). At the condition of oxygen-free, the corrosion of one CdS molecules is
caused by the removal of only two holes instead of four with the presence of oxygen.
CdS + 4h+ + O2 + 2H2O  Cd2+ + SO42- + 4H+
(5.1)
CdS + 2h+  Cd2+ + S0
(5.2)
In general, the decomposition of sulphur on catalyst surface could be
prevented in the presence of sacrificial reagents of S2-/S2O32- as hole scavengers.
The sacrificial reagents make up half of the water-splitting reaction and they were
routinely used to make sure that the photocatalyst can reduce H2O to H2. Sacrificial
electron donors worked irreversibly by consuming the photogenerated holes at the
valence band of CdS nanoparticles. Thus the recombination number of electrons and
hole pairs could be minimized. The net valence band process at CdS nanoparticles is
shown in the Equations of 5.3 to 5.7:
2hvb+ + S2- + SO32-  S2O32-
(5.3)
2hvb+ + 2S2O32-  S4O62-
(5.4)
1.5H2O + S4O62-  SO32- + 1.5S2O32- + 3H+
(5.5)
2hvb+ + 1.5H2O + 0.5S2O32-  SO32- + 3H+
(5.6)
1.5H2O + 1.5S2O32-
hυ
2SO32- + S2- + 3H+
(5.7)
The suggested schematic diagram of the energy level and charges separation
mechanism of CdS/ETS-10 is shown in Figure 4.21. Conduction edge of CdS
exhibits more negative than conduction edge of ETS-10 and the conduction edge of
ETS-10 exhibits more negative than redox potential of H+/H2. Based on the band
potential energy of CdS and ETS-10, the charge transfer from conduction band of
CdS to conduction band of ETS-10 is proposed.
In the meanwhile, the
photogenerated holes at VB of CdS will be trapped by hole scavenger. As a result,
strong reduction power is accumulated at conduction of ETS-10 while strong
oxidation power is present at valence band of CdS. This class of catalyst design may
greatly reduce the probability of charge-recombination and improve the conversion
efficiency. Sankar and co-workers (1996) has reported the Ti-O bonds in the –Ti-O–
70
Ti-O-Ti– chains was alternately long and short.
This feature increased the
opportunity and ability of electron delocalization compared to others common
zeolite matrix.
Negative
Potential (V/NHE)
H2
hν
e-
e- e- e-
CB
CB
0
CdS
S2-
h+ h+ h+
ETS-10
H+
VB
S22-
VB
Positive
Figure 4.21: Schematic diagram of the energy level and charges separation
mechanism of CdS/ETS-10.
4.2.3
The Effect of Synthesis Route of CdS
The effect of synthesis method of CdS towards the water photo-splitting
reaction was studied. The amount of hydrogen generated from water by using CdSIS and CdS-RM as photocatalysts was collected by water displacement method.
CdS-IS was found to perform better than CdS-RM up to 2 fold although the band
gap of CdS-IS is greater than CdS-RM by 0.18 eV. The hydrogen generation of bulk
CdS could be negligible since the size of CdS is in micro size region.
Organic substances that remain in CdS-RM such as Triton X-100,
cyclohexane and n-hexanol are probably the main factor in causing the lesser
71
catalytic conversion efficiency. These organic species adsorbed onto the catalyst
surface and reduced the light harvesting process. It was suggested that if surfactants
remain over the CdS nanoparticles thus obtained, the photocatalytic activity would
be decreased because the remaining surfactants hinder the reaction of
photogenerated electrons or holes with reactants.
This hydrogen generation
conversion proved that the capability of water ionization and the particles size were
the main priority when compared with the band gap value of as-synthesized CdS.
For 0.1 g of catalyst, the average rate of reaction for the first 5 hours was
found to be 68.69 μmol/hr and 49.05 μmol/hr for CdS-IS and CdS-RM respectively.
The rate of hydrogen liberated decreased with the increasing of time due to the lower
concentration of S2-/S2O32- as hole scavengers. The amount of hydrogen generated
for the first 24 hours 0.1 g of catalyst is shown in Figure 4.22.
CdS-IS vs CdS-RM
900
(a)
H2 Volume (μmol/0.1g)
800
700
600
500
(b)
400
300
200
100
0
0
5
10
Time (hr)
15
20
25
Figure 4.22: Amount of hydrogen generated by the catalysts (a) CdS-IS (b) CdSRM
72
Schematic energy diagram of water photo-splitting reaction over assynthesized CdS nanoparticles as photocatalyst is shown in Figure 4.23.
The
electrons at the valence band of CdS will be promoted to the conduction band under
light illumination. The hydrogen generation occurred at the redox potential of zero
(H+/H2) by reduction of water molecules or hydrogen ions. The strong reduction
power supplied from conduction band of CdS in which the photogenerated electrons
were diffused into the liquid-solid interfacial.
Negative
e-
Potential (V/NHE)
Potential (V/NHE)
Negative
CB
0
(a) Positive
2.58 eV
H+
H2
VB
CdS-IS
eCB
H+
H2
0
2.40 eV
VB
CdS-RM
(b) Positive
Figure 4.23: Schematic energy diagram of (a) CdS-IS (b) CdS-RM
4.2.4
The Effect of CdS Loading on ETS-10
The band gap energy of CdS in the CdS/ETS-10 was slightly reduced,
whereas the band gap energy of ETS-10 in the CdS/ETS-10 remains unchanged.
The shiftness of the band edge of CdS-RM and CdS-IS is shown by the dots line as
illustrated in Figure 4.24. With the presence of ETS-10 as co-catalyst, the charge
separation between the electron and hole pairs was more efficient. This is due to the
photogenerated electrons in the conduction band of CdS being transferred to
73
conduction band of ETS-10 due to the natural behavior of electron. These electrons
were then received by water molecules or hydrogen ion and lead to the formation of
hydrogen gas.
Potential (V/NHE)
eCB
CB
0
VB
CdS-IS
Positive
(a)
2.56 eV
4.03 eV
VB
ETS-10
+
H
H2
Potential (V/NHE)
Negative
Negative
eCB
CB
0
2.38 eV
H+
H2
VB
Positive
(b)
CdS-RM
4.03 eV
VB
ETS-10
Figure 4.24: Schematic energy diagram of (a) 10CdS-IS/ETS-10 (b) 10CdSRM/ETS-10 (straight lines represent the original band edge and dotted lines
represent new band edge)
The photocatalytic activity for hydrogen generation under visible light
irradiation is enhanced remarkably by attaching as-synthesized CdS nanoparticles on
ETS-10. As expected, a pure ETS-10 does not exhibit its catalytic property under
visible light irradiation due to its wide band gap energy. The volume of hydrogen
generated for the ETS-10 supports is shown in Table 4.4.
74
Table 4.4: Hydrogen generation of CdS/ETS-10 for the first 24 hours.
Hydrogen generation over ETS-10 supports (μmol/0.1g)
Time
(hr)
ETS-10
CdS-IS (%)
CdS-RM (%)
5
10
15
20
5
10
15
20
0
0
0
0
0
0
0
0
0
0
1
0
6
12
16
33
0
27
14
16
3
0
41
74
90
159
10
65
88
127
5
0
94
147
186
388
29
114
159
227
7
0
133
233
284
560
45
143
235
305
24
0
184
448
815
1478
86
268
646
1059
The higher percentage of CdS loading on ETS-10, the higher amount of
hydrogen gas was liberated. This is due to the increase in light absorption ability
thus lead to more active side. ETS-10 supported 20wt% of CdS-IS, namely 20CdSIS/ETS-10, is the best coupling catalyst and generated 1477 μmol of hydrogen gas
over 0.1 g of catalyst for the first 24 hours. The amount of hydrogen generated by
the CdS-IS catalyst series is shown in Figure 4.25. The average rate of reaction for
the first 5 hours was found to be 77.66 μmol/hr. This number was greater than CdSIS solely which is only 68.69 μmol/hr.
75
1600
1478
H2 generation (μmol/0.1g)
1400
1200
1000
815
800
560
600
388
159
400
0
0
200
0
01
6
12
90
16
41
33
Time (hr)
94
54
448
284
186
233
147
74
12
0
0
33
133
57
184
S4
(d)
S3
(c)
S2 Photocatalysts
Photocatalysts
(b)
S1
(a)
24
6
Figure 4.25: The amount of hydrogen generated by the catalysts: (a) 5CdSIS/ETS-10 (b) 10CdS-IS/ETS-10 (c) 15CdS-IS/ETS-10 (d) 20CdS-IS/ETS-10
Simlar patterns of hydrogen generation were observed the CdS-RM catalysts
series. Higher amount of hydrogen was produced with the increase of percent CdSRM loaded on ETS-10.
20 wt% of CdS-RM loaded on ETS-10 was the best
performed catalyst, by generating 1058 μmol for the first 24 hours with an average
rate of reaction of 45.37 μmol/hr for the first 5 hours. The amount of hydrogen
generated by the CdS-RM catalyst series is illustrated in the Figure 4.26.
76
1200
1059
H2 generation (μmol/0.1g)
1000
800
646
600
400
305
227
127
0
200
0
0
01
14
27
0
0
16
0
12
88
114
65
10
33
Time (hr)
29
54
235
159
268
143
45
75
86
S4
(d)
S3
(c)
S2 Photocatalysts
Photocatalysts
(b)
S1
(a)
24
6
Figure 4.26: The amount of hydrogen generated by the catalysts: (a) 5CdSRM/ETS-10 (b) 10CdS-RM/ETS-10 (c) 15CdS-RM/ETS-10 (d) 20CdS-RM/ETS-10.
ETS-10 supported CdS provides a high crystalline surface that can prevent
the fast deactivation of catalyst. An effective charge separation was found to greatly
reduce the possibility of charge recombination of photo-generated electron and hole
pairs. Besides that, 20CdS-IS/ETS-10 gave higher surface area up to two fold than
pure CdS-IS. Similar pattern was observed in CdS-RM and its derivatives. The
benefit of coupling materials was also reported by Jang and co-workers (2007) who
studied the optimization CdS on TiO2 for hydrogen generation.
It should be noted that although the coupling method was able to reduce
photo-recombination to some extent, however, hydrogen production from pure
water-splitting is difficult to achieve.
This is because the photo-recombination
cannot be completely eliminated and backward reaction of H2 and O2 to form H2O is
more thermodynamically favorable.
77
4.2.5
The Effect of CdS Loading on METS-10
An attempt to reduce band gap energy of ETS-10 was conducted by chemical
treatment. However, the band gap energy of modified ETS-10 (METS-10) was
calculated to be 3.41 eV. This value is considered high and expected would only
function under UV light. METS-10 solely was not active in water photo-splitting
reaction under visible light irradiation. This is due to more energy needed in order to
promote the electrons from the valence band to the conduction band. After the
chemical treatment, the valance edge of ETS-10 has been shifted to more negative
whereas, the conduction edge of ETS-10 shifted to more positive. The schematic
energy diagram of the METS-10 supported CdS photocatalyst is shown in Figure
4.27.
Potential (V/NHE)
eCB
CB
0
VB
CdS-IS
Positive
(a)
2.56 eV
3.41 eV
VB
METS-10
+
H
H2
Potential (V/NHE)
Negative
Negative
eCB
CB
0
Positive
(b)
2.38 eV
H+
H2
VB
CdS-RM
3.41 eV
VB
METS-10
Figure 4.27: Schematic energy diagram of (a) 10CdS-IS/METS-10 (b) 10CdSRM/METS-10 (straight lines represent the original band edge and dotted lines
represent new band edge).
The activity of hydrogen generation was also tested by CdS loaded METS-10
as photocatalysts. The ETS-10 supported CdS performed better than METS-10
supported CdS in water photo-splitting reaction. Although the modification of ETS10 successfully reduced its band gap energy, however, this approach also reduced it
78
crystallinity and affected more structural defect on its surface. The structural defect
in the CdS/METS-10 catalyst, promotes the fast deactivation of the catalyst and
reduced the solar energy conversion efficiency.
Table 4.5: Hydrogen generation of CdS/METS-10 for the first 24 hours.
Hydrogen generation over METS-10 supports (μmol/0.1g)
Time
(hr)
METS-10
CdS-IS (%)
CdS-RM (%)
5
10
15
20
5
10
15
20
0
0
0
0
0
0
0
0
0
0
1
0
6
12
29
29
0
10
14
35
3
0
20
29
61
129
14
27
88
121
5
0
53
65
139
204
25
67
137
204
7
0
106
135
198
315
39
100
178
280
24
0
135
274
601
1053
69
202
503
993
The hydrogen generated pattern for the series of ETS-10 and METS-10
photocatalyst was found similar. With the higher percent loading of CdS, the more
hydrogen could be collected. In the series catalysts of CdS-IS/METS-10, 20 wt% of
CdS-IS, namely 20CdS-IS/ETS-10, is the best coupling catalyst and generated 1053
μmol of hydrogen gas over 0.1 g of catalyst for the first 24 hours. Similar results
were observed for the series catalysts of CdS-RM/METS-10 that generated 993 μmol
of hydrogen gas for the first 24 hours. The average rate of reaction for the first 5
hours for both 20 wt% CdS were found to be 40.87 μmol/hr. The amounts of
hydrogen generated by CdS supported on METS-10 are shown in Figure 4.28 and
Figure 4.29.
79
1200
1053
H2 generation (μmol/0.1g)
1000
800
601
600
315
400
204
129
0
200
0
0
29
6
01
20
12
106
53
33
54
S4
(d)
S3
(c)
S2 Photocatalysts
Photocatalysts
(b)
S1
135
(a)
57
Time (hr)
274
135
65
29
198
139
61
12
0
0
29
24
6
Figure 4.28: The amount of hydrogen generated by the catalysts: (a) 5CdSIS/METS-10 (b) 10CdS-IS/METS-10 (c) 15CdS-IS/METS-10 (d) 20CdSIS/METS-10
993
1000
H2 generation (μmol/0.1g)
900
800
700
600
503
500
400
280
204
300
121
35
0
200
0
100
0
10
0
0
01
14
0
12
88
14
33
Time (hr)
202
100
67
27
25
54
178
137
39
75
69
S4
(d)
S3
(c)
S2 Photocatalysts
Photocatalysts
(b)
S1
(a)
24
6
Figure 4.29: The amount of hydrogen generated by the catalysts: (a)5CdSRM/METS-10 (b)10CdS-RM/METS-10 (c)15CdS-RM/METS-10 (d)20CdSRM/METS-10.
80
4.2.6
Reusability Test
At the beginning of reaction, CdS gave a higher rate of reaction compared to
CdS/ETS-10 since light absorption power of CdS is stronger than CdS/ETS-10. The
consumption of hole scavengers was able to trap the hole at valence band of CdS.
The presence of S2-/S2O32- as hole scavenger totally inhibit the formation of oxygen
gas from water where sulfide ions was chose to discharge rather than oxide ions.
The rate of hydrogen generation drops drastically after first 5 hours since the
concentration of hole scavengers getting lesser. At this moment, the photocorrosion
occurred at high rate, in which the photo-generated electron easily recombines with
the hole at the valance band of CdS nanoparticles. Nevertheless, the activity could
be recovered after a fresh load of S2-/S2O32- solution. Generally, the efficiency of the
catalysts decreases with the increase frequency of recovery (Figure 4.30). The data
of the hydrogen production is shown in Table 4.6.
Cycle 1
Cycle 2
Cycle 3
1600
(a)
1400
H2 Volume (μmol/0.1g)
1200
(a)
(b)
1000
(a)
(b)
(b)
800
(c)
(c)
600
400
(d)
(c)
(d)
(d)
200
0
0
5
10
15
Time (hr)
20
25 0
5
10
15
20
25 0
5
Time (hr)
10
15
tim e (hr)
Figure 4.30: The amount of hydrogen generated in 3 cycles by the catalysts (a)
20CdS-IS/ETS-10 (b) 20CdS-IS/METS-10 (c) CdS-IS (d) CdS-RM.
20
25
Table 4.6: Data of hydrogen generation of the catalysts in 3 cycles.
Hydrogen generation over 0.1 g of catalyst (μmol/0.1g)
CdS-RM
Time (hr)
CdS-IS
20CdS/ETS-10
20CdS/METS-10
Cycle 1 Cycle 2 Cycle 3 Cycle 1 Cycle 2 Cycle 3 Cycle 1 Cycle 2 Cycle 3 Cycle 1 Cycle 2 Cycle 3
0
0
0
0
0
0
0
0
0
0
0
0
0
1
6
4
35
76
57
16
33
49
29
29
33
39
3
145
121
129
245
163
168
159
174
131
129
104
133
5
245
221
213
343
286
290
388
339
229
204
170
204
7
298
305
266
466
368
409
560
450
354
315
264
305
24
392
374
341
768
723
695
1478
1214
995
1053
993
850
81
CHAPTER 5
CONCLUSION AND RECOMMENDATIONS
5.1
Conclusion
As a conclusion, the CdS nanoparticle was successfully synthesized by both
in-situ reduction and reverse micelle method. Both of as-synthesized CdS samples
have similar characteristics in term of crystallinity, morphology and particles size of
8-10 nm. Both of as-synthesized CdS appear to be blue-shifted and have higher
band gap energy compared to bulk CdS. The ETS-10 was successfully synthesized
by hydrothermal method when Ludox-30 and RHA was used as silica source. Band
gap energy of ETS-10 was successfully reduced by chemical treatment with H2O2
from 4.03 to 3.41 eV. From EDAX elemental analysis and TEM micrographs, CdS
was observed to present and seated adjacent to ETS-10. The band gap energy of
CdS in its CdS composite was found to be slightly lesser compared to CdS.
Conduction edge of CdS/ETS-10 and CdS/METS-10 are more negative then
the redox potential of H+/H2. The CdS supported ETS-10 was found to function well
to generate hydrogen from water under visible light irradiation. The photogenerated
electrons at the CB of CdS are being transferred to ETS-10 during the reaction. In
the meanwhile, the photogenerated hole at VB of CdS will be trapped by hole
scavengers.
This class of catalyst design can probability reduces the charge-
recombination and improves the conversion efficiency.
83
The presence of S2-/S2O32- ions additive in the solution greatly enhance the
photocatalytic activities.
In a water photo-splitting reaction by CdS/ETS-10
catalysts, inorganic sacrificing agent was found to give better results compared to
organic sacrificing agent. CdS-IS performed better than CdS-RM up to two fold,
whereas the CdS composite performed better than CdS alone. Besides, ETS-10
supported CdS performed better than METS-10 supported CdS. In both of series of
catalysts, the higher percent loading of CdS, the more quantity of hydrogen gas can
be collected.
The photocatalytic activity of CdS/ETS-10 catalyst was found to decrease
with the reaction time.
However, the activity of the photocatalysts could be
recovered by a fresh load of S2-/S2O32- ions. It was found that CdS/ETS-10 catalysts
could be reused up to 3 times. The water photo-splitting activities of the catalysts
reduced with the frequency of the recovery.
5.2
Recommendations
Since hydrogen gas has been proposed as the clean fuel and will be widely
applied in industries, thus, the hydrogen generation from a renewable resources is
getting much more attention. The study of photocatalyst with high solar energy
conversion efficiency is the current main interest. CdS is one of the promising
visible-light driven photocatalyst due to its conduction edge is more negatively than
H+/H2 redox potential. The CdS incorporated in other porous semiconductors could
be studied and optimize. The other type of mixed semiconductors could also be
studied.
Extensive research works are necessary to study the application of mixed
semiconductor. For example, the application of these photocatalysts could be tested
for other types of reaction including: waste water treatment, dye or organic
decomposition and oxidation or reduction of hazardous waste. These groups of solid
84
photocatalyst have high potential to be reused due to easy separation from the
reaction solution.
85
REFERENCES
Abe, R., Sayama, K., and Arakawa, H. (2002). Efficient Hydrogen Evolution from
Aqueous Mixture of I- and Acetonitrile using a Merocyanine Dye-Sensitized Pt/TiO2
Photocatalyst under Visible Light Irradiation. Chem Phys Lett. 362: 441–444.
Anderson, M.W., Terasaki, O., Philippou, A., Mackay, S.P., Ferreira, A., Rocha, J. and
Lidin, S. (1994). Structure of the Microporous Titanosilicate ETS-10. Nature. 367:
347-351.
Ansón, A., Kuznicki, S.M., Kuznicki, T., Haastrup, T., Wang, Y., Lin, C.C.H., Sawada,
J.A., Eyring, E.M. and Hunter., D. (2008) Adsorption of Argon, Oxygen, and
Nitrogen on Silver Exchanged ETS-10 Molecular Sieve. Micropor. Mesopor. Mater.
109: 557-580.
Bard, A.J. (1979). Photoelectrochemistry and Heterogenous Photocatalysis at
Semiconductor. J. Photochem. 10 (1): 59-75.
Bard, A.J. and Fox, M.A (1995). Artificial Photosynthesis: Solar Splitting of Water to
Hydrogen and Oxygen. Acc. Chem. Res. 28: 141-145.
BP Statistical Review of World Energy June 2008. (2008) London, UK. Beyond
Petroleum.
British Standard, (1991). Classification of Environmental Conditions. BS 75272.4:1991. BSI. London.
86
Casal, B., Ruiz-Hitzky, E., Bergaya, F., Challal, D., Pripiat, J. and Damme H.V. (1985).
Photo-Oxidation of Water Mediated by a Clay-Anchored Os Catalyst. J. Mol. Catal.
33(1): 83-86.
Chang, F.W., Yu, H.Y., Roselin, L.S., Yang, H.C. (2005). Production of Hydrogen via
Partial Oxidation of Methanol over Au/TiO2 Catalysts. Appl. Catal. A: General.
290: 138–147.
Chapman, D.M. (1990). Crystalline Group IVA Metal-containing Molecular Sieve
Compositions. US Patent. 5,015,453.
Chitakunye, R and MacDonnell, F.M. (2006). Investigation of the Photochemistry of
Ruthenium Complexes Containing Unusual Acceptor Ligands. Master of Science.
The University of Texas. Texas.
Choi, J.H., Kim, S.D., Kwon, Y.J. and Kim, W.J. (2006). Adsoprtion Behaviors of ETS10 and its Variant, ETAS-10 on the Removal of Heavy Metals Cu2+, Co2+, Mn2+ and
Zn2+ from a Waster Water. Micropor. Mesopor. Mater. 97: 157-167.
Choi, J.H., Kim, S.D., Noh, S.H., Oh, S.J. and Kim, W.J. (2006). Adsoprtion Behaviors
of Nano-Sized and Al-Substituted-ETAS-10 in Removing Heavy Metal Ions Pb2+
and Cd2+. Micropor. Mesopor. Mater. 87: 163-169.
Christofoletti, T., Assaf, J.M. and Assaf, E.M. (2005). Methane Steam Reforming on
Supported and Non-supported Molybdenum Carbides. Chem. Eng. J. 106: 97–103.
Darwent, J.R. and Porter, G. (1981). Photochemical Hydrogen Production using
Cadmium Sulphide Suspensions in Aerated Water. J. Chem. Soc., Chem. Commun.
145-146.
Das, T.K., Chandwadkar, A.J., Budhkar, A.P. and Sivasanker, S. (1996). A Rapid
Method of Synthesizing the Titanium Silicate ETS-10. Chem. Commun. 1105-1106.
87
Das, T.K., Chandwadkar, A.J., Budhkar, A.P., Belhekar, A.A. and Sivasanker, S.
(1995). Studies on the Synthesis of ETS-10 I: Influence of the Synthesis Parameters
and Seed Content. Micropor. Mater. 4: 195-203.
Dhanalakshmi, K.B., Latha, S., Anandan, S. and Maruthamuthu, P. (2001). Dye
Sensitized Hydrogen Evolution from Water. Int. J. Hydrogen Energy. 26: 669–674.
Doong, R.A., Chen, C.H., Maithreepala, R.A. and Chang, S.M. (2001). The Influence of
pH and Cadmium Sulfide on the Photocatalytic Degradation of 2-Chlorophenol in
Titanium Dioxide Suspensions. Water Res. 35(12): 2873–2880.
Fox, M.A. and Dulay, M.T. (1993). Heterogeneous Photocatalysis. Chem. Rev. 93: 341357
Fujishima and Honda K. (1972). Electrochemical Photolysis of Water at a
Semiconductor Electrode. Nature. 238: 37-38.
Goa, Y., Wu, P. and Tatsumi, T (2004). Liquid-phase Knoevenagel Reactions over
Modified Basic Microporous Titanosilicate ETS-10. J. Catal. 224: 107-114.
Grahn, H.T. (1999). Introduction to Semiconductor Physics. Singapore. World Scientific
Publishing Co. Pte. Ltd.
Guan, G.Q., Kida, T., Kusakabe, K., Kimura, K., Abe, E. and Yoshida, A. (2005).
Photocatalytic Activity of CdS Nanoparticles Incorporated in Titanium Silicate
Molecular Sieves of ETS-4 and ETS-10. Appl. Catal. A: General. 295: 71-78.
Guan, G.Q., Kida, T., Kusakabe, K., Kimura, K., Fang, X.M., Ma, T.L., Abe, E. and
Yoshida, A. (2004). Photocatalytic H2 Evolution under Visible Light Irradiation on
CdS/ETS-4 Composite. Chem. Phys. Lett.. 385: 319-322.
Gueymard, C.A, Myers, D and Emery, K. (2002). Proposed Reference Irradiance
Spectra for Solar Energy Systems Testing. Solar Energy. 73(6): 443-467.
88
Hadi, P. and Tomohiro, A. (2006). Hydrogen Production from Biogas using Hot Slag.
Int. J. Hydrogen Energy. 31: 491–495.
Hirai, T. and Bando, Y. (2005). Immobilization of CdS Nanoparticles formed in Reverse
Micelles onto Aluminosislicates Support and their Photocatalytic Properties. J.
Colloid Interface Sci. 288: 513-516.
Hu, X. and Lu, G.X. (2007). Investigation of Steam Reforming of Acetic Acid to
Hydrogen over Ni–Co Metal Catalyst. J. Mol. Catal. A: Chem. 261: 43–48.
Interface Sci. 305: 218-225.
Izumi, I., Fan, F-R.F., and Bard, A.J (1981). Heterogenous Phototcatalytic
Decomposition of Benzoic Acid and Adipic Acid on Platinized Titanium Oxide
Powder. The Photo-Kolbe Decarboxylative Route to the Breakdown of the Benzene
Ring and to the Production of Butane. J. Phys. Chem. 85(3): 218-223.
Jacovides, C.P., Steven, M.D. and Asimakopoulos, D.N. (2000). Spectral Solar
Irradiance and some Optical Properties for Various Polluted Atmospheres. Solar
energy. 69(3): 215-227.
James, R.L. and Michael, P.H. (2005). Comparison of Methanol-Based Fuel Processors
for PEM Fuel Cell Systems. Applied Catal. B: Environmental. 56: 149–169.
Jang, J.S., Ji, S.M., Bae, S.W., Son, H.S. and Lee, J.S. (2007). Optimization of
CdS/TiO2 Nano-Bulk Composite Photocatalysts for Hydrogen Production from
Na2S/Na2SO3 Aqueous Electrolyte Solution under Visible Light (λ≥420 nm). J.
Photochem. Photobiol. A: Chemistry. 188(1): 112-119.
Jang, J.S., Kim, H.G., Joshi, U.A., Jang, J.W. and Lee, J.S. (2008). Fabrication of CdS
Nanowires Decorated with TiO2 Nanoparticles for Photocatalytic Hydrogen
Production under Visible Light Irradiation. Int. J. Hydrogen Energy. 33: 5975-5980.
89
Jei, C.Y., Sulaiman, A. and Shamsuddin, M. (2008). Synthesis, Characterization and
Sorption Properties of Microporous Titanosilicate ETS-10 from Rice Husk Ash.
Master of science. Universiti Teknologi Malaysia. Skudai.
Jeong, H., Kim, T., Kim, D. and Kim, K. (2006). Hydrogen Production by the
Photocatalytic Overall Water Splitting Ni/Sr3Ti2O7: Effect of Preparation Method.
Int. J. Hydrogen Energy. 31: 1142-1146.
Jin, S. and Shiraishi, F. (2004). Photocatalytic Activities Enhanced for Decompositions
of Organic Compounds over Metal-Photodepositing Titanium Dioxide. Chem. Eng.
J. 97: 203–211.
Kanade, K.G., Baeg, J.O., Mulik, U.P., Amalnerkar, D.P. and Kale, B.B. (2006). NanoCdS
by
Polymer-Inorganic
Solid-State
Reaction:
Visible
Light
Pristine
Photocatalyst for Hydrogen Generation. Mater. Res. Bull. 41(12):2219-2225.
Kang, M.G., Han, H.E. and Kim, K.J. (1999). Enhanced Photodecomposition of 4Chlorophenol in Aqueous Solution by Deposition of CdS on TiO2. J. Photochem.
Photobiol. A: Chem. 125: 119–125.
Karsten, W., Hao, C.L., Rodrigo, J.W., Sotiris, E.P. and Aldo, S. (2006). In situ
Formation and Hydrolysis of Zn Nanoparticles for H2 Production by the 2-step
ZnO/Zn Water-Splitting Thermochemical Cycle. Int. J. Hydrogen Energy. 31: 5561.
Khan, A.M., Akhtar, M.S., Woo, S.I. and Yang, O-B. (2008). Enhanced Photoresponse
Under Visible Light in Pt Ionized TiO2 Nanotube for the Photocatalytic Splitting of
Water. Catal. Commun. 10(1): 1-5.
Khanna, P.K. and. Subbarao, V.V.V.S. (2004). Synthesis of Fine CdS Powder from
Direct In-Situ Reduction of Sulphur and Cadmium Salts in Aqueous N,N‟Dimethylformamide. Mater. Lett. 58: 2801-2804.
90
Kida, T., Guan, G.Q., Yamada, N., Ma, T., Kimura, K. and Yoshida, A. (2004).
Hydrogen Production from Sewage Sludge Solubilized in Hot-Compressed Water
using Photocatalyst under Light Irradiation. Int. J. Hydrogen Energy. 29(3): 269–74.
Kida,T., Guan, G.Q. and Yoshida. A. (2003). LaMnO3/CdS Nanocomposite: A New
Photocatalyst for Hydrogen Production from Water under Visible Light Irradiation.
Chem. Phys. Lett. 371: 563–567.
Kim, W.J., Lee, M.C., Yoo, J.C. and Hayhurst, D.T. (2000) Study on the Rapid
Crystallization of ETS-4 and ETS-10. Micropor. Mesopor. Mater. 41: 79-88.
Kitamura, N., Kawanishi, Y. and Tazuke, S. (1983). Highly Efficient Photoreduction of
Methylviologen by Tris (bis-diazadiimine) Ruthenium(II) Complexes. Chemistry
lett. 12(8): 1185-1188.
Krisnandi, Y.K. and Howe. R.F. (2006). Effect of Ion-Exchange on the Photoreactivity
of ETS-10. Appl. Catal. A: General. 307: 62-69.
Kutty, T.R.N. and Avudaithai, M. (1988). Sacrificial Photolysis of Water on TiO2 Fine
Powders Prepared by the Hydrothermal Method. Mater. Res. Bull. 23: 725-734.
Kutty, T.R.N. and Avudaithai, M. (1990). The Role of Trap-Centres on SrTiO3 Fine
Particles during Water Photolysis. Mater. Res. Bull. 25: 821-828.
Kuznicki, S.M. (1989). Large-Pored Crystalline Titanium Molecular Sieve Zeolites. U.S.
Patent 4,853,202.
Kuznicki, S.M. (1990). Preparation of Small-pored Crystalline Titanium Molecular
Sieve Zeolites. US Patent 4,938,939.
Kuznicki, S.M. and Trush, K.A. (1991). Removal of Heavy Metals, especially Lead,
from Aqueous Systems Containing Competing Ions Utilizing Wide-pored Molecular
Sieves of the ETS-10 Type. US Patent 4,994,191.
91
Leonard, J.J (1981) Preparation of zeolite A by hydrothermal treatment of clinoptilolite.
US Patent 4,247,524.
Lindström, B., Agrell, J. and Pettersson, L.J. (2003). Combined Methanol Reforming for
Hydrogen Generation over Monolithic Catalysts. Chem. Engineering J. 93: 91–101.
Liu, X.S. and Thomas, J.K. (1996). Synthesis of Microporous Titanosilicates ETS-10
and ETS-4 using Solid TiO2 as the Source of Titanium. Chem. Commun. 12: 14351436.
Lo, C.T., Davis, B. and Karan, K. (2006). Hydrogen Generation Process Development
via Methanolysis and Hydrolysis of Sodium Borohydride. Master of Science.
Queen‟s University, Kingston, Ontario, Canada.
Lu Lv, Wang, K. and Zhao, X.S. (2007). Effect of Operating Conditions on the Removal
of Pb2+ by Microporous Titanosilicate ETS-10 in a Fixed-Bed Column. J. Colloid
Lόpez, D.E., Goodwin, Jr.J.G., Bruce, D.A. and Lotero, E. (2005). Transeterification of
Triacetin with Methanol on Solid Acid and Base Catalysts. Appl. Catal. A: General.
295: 97-105.
Maldotti, A., Molinari, A. and Amadelli, R. (2002). Photocatalysis with Organized
Systems for the Oxofunctionalization of Hydrocarbons by O2. Chem. Rev. 102:
3811-3836.
Marty, P. and Grouset, D. (2003). High Temperature Hybrid Steam-Reforming for
Hydrogen Generation without Catalyst. J. Power Sources. 118: 66–70.
Matsumura, M., Furukawa, S., Saho. Y. and Tsubomura, H. (1985). Cadmium Sulfide
Photocatalyzed Hydrogen Production from Aqueous Solution of Sulfite: Effect of
Crystal Structure and Preparation Method of the Catalyst. J. Phys. Chem. 89(8):
1327-1329.
92
Meissner, D., Memming, R., Kastening, B. and Bahnemann, D. (1986). Fundamental
Problems of Water Splitting at Cadmium Sulfide. Chem. Phys. Lett. 127(5): 419423.
Meng Ni, Leung, M.K.H., Leung, D.Y.C. and Sumathy, K. (2007). A Review and
Recent Developments in Photocatalytic Water-Splitting using TiO2 for Hydrogen
Production. Renew. Sustain. Energy rev. 11: 401-425.
Mohammad, A.K and Chen, Y.T. (2005). Process Analysis and Simulation of Solar
Thermochemical Hydrogen Generation. Master of Science. University of Nevada,
Las Vegas.
Nash, M.J., Lobo, R.F. and Doren, D.J. (2008). Photocatalytic Oxidation of Ethylene by
Ammonium Exchanged ETS-10 and AM-6. Appl. Catal. B: Environment. In press.
Osamu, Y. (2006). Generation of Hydrogen Gas by Reforming Biomass with
Superheated Steam. Thin Solid Films. 509:207–211.
Panini, K.K. and Srinivas, P. (2006). Analysis and Control of an in-situ Hydrogen
Generation and Fuel Cell Power System for Automotive Applications. Philosophy
Doctorate. The Florida State University. Florida.
Peral, J. and Mills, A, (1993). Factors Affecting the Kinetics of Methyl Orange
Reduction Photosensitized by Colloidal CdS. J. photochem. Photobiol. A: Chem, 73:
47-52.
Philippou, A., Naderi, M., Pervaiz, N., Rocha, J. and Anderson, M.W. (1998). n-Hexane
Reforming Reactions over Basic Pt-ETS-10 and Pt-ETAS-10. J. Catal. 178: 174181.
Philippou, A., Rocha, J. and Anderson, M.W. (2000) Aldol-Type Reactions over Basic
Microporous Titanosilicate ETS-10 Type Catalysts. J. Catal. 189: 395-400.
93
Praveen, K.C. and Alan, M.L. (2005). Catalytic Autothermal Reforming of Diesel Fuel
for Hydrogen Generation in Fuel Cells I. Activity Tests and Sulfur Poisoning. J.
Power Sources. 152: 256–263.
Praveen, K.C. and William, J.T. (2006). Hydrogen Generation from 2,2,4-trimethyl
pentane Reforming over Molybdenum Carbide at Low Steam-to-Carbon Ratios. J.
Power Sources. 156: 520–524.
Raymond, L.Lee (1998). Mie Theory, Airy Theory, and the Natural Rainbow. Applied
Optics. 37(9): 1506-1519.
Rivka, E., Shimon, V. and Gary, H. (2001). Preparation and Surface Structure of
Nanocrystalline Cadmium Sulfide (Sulfoselenide) Precipitated from Dimethyl
Sulfoxide Solution. Chem. Mater. 13: 2272-2280.
Sankar, G., Bell, R.G., Thomas, J.M., Anderson, M.W., Wright, P.A. and Rocha, J.
(1996). Determination of the Structure of Distorted TiO6 Units in the Titanosilicate
ETS-10 by a Combination of X-ray Absorption Spectroscopy and Computer
Modeling. J. Phys. Chem. 100: 449-452.
Sathish, M., Viswanathan, B. and Viswanath, R.P. (2006). Alternate Synthetic Strategy
for the Preparation of CdS Nanoparticles and its Exploitation for Water Splitting.
Int. J. hydrogen energy. 31:891-898.
Sayama, K., Yase, K., Arakawa, H., Asakura, K., Tanaka, A. and Domen K, (1998).
Photocatalytic Activity and Reaction Mechanism of Pt-Intercalated K4Nb6O17
Catalyst on the Water-Splitting in Carbonate Salt Aqueous Solution. J. Photochem.
Photobiol. A: Chemistry. 114: 125–35.
Schiavello, M. (1985). Fundamentals and Developments of Photocatalytic and
Photoelectrochemical Processes. Holland D.Reidel Publishing Company.
94
Serpone, N, Texier, I., Ameline, A.V., Pichat, P., Hidaka, K. and Zhao, J. (2000). PostIrradiation Effect and Reductive Dechlorination of Chlorophenols at Oxygen-Free
TiO2 / Water Interface in the Presence of Prominent Hole Scavengers. J. Photochem.
Photobiol. A: Chemistry. 136: 145-155.
Serpone, N. (1989). Photocatalysis: Fundamentals and Application. Canada: John
Wiley& Sons. Inc.
Shangguan, W. (2007). Hydrogen Generation from Water Splitting on Nanocomposite
Photocatalysts. Sci. Technol. Adv. Mater. 8: 76-81.
Sitharamam, U., Shalini, R., Igor, N.M. and Kenneth, J.K. (2004) Exploration of
Photocatalytic Activities of Titanosilicate ETS-10 and Transition Metal Incorporated
ETS-10. Micropor. Mesopor. Mater. 67:181-187.
So,W.W., Kim, K.J., Moon, S.J. (2004). Photo-Production of Hydrogen over the CdSTiO2 Nano-Composite Particulate Films Treated with TiCl4. Int. J. Hydrogen Energy
29: 229–234.
Strasser, K and Siemens, AG. (1995). PEM Fuel Cells for Energy Storage Systems.
Erlangen, Germany. Power Generation Group.
Suppes, G.J., Dasari, M.A., Doskocil, E., Mandiky, P.J. and Golf, M.J. (2004)
Tranestrification of Soyabean Oil with Zeolite and Metal Catalysts. Appl. Catal. A:
General. 257: 213-223.
Taramasso, M., Perego, G. and Notari, B. (1983). Preparation of Porous Crystalline
Synthetic Material Comprised of Silicon and Titanium Oxide. US Patent. 4,410,501.
Thring, R.H. (2004). Fuel Cells for Automotive Applications. Bury St Edmunds and
London, UK, Professional Engineering Publishing.
Turner, R.P. (1961). Semiconductor Devices. New York. Holt, Rinehart and Winston
INC.
95
Uchihara, T., Matsumura, M., Ono, J. and Tsubomura, H. (1990). The Effect of
Ethylenediaminetetraacetic acid on the Photocatalytic Activities and the Flat-Band
Potentials of Cadmium Sulfide and Cadmium Selenide. J. Phys. Chem. 94(1): 415418.
Uma, S., Rodrigues, S., Martyanov, I.N. and Klabunde, K.J. (2004). Exploration of
Photocatalytic Activities of Titanosilicates ETS-10 and Transition Metal
Incorporated ETS-10. Micropor. Mesopor. Mater. 67: 181-187.
Wang, L.C., Liu, Y.M., Chen, M., Cao, Y., He, H.Y., Wu, G.S., Dai, W.L. and Fan,
K.N. (2007). Production of Hydrogen by Steam Reforming of Methanol over
Cu/ZnO Catalysts Prepared via a Practical Soft Reactive Grinding Route based on
Dry Oxalate-Precursor Synthesis. J. Catal. 246: 193–204.
Williams, P.T and Nugranad, N. (2000). Comparison of Products from the Pyrolysis and
Catalytic Pyrolysis of Rice Husks. Energy. 25:493-513.
Yahiro, H., Kyakuno, T. and Okada, G (2002). New Preparation Method of CdS
Clusters encapsulated in Y-Type zeolites. Topics Catal. 19(2): 193-195.
Yang, X.H., Wu, Q.S., LiLi, Ding, Y.P. and Zhang, G.X.. (2005). Controlled Synthesis
of the Semiconductor CdS Quasi-Nanospheres, Nanoshuttles, Nanowires and
Nanotubes by the Reverse Micelle Systems with Different Surfactants. Colloids and
surfaces A: Physicochem. Eng. Aspects. 264: 172-178.
Yu, J.Y., Schreiner, S. and Vaska, L. (1990). Homogeneous Catalytic Production of
Hydrogen and other Molecules from Water-DMF Solutions. Inorg. Chim. Acta.
170(2): 145-147.
Zhang, Y.J., Yan, W., Wu, Y. P. and Wang, Z.H. (2008). Synthesis of TiO2 Nanotubes
Coupled with CdS Nanoparticles and Production of hydrogen by Photocatalytic
Water Decomposition. Mater. Lett. 62: 3846-3848.
96
Zhu, W.L., Han, W., Xiong, G. X. and Yang, W.S. (2005). Mixed Reforming of
Heptane to Syngas in the Ba0.5Sr0.5Co0.8Fe0.2O3 Membrane Reactor. Catal. Today.
104: 149–153.