PSZ 19:16 (Pind. 1/97)
UNIVERSITI TEKNOLOGI MALAYSIA
♦
JUDUL : DESIGN AND TAILORING OF A NOVEL HETEROGENEOUS
CATALYST FOR EXPOXIDATION OF 1-OCTENE
SESI PENGAJIAN: 2004/2005
Saya : NG YUN HAU
( HURUF BESAR ) mengaku membenarkan tesis ( PSM / Sarjana / Doktor Falsafah )* ini disimpan di
Perpustakaan Universiti Teknologi Malaysia dengan syarat-syarat kegunaan seperti berikut:
1. Tesis adalah hakmilik Universiti Teknologi Malaysia.
2. Perpustakaan Universiti Teknologi Malaysia dibenarkan membuat salinan untuk tujuan pengajian sahaja.
3. Perpustakaan dibenarkan membuat salinan tesis ini sebagai bahan pertukaran antara institusi pengajian tinggi.
4. **Sila tandakan (
√
)
SULIT
TERHAD
√
TIDAK TERHAD
(Mengandungi maklumat yang berdarjah keselamatan atau kepentingan Malaysia seperti yang termaktub dalam AKTA
RAHSIA RASMI 1972)
(Mengandungi maklumat TERHAD yang telah ditentukan oleh organisasi/badan di mana penyelidikan dijalankan)
Disahkan oleh
______________________________ ________________________________
(TANDATANGAN PENULIS) (TANDATANGAN PENYELIA)
Alamat Tetap:
272 Jalan Temiang,
PROF.
DR HALIMATON HAMDAN
70200 Seremban,
Negeri Sembilan
Tarikh: 8 AUGUST 2005 8 AUGUST 2005
CATATAN: * Potong yang tidak berkenaan.
** Jika tesis ini SULIT atau TERHAD, sila lampirkan surat daripada pihak berkuasa/organisasi berkenaan dengan menyatakan sekali sebab dan tempoh tesis ini perlu dikelaskan sebagai SULIT atau TERHAD.
♦
Tesis dimaksudkan sebagai tesis bagi Ijazah Doktor Falsafah dan Sarjana secara penyelidikan, atau disertasi bagi pengajian secara kerja kursus dan penyelidikan, atau Laporan Projek Sarjana Muda (PSM).

“We hereby declare that we have read this thesis and in our 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 Halimaton Hamdan
Signature :………………………….
Name of Supervisor : Assoc. Prof. Mohd Nazlan Mohd Muhid
Signature :………………………….
Name of Supervisor : Dr. Hadi Nur

BAHAGIAN A
⎯
Pengesahan Kerjasama*
Adalah disahkan bahawa projek penyelidikan tesis ini telah dilaksanalan melalui kerjasama antara _______________________ dengan ________________________
Disahkan oleh:
Tandatangan : _________________________________ Tarikh: ______________
Nama : _________________________________
(Cop rasmi)
* Jika penyelikan tesis/projek melibatkan kerjasama.
BAHAGIAN B
⎯
Untuk Kegunaan Pejabat Sekolah Pengajian Siswazah
Tesis ini telah diperiksa dan diakui oleh:
Nama dan Alamat Pemeriksa Luar : _____________________________________
_____________________________________
Nama Penyelia Lain (jika ada) : _____________________________________
_____________________________________
_____________________________________
_____________________________________
Disahkan oleh Penolong Pendaftar di SPS:
Tandatangan : _________________________________ Tarikh: ______________
Nama

DESIGN AND TAILORING OF A NOVEL HETEROGENEOUS CATALYST
FOR EPOXIDATION OF 1-OCTENE
NG YUN HAU
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 2005

ii
I declared that this thesis entitled “DESIGN AND TAILORING OF A NOVEL
HETEROGENEOUS CATALYST FOR EPOXIDATION OF 1-OCTENE” 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.
Name : N G YUN HAU

iii

iv
ACKNOWLEDGEMENTS
First of all, I would like to thank Professor Dr. Halimaton Hamdan, Associate
Professor Mohd Nazlan Mohd Muhid and Dr. Hadi Nur for their patience in supervising and giving thoughtful guidance with knowledge towards the completion of this research. Designing and tailoring a catalyst for special application has been an attractive topic for me as investigated intensively by the Zeolite and Porous Materials
Group (ZPMG) of Universiti Teknologi Malaysia. For what has been done, as a member, I am very pleased that I was involved in the research and group weekly meeting.
A special gratitude goes to my parents, Mr. Ng Voon Pin and Mrs Chan Moo
Moy. This thesis is a labour of love that evolved over the 20 months prior to completion. Their time, dedication, energy, love and support are appreciated more than they will ever know.
This work would not have been possible without the assistances and insightful comments from my lecturers, seniors and friends. Special thanks to Dr.
Sugeng, Didik Prasetyoko, Lim Kheng Wei, Lau Chin Guan, Azmi Mohamed,
Carmen Wong etc for their support that has been given to me personally. Finally, I am particularly grateful to MOSTI for the National Science Fellowship.

v
ABSTRACT
A series of alkylsilylated fluorinated zirconia and zeolite containing titanium were prepared by titanium impregnation followed by fluorination and alkylsilylation of zirconium oxide and NaY zeolite. These samples were characterized by
13
C and
29
Si magic angle spinning nuclear magnetic resonance spectroscopy (MAS NMR), xray diffraction analysis (XRD), Fourier transform infrared spectroscopy (FTIR), ultraviolet-visible spectroscopy (UV-VIS), hydrophobicity measurement, Brunner,
Emmet and Teller (BET) surface area analysis and energy dispersive x-ray analysis
(EDAX). The materials have been designed for epoxidation of 1-octene with aqueous hydrogen peroxide as oxidant. Epoxidation of alkenes is one of the most important methods of functionalizing simple hydrocarbons. Analysis by gas chromatographymass spectrometry (GC-MS) indicates that 1,2-epoxyoctane is the sole product of the reaction with 100% selectivity. The alkylsilylated fluorinated catalysts were more active and more efficient than the conventional titania-silica and zirconia-silica mixed oxides in linear alkene epoxidation; enhanced by the presence of alkylsilane and fluorine groups in the catalysts. Modification with alkylsilane was aimed to induce the hydrophobic behaviour of zirconia and zeolite which are hydrophilic in nature; whereas fluorine was chosen for its electron-drawing effect which further activates the titanium active sites.

vi
ABSTRAK
Satu siri mangkin baru zirkonia dan zeolite yang mengandungi titanium telah disediakan melalui pengfluorinan dan pengalkilsililan zirkonium oxida dan zeolite
NaY. Sampel mangkin ini telah dicirikan dengan menggunakkan teknik spektroskopi
13
C dan
29
Si MAS NMR, teknik pembelauan XRD, spektroskopi inframerah, spektroskopi UV-VIS, pengukuran hidrofobik, analisis luas permukaan BET dan analisis EDAX. Mangkin ini diubahsuai dengan tujuan untuk tindakbalas pengepoksidaan 1-octena dengan menggunakan hidrogen peroksida sebagai ejen pengoksidaan. Secara amnya, pengepoksidaan alkena merupakan salah satu tindakbalas yang paling penting dalam mengfungsikan hidrokarbon. Analisis dengan kromatografi gas-spektrometri jisim (GC-MS) menunjukkan 1, 2-epoksioktana adalah satu-satunya produk yang terbentuk. Mangkin yang diubahsuai dengan penfluorinan dan pengalkilsililan adalah lebih aktif dan lebih efisien daripada titaniasilika dan zirkonia-silika biasa dalam pengepoksidaan alkena. Hal ini menunjukkan kesan positif daripada kumpulan alkilsilana dan fluorin dalam tindakbalas ini.
Modifikasi dengan kumpulan alkilsilana bertujuan untuk menambahkan sifat hidrofobik zirkonia dan zeolite yang bersifat hidrofilik secara semulajadi manakala fluorin dipilih kerana ia mempunyai kesan penarikan-elektron yang boleh mengaktifkan lagi titanium sebagai tapak aktif.

TABLE OF CONTENTS
CHAPTER
TITLE
TITLE
STATEMENT
DEDICATION
ACKNOWLEDGEMENTS
ABSTRACT
ABSTRAK
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
ABBREVIATIONS
LIST OF APPENDICES
1 INTRODUCTION
1.1
Research Background and Problem Statement
1.2
Research Objectives
1.3
Scope of Study
1.4
Thesis Outline
2.1
Catalyst
2.2
Zirconia-Based Catalyst
2.3
Zeolite
2.4
Catalytic Epoxidation
2.5
Hydrogen Peroxide vii
1
5
5
6
7
9
10
13
15
PAGE i vi vii x xi ii iii iv v xiv xv

viii
2.6
Titanium Oxide
2.7
Alkylsilylation of Catalyst
2.8
Fluorination of Catalyst
2.9
Characterization Techniques for Alkylsilylate
Fluorinated Catalyst
16
18
19
20
2.9.1
Powder X-Ray Diffraction (XRD)
2.9.2
Nuclear Magnetic Resonance (NMR)
20
22
2.9.3
Ultra-Violet Visible Spectroscopy (UV-VIS) 24
2.9.4
Fourier Transformed Infrared Spectroscopy (FTIR) 25
2.9.5
EDAX Microscopy
2.9.6
BET Surface Area Analysis
27
28
2.9.7
Gas Chromatography-Mass Spectrometry Analysis
(GC-MS) 30
4
3.1
Preparation of Ti-ZrO
2
and Ti-NaY
3.2
Alkylsilylation and Fluorination of Ti-ZrO
2
and
Ti-NaY
3.3
Characterization of Catalysts
3.3.1
Powder X-Ray Diffraction
3.3.2
Solid State NMR Spectroscopy
3.3.3
UV-VIS Sepctroscopy
3.3.4
Fourier Transformed Infrared Spectroscopy
33
33
38
38
38
39
39
3.3.5
EDAX Analysis
3.3.6
BET surface Area Analysis
39
40
3.4
Epoxidation of 1-Octene with Aqueous Hydrogen Peroxide 40
RESULTS AND DISCUSSION
4.1
Alkylsilylated Fluorinated Catalysts
4.2
The Effect of Alkylsilane Coverage on Catalyst
– NMR
43
46

ix
4.3
Vibrational Spectroscopy of Alkylsilylated
Fluorinated Zeolite – FTIR
4.4
XRD Analysis on the Structure of Catalyst with
Various Fluorine Concentrations
4.5
Water Adsorption Study on Hydrophobicity of
Alkylsilylated Fluorinated Catalysts
47
51
54
4.6
Titanium Species in Fluorinated Phase-Boundary
Catalysts by UV-VIS Spectroscopy
4.7
Structural Characterization of Alkylsilylated Fluorinated
56
Catalysts by XRD, BET Surface Area and EDAX Analysis 60
4.8
Catalytic Reaction: Epoxidation of 1-Octene with
Hydrogen Peroxide
4.8.1
The Effects of Alkylsilylation and
65
4.8.2
Fluorination
The Effect of Various Alkylsilane Groups
4.8.3
The Effect of Partial and Fully Alkylsilylated
NaY Catalysts
4.8.4
The Effect of Washing on Fluorinated
69
70
71
Catalyst
4.8.5
The Effect of Stirring
73
75
4.8.6
The Effect of Reaction Duration
4.8.7
Leaching Test
78
80
4.8.8
Postulated Epoxidation Mechanism for
Alkylsilylated Fluorinated Catalyst 81
5 CONCLUSION 84
REFERENCES
APPENDICES
86
95

x
LIST OF TABLES
TABLE
3.2
3.3
3.4
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
TITLE PAGE
Modified Ti-NaY series catalysts with description.
GC-FID temperature-programmed setup for 1, 2-epoxyoctane
35
35
41
GC-MSD temperature-programmed setup for verifying
1, 2-epoxyoctane product. 42
BET surface area and EDAX analysis of Ti-ZrO
2
series catalysts. 63
BET surface area and EDAX analysis of Ti-NaY series catalysts. 66
Epoxidation of 1-Octene by Variously Modified Zirconia and NaY.
The yield of 1, 2-epoxyoctane obtained from partially alkylsilylated and fully alkylsilylated Ti-NaY series.
The yield of 1, 2-epoxyoctane produced by the pre-washed and post-washed fluorinated Ti-NaY series catalysts.
The yield of 1, 2-epoxyoctane using Ti-ZrO
2
series catalysts under stirring and static conditions.
The yield of 1, 2-epoxyoctane using Ti-NaY series catalysts under stirring and static conditions.
Yield of 1, 2-epoxyoctane over Ti-ZrO
2
series catalysts.
Yield of 1, 2-epoxyoctane over Ti-NaY series catalysts.
Leaching test for CTMS-F -Ti-ZrO
2
sample.
Leaching test for CTMS-F -Ti-NaY sample.
71
75
76
78
79
81
82
83
83

xi
LIST OF FIGURES
FIGURE
2.1
TITLE PAGE
Important industrial organic chemicals produced by heterogeneous
Oxidation 14
2.2
2.3
2.4
Five-membered ring structure of the hydroperoxo in titanium- catalyzed epoxidation of lower alkenes
Derivation of Bragg’s law for X-ray diffraction
A schematic diagram of electrons excitation in UV-VIS
17
21
2.5
2.6
3.1
Spectroscopy 25
Linear plots of Brunauer, Emmett, Teller equation for nitrogen
Adsorption 29
Schematic representation of GC-MS 32
Sequences of alkylsilylated fluorinated Ti-ZrO
2
series catalysts
Preparation 36
3.2
4.1
Sequences of alkylsilylated fluorinated Ti-NaY series catalysts
Preparation 37
Photographs of modified Ti-ZrO
2
catalysts: Ti-ZrO
2
, OTS -Ti-ZrO
2
,
F -Ti-ZrO
2
, OTS-F -Ti-ZrO
2
and CTMS-F -Ti-ZrO
2
(from left to right) 44 modified OTS -Ti-NaY,
4.3
4.4
4.5
F -Ti-NaY, OTS-F -Ti-NaY and CTMS-F -Ti-NaY (from left to right) 45
13
C CP/MAS NMR of (a) fully OTS -HZSM-5 and (b) OTS -HZSM-5 47
IR spectra of a) NaY, b) Ti-NaY, c) OTS -Ti-NaY and d) F -Ti-NaY 49
IR spectra of a) Ti-ZrO
2
, b) OTS -Ti-ZrO
2
, c) F -Ti-ZrO
2
and d)
OTS-F -Ti-ZrO
2
50

xii
4.6
4.7
4.8
4.9
4.10
4.11
52
53
55
56
57
59
4.12 X-ray diffractograms of a) Ti-ZrO
2
, b) OTS -Ti-ZrO
2
, c) F -Ti-ZrO
2
,
d)
2
and e) CTMS-F -Ti-ZrO
2
61
4.13 X-ray diffractograms of a) Ti-NaY, b) OTS -Ti-NaY, c) F -Ti-NaY,
d) CTMS-F -Ti-NaY 63
4.14
4.15
4.16
Chromatogram of 1, 2-epoxyoctane standard
Chromatogram of blank reaction (without catalyst)
4.17
4.18 Mass spectrum of 1, 2-epoxyoctane from (a) standard sample, (b) reaction yield with CTMS-F -Ti-NaY catalyst and (c ) reaction yield
66
66
Chromatogram of the epoxidation of 1-octene with H
2
O
2
using
CTMS-F -Ti-ZrO
2
67
Chromatogram of the epoxidation of 1-octene with H
2
O
2
using
CTMS-F -Ti-NaY 67 with CTMS-FTi-ZrO
2
68
4.19 The effect of various alkylsilane groups on the yield of 1, 2-epoxide in the alkylsilylated fluorinated catalytic system
4.20 The yield of 1, 2-epoxyoctane obtained from (a) OTS-F -Ti-NaY,
71
(b) fully-OTS-F -Ti-NaY, (c ) CTMS-F -Ti-NaY and (d) fully-
CTMS-F -Ti-NaY 72
4.21 The yield of 1, 2-epoxyoctane produced by the pre-washed and
Post-washed fluorinated Ti-NaY catalysts
4.22 The yield of 1, 2-epoxyoctane using Ti-ZrO
2
series catalysts under stirring and static conditions
74
76
X-ray diffractograms to demonstrate the effect of different fluorine concentrations on the Ti-ZrO
2
structure
X-ray diffractograms to demonstrate the effect of different fluorine concentrations on the Ti-NaY structure
Water adsorption capacity (w/w%) of various modified Ti-ZrO
2
Water adsorption capacity (w/w%) of various modified Ti-NaY
UV-VIS spectra of a series of Ti-ZrO
2
catalysts
UV-VIS spectra of a series of Ti-NaY catalysts

xiii
4.23 The yield of 1, 2-epoxyoctane using Ti-NaY series catalysts under stirring and static conditions
4.24 The effect of reaction time on the yield of 1, 2-epoxyoctane over
Ti-ZrO
2
series catalysts
4.25 The effect of reaction time on the yield of 1, 2-epoxyoctane over
4.26
4.27
Ti-NaY series catalysts
Structure of hydroperoxo intermediate, Ti-OOH
Postulated epoxidation mechanism for alkylsilylated fluorinated
77
78
79
82
Catalyst 83

ABBREVIATIONS
λ Wavelength
2 θ angle
BET
Cu K
α
Brunner, Emmet and Teller
X-ray diffraction from copper K energy level h hour xiv
N
2
O
2
Nitrogen
Oxygen
P/P o relative pressure; obtained by forming the ration of the equilibrium pressure and the vapour pressure,
P o
of the adsorbate at temperature where the isotherm is measured t
R
Retention time
UV-VIS Ultraviolet-Visible
NMR Nuclear Magnetic Resonance Spectroscopy
FTIR
EDAX
Fourier Transform Infrared
Energy Dispersive X-ray Analysis

xv
LIST OF APPENDICES
APPENDIX
A
B
C
TITLE PAGE
EDAX Analysis of Ti-NaY series Catalysts
Water Adsorption Data
Chromatogram of the Reaction Mixture
95
100
101

CHAPTER 1
INTRODUCTION
1.1
Research Background and Problem Statement
Epoxides are substances of important commercial value and have found use in diverse areas. For example, polymerization of propylene oxides with alcohols as initiators afford polyether polyols. This polymerization which is catalyzed by ferric chloride, for instance, yields poly(propylene oxide) polymers with molecular weight of around 100,000 or more. Reaction with water furnishes a broad spectrum of propylene glycols, including monopropylene glycol, dipropylene glycol, tripropylene glycol and so forth. On the other hand, reaction between epoxides and ammonia or amines generally gives aminoalcohols. Each of the foregoing is an important niche commodity, for example, as the reactive components in the manufacture of polyurethane [1].
Currently, one of the most intensively researched areas of chemistry is the search for novel epoxidation methods. Despite numerous reports in the literature, the epoxidation of terminal alkenes remains a challenge in petrochemistry. Many different methods for preparation of epoxides have been developed. To date, liquidphase epoxidation with hydrogen peroxide catalyzed by transition metals has been largely dominated by the use of complexes in solution [2]. A wide variety of epoxidation and oxidation reactions have been performed with unsurpassed activity and selectivity (even enantioselectivity). However, while these homogeneous systems may be suitable for the preparation of fine chemicals or pharmaceuticals, the obvious problem of the catalyst separation and recovery has so far hampered their

2 use in larger scale operations. In most cases, separation difficulties have been the key barrier to commercialization of delicate synthetic chemical methodologies. Hence, the heterogeneous catalytic system offers a better alternative nowadays since it is safer, possesses higher efficiency and environmentally acceptable which eventually reduces plant maintenance and minimizes environmental problems.
The breakthrough invention of titanium-silicalite-1 (TS-1) by Taramasso et al. opened a new route for the synthesis of epoxides heterogeneously[3]. TS-1 catalyses the epoxidation of alkenes with diluted hydrogen peroxide with high conversion rates and selectivity under mild conditions in the liquid phase. TS-1 has unique catalytic properties, being effective in oxidation of a variety of organic compounds at low temperature, using diluted hydrogen peroxide as oxidant for the epoxidation of alkenes [4], epoxidation of allylic alcohols [5], oxidation of alkanes [6], hydroxylation of aromatics [7], and ammoximation of ketones [8]. Although TS-1 exhibits remarkable reactivity, by-products are formed in consecutive reactions such as the solvolysis of epoxides. This is quite a serious problem where the epoxides are the desired products. Moreover, one of the main difficulties in the use of TS-1 is the complexity in its synthesis process.
Subsequent to the weaknesses of TS-1, a large volume of works has been carried out on titanium-incorporated ordered mesoporous silica [9, 10], titanium silica-supported catalysts [11, 12], and amorphous titania-silica aerogels. The epoxidation of a range of alkenes was shown to proceed in excellent conversions and almost complete selectivity. A common and major claim of all these works is the excellent performance of these catalysts in the epoxidation reactions of the olefinic substrates with alkyl hydroperoxide as oxidant, for instance, tert -butyl hydroperoxide
(TBHP). Despite good activity with alkyl hydroperoxides, however, they are in general fail to promote epoxidations with hydrogen peroxide because of their surface hydrophilicity properties compared to the hydrophobic silicalites. Hydrophilicity of catalyst hampers the diffusion of organic substrates inside the pores or surface of the solid. This becomes a major hurdle that prevents the practical use of aqueous hydrogen peroxide. In order to modify the surface hydrophobicity/ hydrophilicity, several reports have been documented: methylation by Klein and Maier [13],

3 modification with covalent chloride ligands by Neumann and Levin-Elad [14], and partial octadecylsilylation by Ohtani, Ikeda and Hadi [15, 16, 17].
In our study, however, zirconia has been employed to act as the host for the epoxidations. The zirconia proved to provide a suitable environment for the reaction to take place without the incursion of significant side-reactions. One of the main reasons that zirconia is selected as the host material is its ultra-stability towards chemical modification. Zirconia structure remains intact even under modification with strong acid, for instance, hydrochloric acid. This is the additional strength which zeolite-based catalyst does not possess. Besides that, the surface of zirconia is less hydrophilic than zeolite. The high surface hydrophilicity of zeolite is due to the existence of hydroxyl groups in large quantity. Hence, zirconia offers a better possibility of using aqueous hydrogen peroxide as the oxidant in epoxidation.
Particularly, the epoxidation of alkenes with aqueous hydrogen peroxide is one of the main objectives in the areas of industrial and academia chemistry. The need for more environmentally friendly methods in fine chemistry that allow removal of commonly hazardous oxidants, such as organic peroxyacids, is pushing forward the use of cleaner oxidants like hydroperoxides (particularly hydrogen peroxide) under catalytic conditions. Hydrogen peroxide is one of the most promising oxidants for a clean oxidation process because it gives only water as a product in a wide range of oxidation reactions. Moreover, it is comparatively less expensive and more accessible than other oxidizing agents, such as organic peracids or alkyl hydroperoxides. Therefore, heterogeneous catalytic epoxidation using aqueous hydrogen peroxide is desired.
There were many researches on the feasibility of various transition metals as the catalysts in the application of alkenes epoxidation, such as Co, Fe and Ni. In this study, titanium transition metal was used as the active site to catalyze the epoxidation of terminal alkene, 1-octene. The selection of titanium as the active site is because titanium complexes are much more efficient than the other elements in the oxidation of organic compounds with hydroperoxides as oxidants [18]. In a titanium-containing catalyst matrix, it is generally believed that the tetrahedrally coordinated titanium is the real active site in epoxidation reaction. It is responsible for the production of

4 epoxides through the formation of hydroperoxo species with hydrogen peroxide.
Thus, most of current studies are concerned about the method to generate more tetrahedral titanium rather than octahedral titanium, polymerization of titanium or clustering of titanium in the catalyst matrix. Octahedral titanium is regarded to diminish the catalytic performance because they fail to form the hydroperoxo species.
When most researchers in the world adjust their direction towards generation of more tetrahedrally coordinated titanium, a few are concerned on the method to further activate the existing tetrahedrally coordinated titanium. Therefore, our attention is on the extended activation of tetrahedral titanium species instead of generation of more tetrahedral titanium sites.
Fluorine is the most electronegative elements on earth which possesses high affinity towards electron to stabilize its electron configuration. It tends to attract electrons from its nearby element, thus strengthen the bond between them. After the fluorination of catalysts, it is assumed that electrons in the titanium element have the inclination to move towards the direction where the fluorine is located. The bond between titanium and fluorine is thus reinforced resulting in the weakening of the bond between titanium and the oxygen. Consequently, that particular oxygen is vulnerable to the attack of alkenes. This chain of occurrences indicates that with the presence of fluorine, epoxidation may be made much easier. Therefore, it is hypothetical that the titanium active sites are further activated by fluorine. Besides that, fluorination has been used as a means to introduce hydrophobicity in a catalyst especially in zeolite-based catalyst [19].
In addition, with the activation of titanium species, it is expected that the epoxidation reaction could be carried out under ambient temperature. Currently, epoxidation reactions are mostly carried out at 60
°
C, 80
°
C or 100
°
C because they require higher activation energy to proceed. However, most of the epoxides is unstable at high temperature; decomposes or converts to other components. In the liquid-phase epoxidation, however, suffers from the lost of epoxides occurred since the epoxides have low boiling point, for instance, 1,2-epoxyoctane’s boiling point is
64
°
C. Hence, an epoxidation reaction performed at room temperature would be appreciated.

5
In this study, the strategy to design and develop an effective catalyst for epoxidation reaction is by employing partial alkylsilylation and fluorination onto the titanium-containing zirconia and titanium-containing zeolite. Partial alkylsilylation is aimed at modifying the catalyst surface polarity towards hydrophobic whereas fluorination is targeted to further activate the titanium species as reiterated above.
The combination of partial alkylsilylation and fluorination of catalysts may give a synergy effect on the preparation of epoxides. Zirconia and zeolite are chosen as the host materials as they possess several advantages over other materials that will be discussed in Chapter II. This approach to the development of new catalysts also has been chosen in order to widen the knowledge of fine chemical synthesis as part of the thesis strategy and intention. To the best of our knowledge, there is no other system that is similar to ours have been reported up to now.
1.2
Research Objectives
The objectives of this research are:
1.
To synthesize and modify Ti- ZrO
2
and Ti-NaY with partial alkylsilylation and fluorination.
2.
To characterize the modified Ti- ZrO characterization methods.
2
and Ti-NaY with various
3.
To carry out the epoxidation of 1-octene using aqueous hydrogen peroxide as
4.
oxidant over modified Ti- ZrO
2
and Ti-NaY catalysts.
To study and analyze the effects of partial alkylsilylated and fluorinated
5.
catalysts on epoxidation of 1-octene.
To study and analyze the parameters that affect the reaction such as various alkylsilane groups, stirring effect and reaction duration.
1.3
Scope of Research
The scope of this research is to develop a new oxidative catalyst which is able to carry out the epoxidation of terminal alkene, that is 1-octene with aqueous hydrogen peroxide as oxidant at ambient temperature. In this study, titanium-

6 containing zirconia and titanium-containing NaY were prepared by impregnating titanium isopropoxide into zirconia and NaY. Alkylsilane and fluorosilicate were attached to the support materials through vigorous stirring with the support particles in suitable solvents. As partial alkylsilylation introduces hydrophobic behaviour, the hydrophobicity of catalysts was measured by water adsorption testing. The solid state
Nuclear Magnetic Resonance (NMR) was employed to characterize the arrangement of alkylsilane groups on the catalyst surface. The research was also extended to the characterization of the physicochemical properties of the materials by employing appropriate techniques which include powder X-ray diffraction (XRD), UV-VIS spectroscopy, Fourier Transformed Infrared spectroscopy (FTIR), BET surface area analysis and Energy Dispersive X-ray microscopy (EDAX). Epoxides yield from the heterogeneous reaction was qualitatively and quantitatively measured by gas chromatography (GC) and gas chromatography-mass spectrometry (GC-MS).
Detailed interpretation and characterization of the materials will provide insight information on the structure identification, nature of active sites, surface properties and composition of the materials. Lastly, the most preeminent material among all the modified catalysts in the epoxidation of 1-octene was chosen.
1.4
Thesis Outline
This thesis brings together information concerning the synthesis, modification, characterization and potential catalytic application of alkylsilylated fluorinated titanium-containing zirconia and zeolite. Chapter I explains the research background and contemporary issues which defines the problem statement and resolution strategy.
Chapter II is the literature reviews on the chemistry of titanium-containing catalysts and the fundamental aspect of epoxidation reactions. Subsequently, Chapter III covers the experimental methodology, while Chapter IV reveals the results from the investigation and discusses the factors that influence the process. Finally, Chapter V summarizes the results obtained with recommendations for future work.

CHAPTER 2
LITERATURE REVIEW
2.1 Catalysts
The basic concept of a catalyst is that it is a substance that increases the rate of reaction without being appreciably consumed in the process. It is a reaction involving a cyclic process in which a site on a catalyst forms a complex with reactants, from which products are then desorbed, thereby restoring the original site.
Catalytic technologies have played a vital role in the economic development of the chemicals industry in the 20 th century, with a total contribution of ~20% of world GNP. In the 21 st century, the drive toward cleaner technologies brought about by public, legislative, and corporate pressure is expected to provide new and exciting opportunities for catalysis and catalytic processes [20].
A rapidly growing area of heterogeneous catalysis is for environmental pollution control. Although heterogeneously catalyzed processes are widely used in large scale petrochemical processes, the majority of fine, specialty, and pharmaceutical chemicals manufacturing processes still rely on homogeneous reagents and catalyst. Many of these processes were developed about 100 years ago, simply to maximize product yield, disregarding the environmental impact of inorganic waste and toxic byproducts formed during the reaction. Most of the waste is generated during the separation stage of the process when a typical water quench and neutralization (for acidic or alkaline systems) results in the formation of large volumes of hazardous waste. Heterogeneous catalysts are environmentally friendly

8 with respect to safety, corrosiveness, and ease of separation and recovery. Product isolation is simplified and reactions often run under milder conditions and give higher selectivity. The process is simpler, precious raw materials used in the manufacture of the catalyst are given increased lifetime (through reuse), and the volume of waste is significantly reduced. Thus, replacement of the homogeneous catalysts with heterogeneous catalysts is desirable in the chemical industry.
Most commercial heterogeneous catalysts are composed of active catalytic components supported or blended with refractory oxides, zeolites and carbons, while others are bulk metals. With a few exceptions, they all have pore structures through which reactant and product molecules must pass in order to reach the majority of active sites. The optimization of catalyst preparations involves maximizing the accessible active sites by efficiently dispersing them on a high surface area carrier, with pore structures sufficiently large to minimize diffusion resistances. A delicate balance of these chemical and physical properties must be achieved to produce a satisfactory catalyst. Molecules differ in size and chemical structure, so different catalysts are necessary for different reaction processes. In this respect the optimum catalyst is specific for each reaction and operating condition.
In a catalytic process, reactants must interact with the “active site” on the catalyst. The term “active site” is thought to be sites on the catalyst surface in which chemisorption and reaction occur. At least one molecule involved in the catalytic process must
(1) diffuse to the surface of the catalyst particle,
(2) diffuse through the pore network to the active sites,
(3) chemisorb,
(4) form an activated complex,
(5) convert to the product while retaining its chemisorbed state,
(6) desorb from the active site,
(7) diffuse through the pore structure and finally
(8) diffuse into the bulk fluid.

Steps (2) and (7) are controlled by the molecular size and pore structure of the catalyst while steps (3) to (6) depend on the chemical compatibility of reactant and product molecules with the active sites. The mass transfer steps (1) and (8) depend on reactant/product concentrations as well as catalyst amount and particle size. One of these steps will control the overall reaction rate. The slowest step determines the reaction rate. Therefore, it is essential to understand which is controlling so that the catalyst or conditions of operation can be varied.
9 wide variety of reasons. The causes may be grouped into poisoning, fouling, reduction of active area by sintering and loss of active species. A catalyst is poisoned when an impurity is present in the feed stream. In a complex reaction it may affect one reaction step more than another; hence the selectivity towards a desired reaction may be improved by deliberately adding a poison. It adsorbs on active sites of the catalyst, and if not adsorbed too strongly, it is gradually desorbed when the poison is eliminated from the feed stream. The phenomenon is then temporary. If adsorption is strong, the effect is permanent. On the other hand, fouling is generally describing a physical blockage such as the deposit of dust or fine powder or carbonaceous deposits (coke). In the coking case, the activity can usually be restored by removal of the coke by burning. Sintering is an irreversible physical process leading to a reduction of effective catalytic area. It may consist of growth of crystallites on a support. Besides that, the particular active species may also be converted to another form less active. An amorphous catalyst may crystallize, or a compound active in one crystal habit may be converted into a less active crystalline form. A supported metal catalyst may be reduced in activity by reaction with support; for example, a nickel/alumina catalyst may be converted to a nickel aluminate.
2.2
Zirconia-Based Catalyst
Zirconium oxide has been used as catalyst, host material or mixed oxide with other transition metals in many chemical reactions [21, 22, 23]. Okuhara et al. reported that MoO
3
-ZrO
2
was highly active in the hydrolysis of ethyl acetate in excess water and esterification of acetic acid with ethanol [24]. For ethylene

10 hydration, amorphous zirconium tungstate ZrW
2
O
0.5
has shown high catalytic activity. In this reaction, because it is insoluble in water; zirconium tungstate holds promise as an alternative for the problematic phosphoric acid catalyst. On the other hand, Segawa et al. have synthesized several types of zirconium phosphonates,
Zr(O
3
PR)
2
also to catalyze the hydrolysis of ethyl acetate in water and esterification of acetic acid with ethanol [25]. However, it is facing problems in terms of swelling or dissolution.
Strukul et al. have reported that some ZrO
2
-SiO
2
mixed oxides have been successfully employed for the oxidation of a variety of olefins with hydrogen peroxide to yield diols as the main product. Zirconium as the active site for oxidation reaction has been far less investigated with respect to titanium. The reason probably relies on Zr(IV) being far less reducible than Ti(IV), a factor that limit the decomposition of the oxidant to oxidize olefins. On the other hand, zirconium complexes are much less efficient than the corresponding titanium spesies in the oxidation of organic compounds with hydroperoxides as oxidants [26]. Control of site isolation has been achieved by Quignard et al. by grafting the zirconium on the silica surface through the use of tetraneopentylzirconium followed by hydrolysis of the residual organic ligands [27]. Besides that, sol-gel methods have been employed by Tuel and co-workers for the preparation of zirconium-containing mesoporous silicas [28]. These types of catalysts have proved capable of epoxidizing cyclohexene. Since the zirconium and titanium could be used as the active sites for epoxidation reaction with different efficiency respectively, the combination of zirconium and titanium (zirconia-titania mixed oxide or titanium impregnated on zirconia support) would probably yield a result different from common observation.
2.3
Zeolite
Zeolites are highly crystalline, hydrated aluminosilicates that upon dehydration develop in the ideal crystal with a uniform pore structure having minimum channel diameters (apertures) of from about 0.3 to 1.0 nm. The size depends primarily upon the type of zeolite and secondarily upon the cations present and the nature of treatments such as calcinations. Zeolites have been potential

11 catalysts for a few decades because of the high activity and unusual selectivity they provide in a variety of reactions. In many cases, but not all, the unusual selectivity is associated with the extremely fine pore structure. This permits only certain molecules to penetrate into the interior of the catalyst particles, or only certain products to escape from the interior.
The structure of a zeolite consists of a three-dimensional framework of SiO
4 and AlO
4
tetrahedra, each of which contains a silicon or aluminum atom in the center. The oxygen atoms are shared between adjoining tetrahedra, which can be presented in various ratios and arranged in a variety of ways. Zeolite may be represented by the empirical formula
M
2/n
•Al
2
O
3
• x SiO
2
• y H
2
O
The metal cation (of valences n ) is present to produce electrical neutrality since for each aluminum tetrahedron in the lattice there is an overall charge of -1. Access to the channels is limited by apertures consisting of a ring of oxygen atoms of connected tetrahedra. They may be 4, 5, 6, 8, or 12 oxygen atoms in the ring. The largest apertures occur in the faujasite-type zeolites (types X and Y) and mordenite, which are of high current interest as catalyst. Zeolites are often prepared in the sodium form, and this can be replaced by various other cations or by a hydrogen ion.
At least 34 species of zeolite minerals and over 100 types of synthetic zeolites are known.
Naturally occurring materials are normally described by a mineral name, e.g., mordenite. New synthetic types are usually designated by a letter or group of letters assigned by the original investigators, for example, A, X, Y, and ZSM. Types X and
Y are structurally and topologically related to the mineral faujasite and are frequently referred to as faujasite-type zeolites. The atomic ratio of Si/Al in the zeolites as originally prepared varies between 1 and 5, so bonds of the type Al – O –Al are not formed. Alumina can be selectively removed from some of the zeolites such as mordenite, which has an original Si/Al ratio of about 5, to produce a stable structure having a much higher Si/Al ratio.

12
Zeolites are of practical interest for a variety of reasons. The fine pore structure allows adsorption separations to be carried out on the basis of molecular size and shape, so-called molecular sieving, as in the separation of n -paraffins from isoparaffins. The ability to alter zeolite properties by ion exchange permits the synthesis of adsorbents of unusual selectivity, even when all molecules have free access to the interior pores of the zeolite. The ion-exchange properties also allow a high degree of flexibility in synthesizing catalysts, for example, the ability to produce a highly dispersed metal. From the catalytic point of view, zeolites are of interest in that they exhibit unusually high activity for various reactions such as cracking, the ability to combine a molecular-sieving property with catalysis, and unusual selectivity behavior.
However, there is a number of other materials that have a fine and more or less regular pore structure, such as some fine-pore silicas, porous glass, montmorillonite and other clays, and porous carbons prepared by controlled pyrolysis of synthetic polymers. These may therefore also exhibit molecular-sieving properties. The term ‘molecular-sieve’ zeolites is sometimes used to distinguish zeolites from this broader group of materials, although the term ‘molecular-sieve’ is sometimes used loosely and perhaps misleadingly as a synonym for a zeolite even when the molecular-sieving property of a zeolite is not being utilized [29]. The zeolite must have sufficiently large pores to accommodate most feed molecules, and have good stability to the high temperatures encountered in reaction and regeneration. Stability increases roughly with Si/Al ratio. Type Y is more hydrothermally stable and has a higher Si/Al atomic ratio than X (about 2.5 versus
1.25). An enormous amount of detailed attention has been devoted to the best economical procedures for obtaining an active and highly stable catalyst.
The Y zeolite (faujasite-type) which was used in this research has threedimensional intersecting channels in which the minimum free diameter is the same in each direction. The Y type consists of an array of cavities having internal diameters of about 1.2 nm. Access to each cavity (also termed a supercage) is through six equispaced necks having a diameter of about 0.74 nm. The Y zeolite has among the highest void fractions.

13
The Y zeolite is synthesized in the Na form, but this may be replaced with a wide variety of cations and/or with H + . In many cases, rare earth ions are incorporated into zeolite Y, replacing from 80 to 97+ percent of the equivalent Na, which provides additional structural strength [29]. The hydrogen Y form cannot be obtained by direct treatment of NaY with acid since the structure is attacked. Instead, the Na + is firstly replaced by NH
4
+ . Upon heating, ammonia and water are driven off to form hydrogen Y zeolite. Not all Na ions are identical in the crystal structure, and some are more difficult to exchange than others. Various procedures have been utilized to achieve a high degree of removal of Na + , desired for increased temperature stability. So-called ultrastable forms of Y (USY) may be prepared by these various means, in which the crystal form may be retained at temperatures as high as 1000 ºC. Zeolite stability and ultrastable zeolites are reviewed by McDaniel and Maher [30]. The first significant industrial application of Y zeolite in catalysis was the incorporation into a silica-alumina matrix of up to 15 percent of Y zeolite to form a new type of cracking catalyst. This was first introduced commercially in about 1962.
2.4
Catalytic Epoxidation
An intimate relationship exists between the petroleum and chemical industries since about 85 percent of the primary organic chemicals produced today are derived from petroleum and natural gas sources. Oxidation reactions in the chemical sector are made much more selective in nature by means of a catalyst which lowers the activation energy for the selected process and produces the useful products.
Figure 2.1 shows the important industrial organic chemicals produced by heterogeneous oxidation. One such class of processes which is a great industrial importance is epoxidation by heterogeneous catalysis [31]. The epoxidation of alkenes is one of the most important methods of functionalizing simple hydrocarbons.

The search for novel epoxidation methods is one of the most intensively researched areas in chemistry. Many exist, but few are genuinely environmentally acceptable. An example of clean epoxidation is given by the group led by Graham
Hutching at Cardiff University [32]. They have published results on the heterogenisation of Mn Salen complexes, and their use in epoxidations. They used
Mn-exchanged Al-MCM-41 as support, and adsorbed the chiral complex onto the support. The resultant material was an efficient catalyst for the conversion of ( Z )stilbene to the corresponding epoxide. Enantioselectivity under the best conditions was approximately equal to the homogeneous system.
14
12%
Aromatic
Oxidation
17%
Allylic
Oxidation
46%
Epoxidation
24%
Methanol
1%
Paraffinic
Oxidation
Figure 2.1
Important industrial organic chemicals produced by heterogeneous oxidation
Another contribution to this area is from Oliver Weichold and his colleagues at the University of Wurzburg [33]. Their system is based on the well known combination of methyl rhenium trioxide (MTO) and hydrogen peroxide. This combination can be used to carry out a number of oxidations, but often suffers from lack of selectivity, and typically requires mild base buffers to control side reactions.

15
If hydrogen peroxide is replaced with the urea-hydrogen peroxide complex (UHP) then these problems can be ameliorated. However, this generates urea as by-product as well as the water from the peroxide.
Weichold and co-workers have overcome this barrier by employing a NaY zeolite to act as a host for the oxidations. The NaY provides a suitable environment for the reaction to take place without the appearance of significant side-reactions, and without the use of soluble basic buffer or significant amounts of urea. However, one of the problems when dealing with zeolites in the liquid phase reactions is their generally high hydrophilicity, which hampers the diffusion of organic molecules inside the pores of the solid. This becomes a major hurdle that prevents the practical use of aqueous oxidants like hydrogen peroxide.
2.5
Hydrogen Peroxide
Oxidation processes play a significant role in chemical industry, being the basis for the production of important compounds. Benefits deriving from availability, low cost and absence of wastes, elect molecular oxygen as the oxidant of choice in the manufacture of bulk chemicals. However, selectivity problems, due to the severe reaction conditions required or to the radical nature of the reaction involved, are common obstacles to a wider use of it in synthetic chemistry. As an alternative, monooxygen donors are available as milder oxidants: hydrogen peroxide, peracids, organic peroxyacids, organic hydroperoxides, inorganic and metallorganic peroxides, sodium hypochlorite, iodosobenzene, and nitrous oxide. In catalytic oxidations, these are generally characterized by good activity at moderate temperatures and, often, by good selectivity as well. However, some of these monooxygen donors are expensive or produce hazardous by-product [34].
The demand for more environmentally acceptable methods in the fine chemistry that allow elimination of hazardous or costly oxidants, such as organic peroxyacids, is pushing forward the use of cleaner oxidants (particularly hydrogen peroxide). Recently, the use of hydrogen peroxide in the oxidation of organic molecules is a major goal, both in academia and in industry, because of the

16 environmental acceptability of this oxidant, which depends mainly on the nature of its by-product, water [35]. To date, liquid-phase oxidation with hydrogen peroxide catalyzed by transition metals has been largely dominated by the use of complexes in solution [2]. A wide variety of oxidation reactions (epoxidation of olefins, oxidation of alcohols, hydroxylation of aromatics, etc.) with unsurpassed activity and selectivity (even enantioselectivity) have been conducted. However, while these systems may be suitable for the preparation of fine chemicals and pharmaceuticals, the obvious problem of the catalyst separation and recovery has so far hampered their use in larger scale operations.
The major breakthrough in the use of the highly desirable H
2
O
2
oxidant in industry has been the discovery about 20 years ago of titanium silicalite TS-1. Other interesting results have been more recently obtained with V, Cr, and Sn containing zeolites or aluminophosphates. However, the major limitation of these crystalline materials rests in the limited number of heteroelements that believed to be the real active sites for oxidation reaction. Moreover, the most extensively studied zeolitebased in general fail to promote oxidations with hydrogen peroxide because of their surface hydrophilicity properties. This important point has been clearly demonstrated by the recent work of Ohtani and Hadi [16] who achieved interesting results with hydrogen peroxide as the oxidant by using zeolites in which the surface polarity was modified by partial alkylsilylation of the surface of silica through the use of octadecyltrichlorosilane precursor. It seems therefore that this surface property may be as important as the presence of catalytically active sites, as a wrong surface polarity may mask the catalytic process.
2.6
Titanium Oxide
The processes of catalytic epoxidations have been intensively investigated over the past three decades. The reasons for this are many but specifically because of the technological importance of the conversion of raw materials and petrochemicals like alkenes, and aromatics into a variety of bulk industrial chemicals. In this way, chemists have used different kinds of metals salt or transition metal elements or oxides in the form of homogeneous catalyst, or supported metals as heterogeneous

17 systems. Supported metals and zeolites which are significantly important due to their ion exchange properties, crystallinity, thermal stability and cage structure which change reactivities and especially selectivities towards a desirable product.
Transition metals elements like Ti, Cr, Mn, Fe, Co, Ni, and Cu can be introduced in different states and coordinations. These materials catalyze the decomposition of oxidants in the presence of olefins, thus oxidation of double bond occurs. Among these transition metals, titanium oxide is the most extensively studied due to its outstanding catalytic performance in epoxidation reaction.
Area of titanium catalyzed epoxidation of olefins with hydroperoxides is largely studied because of the discovery of TS-1 where Ti has been substituted for Si in the MFI framework. Other research has been directed at homogeneous analogs of
TS-1, such as CpTi-silsesquioxane, and immobilized Ti in the mesopores of MCM-
41. It is commonly accepted that isolated tetrahedral Ti in the catalyst matrix is the most active site in titania-silica epoxidation catalysts, including silicalites and mixed oxides [36]. Tetrahedrally coordinated titanium is necessary to form the active site for epoxidation, it is believed to be the titanium hydroperoxo species as shown in
Figure 2.2.
H
O
O Ti
O
H
O
O
O
H
Figure 2.2
Five-membered ring structure of the hydroperoxo in titanium-catalyzed epoxidation of lower alkenes.

18
2.7
Alkylsilylation of Catalyst
One of the key issues for successful use of H
2
O
2
with TS-1 is its hydrophobicity. Since then, extensive studies have been carried out on Ti containing zeolite and Ti containing metal oxide in comparison to TS-1. In general, the main problem when dealing with zeolites or metal oxides in the liquid phase is their generally high hydrophilicity, which hampers the diffusion of organics onto the surface of the solid.
Some years ago, Klein and Maier reported that the reactivity of some mixed oxides could be suitably controlled through appropriate methylation of the surface
[13]. Other efforts to explore different surface modifiers were subsequently carried out by Mallat and Baiker and co-workers [37], and more recently by Deng and Maier
[38], who thoroughly investigated the effect of methylation on reactivity.
Unfortunately, all studies dealt mainly with t -butylhydroperoxide as the oxidant, because the activity observed with H
2
O
2
was rather modest. Besides that, Takashi
Tatsumi reported that hydrophilic/hydrophobic property of Ti zeolites plays an important role in their activity for liquid phase oxidation [39]. They conducted trimethylsilylation of Ti-MCM-41 and Ti-MCM-48 in order to enhance activity in oxidation with dilute H
2
O
2
through increasing their hydrophobicity. They reported that trimethylsilylation of Ti-containing mesoporous molecular sieves resulted in a remarkable enhancement of catalytic activity in oxidation of alkenes and alkanes with H
2
O
2
. Furthermore, Ohtani, Hadi and Shigeru have performed the epoxidation reaction of alkene by partially octadecyltrichlorosilylated titanium containing zeolites with H
2
O
2
[16]. They found that alkylsilylation can modify the hydrophobicity and catalytic properties of titanium containing zeolite.
Alkylsilanes such as chloro-trimetylsilane and octadecyl-trichlorosilane possess hydrophobic characteristic which is attributed from their alkyl chain. In the case of octadecyltrichlorosilane (C
18
H
37
Cl
3
Si), the long C-18 methylene chain induced a very strong hydrophobicity while attached to the surface of catalyst.
However, chloro-trimethylsilane (C
3
H
9
ClSi) provides a sufficient hydrophobic coverage to the catalyst surface from its three methyl groups.

19
2.8
Fluorination of Catalyst
In the conventional hydrothermal zeolite synthesis method, a small amount of
Al is generally necessary as a crystallization-promoting element [40]. The introduction of Al into the zeolite frameworks leads to the formation of acid sites.
Furthermore, the very complex nature of severely intergrown zeolite structure results in a great concentration of silanol groups [41]. These characteristics lead to the hydrophilicity of zeolite walls, which decreases the oxidation activity for hydrophobic substrate because the pore walls attract water to prevent the contact of the hydrophobic substrates with the active oxidation sites [42]. To remedy these shortcomings, numerous attempts have been reported: dry-gel conversion [43], cogel
[44], seeding [45] and fluorination [46]. The fluorination gives rise to a very hydrophobic zeolite free of framework Al and silanol defects. Therefore, fluorination can be used as a method of introducing hydrophobicity onto catalysts.
In the literature, it was reported that the fluorination of zeolites can increase its catalytic activity in certain reactions [47, 48]. Different fluorination treatments were proposed. Lok et al. treated different zeolites with F
2
gas and reported structural dealumination and stabilization and, in some cases, observed an increase in catalytic activity in n -butane cracking [49]. Aneka et al. added aluminum fluoride to HY zeolite as a promoter [50]. Penchev et al. modified CaY zeolite with aqueous HF
[51]. Becker and Kowalak observed a considerable increase in catalytic activity in cumene cracking of mordenites and faujasites treated with NH
4
F and CH
3
F [52]. An increase in catalytic activity in pentane isomerization on fluorinated mordenites was observed by Bursian et al. [53].
In addition, fluorine is the most electronegative element. It is the first group
VII element which lacks 1 electron to fulfill a complete electron configuration.
Hence, it has a very strong affinity and tendency to attract electrons from its surrounding. This is called as ‘electron-withdrawing’ effect. Fluorine is able to attract electrons from elements and thus strengthen the bond between these two elements. With this characteristic, fluorine can be used to further activate an active site by making the active site more electrophilic, easier to be attacked by nucleophile substrates like alkenes. In this study, however, during the fluorination of zeolite, the

20 possible presence of HF in low concentration may have an effect toward the structure of Y zeolite.
2.9
Characterization Techniques for Alkylsilylated Fluorinated Catalyst
An enormous variety of instruments are capable of revealing information of value to some aspect of catalysts. In this study, comprehensive characterizations of the catalysts can be divided into three main aspects: structure identification, surface analysis and oxidative properties. Structures of catalysts were identified by employing powder X-ray diffractometer, solid state NMR, UV-VIS spectrometer,
FTIR spectrometer and EDAX analyzer. Besides that, surface area of catalysts was measured with nitrogen adsorption method by BET surface area analyzer. The oxidative properties of catalysts were examined through the epoxidation catalytic testing.
2.9.1
Powder X-Ray Diffraction (XRD)
X-ray diffraction technique is used for fingerprint characterization of crystalline materials and for determination of their structure [54]. It is one of the principal techniques in solid state chemistry. Common compounds can be identified using tabulations of reference patterns. Each crystalline solid has its own characteristic X-ray powder pattern which may be used as a ‘fingerprint’ for its identification. The powder patterns of most known inorganic solids are included in an updated version of the Powder Diffraction File [55].
X-rays are electromagnetic radiation of wavelength ~1 Å (10 -10 m), sufficiently energetic to penetrate solids and are well-suited to probe their internal structure. In generation of X-rays, the high-energy electron beam, provided by a heated tungsten filament, is accelerated to collide with a metal target, often copper.
The incident electrons have sufficient energy to knock out some of the copper K shell
(1s) electrons. An electron in an outer orbital, L shell (2p) or M shell (3p) immediately drops down to occupy the vacant K shell level and the energy released

21 in the transition appears as X-ray. The transition energies have fixed values. For copper, the L → K transition, called K
α
, has a wavelength of 1.5418 Å and the M →
K transition, K
β
, 1.3922 Å. The K
α
transition occurs much more frequently than the
K
β
transition and it is the more intense K
α
radiation which is used in diffraction experiments.
The Bragg approach to diffraction is to regard crystals as built up in layers of planes such that each acts as a semi-transparent mirror. Some of the X-rays are reflected off a plane with the angle of reflection equal to the angle of incidence, but the rest are transmitted to be subsequently reflected by succeeding planes [56]. The derivation of Bragg’s law is shown in Figure 2.3. Diffraction generally occurs when the wavelength of the wave motion is of the same order of magnitude as the repeat distance between scattering centers, with: n λ = 2 d sin θ where n
λ
= diffraction order (1, 2, …)
= wavelength of the X-rays d = interplanar distance on a set of planes
θ = angle between incident beam and lattice plane
Incoming beam
Diffracted beam
θ d
Figure 2.3
Derivation of Bragg’s law for X-ray diffraction
When Bragg’s law is satisfied, the reflected beams are interfering constructively. At angles of incidence other than the Bragg angle, reflected beams are out of phase and destructive interference or cancellation occurs. In real crystals, which contain thousands of planes and not just as shown in Figure 2.3, Bragg’s law

22 imposes a stringent condition on the angles at which reflection may occur. If the incident angle is incorrect by more than a few tenths of a degree, cancellation of the reflected beam is usually complete.
The XRD pattern has two characteristic features: the d-spacing of the lines and their intensity. Of the two, d-spacing is far more useful and capable of precise measurement. The d-spacing should be reproducible from sample to sample unless impurities are present or the material is in some stressed, disordered or metastable condition. On the other hand, intensities are more difficult to measure quantitatively and often vary from sample to sample (much more if preferred crystal orientation is present).
2.9.2
Nuclear Magnetic Resonance (NMR)
NMR arises from the adsorption and emission of radio-frequency energy by nuclear spins as they oscillate and reorient under internal and external magnetic fields. The local environments around the observed nuclei and the mobility of the nearby atoms or molecules greatly influence the oscillation frequency and reorientation of the observed nuclei. Due to its extreme sensitivity to the local environment and the mobility of the molecules in the sample, NMR spectroscopy allows obtaining detailed structural and dynamic information on many different types of materials [57]. Based on the fact that the NMR signal intensity is proportional to the number of contributing nuclei, the NMR method can also be used for quantitative compositional analysis of unknown materials.
There are three basic interactions in solid-state NMR and the way they yield specific information about local symmetry and bonding is briefly discussed [58].
(a) Chemical Shift Interactions
The chemical shift interaction arises from the magnetic electron-nucleus interaction (electrons shielding nuclei from the applied magnetic field), which alters the local field experienced by the nucleus and therefore affects its oscillation

23 frequency. The change in the oscillation frequency is measured as the chemical shift of a specific nucleus. The magnitude of the chemical shift is proportional to the applied magnetic field and the shielding constant determined by the electron distribution near the specific nucleus. For example, the methyl proton has a different chemical shift from the methylene proton, due to a difference in shielding constants.
Because the electron distribution near the nuclei in general is not symmetric, the chemical shift has directional properties. The chemical shifts seen in NMR spectra depend largely on the microscopic orientation of the atomic or molecular species with respect to the applied magnetic field [59, 60]. For molecules randomly oriented in a rigid lattice, the distribution of orientations produces a range of chemical shifts, hence a broad spectral line. In contrast, the rapid and isotropic molecular motion in liquids averages the anisotropic interactions and thus narrows the spectral lines. To reduce the line broadening in the solid sample due to the anisotropic effect, and to increase the sensitivity, magic angle spinning (MAS) is usually used. In MAS, the sample is rotated about an axis inclined at 54.7º with respect to the applied magnetic field. At 54.7º, the molecular orientation of solid sample will behave as liquid sample thus minimizes the anisotropic effect. This technique has been widely used for 13 C,
15 N, 19 F, 29 Si and 31 P. From the MAS NMR spectra, the isotropic chemical shift corresponds to the values that would be obtained in isotropic solution. These chemical shifts are fingerprints of specific functional groups or species and can be used to identify the functional groups or species in solid materials. The analysis can be used to obtain structural information, including the coordination and symmetry of the molecules.
(b) Dipole-dipole Interaction
The dipole-dipole interaction is due to the interaction of magnetic moments between two nuclei, and is dependent upon the magnitude of the magnetic moment of the nearby nuclei, the distance between two nuclei, and the angle between the internuclear vector and the applied magnetic field. For certain solids containing 1 H or
19 F, the spectral linewidth are significantly broadened, due to the large magnetic moments. In order to remove the large dipole interaction from the nearby protons, proton dipolar decoupling is used. This decoupling is accomplished by applying a perturbing field at the proton resonance frequency that results in a reduced linewidth.

24
Although dipole-dipole interactions can significantly broaden spectral lines, the dipole-dipole interaction is the basis for the commonly used cross-polarization
(CP) technique. Due to the difference in nuclear spin temperatures, CP is widely used to enhance the sensitivity of the observed nuclei by transferring magnetization from an abundant spin system (usually protons) to the spin system of the observed nucleus. Cross-polarization occurs if the nuclei can interact with each other via moderately strong dipole-dipole couplings. The stronger the dipole coupling, the faster the cross-polarization process will occur. CP-MAS and proton decoupling are often used together to obtain high-resolution solid-state NMR spectra.
2.9.3
Ultraviolet-Visible Spectroscopy (UV-VIS)
Ultraviolet-visible spectroscopy (uv = 200-400 nm, visible = 400-800 nm) corresponds to electronic excitations between the energy levels that associated with the molecular orbital of the system. In particular, transitions involving π orbitals and lone pairs are important and so UV-VIS spectroscopy is of most use for identifying conjugated systems which tend to have stronger absorptions.
The lowest energy transition is that between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) in the ground state. The absorption of the electromagnet radiation excites an electron to the
LUMO and creates an excited state. The more highly conjugated the system, the smaller the HOMO-LUMO gap, ∆ E, and therefore the lower the frequency and longer the wavelength, λ (see figure 2.4). The unit of the molecule that is responsible for the absorption is called the chromophore, of which the most common are C=C ( π to π *) and C=O (n to π *) systems.
UV-VIS spectroscopy has a variety of applications associated with the local structure of materials. This is because the positions of the absorption bands are sensitive to coordination environment and bond character. In this study, UV-VIS spectroscopy is utmost important for the characterization of titanium species

25 incorporated in the catalyst. The local environment of Ti has been studied in depth by using UV-VIS technique [61,62].
Figure 2.4
A schematic diagram of electrons excitation in UV-VIS spectroscopy
2.9.4
Fourier Transform Infrared Spectroscopy (FTIR)
Fourier Transform Infrared (FTIR) spectroscopy is a powerful tool for identifying types of chemical bonds in a molecule by producing an infrared absorption spectrum that is like a molecular ‘fingerprint’ [63]. It offers a rapid and useful means for framework structural content determination. Basically, infrared spectroscopy provides information on short range and long range bond order caused by electrostatic, lattice coupling, and other effects [64]. This technique is based on the possibility to separate the energy of a molecule into three additive components associated with:
(i) the rotation of the molecule as a whole
(ii) the vibration of the constituent atoms and
(iii) the motion of the electrons in the molecule.
Molecular bonds vibrate in various frequencies depending on the elements and the type of bonds. For any given bond, there are several specific frequencies at which it can vibrate. According to quantum mechanics, these frequencies correspond

26 to the ground state (lowest frequency) and several excited states (higher frequency).
One way to cause the frequency of a molecular vibration to increase is to excite the bond by having it absorb light energy. For any given transition between two states the light energy (determined by the wavelength) must exactly equal the difference in the energy between the two states [usually ground state (E o
) and the first excited state
(E
1
)] [65].
Difference in Energy States = Energy of Light Absorbed
E
1
– E o
= h c/ λ
Where h = Planck constant c = Speed of light
λ = the wavelength of light
The energy corresponding to these transitions between molecular vibrational states is generally 1-10 kilocalories/mole which corresponds to the infrared portion of the electromagnetic spectrum.
In the infrared spectroscopy, frequencies can range from 400 to 4000cm -1 but frequencies of certain framework vibrations such as tetrahedral SiO
4
or AlO
4 linkages are in the mid-infrared region of 400-1400 cm -1 . The IR spectra of the zeolite-based catalysts show typical absorption band in certain regions, each of which corresponds to a specific kind of vibrational mode. The assignment of each region is as follow:
(i) Stretching vibrations of hydroxyl groups (3000-3700 cm -1 )
This region is associated with H-bridging hydroxyl (- Si – OH…O – Si -) groups and isolated silanol (-Si – OH) groups.
(ii) Asymmetric O-T-O stretch (1000-1300 cm -1 ) (T being the atom at the centre of the tetrahedron, e.g. Si and Al)
(iii) Symmetric O-T-O stretch (700-850 cm -1 )

27
(iv) Vibrations of Si-OH (980-1000 cm -1 )
However, in Ti-containing zeolites, there is another characteristic peak observed in the region 910-960 cm -1 . The 960 cm -1 vibrational band can be assigned to asymmetric Ti-O-Si stretching modes of corner-sharing TiO
4
and SiO
4
tetrahedral.
In the zeolites or purely silicious species, the T-O-T asymmetric peak appears around
1000 cm -1 . The downward shift of the frequency on Ti substitution has been interpreted as a result of the Ti-O bonds being about 20 pm longer than the Si-O bonds [66]. Thus, longer bond length results in lower frequency of Ti-O-Si vibration.
It is stressed that this region is in general vulnerable to be overshadowed by the Si-
OH vibration (rocking mode) at approximately 980 cm -1 , which hinders precise identification of the band at 910-960 cm -1 [67].
2.9.5
EDAX Microscopy
EDAX analysis stands for Energy Dispersive X-ray analysis. It is a technique used for identifying the elemental composition of a sample. The EDAX analysis system works as an integrated feature of a scanning electron microscope. During analysis, the sample is bombarded with an electron beam inside the scanning electron microscope. The bombarding electrons collide with the sample atoms’ own electrons, knocking some of them off in the process. A position vacated by an ejected inner shell electron is eventually occupied by a higher-energy electron from an outer shell.
However, to enable to do so, the transferring outer electron must release some of its energy by emitting an X-ray.
The amount of energy released by the transferring electron depends on which shell it is transferring from, as well as which shell it is transferring to. Furthermore, the atom of every element releases X-rays with unique amounts of energy during the transferring process. Thus, by measuring the amount of energy present in the X-rays being released by a sample during electron beam bombardment, the identity of the atom from which the X-ray was emitted can be established.

28
2.9.6
BET Surface Area Analysis
In comparing different catalysts or the effect of various treatments on catalytic activity, it is necessary to know the extent to which a change in activity is caused by a change in the surface area of a catalyst. Methods of measuring surface area are of concern in many fields of science and technology and have received wide and detailed study. The principle method of measuring surface area of porous structures is by adsorption of a particular molecular species from a gas or liquid on the surface. If the condition under which a complete adsorbed layer, averaging the molecule thickness, can be established and the area covered per molecule is known, then the quantity of adsorbed material gives directly the surface area of the sample
[68].
One of the most common methods of measuring surface area, and used routinely in most catalyst studies, is developed by Brunauer, Emmett, and Teller
(BET) [68]. The BET treatment is based on a kinetic model of the adsorption process by Langmuir in which the surface of the solid was regarded as an array of adsorption sites. The starting point for measurement is the determination of the adsorption isotherm that is a plot of an amount of gas adsorbed at equilibrium as a function of the partial pressure P/P o
at constant temperature. The quantity of gas adsorbed is mainly expressed as the mass of gas or the volume of gas reduced to Standard
Temperature and Pressure (STP). By considering the linear section of the BET parameter versus P/P o
, the monolayer gas uptake and the C parameter can be determined. The BET equation is generally expressed as follows: p
V ads
( p
°
p )
1
V m c
c
1
V m c p p
°
Where V ads
= volume of gas adsorbed at pressure P
V m
= volume of gas adsorbed in monolayer, same units as V
P o
= saturation pressure of adsorbate gas at the experimental temperature
C = a constant related to the enthalphy of adsorption in the first adsorption layer.

29
If BET equation is obeyed, a graph of P / V ( P o
P ) versus P / Po should give a straight line whose slope and intercept can be used to calculate V m
and C . Many adsorption data show very good agreement with the BET (Figure 2.5) equation over values of the relative pressure P / P o
between approximately 0.05 and 0.03, and this range is usually used for surface area measurements. At higher P / P o
values, complexities associated with the realities of multilayer adsorption and/or pore condensation causes increasing deviation. At values of P / P o
much below about 0.05 the amount adsorbed in many cases is so low that the data become less accurate.
16
14
12
10
8
6
4
2
0
0 0.05 0.1
0.15 0.2
0.25 0.3
0.35 0.4
P/Po
Figure 2.5
Linear plots of the Brunauer, Emmett, Teller equation for nitrogen adsorption.
After calculation of V m
and C, the specific surface area (A
BET
) can be calculated by:
A
BET
= V m
.N
A
.a
m or
A
BET
= V m
(4.53) m 2 /g

30
Where N
A
is Avogadro number (6.023 x 10 23 mol -1 ), a m
is cross sectional area of the adsorbate molecule (0.162 nm 2 /molecule for nitrogen at 77 K).
Any condensable inert vapour can be used in the BET method, but for the most reliable measurements, the molecules should be small. Krypton, argon, and nitrogen are suitable choices in view of their commercial availability. Liquid nitrogen is a readily available coolant, but argon and krypton are expensive relative to nitrogen and must be highly purified. Consequently, nitrogen is usually used since it is relatively cheap and readily available in high purity.
2.9.7
Gas Chromatography-Mass Spectrometry Analysis (GC-MS)
Chromatography is a broad range of physical methods used to separate and/or to analyze complex mixtures [69, 70]. The components to be separated are distributed between two phases: a stationary phase bed and a mobile phase which flushes through the stationary bed [71]. A mixture of various components enters a chromatography process, and the different components are percolated through the system at different rates. These differential rates of migration as the mixture moves over stationary phase provide separation. Repeated sorption/desorption acts that take place during the movement of the sample over the stationary bed determine the rates.
The smaller the affinity a molecule has for the stationary phase, the shorter the time spent in a column, whereas a molecule with high affinity for the stationary phase will move through the system very slowly. Gas chromatography (GC) often called Gas
Solid Chromatography (GSC) is one of the chromatographic methods in which the mobile phase is a gas. GC analytical technique is well-established and routinely used to identify compounds which are manufactured in most industrial and academic laboratories, is attributed to its capability of high resolution, selectivity and sensitivity.
In GC, the sample (usually a gas or a liquid) is injected, directly or indirectly, onto the column. The column is contained within an oven and the temperature is being closely controlled by temperature-programmed mode in order to achieve the required chromatographic separations. The carrier gas – commonly nitrogen or

31 helium – elutes the compounds from the column at times which are related to their distribution coefficients between the stationary phase and the gas; the higher the distribution coefficient, the longer it takes for the compound to be eluted. On emerging from the column, the gas passes through a detector unit sensitive to the eluted compounds, the response of which is fed continuously to an amplifier and then to the recording system. When a compound is eluted, its presence is revealed as a peak, ideally of Gaussian shape. Factors which should be taken into consideration when operating a GC are type of carrier gas (mobile phase), sample injection system, type of column (stationary phase) and type of detector. The core requirements for sample are that the sample must be thermally stable and has a reasonable vapour pressure at the column temperature. This allows the sample analytes to be vaporized in and move with the gaseous mobile phase.
With proper calibration, the amounts of the components of a mixture can be measured accurately; both qualitative and quantitative data can be obtained.
Identification if qualitative analysis is based upon retention time of the peak of interest compared with the retention time of an authentic compound. Moreover, the alternative confirmation of the interest peak can be done by making relative retention. Relative measurement is based on comparing the corrected retention time of an added standard. The use of corrected retention time recognizes the observation made earlier assuming that all components spend the same time in the gas phase and that differences in retention between compounds are due to their relative affinities for the stationary phase.
When a GC is coupled to a mass spectrometer as its detector, this powerful analytical technique belongs to the class of hyphenated analytical instrumentation
(since each part had a different beginning and can exist independently) and is called gas chromatography-mass spectrometry (GC-MS) [72, 73]. Figure 2.6 shows the principal elements of a GC-MS system.

Gas Chromatography
Carrier Gas Supply
Sample inlet/
Injector
Interface Mass Spectrometer
32 column
Oven
Ionizer
Analyzer
Detector
Vacuum System
Data Storage Computer Control and
Data Acquisition
Display System
Figure 2.6
Schematic representation of GC-MS
Basically, MS consists of three stages: (1) the volatilization and ionization of the substances of interest, (2) the separation of the ions on the basis of their mass/charge (m/z) ratios and (3) the detection and recording of the separated ions.
When a MS is placed at the end of a chromatographic column in a manner similar to the other GC detectors, the mass detector is more complicated than, for instance, the
FID because of the mass spectrometer's complex requirements for the process of creation, separation, and detection of gas phase ions. A capillary column most often used in the chromatograph because the entire MS process must be carried out at very low pressures (~10 -5 torr) and in order to meet this requirement a vacuum is maintained via constant pumping using a vacuum pump. It is difficult for packed GC columns to be interfaced to an MS detector because they have carrier gas flow rates that cannot be pumped away by normal vacuum pumps; however, capillary columns' carrier flow is 25 or 30 times less and therefore easier to "pump down."

CHAPTER 3
RESEARCH METHODOLOGY
3.1
Preparation of Ti-ZrO
2
and Ti-NaY
Zirconium (IV) hydroxide (Zr(OH)
4
, Aldrich) was used as obtained. First, tetrapropyl orthotitanate (Ti(OPr i )
4
, Fluka Chemika) was impregnated from cyclohexanol solution into zirconium hydroxide powder by vigorous stirring at 343
K until the cyclohexanol was almost dried. The powder product was heated at 383 K overnight. The molar amount of Ti was 500 µmol g -1 of zirconium hydroxide.
Finally, the dried sample powder was calcined at 873 K for two hours to produce Ti-
ZrO
2
.
For preparation of Ti-NaY, NaY zeolite powder (Zeolyst International) and tetrapropyl orthotitanate were used as starting materials. NaY powder was immersed into tetrapropyl orthotitanate solution and stirred until it was nearly dried. The molar amount of Ti was 500 µmol g -1 of NaY, same as the preparation of Ti-ZrO
2
. The powder was then dried overnight at 383 K. The catalyst is labelled Ti-NaY.
3.2
Alkylsilylation and Fluorination of Ti-ZrO
2
and Ti-NaY
For alkylsilylation, two alkylsilane were used. They were octadecyltrichlorosilane, herein referred as OTS (C
18
H
37
SiCl
3
, Aldrich), and chlorotrimethylsilane, herein called as CTMS (C
3
H
9
SiCl, Aldrich). Firstly, a small amount of water (~50 w/w %) was added to the Ti-ZrO
2
to lead the aggregation

34 among powder particles [16]. In the second step, the wetted Ti-ZrO
2
was immersed in 10 cm 3 toluene containing 500 µmol of OTS, and the suspension was shaken for
15 mins at room temperature. Then, the powder was collected by centrifugation and washed with excessive toluene and ethanol. The catalyst powder was dried at 383 K overnight. Aggregation due to addition of water was expected at the outer surface of aggregates that are in contact with the organic OTS, hence only partial surface could be modified with OTS. This catalyst was labeled as OTS -Ti-ZrO
2
.
In the fluorination of catalyst, ammonium hexafluorosilicate ((NH
4
)
2
SiF
6
,
Aldrich) was used as the source material. A series of ammonium hexafluorosilicate solution with different concentrations were prepared: 0.2 M, 0.4 M, 0.6 M, 0.8 M and
1.0 M. 1 g of the Ti-ZrO
2
prepared above was immersed in 10 cm 3 of each solution and was shaken for 15 mins at room temperature. Different concentration of fluorine solution may result in different hydrophobicity of sample. Thus, this was aimed to determine the appropriate concentration to be used to prepare the rest of the sample.
Afterwards, 1.0 M solution was chosen. 1 g of Ti-ZrO
2
was immersed into 10 cm 3 of
1.0 M ammonium hexafluorosilicate solution and the suspension was shaken for 15 mins at room temperature. The particles were then collected through centrifugation and dried at 383 K overnight. This catalyst was labelled as F -Ti-ZrO
2
.
Based on the F -Ti-ZrO
2
, the alkylsilylation was repeated in order to complete the combination of these two modifications: alkylsilylation and fluorination. The alkylsilylation procedures described above were used. This modification provided the OTS-F -Ti-ZrO
2
. In the alkylsilylation of F -Ti-ZrO
2
catalyst, CTMS was used as well as OTS. The sample prepared by CTMS was labeled CTMS-F -Ti-ZrO
2
. As mentioned, alkylsilane groups only partially covered the surface of catalyst.
Therefore, fully modified catalyst where the surface were fully covered by alkylsilane groups were prepared using similar procedures but without the addition of water. These supplied us with the fully-OTS-F -Ti-ZrO
2
and fully-CTMS-F -Ti-ZrO
2
.
In order to prepare a series of alkylsilylated fluorinated Ti-NaY, the similar procedures as for the preparation of alkylsilylated fluorinated Ti-ZrO
2
series were repeated. Table 3.1 and Table 3.2 explain the modified catalysts that have been

35 prepared whereas Figure 3.1 and Figure 3.2 show the flow sequences of the catalyst preparation.
Table 3.1
Modified Ti- ZrO
2
series catalysts with description.
No Sample Description
1 Ti-ZrO
2
2 OTS -Ti-ZrO
2
3 F -Ti-ZrO
2 unmodified catalyst partial alkylsilylated with OTS fluorinated catalyst
4 OTS-F -Ti-ZrO
2
5 CTMS-F -Ti-ZrO
2 fluorinated and partial alkylsilylated with OTS fluorinated and partial alkylsilylated with CTMS
Table 3.2
Modified Ti-NaY series catalysts with description.
No Sample Description
1 Ti-NaY
2
4
OTS -Ti-NaY
3 F -Ti-NaY
OTS-F -Ti-NaY partial alkylsilylated with OTS fluorinated catalyst fluorinated and partial alkylsilylated with OTS
5 CTMS-F -Ti-NaY fluorinated and partial alkylsilylated with CTMS
6 fully-OTS-F -Ti-NaY fluorinated and fully alkylsilylated with OTS
7 fully-CTMS-F -Ti-NaY fluorinated and fully alkylsilylated with CTMS

Alkylsilylation with OTS
OTS -Ti- ZrO
2
Zirconium Hydroxide
+ titanium isopropoxide
Impregnation followed by calcinations at 873 K
Ti-ZrO
2
Fluorination with
(NH
4
)
2
SiF
6
F -Ti- ZrO
2
36
Alkylsilylation with OTS
OTS-F -Ti- ZrO
2
Alkylsilylation with CTMS
CTMS-F -Ti- ZrO
2
Figure 3.1
Sequences of alkylsilylated fluorinated Ti-ZrO
2
series catalysts preparation

Alkylsilylation with OTS
OTS -Ti-NaY
NaY zeolite
+ titanium isopropoxide
Impregnation
Ti-NaY
Fluorination with
(NH
4
)
2
SiF
6
F -Ti-NaY
37
Alkylsilylation with OTS
Alkylsilylation with CTMS
OTS-F -Ti-NaY
Fully-OTS-F -Ti-NaY
CTMS-F -Ti-NaY
Fully-CTMS-F -Ti-NaY
Figure 3.2
Sequences of alkylsilylated fluorinated Ti-NaY series catalysts preparation.

38
3.3
Characterization of catalysts
In this study, a thorough investigation of catalysts is carried out by employing a wide range of characterization techniques: powder X-ray diffraction (XRD), solid state nuclear magnetic resonance (NMR), ultraviolet-visible spectroscopy (UV-VIS),
Fourier transform infrared spectroscopy (FTIR), energy dispersive X-ray analysis
(EDAX) and BET surface area analysis.
3.3.1
Powder X-ray Diffraction (XRD)
Powder X-ray diffraction (XRD) patterns were collected on a Bruker D4 series diffractometer; operated at 40 kV and 40 mA and using Cu K
α
radiation ( λ =
1.5418 Å). It was used to characterize the samples crystallinity, structures and phases. Approximately 1 g of sample was carefully ground to a fine powder by a mortar. The powder sample was then put into a sample holder followed by pressing it softly between two glass slides to obtain a thin layer. After locating and locking it properly in the correct part of the diffractometer, samples were measured in the appropriate 2 θ scale according to the nature of samples. For ZrO
2
series, the samples were scanned from 20º to 40º. The NaY series, however, 2 θ ranged from 2º to 50º was used. For both types of samples, a step interval of 0.02º with counting time of 1 s per step was applied.
3.3.2
Solid State Nuclear Magnetic Resonance (NMR)
The solid state MAS NMR experiments were performed using Bruker
Avance 400 MHz 9.4 T instrument. The 29 Si MAS NMR spectra were recorded at
79.44 MHz using 5.5 µs radio frequency pulses, a recycle delay of 20 s and spinning rate of 7.0 kHz. The 13 C NMR, spectra were collected by a Cross Polarization (CP)
MAS method with a 1000 µs 13 C pulse and 5 s recycle delay. Both 29 Si and 13 C NMR chemical shifts were referred to TMS at 0 ppm.

39
3.3.3
UV-VIS Spectroscopy
Ultraviolet-visible spectra were measured on a Perkin Elmer Lambda 900
UV/VIS/NIR spectrometer. It was used to identify the local structure of materials where the absorption bands are corresponded to coordination environment. A BaSO
4 was used as reference. The catalyst powder was spread over the sample holder window flatly. The sample was analyzed from 190-600 nm and the signal was detected in absorbance rather than reflectance. All spectra were acquired under ambient conditions.
3.3.4
Fourier Transform Infrared Spectroscopy (FTIR)
Fourier Transform Infrared (IR) spectra were recorded on a Perkin Elmer
Spectrum One spectrometer. Information regarding types of chemical bonds which exist in the sample can be obtained. The sample was ground into fine powder. After that, the sample powder was mixed with KBr (10 % w/w sample) and then pressed into a thin pellet under vacuum ca. 10 tons. The KBr pellet was then put into a sample holder to record the spectrum. The sample was scanned from 400 cm -1 to
4000 cm -1 and the signal was detected in % transmittance recorded at room temperature.
3.3.5
Energy Dispersive X-ray Analysis (EDAX)
EDAX analysis was performed on a Philips X400 instrument with 40 kV. By scanning the energy (energy dispersive, ED) of the emitted X-rays from samples, it can be used to identify the elements present. With adoption of the internal calibration of the Philips instrument, a semi quantitative elemental analysis was made. In the sample preparation stage, the catalyst was dried at 60 ºC in an oven for 24 hours as moisture will cause disturbance in the measurement. Subsequently, the dried sample powder is attached onto a sample stud with adhesive double-sided tape. The sample was then kept in a desiccator filled with silica gel. One of the requirements for the sample to be analyzed is conductivity. Therefore, our sample needed to be coated

40 with gold before the examination. After coating, the sample was analyzed on different spots and the averaged results were obtained.
3.3.6
BET Surface Area Analysis
Surface area of all catalysts was analyzed using BET surface area analyzer
(Thermo Finnigan Qsurf Series). This instrument allows the use of a gas mixture, typically 30% nitrogen balance and results in determination of surface area.
Approximately 0.1 g of sample powder was put into a sample tube holder followed by preheating under 473 K in the oven attached to the instrument and connected to a supply of gas mixture (nitrogen and helium). In this stage, contaminants such as moisture and volatile compound which physically adsorbed on the catalysts surface were desorbed and being carried away with the flow of gas mixture. This prepared a clean surface which represents the real uncontaminated surface of a catalyst. After preheating for 20 minutes, the sample tube holder was immersed into the liquid nitrogen at 77 K to allow adsorption of nitrogen gas onto the surface occurred.
Desorption which took place after adsorption led to the quantification of adsorbed nitrogen. Therefore, this allowed the calculation of the catalyst surface area based on the BET equation.
3.4
Epoxidation of 1-Octene with Aqueous Hydrogen Peroxide
The activation of catalysts was carried out in a furnace before putting the catalysts into the reactant mixtures. 50 mg of all catalysts was preheated at 373 K for half an hour in order to remove the physically adsorbed moisture in the catalysts.
Typically, 1-octene (4 cm 3 , Aldrich), 30 % aqueous hydrogen peroxide (1 cm 3 ,
HANNS) and activated catalysts powder (50 mg) were placed in a glass tube and reacted at room temperature for 24 hours. In some experiments, the mixtures were stirred magnetically. The glass tube that contained reaction mixtures and catalyst powder was wrapped by aluminum foil to prevent sunlight from prompting the decomposition of aqueous hydrogen peroxide. The resulting product samples were withdrawn and analyzed at certain interval with gas chromatography (GC). Gas

41 chromatography-mass spectrometer (GC-MS) was used to verify the resulting product.
GC (Thermo Finnigan Trace One) equipped with a flame ionization detector
(FID) and a polar capillary column (Equity-1) was used to quantitatively measure the product. A calibration curve was plotted by using different concentration of authentic sample, that is 1,2-epoxyoctane (Fluka Chemika). Besides that, 1-dodecene (Aldrich) was used as the internal standard. The GC temperature programme was setup as tabulated in Table 3.3.
Table 3.3
: GC-FID temperature-programmed setup for 1,2-epoxyoctane product measurement
Initial Temperature
Hold Time
Increasing Rate
Final Temperature
Hold Time
40 ºC
5 min
10 ºC/min
150 ºC
1 min
GC-MS (Agilent 6890N-5973 Network Mass Selective Detector) is equipped with HP-5 MS column (30 m x 0.251 mm x 0.25 µm), diffusion pump and turbomolecular pump. Samples were analyzed on splitless method with helium (He) as the carrier gas. The temperature-programmed setup was same as GC-FID temperature setup and can be seen in Table 3.4.

42
Table 3.4
: GC-MSD temperature-programmed setup for verifying 1,2-epoxyoctane product.
Initial Temperature
Hold Time
Increasing Rate
Final Temperature
Hold Time
40 ºC
5 min
10 ºC/min
150 ºC
1 min

CHAPTER 4
RESULTS AND DISCUSSION
4.1
Alkylsilylated Fluorinated Catalysts
Figure 4.1 shows the apparent distribution of a series of modified Ti-ZrO
2 catalysts suspended in hydrogen peroxide and 1-octene mixture. Only the unmodified
Ti-ZrO
2
with naturally hydrophilic behaviour was dispersed well in the aqueous phase. When the Ti-ZrO
2
hydrophilic surface was partially modified with OTS, these particles became more hydrophobic due to the coverage of alkylsilyl groups. A small amount of water was added to the dried Ti-ZrO
2
powder before the reaction with
OTS in order to adjust the surface coverage of alkylsilyl groups on the surface. Since the capillary force of water induces tight adhesion between the catalyst particles, the particles were aggregated. The binding water inhibits the attachment of hydrophobic
OTS groups into the aggregates; OTS molecules can only react with the outer surface of the aggregates. Therefore, the surface of
OTS
-Ti-ZrO
2
is expected to be partially hydrophobic, having both the hydroxyl and alkylsilyl faces. Thus, the
OTS
-Ti-ZrO
2 particles were located at the aqueous-organic interphase. The rest of the catalysts were all retained at the interphase between the two immiscible phases as they were modified with hydrophobic-inducing agent. The hydrophobic-inducing agents used were octadecyltrichlorosilane (OTS), trimethylsilane (CTMS) or fluorosilicate (F).
Although the fluorinated Ti-ZrO
2
(
F
-Ti-ZrO
2
) was slightly hydrophobic and stayed at the interphase, its stability at the interphase was not maintained. The
F
- Ti-
ZrO
2
dropped to the bottom of the aqueous phase by gravity after a few hours. Even the fluorosilicate functional groups possess hydrophobicity, its hydrophobicity is not

44 strong enough to sustain the support material at the interphase for a long period of time. This might be due to a low surface coverage of fluorosilicate on the Ti-ZrO
2
.
Thus, there is a need to enhance the hydrophobic behaviour of
F
-Ti-ZrO
2
using OTS or CTMS to give OTS-F -Ti-ZrO
2
and CTMS-F -Ti-ZrO
2
catalysts. The OTS-F -Ti-
ZrO
2
and CTMS-F -Ti-ZrO
2
catalysts are hydrophobic and stable at the interphase, as expected.
Figure 4.1: Photographs of modified Ti-ZrO
2
catalysts: Ti-ZrO
2
, OTS-Ti-ZrO
2
,
F-Ti-ZrO
2
, OTS-F-Ti-ZrO
2
and CTMS-F- Ti-ZrO
2
(from left to right).
The modified Ti-NaY series catalysts suspension in hydrogen peroxide and 1octene mixture are shown in Figure 4.2. A similar trend in the particles distribution of the Ti-NaY series and Ti-ZrO
2
series catalysts in this immiscible liquid-liquid system was observed. All modified Ti-NaY samples were retained at the interphase except for the unmodified Ti-NaY which was dispersed in the aqueous phase. All

45 zeolites are hydrophilic in nature. When some Si 4+ ions become partly substituted by
Al 3+ ions in the tetrahedral SiO
4
framework, a negative charge would result on the Al atom. Ordinarily, counter ions such as alkaline and alkaline earth ions, as well as protons, can compensate for the charge and induces affinity for water. This is the reason why zeolites are originally hydrophilic. In comparison, zirconia is less hydrophilic than zeolite. Due to the intrinsic hydrophilicity of the NaY zeolite and titanium oxide spesies, Ti-NaY particles were well dispersed in the aqueous phase, as expected.
The Ti-NaY series catalysts were modified with the same hydrophobicinducing agent as Ti-ZrO
2
series; they were octadecyltrichlorosilane (OTS), trimethylsilane (CTMS) or fluorosilicate (F). In contrast to the fluorinated Ti-ZrO
2
(
F-
Ti-ZrO
2
) which stayed for a few hours at the phase boundary, the fluorinated Ti-
NaY (
F
-Ti-NaY) seems to be less stable at the interface. The
F
-Ti-NaY particles stayed at the phase boundary only for about half an hour compared to the
F-
Ti-ZrO
2 which managed to maintain for a few hours. Due to the same reason as for
F
-Ti-ZrO
2
, the hydrophobicity of F-Ti-NaY was enhanced by OTS and CTMS.
Figure 4.2: Photographs of modified Ti-NaY catalysts: Ti- NaY, OTS-Ti- NaY,
F-Ti-NaY, OTS-F-Ti-NaY and CTMS-F-Ti-NaY (from left to right).

46
4.2
The Effect of Alkylsilane Coverage on Catalyst - NMR
NMR is very sensitive to the local chemical environment and the molecular conformation. The relative concentration of trans and gauche conformations influence the 13 C chemical shift of the alkyl chains. Normally it is difficult to obtain detailed information on local molecular conformation by using imaging and scattering techniques.
The carbon atoms of n-alkanes gave a resonance at 30 ppm in solution where equilibrium concentration of gauche and trans conformations exist [58]. However, in the crystalline state, a resonance peak at 34 ppm is observed for well-oriented methylene carbons in the all-trans conformation. Figure 4.3 shows the 13 C CP/MAS spectra of
OTS
-HZSM-5 and fully-OTS
-HZSM-5. In this work the zeolite support used is HZSM-5 instead of NaY zeolite. Since the objective of the 13 C CP/MAS
NMR study is to elucidate the molecular conformation of the alkyl chains, there is no influence expected from the support. Moreover, HZSM-5 is also microporous; analogous to NaY and contains no carbon. Therefore, HZM-5 can be used as the host material in the study of the arrangement of alkylsilane groups on the catalyst surface.
The 13 C CP/MAS NMR spectra in Figure 4.3 shows that there is a resonance peak at
34 ppm (I) for both catalysts which corresponds to the well-oriented methylene carbons. The peak at 32 ppm (II) which appears as a shoulder in the spectrum is assigned to the less-oriented methylene chain [58]. Thus, the chemical-shift position of the resonance associated with the internal methylene can be used to measure the degree of order in the long-chain molecules. By comparison of the peak at 34 ppm
(highly ordered methylene chains) and 32 ppm (less ordered methylene chains), it suggests that the chain conformation of fully-OTS -HZSM-5 is more symmetrical and compact than the OTS -HZSM-5 which is more irregular and loosely packed.

47
I
I.
“ordered packing of methylene chains”
(a)
I
II zeolite
II. “disordered packing of methylene chains”
II
(b)
40 35 30 ppm
Figure 4.3
13
C CP/MAS NMR of (a) fully-OTS-HZSM-5 and (b) OTS-HZSM-5.
4.3
Vibrational Spectroscopy of Alkylsilylated Fluorinated Zeolite - FTIR
The infrared spectra in Figure 4.4 show the characteristic vibration bands of
NaY zeolite, Ti-NaY, OTS -Ti-NaY and F -Ti-NaY. Firstly, in the 4000-3000 cm -1 spectra region, where stretching vibrations of hydroxyl groups appear, a broad band extending between 3000 and 3500 cm -1 can be discerned for all samples. This band

48 comes from H-bridging hydroxyl groups (-Si-OH…O-Si). The high intensity of this band corresponds to the large amount of silanol groups present in the zeolite.
Secondly, samples exhibit strong vibration in the wavenumber range of 1020 and
1100 cm -1 , assigned to asymmetric Si-O-Si stretching vibration [67]. Meanwhile, the symmetric stretching of Si-O-Si and Si-O-Al bands occur at the region between 740 and 830 cm -1 are observed. The spectra of OTS -Ti-NaY demonstrate the various C-H stretching vibrations at 2920 and 2847 cm -1 . These C-H groups are from the hydrophobic-inducing agent, i.e. the OTS. However, the spectra observed for F -Ti-
NaY is different; indicating that the structure of NaY has been altered and collapsed as observed by the XRD technique later. The band observed in
F
-Ti-NaY around
1400 cm -1 may be due to the Si-F bonding [74]. The hydrofluoric acid that may have been formed during the process of fluorination attacked the aluminosilicate structure and led to collapse.
All infrared spectra of modified Ti-NaY samples shows a distinct peak which is absent in the spectra of pure NaY sample. This peak is found at 910 - 950 cm -1 region and is an indication of the isomorphous substitution of Si by Ti atoms in the lattice. This 950 cm -1 vibrational band can be assigned to asymmetric Ti-O-Si stretching mode [67]. In zeolite, Si-O-Si peak appears around 1060 cm -1 . The downward shift of the frequency on Ti substitution is due to two possible reasons.
Firstly, because of the higher atomic mass of the Ti atoms, the vibration peak shifts to lower frequency. The second reason suggests an inequivalence of the Si-O and Ti-
O bonds length in the structure. Ti-O bonds are about 20 pm longer than the Si-O bonds [18].
Comparatively, there is not much information could be obtained from the infrared spectra of metal oxides. This is due to metal oxides are not sensitive to infrared energy. They are usually widely investigated through the UV-VIS spectrometry. Figure 4.5 shows the infrared spectra of Ti-ZrO
2
, OTS -Ti-ZrO
2
, F -Ti-
ZrO
2
and OTS-F -Ti-ZrO
2
.
OTS -Ti-ZrO
2
and OTS-F -Ti-ZrO
2 exhibit peaks at 2919 and 2846 cm -1 which attributed to various C-H groups stretching. These C-H groups are from the alkylsilane. On the other hand, a peak at around 2300 cm -1 exists in all infrared spectra of the zirconia series. This peak is commonly related to the CO
2 from the air.

1638
49
(a)
786
576.
3456
3442
1638
1024
(b)
944
789
594.
2847
3427 2920
1635
1082.
1031
930
(a)
(c)
(d)
941
1635
3170
1411 1093
4000.0 3000 2000 1500 1000 cm -1
Figure 4.4: IR spectra of (a) NaY, (b) Ti-NaY, (c) OTS-Ti-NaY and
(d) F-Ti- NaY
450.0

50
(a)
%T
(b)
(c)
2846
2330
2919
(d)
4000.0
3000 2000 cm -1
1500 1000 450.0
Figure 4.5: IR spectra of (a) Ti-ZrO
2
, (b) OTS-Ti-ZrO
2
, (c) F-Ti-ZrO
2
and
(d) OTS-F-Ti-ZrO
2

51
4.4
XRD Analysis on the Structure of Catalyst with Various Fluorine
Concentrations
Figure 4.6 present the X-ray diffractograms which demonstrates the effect of different fluorine concentration onto the zirconia structure as the support material. It is apparent that zirconia structure remained intact regardless of the concentration of fluorine used in the fluorination modification. As we know, zirconium oxide is one of the most stable structures ever known [75]. It is unlikely to be dissolved or digested by any chemical means. For that reason, the structures of zirconia were observed to be unaltered.
However, for the support materials made up of zeolite Y, it can be clearly seen that the NaY structure remained intact only when the concentration of fluorosilicate solution used was 0.2 M (see Figure 4.7). As from 0.6 M until 1.0 M, all the structures collapsed and new phases were formed. They are Malladrite
(Na
2
SiF
6
) and Rosenbergite (AlF
3
.H
2
O) [76, 77]. Being water soluble, these new phases were removed after washing with excessive water. Although the structure of sample with 0.2 M remained undamaged, it showed no hydrophobicity at all as confirmed by hydrophobicity testing. The samples with 0.2 M, 0.4 M and 0.6 M fluorine were hydrophilic. When powder samples of these were put into the mixture of aqueous hydrogen peroxide and 1-octene, these powders dropped to the aqueous phase immediately without any delay in the organic phase. Thus, the concentration used to introduce hydrophobicity was inadequate although the structure of NaY was collapsed. Only samples with 0.8 M and 1.0 M fluorine were retained in the organic phase. The sample with 1.0 M fluorine showed the longest withholding time in the organic phase. Therefore, the concentration of 1.0 M was significant to introduce a certain degree of hydrophobicity to the sample.
The reason leading to the collapse of zeolite samples may probably be due to the formation of hydrofluoric acid during the fluorination process. While aluminum fluorosilicate as the source of fluorine was dissolved in water, it would most likely form a low concentration of HF. Hydrofluoric acid is able to dissolve silica.
Consequently, even in low concentration of HF, the structure of zeolite could be affected during the 15 minutes contact time during fluorination.

52 a) 0.0 M b) 0.2 M c) 0.4 M d) 0.6 M e) 0.8 M f) 1.0 M
20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
2-Theta - Scale
40
Figure 4.6: X-ray diffractograms to demonstrate the effect of different fluorine concentrations on the Ti-ZrO
2
structure.

53 a) 0.0 M b) 0.2 M c) 0.4 M d) 0.6 M e) 0.8 M f) 1.0 M
2 10 20
2-Theta - Scale
30 40
Figure 4.7
:
X-ray diffractograms to demonstrate the effect of different fluorine concentrations on the Ti-NaY structure.
50

54
4.5
Water Adsorption Study on Hydrophobicity of Alkylsilylated
Fluorinated Catalysts
Figure 4.8 demonstrates the water adsorption capacity of alkylsilylated fluorinated Ti-ZrO
2
series catalysts at ambient temperature. As we can see, only unmodified Ti-ZrO
2
and OTS -Ti-ZrO
2
samples adsorbed water to a slight extent, with 2.64 wt% and 1.10 wt% respectively. The rest of the samples did not adsorb water at all. They are highly hydrophobic. As mentioned before, ZrO
2
itself is not as hydrophilic as zeolite materials. In ZrO
2
framework, Zr-O-Zr bonds are abundant.
These bonds do not have affinity towards water molecules [78]. However, water molecules can always be physically adsorbed in the pores or surface of ZrO
2
. This explains why the Ti-ZrO
2
gives such low water adsorption capacity. After being modified with OTS, the
OTS
-Ti-ZrO
2
shows a decrease in water adsorption indicating that the OTS is an effective hydrophobic-inducing agent. On the other hand, the sample shows no water was adsorbed after fluorination with fluorosilicate.
This result gives the insight that the F-modified catalyst is more hydrophobic.
However, in the previous part, we observed that although
F
-Ti-NaY is hydrophobic, its stability at the interphase was not maintained for a long period of time. As we know, OTS consists of 18-methylene carbon chain and this long carbon chain is highly water resistant. Therefore, when the OTS-modified catalyst stays at the interphase, the long carbon chain is capable to support the host materials at the interphase. However, due to the bulk molecular size of OTS, the Ti-ZrO
2
surface is not fully covered by the OTS. The large molecular size of OTS makes them difficult to pack too close to each other on the Ti-ZrO
2
surface. Under this condition, water molecules still have available spaces to adsorb during the water adsorption experiment.

55
4
3
2
Ti-ZrO
2
OTS -Ti-ZrO
2
F -Ti-ZrO
2
OTS-F -Ti-ZrO
2
CTMS-F -Ti-ZrO
2
0
0 5 10 15 20
Time (hrs)
25 30 35
Figure 4.8: Water adsorption capacity (w/w%) of various modified Ti-ZrO
2
.
As shown in Figure 4.9, in the water adsorption experiment, it was observed that all the Ti-NaY catalysts which were modified with octadecyltrichlorosilane
(OTS), trimethylsilane (CTMS) or fluorosilicate (F) show lower water adsorption capacity than that of the parent Ti-NaY zeolite at ambient temperature. The hydrophilic Ti-NaY shows the highest adsorption capacity with 6.27% water adsorbed by weight. In the hydrophobicity comparison between the alkylsilanemodified and the F-modified catalysts, the F -Ti-NaY presents a lower adsorption capacity (3.63%) than the OTS -Ti-NaY (4.45%). This is once again evident that the fluorosilicate is more hydrophobic than OTS. Fluorosilicate has a much smaller molecular size compared to OTS. Consequently, fluorosilicate has a larger surface coverage on the Ti-NaY surface than
OTS
-Ti-NaY; resulting in
F
-Ti-NaY being more hydrophobic. However, this hydrophobic behaviour of
F
-Ti-NaY is not strong enough to support the Ti-NaY particles at the interphase of these immiscible liquids.
Therefore, the
F
-Ti-NaY only managed to retain at the interphase for about 30 minutes followed by dropping gradually to the bottom of glass tube by gravity.
Besides that, the adsorption capacity of
OTS-F
-Ti-NaY (2.54%) is not so different

from that of the
CTMS-F
-Ti-NaY (2.58%), suggesting that these two hydrophobicinducing agents (OTS and CTMS) are similar in terms of hydrophobicity.
56
4
3
2
1
0
0
7
6
5
Ti-NaY
OTS-Ti-NaY
F-Ti-NaY
OTS-F-Ti-NaY
CTMS-F-Ti-NaY
5 10 15 20
25 30 35
Figure 4.9 Water adsorption capacity (w/w%) of various modified Ti-NaY
4.6
Titanium Species in Fluorinated Phase-Boundary Catalysts by UV-VIS
Spectroscopy
UV-Visible spectroscopy is commonly used to characterize the titanium species in Ti-containing inorganic materials. The UV-VIS wavelength at which the transition occurs is highly sensitive to the coordination of titanium sites, and in the literature this has been proposed as a probe to test titanium coordination [66]. Thus, the spectra indicate the coordination state of Ti in a catalyst. This technique is more sensitive than IR and Raman techniques in the detection of extra-framework TiO
2
.
UV-Vis spectra of fluorinated Ti-ZrO
2
series catalysts are shown in Figure
4.10. An absorption band at 228 nm is observed for all samples. It is well established that the charge transfer at around 210-230 nm characterizes isolated, tetrahedrally coordinated titanium species. This electronic transition has been assigned to a O 2→
Ti 4+ charge transfer transition of tetrahedrally coordinated titanium. The Ti(IV)

228 nm 304 nm
57
Ti-ZrO
2
OTS -Ti-ZrO
2
F
-Ti-ZrO
2
OTS-F-
Ti-ZrO
2
200 250 350 400 450 500
Wavelength (nm)
Figure 4.10: UV-Vis Spectra of a series of Ti-ZrO
2
catalysts.

58 species in tetrahedral coordination is considered to be the real active species in olefin epoxidation reactions which were carried out in this study [18]. As stated under the
Experimental section, the samples were prepared by dispersing titanium isopropoxide in cyclohexanol (solvent), followed by mixing the solution with the Ti-
ZrO
2
catalyst. Cyclohexanol was selected because it has a very strong interaction with Ti precursor in solution. This gives a protective effect on the Ti precursor, hindering its hydrolysis to form octahedrally coordinated titanium species in Ti-ZrO
2 catalyst.
In all catalysts, the presence of TiO
2
clustering can be ruled out because no absorption at 370-410 nm was detected. As can be seen, all spectra exhibit an absorption band at 304 nm, indicating that the octahedrally coordinated titanium or the polymeric titanium species were formed on zirconia surface. However, there is no band observed at around 265 nm as Ti-NaY catalysts. This indicates that the hydrated tetrahedral titanium species was not formed in zirconia surface. This is because zirconia surface is relatively less hydrophilic compared to NaY zeolite.
After the fluorination of Ti-ZrO
2 catalysts, intensity of absorption band at 304 nm decreased dramatically and the band at 228 nm became more obvious and sharp.
It indicates that the titanium species in Ti-ZrO
2
tend to form tetrahedral coordination instead of octahedral coordination. This suggests that the addition of fluorine groups towards the Ti-ZrO
2
catalysts prompted the formation of tetrahedrally coordinated titanium species. Tetrahedral titanium is the desired species acting as the active sites for the epoxidation of alkenes. Hence, the modification of Ti-ZrO
2
catalysts with fluorine is favorable in enhancing the epoxidation reactions.
Figure 4.11 shows the UV-Vis spectra of alkylsilylated fluorinated Ti-NaY series catalysts. Similar to Ti-ZrO
2
series, all Ti-NaY catalysts exhibit an absorption band at 228 nm which is attributed to the charge transfer of tetrahedrally coordinated titanium in the NaY framework. Once again, the presence of TiO
2
clustering can be ruled out because there is no absorption at 370-410 nm.
For Ti-NaY and
OTS
-Ti-NaY samples, there is an absorption band at 268 nm which is associated with hydrated titanium species, also in tetrahedral coordination.

59
However, the
F
-Ti-NaY and
OTS-F
-Ti-NaY samples show a band at 296 nm, due to the octahedrally coordinated or polymeric titanium species grown on the NaY zeolite surface.
The hydrated titanium sites are formed in NaY zeolite but absent in ZrO
2 catalyst. This finding is attributed to the more hydrophilic character of the zeolite surface, facilitating the hydration of titanium. The hydration of Ti(IV) is favoured as a consequence of the hydrophilic character of NaY zeolite. In contrast, ZrO
2
surface has no affinity towards water, thus hydration of titanium sites is not likely to happen.
As discussed in the water adsorption section, we found that fluorosilicate is more hydrophobic than OTS. Due to this, hydration of titanium site of the fluorinated Ti-
NaY is more difficult. In NaY, with the presence of fluorine element, the titanium species has the tendency to form more octahedrally coordinated titanium sites.
Therefore, fluorination effects the titanium coordination formed in NaY zeolite.
228 nm
268 nm
296 nm
Ti-NaY
OTS
-Ti-NaY
F
-Ti-NaY
OTS-F -Ti-NaY
200.0 250 300 350 400 450 500 550 600.0
Wavelength (nm)
Figure 4.11: UV-Vis Spectra of Ti-NaY series catalysts.

60
4.7
Structural Characterization of Alkylsilylated Fluorinated Catalysts by
XRD, BET Surface Area and EDAX Analysis
Figure 4.12 shows the XRD diffractograms of alkylsilylated fluorinated Ti-
ZrO
2
series catalysts calcined at 873 K for 2 hours, with absolute intensity scale having similar range from 0 to 3000 Cps. The XRD pattern of Ti-ZrO
2
in Figure 4.12 proved to be the monoclinic phase. The monoclinic phase is characterized by 2 peaks at 2 θ = 28 º and 31.5 º, which correspond to the (111) and (11-1) lines. However, a small amount of tetragonal phase is observed for all samples as well. The tetragonal form is detected by the presence of peak at 2 θ = 30 º, which corresponds to the (111) reflection. Calcined-temperature factor determines the formation of monoclinic or tetragonal phase. There were some reports on the effects of calcined-temperature in the formation of monoclinic or tetragonal phase [79]. ZrO
2
is chosen as the support host for the fluorinated phase-boundary catalyst because it is known as a very robust material. Its resistance towards most strong acid makes its structure stable for almost all environments. Furthermore, ZrO
2
is microporous, like NaY zeolite. Therefore, the comparison between the ZrO
2
and NaY zeolite as support materials is valid and mutually complemented. Detailed inspection on the XRD patterns in Figure 4.13 demonstrates that the structure of Ti-ZrO
2
is not affected by alkylsilylation as well as fluorination. All samples exhibit the monoclinic peaks at 28 º and 31.5 º as their main reflections. These findings are further supported by the BET surface area analysis.
Monoclinic ZrO
2
has surface area in the range from 30-50 m 2 /g. Any modifications done to it will give impact to its surface area. In this study, the repeatability of BET results has been confirmed by three times measurement. Table
4.1 shows the summary of BET surface area analysis of Ti-ZrO
2
series catalysts. The surface area differences between all samples are small, establishing that the monoclinic structure of ZrO
2
remained intact. The ratio of alkylsilane and fluorine groups that attached on the zirconia surface is relatively small. Therefore, the surface area of zirconia series is not much altered. On the other hand, the presence of fluorine components in the samples after fluorination is proven in the EDAX result.
The presence of trace amount of fluorine in Ti-ZrO
2
and
OTS
-Ti-ZrO
2
might be attributed to instrumental error.

61
Table 4.1 BET surface area and EDAX analysis of Ti-ZrO
2
series catalysts
Sample
Ti-ZrO
2
OTS
-Ti-ZrO
2
F-
Ti-ZrO
2
OTS-F -Ti-ZrO
2
CTMS-FTi-ZrO
2
BET surface area, m
2
/g EDAX (Fluorine, wt %)
31.65 0.24
28.86 0..35
28.39 1.22
28.2 1.14
23.08 1.25
Monoclinic
Tetragonal
Monoclinic a) Ti-ZrO
2 b)
OTS
-Ti-ZrO
2 c) F -Ti-ZrO
2 d)
OTS-F-
Ti-ZrO
2 e) C TMS-FTi-ZrO
2
20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
2-Theta
40
Figure 4.12 X-ray diffractograms of (a) Ti-ZrO
2
, (b) OTS-Ti-ZrO
2
,
(c) F-Ti-ZrO
2
, (d) OTS-F-Ti-ZrO
2
and (e) CTMS-F-Ti-ZrO
2

62
Ti-NaY series catalysts were examined by XRD technique (see Figure 4.13).
The X-ray diffraction patterns of the Ti-NaY and
OTS
-Ti-NaY zeolite exhibit intense peaks at 6º, 15.6º, 23.5º, 27º, 31.4º and some other low intensity peaks. These strong reflection peaks are characteristics of a NaY zeolite structure. As can be seen, the structure of the NaY zeolite framework remained intact after the titanium impregnation and the attachment of octadecyltrichlorosilane groups. However, the attachment of OTS groups leads to a decline in all peaks intensity indicating that the
OTS groups might have affected the crystallinity of NaY zeolite. Besides that, there is a small peak at 31.9º which represents the formation of cubic NaCl. NaCl was formed when the chlorine in OTS interacted with sodium from zeolite framework.
A remarkable finding here is the fluorination effect on the crystallinity of
NaY zeolite. As shown in Figure 4.13, all main characteristic reflection peaks of
NaY zeolite disappeared after the process of fluorination. This is observed for all fluorinated Ti-NaY, including
F
-Ti-NaY,
OTS-F-
Ti-NaY and
CTMS-F
-TiNaY samples which implied collapse of the NaY zeolite frameworks. This occurrence is due to the formation of a portion of hydrofluoric acid (HF) in the source of fluorine that was used. As discussed earlier, we used the ammonium hexafluorosilicate
((NH
4
)
2
SiF
6
) as the source of fluorine. The ammonium hexafluorosilicate dissociated in water to form ammonium ions and SiF
6
2 ions. However, a small portion of SiF
6
2ions further dissociated in water to form Si(OH)
4
and HF. As we know, HF is an effective agent to digest silica in zeolite [80]. Therefore, in the presence of HF, the
NaY crystal structure turned amorphous.
However, new phases were detected in these fluorinated Ti-NaY. The peaks at 16.6º, 20.2º, 21.5º, 27.1º and 39.9º indicate the formation of Malladrite (Na
2
SiF
6
) and Rosenbergite (AlF
3
.H
2
O) which were formed from the interaction between fluorosilicate and sodium in zeolite [76, 77]. However, these phases are water soluble, during the washing of these samples with water under stirring condition for 24 hours, these phases were removed. The effects of the removal of these phases will be discussed in the next section.

63 a) Ti-NaY b) OTS -Ti-NaY c)
F
-Ti-NaY d) OTS-F -Ti-NaY e) CTMS-F -Ti-NaY
2 10 20
2-Theta
30 40
Figure 4.13 X-ray diffractograms of (a) Ti-NaY, (b) OTS-Ti-NaY, (c) F-TiNaY ,
(d) OTS-F-Ti-NaY and (e) CTMS-F-Ti-NaY
NaY zeolite is a microporous material with fairly high surface area, ~800 m 2 /g [81]. Any modifications done to the zeolite will give impact to its surface area.
Surface area is crucial because it plays a key factor in certain catalytic reactions.
50

64
Large surface area increases the surface contact between the substrates and the catalyst surface. It also provides a medium for active site dispersion on its external surface. In the BET surface area analysis, the effect of titanium impregnation, alkylsilylation and fluorination can be observed. Table 4.2 summarizes the BET surface area and EDAX analysis of the Ti-NaY series catalysts. The presence of fluorine elements in the catalysts after fluorination is confirmed.
Table 4.2 BET surface area and EDAX analysis of Ti-NaY series catalysts
Sample BET surface area, m
2
/g EDAX (Fluorine, wt%)
Ti-NaY 718.46 0.26
OTS
-Ti-NaY 118.04 0.27
F-
Ti-NaY
OTS-F
-Ti-NaY
CTMS-FTi-NaY
22.37 6.58
14.2 4.59
16.88 3.91
The surface area of Ti-NaY remained high with 718.46 m 2 /g. OTS groups may obstruct the zeolite pores or lead to a less-ordered form of NaY zeolite structure.
Data in Table 4.2 indicates that the
OTS
-Ti-NaY possesses a smaller surface area
(118.04 m 2 /g) than that of Ti-NaY. Similarly, a drastically reduced surface area was observed for all fluorinated Ti-NaY samples. They are
F-
Ti-NaY (22.37 m 2 /g),
OTS-
F-
Ti-NaY (14.20 m 2 /g) and
CTMS-F-
Ti-NaY (16.88 m 2 /g). The marked decrease in surface area suggests that zeolite framework has turned amorphous. Once a zeolite framework collapsed from its crystalline structure to amorphous phase, the surface area is decreased significantly. The BET surface area analysis of fluorinated Ti-NaY samples agrees well with the findings observed by XRD analysis shown in Figure
4.13.

65
4.8
Catalytic Reaction: Epoxidation of 1-Octene with Hydrogen Peroxide.
Epoxidation of 1-octene with hydrogen peroxide (H
2
O
2
) as an oxidant was chosen to investigate the activity of alkylsilylated fluorinated catalysts. The alkylsilylated fluorinated catalysts in this study consist of ZrO
2
metal oxide and NaY zeolite as the support material. The reaction over these fluorinated phase-boundary catalysts is expected to yield 1,2-epoxyoctane. Primarily, the 1,2-epoxyoctane compound is verified by gas chromatography (GC) and gas chromatography-mass spectrometry (GC-MS). GC was selected for qualitative and quantitative measurement in which comparison was based on similar retention time between the product compound and 1,2-epoxyoctane standard. On the other hand, GC-MS was used in order to verify the molecular weight characteristic of the desired compound.
GC uses flame ionization detector (FID) while GC-MS uses the more sensitive mass selective detector (MSD). Therefore, GC technique can be regarded as an initial verification method while GC-MS confirms the product based on the molecular weight of the compound.
In the catalytic reaction with a duration of 24 hours under ambient temperature over the alkylsilylated fluorinated Ti-ZrO
2
series catalysts and alkylsilylated fluorinated Ti-NaY series catalysts, 1,2-epoxyoctane was identified as the sole product having the retention time ( t
R
) of 10.67 minutes similar to the standard sample. Figure 4.14 to Figure 4.17 show the chromatograms of 1,2epoxyoctane standard, blank reaction (without catalyst), reaction of
CTMS-F
-Ti-
ZrO
2
and reaction of
CTMS-F
-Ti-NaY. The chromatograms indicate that the product from the reaction is the desired compound as compared to the standard 1,2epoxyoctane standard. The blank reaction was carried out as well, under the same conditions. The blank reaction was used to confirm that the catalysts were needed to catalyze the reaction.
The verification with GC-MS instrument shows that each fragment from the mass spectrum is similar to the 1,2-epoxyoctane standard (Figure 4.18). The GC-MS technique is believed to be the most accurate verification for such compound since it shows the characteristic fingerprint of mass spectrum.

66
Ethanol
1,2-Epoxyoctane
1-Dodecene
(Internal Standard)
0.00 3.40 6.80 10.20 13.60 17.00
Time (min)
Figure 4.14 Chromatograms of 1,2-epoxyoctane standard.
Ethanol 1-Octene
Decene
1-Dodecene
(Internal Standard)
0.00 3.40 6.80
Time (min)
10.20
13.60 17.00
Figure 4.15 Chromatogram of blank reaction (without catalyst).

67
Ethanol 1-Octene
1,2-Epoxyoctane
Decene
1-Dodecene
(Internal Standard)
0.00 3.40 6.80
Time (min)
10.20
13.60 17.00
Figure 4.16 Chromatogram of the epoxidation of 1-octene with H
2
O
2
using
CTMS-F-Ti- ZrO
2
Ethanol 1-Octene
1-Dodecene
1,2-Epoxyoctane
(Internal Standard)
Decene
0.00 3.40 6.80
Time (min)
10.20
13.60 17.00
Figure 4.17 Chromatogram of the epoxidation of 1-octene with H
2
O
2
using
CTMS-F-Ti-NaY

68
(a)
(b)
(c)
Figure 4.18: Mass spectrum of 1,2-epoxyoctane from (a) standard sample, (b) reaction yield with CTMS-F-Ti-NaY catalyst and (c) reaction yield with CTMS-
F-Ti-ZrO
2
.

69
4.8.1
The Effects of Alkylsilylation and Fluorination
Table 4.3 summarizes the yields of 1,2-epoxyoctane and turnover number
(TON) for Ti atom per specific surface area in epoxidation of 1-octene with aqueous
H
2
O
2
by using Ti-ZrO
2
series and Ti-NaY series catalysts at room temperature under the condition of stirring for 24 hours. All of the catalysts showed activity for epoxidation of 1-octene to give 1,2-epoxyoctane, and the reaction did not occur without catalyst (blank reaction). Generally, modification of support materials (Ti-
ZrO
2
and Ti-NaY) with hydrophobic alkylsilane groups (OTS and CTMS) and fluorosilicate led to a remarkable yield enhancement.
Table 4.3: Epoxidation of 1-Octene by Variously Modified NaY and Zirconia
Entry Catalyst
1 None
Zirconia series
2 Ti-ZrO
2
3
OTS
-Ti-ZrO
2
4
F
-Ti-ZrO
2
5
OTS-F
-Ti-ZrO
2
6
CTMS-F
-Ti-ZrO
2
NaY series
7 Ti-NaY
8 OTS -Ti-NaY
9
F
-Ti-NaY
10
OTS-F
-Ti-NaY
11
CTMS-F-
Ti-NaY
Epoxide yield (µmol) BET (m
2
/g) TON TON/BET
0.0
74.9
- for Ti ( x 10
-3
)
- -
718.46 3.0 4.18
In particular, the alkylsilylated catalysts (
OTS
-,
OTS-F
-, C
TMS-F
-) showed much higher activities than the hydrophilic catalysts (Ti-ZrO
2 and Ti-NaY). As discussed earlier, OTS and CTMS are effective hydrophobic-inducing agents.
Through the partial modification with these agents, the catalysts became partially hydrophobic. These partial hydrophobic catalysts particles, containing both hydrophilic and hydrophobic regions, are placed at the interphase of H
2
O
2
and 1-

70 octene mixture with the hydrophobic side facing the octene phase and the hydrophilic side facing the H
2
O
2
phase. As a result, titanium active sites on the catalyst (both ZrO
2
and NaY) can be in contact with both organic octene substrate and aqueous H
2
O
2
. This realized a continuous supply of H
2
O
2
and 1-octene substrates to the titanium active sites on the particles through the hydrophobic and hydrophilic parts of the catalyst particles. Consequently, the major effect of partial hydrophobic behaviour in catalysts is to continuously attract more substrates towards the titanium active sites in order to catalyze the reaction. Hydrophilic catalysts do not have the affinity to attract 1-octene substrates to the titanium active sites hence resulted in a lower yield of product.
Besides the capability to continuously supply substrates to the active sites, another approach to enhance the catalytic activity is to further activate the active sites.
We used fluorination to achieve this purpose. As can be observed in Table 4.3, the product yield of fluorinated catalysts was notably higher than that of unmodified catalysts. The effect of fluorination is apparent. As we know, fluorine is the most electronegative element. Thus, it has a high tendency and affinity to attract electrons from its nearby elements. Based on this nature, fluorine which was introduced to the catalysts by fluorination will withdraw the electron from titanium active sites towards the fluorine sites (Ti 4+ → F ). Consequently, titanium active sites are further activated by fluorination.
Catalysts which have these partial hydrophobic and activated titanium characteristics (
OTS-F-
and
CTMS-F-
) evidently showed considerable enhancement of product yield than other catalysts. Combination of these two modifications raised a synergistic effect in this reaction.
4.8.2
The Effect of Various Alkylsilane Groups
Figure 4.19 shows the effect of various alkylsilane groups on the yield of 1,2epoxide in the alkylsilylated fluorinated catalytic system. Hydrophobic-inducing alkylsilane groups (OTS and CTMS) have the function to attract organic substrates towards the active sites on the catalyst. Although the alkylsilane groups can attract

71 organic substrates, their molecular size did affect the reaction.
Octadecyltrichlorosilane (OTS) consists of 18-carbons chain whereas trimethylsilane
(CTMS) comprises of three methyl groups. OTS which is a bulky molecule has a steric constrain effect on the organic substrates uptake. Organic substrates are relatively more difficult to approach the active sites on the catalyst surface due to the physical barrier formed from the large OTS molecules. Comparatively, CTMS which has a smaller molecular size does not act as a barrier for the organic substrates. Thus,
1-octene can approach titanium active sites more frequently resulting in a higher yield of product.
300
250
OTS-F-
CTMS-F-
200
150
100
50
0
Ti-NaY series Ti-ZrO
2
series
Figure 4.19: The Effect of various alkylsilane groups on the yield of 1,2epoxide in the alkylsilylated fluorinated catalytic system.
4.8.3
The Effect of Partial and Fully Alkylsilylated NaY Catalysts
Surface coverage of alkylsilane groups on catalyst does have effects on the epoxidation reaction. In the preparation of partially alkylsilylated catalyst which possesses both hydrophobic groups and hydrophilic regions, a small amount of water was added in order to prompt the aggregation between catalyst particles. The aggregation of particles only allowed the outer surface to be alkylsilylated, whereas

72 the inner surface still remained in the hydrophilic region. The drying process after the alkylsilylation broke the aggregation and hence, the particles with both hydrophobic alkylsilane and hydrophilic groups were obtained. However, the catalysts which did not undergo aggregation step were fully alkylsilylated in all surfaces.
Figure 4.20 and Table 4.4 exhibit the yield of 1, 2-epoxyoctane obtained from partially alkylsilylated catalyst and fully alkylsilylated catalyst. There is an apparent difference between the partially alkylsilylated and fully alkylsilylated catalysts. The
OTS-F
-Ti-NaY and
CTMS-F
-Ti-NaY produced 172.5 µmol and 246.5 µmol of 1, 2epoxyoctane respectively. For the fully alkylsilylated samples, however, fully-OTS-
F
-Ti-NaY and fully-CTMS-F
-Ti-NaY only generated 88.3 µmol and 91.4 µmol of product respectively under same catalytic conditions. These results evidently show that the titanium active sites have been restricted from contact with both aqueous
H
2
O
2
and organic 1-octene. On the other hand, in fully alkylsilylated catalysts, the surfaces were totally covered by hydrophobic alkylsilane, which prevented aqueous substrates from diffusing into the titanium sites. Partial alkylsilylation was needed since it demonstrated amphiphilic character that attracted both the organic and aqueous substrates to reach the active sites.
300
250
200
150
100
50
0
(a) (b) (c ) (d)
Figure 4.20
The yield of 1, 2-epoxyoctane obtained from (a)
OTS-F
-Ti-NaY, (b) fully-OTS-F
-Ti-NaY, (c)
CTMS-F
-Ti-NaY and (d) fully-CTMS-F
-Ti-NaY

73
Table 4.4:
The yield of 1, 2-epoxyoctane obtained from partially alkylsilylated and fully alkylsilylated Ti-NaY series.
Entry Catalysts Yield
1
OTS-F
-Ti-NaY
2 fully-OTS-F
-Ti-NaY
172.5
88.3
3
4
CTMS-F fully-CTMS-F
-Ti-NaY
-Ti-NaY
246.5
91.4
4.8.4
The Effect of Washing on Fluorinated Catalyst
In the fluorination process, fluorine groups are attached onto the active sites and surface of catalysts. The fluorine groups were impregnated onto the catalyst by mixing the fluorosilicate solution with the sample powders. The fluorinated catalysts were obtained through centrifugation without washing and followed by drying process. It is believed that fluorine groups on the catalysts would probably be hydrolyzed or removed after a long contact with water. With the intention of protecting the fluorine from being removed by water or being dissolved into the aqueous phase of reactants, alkylsilane groups may be able to shield the fluorine species from contacting with water. In order to establish this hypothesis, the fluorinated and the alkylsilylated fluorinated catalysts were stirred vigorously in water for 24 hours at room temperature followed by catalytic reaction.
Table 4.5 and Figure 4.21 present the yield of 1, 2-epoxyoctane produced by the pre-washed and post-washed fluorinated catalysts. From the results, it is proven that alkylsilane groups did play a vital role in preventing the fluorine groups from being removed by water. The fluorinated catalyst which did not contain alkylsilane showed decrease in catalytic activity due to the loss of fluorine groups after the washing process. After washing, the washed-F -Ti-NaY catalyst could only produce
75.8 µmol of 1, 2-epoxyoctane that is similar to the unmodified Ti-NaY which yields
74.9 µmol of product. This gave an insight that the fluorine had been removed and

74
Ti-NaY left. However, the catalysts which encompass both alkylsilane and fluorine maintained their catalytic performance after washing suggesting that the alkylsilane prevented the fluorine from being leached
Table 4.5
The yield of 1, 2-epoxyoctane produced by the pre-washed and postwashed fluorinated Ti-NaY series catalysts
1
2
3
4
5
6
F
-Ti-NaY washed-F
-Ti-NaY
OTS-F -Ti-NaY washed-OTS
-F-Ti-NaY
CTMS-F
-Ti-NaY washed-CTMS-F
-Ti-NaY
111.2
75.8
172.5
171.1
246.5
245.0
300
250
200
150
Pre-washed
Post-washed
100
50
0
F-Ti-NaY OTS-F-Ti-NaY CTMS-F-Ti-NaY
Figure 4.21
The yield of 1, 2-epoxyoctane produced by the pre-washed and postwashed fluorinated Ti-NaY catalysts

75
4.8.5
The Effect of Stirring
Figure 4.22 and Figure 4.23 show the effect of stirring on the yield of 1,2epoxyo ctane. The activity of hydrophilic catalysts (Ti-ZrO
2 and Ti-NaY) was only appreciable under the condition of vigorous stirring. Under static condition, there was no yield observed for these hydrophilic catalysts. However, fluorinated cataly sts
( F) gave the yield even under static condition and the product yield of fluorinated catalysts was increased when the reaction was carried out with stirring. Generally, the purpose of stirring is to improve the mass transfer between the 2 substrates.
On the other hand, the activity of alkylsilylated catalysts (
OTS-, OTS-F
-, and
TMS-F
-) was relatively independent of the stirring factor. The yield difference between the stirring condition and the static condition for these alkylsilylated catalysts was negligible. This is one of the most prominent characteristics of alkylsilylated catalysts. Partial alkylsilylation onto the catalyst particles, induc es affinity for both hydrophilic and hydrophobic compounds, and enables a continuo us supply of H
2
O
2
and organic substrates to the active sites on the particles even without stirring.

76
140
120
100
80
60
200
180
160
40
20
0
Static
Stirring
~ 0
Ti-ZrO
2
OTS -Ti-ZrO
2
F -Ti-ZrO
2
OTS-F -Ti-ZrO
2
CTMS-F -Ti-ZrO
2
Figure 4.22: The yield of 1,2-epoxyoctane using Ti-ZrO
2
series catalysts under stirring and static conditions.
T able 4.6: The yield of 1,2-epoxyoctane using Ti- ZrO
2
series catalysts under stirring and static conditions.
Ti-ZrO
2
OTS-Ti-ZrO
2
F-Ti-ZrO
2
OTS-F-Ti-ZrO
2
CTMS-F-Ti-ZrO
2
Stati c Stirri ng
0.0 32.9
28.1 49.4
51.6 52.3
77.9 96.6
180.3 194.8

77
220
200
180
160
300
280
260
240
140
120
100
80
60
40
20
0
~ 0
Static
Stirring
Ti-NaY OTS -Ti-NaY F -Ti-NaY OTS-F -Ti-NaY CTMS-F -Ti-NaY
Figure 4.23: The yi eld of 1,2-epoxyo ctan e using T i-N a stirring and static conditions.
Table 4.7: The yield of 1,2-epoxyoctane us ing Ti-NaY series catalysts under stirring and static conditions.
OTS-Ti-NaY
F-Ti-NaY
OTS-F-Ti-NaY
CTMS-F-Ti-NaY
Stat ic Stirri ng
77.3 94.1
66.0 111.2
163.3 172.5
213.5 246.5

78
4.8.6
The Effect of Reaction Duration
F igure 4.24 and Figure 4.25 show the yield of 1,2-epoxyoctane over the Ti-
ZrO
2
a nd Ti-NaY series catalysts. The steep increase in the product yield occurs during the first hour. CTMS-F -Ti-ZrO
2
and CTMS-F -Ti-NaY catalysts gave the highest yield of product during the first hour of reaction. This means that these modified catalysts which have the hydrophobic behavior and fluorine factor hav e been the active catalyst for the epoxidation reaction. After the first 5 hours, the increase in the product yield became slower. The reactions stabilized in the first 10 hours for all catalysts, followed by constant yield.
250
Ti-ZrO
2
200
OTS-Ti-ZrO
2
150 F-Ti-ZrO
2
100
OTS-F-Ti-ZrO
2
CTMS-F-Ti-ZrO
2
50
0
0 5 10 15
Time (Hours)
20 25 30
Figure 4.24: The effect of reaction time on the yield of 1,2-epoxyoctane over Ti-
ZrO
2
series catalysts.

79
Table 4.8: Yield of 1,2-epoxyoctane over Ti-ZrO
2
series catalysts
Time
(µmol)
Ti-ZrO
2
OTS
-Ti-ZrO
2
F
-Ti-ZrO
2
OTS-F
-Ti-ZrO
2
CTMS-F
-Ti-ZrO
2
0.50 13.99 23.61 25.40 40.29
1.00 17.72 31.73 32.59 51.63
2.00 25.00 37.44 38.58 65.95
5.00 27.04 40.78 46.31 80.89
39.92
57.08
81.29
147.66
10.00 32.21 46.88 51.52
24.00 32.88 49.35 52.29
91.11
96.60
189.89
194.84
300
250
200
150
100
50
Ti-NaY
OTS
-Ti-NaY
F -Ti-NaY
OTS-F -Ti-NaY
CTMS-F
-Ti-NaY
0
0 5 10 15
Time (Hours)
20 25 30
Figure 4.25: The effect of reaction time on the yield of 1,2-epoxyoctane over Ti-
NaY series catalysts.

Table 4.9: Yield of 1,2-epoxyoctane over Ti-NaY series catalysts
Time
80
Ti-NaY
OTS
-Ti-NaY
F
-Ti-NaY
OTS-F
-Ti-NaY
CTMS-F
-Ti-NaY
0.50 48.05 59.41 61.78 91.18
1.00 57.10 68.32 76.16 121.07
2.00 65.58 76.35 86.54 138.96
5.00 71.37 88.66 100.47 157.36
131.94
164.42
207.82
228.79
10.00 74.20 92.90 109.66 167.87 243.53
24.00 74.94 94.09 111.22 172.45 246.52
4 .8.7
Leaching Test
In order to check the possibility of leaching of Ti species during epoxidation, as recommended in the literature, an experiment performed under similar reaction conditi on was interrupted after 30 minutes. The H
2
O
2
and 1-octene mediums were removed with a syringe and immediately transferred to another glass tube under the same condition. The compositions were then monitored for 24 hours (Table 4.10 an d
Table 4.11). No activity was found after the catalysts were removed from the reaction, suggesting that no active tetrahedral coordinated Ti species had leached out from the catalyst. Therefore, the activity is only attributed to the Ti atoms incorporated onto the surface of the support material.

81
Table 4.10`: Leaching test for CTMS-F-Ti-ZrO
2
sample
Time (Hours) 1,2-Epoxyoctane/ µmol
0.5 39.68
1 39.63
2 39.67
5 39.82
10 39.69
24 40.16
Table 4.11: Leaching test for CTMS-F-Ti-NaY sample
Time (Hours) 1,2-Epoxyoct ane/ mol
0.5 131.94
1 132.00
2 131.96
5 131.96
10 132.09
24 131.99
4 .8.8
Postulated Epoxidation Mechanism for Alkylsilylated Fluorinated
Catalyst
It is comm only accepted that for epoxidation reaction, tetrahedral Ti is most a ctive in titanium-containing catalysts, including mixed oxides and zeolite. The active s ites for epoxidation are believed to be the titanium hydroperoxo species, Ti-
OOH as shown in Figure 4.26. Hydrogen bonding between hydrogen of water an d the terminal OH oxygen, O (2), of the bound hydroperoxo was shown to enhance the stability of the hydroperoxo intermediate. It has been indicated that the oxygen

attached to the Ti, O (1) was the active site for the alkene attack [82]. The ethyle ne interaction with O (2) was shown to be repulsive.
H
O
82
O
H
O Ti
O
(1)
O
(2)
O
H
Figure 4.26: Struc ture of hydroperoxo intermediate, Ti-OOH
In conventional environment, the Ti-O would form species 1 as depicted in
Figure 4.27. Species 1 may be converted to titanium hydroperoxo species and proceeded with the reaction. However, in our alkylsilylated fluorinated catalyti c system, hypothetically, the fluorinated Ti-O would form species 2. Species 2 may be converted to titanium hydroperoxo species as well as species 1 but with an improvement in the stability of hydroperoxo intermediate. The highly electr onegative fluorine would withdraw the electrons from Ti towards the direction where the fluorine is located. This results in oxygen atom (1) to be more vulnerable to the attack of π electron clouds of the 1-octene. Therefore, it is suggested that the fluorination of catalysts via ligand conversion of Ti-O-H to Ti-O-SiF
2
could in crease the activity of epoxidation.

H
O O
H
2
O
Ti
O
Species 1
O
83
H
2
O
2
O
H
O
H
Ti
O
O
O
H
O
H
O O
Ti
O O
F
F 2
6
H
Si
O
(NH
4
)
2
SiF
6
Si
O O
H
2
O
2
O Ti
O
H
Ti O
O
δ +
O
O O
Species 2
Alkene
F igure 4.27: Postulated epoxidation mechanism for alkylsilylated fluorinated catalyst
H

CHAPTER 5
CONCLUSIONS
This thesis describes the molecular design and tailoring of a novel heterogeneous catalysis system by the combination of alkylsilylation and fluorination on zirconia and zeolite materials. This alkylsilylated fluorinated catalyst was designed for the epoxidation of 1-octene using aqueous hydrogen peroxide towards the formation of 1, 2-epoxyoctane. The physicochemical properties of alkylsilylated fluorinated catalysts such as structure, surface and active sites nature were investigated.
The alkylsilylated fluorinated catalysts were successfully prepared through the fluorination of support materials followed by the alkylsilylation with various alkylsilanes. The alkylsilanes used in this research were octadecyltrichlorosilane
(OTS) and chlorotrimethylsilane (CTMS). The principles of this catalyst preparation are based on the hydrophobicity possessed by alkylsilane and ‘electron-withdrawing’ characteristic by fluorine. Zirconia and zeolite are naturally hydrophilic but hydrophobicity is required to attract the organic substrates in epoxidation reaction.
Thus, alkylsilane groups overcome the hydrophilic problem that conventionally hampers the practical use of zeolite and zirconia in epoxidation of alkenes. On the other hand, effects of fluorine groups which possess strong affinity toward electrons in the epoxidation reactions were studied as well.
A comprehensive investigation of alkylsilylated fluorinated catalysts was carried out by employing a wide range of characterization techniques. It was confirmed by XRD and FTIR that zeolites structures were rendered amorphous by

85 fluorination. However, zirconia retained its structure upon fluorination. In contrast, zirconia and zeolite structures were not distorted by alkylsilylation. The surface investigation perfomed by BET method supported the results from XRD, that is the surface area of catalysts decreased drastically when the zeolite structures collapsed.
Besides that, UV-VIS spectroscopy confirmed the presence of tetrahedrally coordinated titanium oxide which was responsible for the epoxidation reaction. In addition, the water adsorption experiment proved that both alkylsilane and fluorosilicate possess hydrophobic behaviour to a different degree.
13
C CP/MAS
NMR studies revealed the arrangement of OTS on the surface of catalyst. It showed that ethylene chains of OTS were polymerized and closely packed on the catalyst surface.
In conclusion, the alkylsilylated fluorinated catalysts are active and selective in the epoxidation of 1-octene with aqueous hydrogen peroxide producing 1, 2epoxyoctane. Various parameters were investigated such as the effects of using various alkylsilane groups, effects of stirring and effect of alkylsilane coverage on the catalyst surface. Leaching of titanium active sites was not observed in this study.
Catalytic study showed that 100 % selectivity of 1, 2-epoxyoctane was obtained.
Alkylsilylated fluorinated NaY with CTMS as alkylsilane group and fluorosilicate as fluorine source gives the highest yield of 1, 2-epoxyoctane within reaction time of 24 hours under stirring condition at ambient temperature. However, in terms of catalysts stability, alkylsilylated fluorinated zirconia appears to be the better catalyst.

86
REFERENCES
1.
G.D.Yadav and A.A.Pujari, (2000), “Epoxidation of Styrene to Styrene Oxide:
Synergism of Heteropoly Acid and Phase-Transfer Catalyst under Ishii-
Venturello Mechanism”,
Organic Research & Development
,
4
, 88-93.
2.
C.Palazzi, L.Oliva and G.Strukul, (2000), “Microporous Zirconia-Silica Mixed
Oxides Made by Sol-Gel as Catalysts for the Liquid-Phase Oxidation of Olefins with Hydrogen Peroxide”,
J. Catal
,
194
, 286-293.
3.
B.Notari, (1993), “Titanium Silicalites”,
Catal. Today
,
18
, 163-172.
4.
M.G.Clerici and P.Ingallina, (1993), “Epoxidation of Lower Olefins with
Hydrogen Peroxide and Titanium Silicalite”,
J. Catal.
,
140
, 71-83.
5.
L.J.Davies, P.Macmom, D.Bethell et.al.
, (2001), “Oxidation of Crotyl Alcohol using Tiβ and Ti-MCM-41 Catalysts”,
J. Mol. Catal. A: Chem
,
165
, 243-247.
6.
M.G.Clerici, (1991), “Oxidation of Saturated Hydrocarbons with Hydrogen
Peroxide, Catalysed by Titanium Silicalite”,
Appl. Catal.
,
68
, 249-261.
7.
J.L.Motz., H.Heinichen and W.F. Holderich, (1998), “Direct Hydroxylation of
Aromatics to Their Corresponding Phenols Catalysed by H-[Al]ZSM-5
Zeolite”, J. Mol. Catal, 136 , 175-184.
8.
M.G.Clerici, G.Bellussi and U.Romano, (1991), “Synthesis of Propylene Oxide from Propylene and Hydrogen Peroxide Catalyzed by Titanium Silicalite”,
J.
Catal
.,
129
, 159-167.

87
9.
J.Q.Yu, Z.C.Feng, C.Li et. al.
, (2001), “Ti-MCM-41 Synthesized from
Colloidal Silica and Titanium Trichloride: Synthesis, Characterization and
Catalysis”,
Chem. Mater
.,
13
, 994-998.
10.
Z.H.Luan and L.Kevan, (2001), “Characterization of Titanium-Containing
Mesoporous Silica Molecular Sieve SBA-15 and Generation of Paramagnetic
Hole and Electron Centers”, Microporous and Mesoporous Mater , 45-45 , 337-
344.
11.
M.C. Capel-Sanchez, J.M.Campos-Martin and J.L.G.Fierro, (2003), “Influence of Solvent in the Synthesis Steps of Titanium-Supported Amorphous Silica
Epoxidation Catalysts”,
J. Catal.
,
217
, 195-202.
12.
A.Corma, P.Esteve and A.Martinez, (1996), “Solvent Effects during the
Oxidation of Olefins and Alcohols with Hydrogen Peroxide on Ti-Beta
Catalyst: The Influence of the Hydrophilicity-hydrophobicity of the Zeolite”,
J.
Catal.
,
161
, 11-19.
13.
S.Klein, S.Thorimbert and W.F.Maier, (1996), “Amorphous Microporous
Titania-Silica Mixed Oxides: Preparation, Characterization, and Catalytic
Redox Properties”,
J. Catal.
,
163
, 476-488.
14.
R.Neumann and M.Levin-Elad, (1997), “Metal Oxide (TiO
2
, MoO
3
, WO
3
)
Substituted Silicate Xerogels as Catalysts for the Oxidation of Hydrocarbons with Hydrogen Peroxide”,
J. Catal
.,
166
, 206-217.
15.
H.Nur, S.Ikeda and B.Ohtani, (2001), “Phase-Boundary Catalysis of Alkene
Epoxidation with Aqueous Hydrogen Peroxide using Amphiphilic Zeolite
Particles Loaded with Titanium Oxide”, J.Catal
., 204 , 402-408.
16.
H.Nur, S.Ikeda and B.Ohtani, (2000), “Phase-Boundary Catalysis: A New
Approach in Alkene Epoxidation with Hydrogen Peroxide by Zeolite Loaded with Alkylsilane-Covered Titanium Oxide”,
Chem. Commun.
, 2235-2236.

88
17.
S.Ikeda, H.Nur, T.Sawadaishi et. al
., (2001), “Direct Observation of Bimodal
Amphiphilic Surface Structures of Zeolite Particles for A Novel Liquid-liquid
Phase Boundary Catalysis”,
Langmuir
,
17
, 7976-7979.
18.
T.Tatsumi, (1997), “Metallozeolites and Applications in Catalysis”, Solid State
& Materials Science , 2 , 76-83.
19.
T.M.Tillotson, K.G.Foster and J.G.Reynolds, (2004), “Fluorine-Induced
Hydrophobicity in Silica Aerogels”,
Journal of Non-Crystalline Solids
,
350
,
202-208.
20.
P.N.Rylander, H.Greenfield and R.L.Augustine, (1988), “Catalysis of Organic
Reactions”,
Marcel Dekker Inc
., pg 177-210, 359-379.
21.
G.D.Yadav, J.J.Nair, (1999), “Review: Sulfated Zirconia and its Modified
Versions as Promising Catalysts for Industrial Processes”,
Microporous
Mesoporous Mater
.,
33
, 1.
22.
J.H.Clark, G.L.Monks, J.F.White and et. al
., (2000), “A New Solid Acid-Based
Route to Linear Alkylbenzenes”,
J. Catal.
,
193
, 348-350.
23.
J.H.Clark, G.L.Monks, J.F.White and et. al., (2000), “Production of Aromatics
Catalyzed by Commercial Sulfated Zirconia Solid Acids”,
Marcel Dekker, New
York
, pg 135-143.
24.
T.Okuhara, (2002), “Water-Tolerant Solid Acid Catalysts”, Chem. Rev ., 102 ,
3641-3666.
25.
K.Segawa, T.Ozawa, (1999), “Two-Dimensional Composite Zirconium
Phosphonates: Preparation and Catalytic Activities”, J. Mol. Catal A: Chem,
141
, 249-255
26.
G.Strukul, M.Morandin and R.Gavagnin, (2002), “Oxidation of Cyclohexene with Hydrogen Peroxide using Zirconia-Silica Mixed Oxides: Control of the

89
Surface Hydrophilicity and Influence on the Activity of the Catalyst and
Hydrogen Peroxide Efficiency”,
J. Catal
.,
212
, 193-200.
27.
S.A.Holmes, F.Quignard and A.Choplin, (1998), “Tetraneopentyltitanium
Derived Silica Supported Complexes: Part 2. Influence of the Hydrophobic
Character of Silica on the Catalytic Properties” J. Catal.
, 76 , 182-191.
28.
S.Gontier and A.Tuel, (1996), “Novel Zirconium Containing Mesoporous
Silica for Oxidation Reactions in the Liquid Phase”,
Appl. Catal. A: General
,
143
, 125-135.
29.
C.N.Satterfield, (1980), “Heterogeneous Catalysis in Practice”,
McGraw-Hill
Inc.
, pg 7-22, 102-105, 164-169.
30.
T.S.Newmeyanova, Y.V.Mirskii, T.V.Goos, (1978), “Preparation of Ultrastable
Faujasites and Investigation of their Catalytic Properties in Cracking Petroleum
Feedstock”,
Kinet. Catal. (USSR)
,
19
, 458-462.
31.
W.R.Moser, (1980), “Catalysis of Organic Reactions”,
Marcel Dekker Inc
., pg
17-37.
32.
G.J.Hutchings, (2001), “Promotion in Heterogeneous Catalysis: A Topic
Requiring a New Approach”,
Catal. Letter
,
75
, 1-12.
33.
W.Adam, C.M.Mitchell and O.Weichold, (2000), “Titanium-Catalysed
Oxidation of Thianthrene 5-Oxide in Heterogeneous (Zeolites) Versus
Homogeneous Media: Electrophilic Versus Template-mediated Oxygen
Transfer”, J. Mol. Catal A: Chem , 154 , 251-255.
34.
V.Conte, F.D.Furia and G.Licini, (1997), “Liquid Phase Oxidation Reactions by Peroxides in the Presence of Vanadium Complexes”,
Appl. Catal. A:
General, 157
, 335-361.

90
35.
P.Ingallina, M.G.Clerici and G.Bellussi, (1995), “Catalysis with TS-1: New
Perspectives for the Industrial Use of Hydrogen Peroxide”,
Elsevier Science
,
31-40.
36.
D.Prasetyoko, Z.Ramli, S.Endud and H.Nur, (2005), “Preparation and
Characterization of Bifunctional Oxidative and Acidic Catalysts Nb
2
O
5
/TS-1 for Synthesis of Diols” Materials Chemistry and Physics , 93 , 443-449.
37.
T.Mallat, A.Baiker, (1999), “Organically Modified Titania-Silica Aerogels for the Epoxidation of Olefins and Allylic Alcohos”
J. Catal.
,
184
, 280-293.
38.
Y.Deng, W.F.Maier, (2001), “Hydrophobic Ti-Si Xerogels: Catalysts for the
Selective Epoxidation of Olefins and Allylic Alcohols”,
J.Catal.
,
199
, 115-122.
39.
T.Tatsumi, K.A.Koyano and N.Igarashi, (1998), “Remarkable Activity
Enhancement by Trimethylsilylation in Oxidation of Alkenes and Alkanes with
H
2
O
2
Catalyzed by Titanium-containing Mesoporous”,
Chem. Commun
., 325-
326.
40.
G.Yasuhide, P.Wu and T.Tatsumi, (2004), “Influence of Fluorine on the
Catalytic Performance of Ti-Beta Zeolite”,
J.Phys. Chem. B
,
108
, 4242-4244.
41.
A.Corma, P.Esteve and S.Valencia, (1995), “Oxidation of Olefins with
Hydrogen Peroxide and Tert-Butyl Hydroperoxide on Ti-Beta Catalyst”,
J.
Catal.
,
152
, 18-24.
42.
A.Bhaumik and T.Tatsumi, (2000), “Organically Modified Titanium-rich Ti-
MCM-41, Efficient Catalysts for Epoxidation Reactions”, J. Catal.
, 189 , 31-39.
43.
M.Ogura, S.Nakata and M.Matsukata, (2001), “Effect of NH +
4
Exchange on
Hydrophobicity and Catalytic Properties of Al-free Ti-Si-Beta Zeolite”,
J.
Catal., 199
, 41-47.

91
44.
M.A.Camblor, M.Constantini, A.Corma and et. al
., (1995), “A New Highly
Efficient Method for the Synthesis of Ti-Beta Zeolite Oxidation Catalyst”,
Appl. Catal.
,
133
, L185-L189
45.
M.A.Camblor, S.Iborra, A.Corma and et. al ., (1997), “Beta Zeolite as a
Catalyst for the Preparation of Alkyl Glucoside Surfactants: The Role of
Crystal Size and Hydrophobicity”, J. Catal ., 172 , 76-84.
46.
T.Blasco, M.A.Camblor, A.Corma and et. al
., (1998), “Direct Synthesis and
Characterization of Hydrophobic Aluminum-free Ti-Beta Zeolite”,
J. Phys.
Chem
.
B
,
102
, 75-88.
47.
A.G.Panov, V.Gruver and J.J.Fripiat, (1997), “Fluorination of USY and
Modification of its Catalytic Properties”,
J. Catal.
,
168
, 321-327.
48.
N.A.Sanchez, J.M.Saniger J.J.Fripiat et. al.
, “Dealumination and Surface
Fluorination of H-ZSM-5 by Molecular Fluorine”,
Microporous and
Mesoporous Mater.
,
50
, 41-52.
49.
B.M.Lok and T.P.J.Izod, (1982), “Modification of Molecular Sieves –Direct
Fluorination”,
Zeolites
,
2
, 66-67.
50.
N.A.Sanchoz, J.M.Saniger and J.J.Fripiat, (2001), “Reaction of HY Zeolite with Molecular Fluorine”,
J. Catal.
,
201
, 80-88.
51.
M.Han, C.Xu, J.Lin, Y.Liang and E.Roduner, (2003), “Alkylation of Benzene with Long-Chain Olefins Catalyzed by Fluorinated β Zeolite”, Catal. Letters ,
86 , 81-86.
52.
S.Kowalak, E.Szymkowiak and K.A.Becker, (1999), “Catalytic Properties of
Zeolite CeY Modified with NH
4
F”,
Journal of Fluorine Chemistry
,
93
, 175-
180.

92
53.
C.N.Satterfield and G.T.Chin, (2004), “Sorption and Diffusivity Characteristics of Acid-Leached Mordenites”,
AICHE Journal
,
20
, 522-529.
54.
B.D.Cullity, (1956), “Elements of X-Ray Diffraction”, 2 nd ed., Adison-Wesley
Publishing Co. Inc.
, pg 84-87 .
55.
W.M.Meier and D.H.Olson, (1987), “Atlas of Zeolite Structure Types, 2 nd
Revised Edition”, Butterworth & Co. London .
56.
Pecharsky and K.Vitalij, (2003), “Fundamentals of Powder Diffraction and
Structural Characterization of Materials”,
Kluwer Academic Pubs
.
57.
C.A.Fyfe, (1983), “Solid State NMR for Chemists”,
CFC Press
, pg 10-17.
58.
L.Q.Wang, G.J.Exarhos and J.Liu, (1999), “Nuclear Magnetic Resonance –
Characterization of Self-Assembled Nanostructured Materials”,
Adv. Mater
.,
11
, 1331-1341.
59.
G.E.Maciel, C.E.Bronnimann and A.Jurkiewicz, (1991), “Recent Advances in
Coal Characterization by 13 C and 1 H NMR”,
Fuel
,
70
, 925-930.
60.
A.Badia, W.Gao and L.Reven, (1996), “Structure and Chain Dynamics of
Alkanethiol-Capped Gold Colloids”,
Langmuir
,
12
, 1262-1269.
61.
T.Trung, W.J.Cho and C.S.Ha, (2003), “Preparation of TiO
2
Nanoparticles in
Glycerol-Containing Solutions”, Materials Letters , 57 , 2746-2750.
62.
R.M.Mohamed, A.A.Ismail and I.A.Ibrahim, (2005), “Preparation of TiO
2
-
ZSM-5 Zeolite for Photodegradation of EDTA”, J. Mol. Catal A: Chem, 238 ,
151-157.
63.
R.Williams, (1996), “Spectroscopy and The Fourier Transform: An Interactive
Tutorial”,
VCH Publishers Inc.
, pg 7-9.

93
64.
M.Manzoli, A.Chiorino and F.Boccuzzi, (2004), “Decomposition and
Combined Reforming of Methanol to Hydrogen: A FTIR and QMS Study on
Cu and Au Catalysts Supported on ZnO and TiO
2
”,
Appl. Catal. B:
Environmental , 57 , 201-209.
65.
K.Nakamoto, (1962), “Infrared and Raman Spectra of Inorganic and
Coordination Compounds”, A Wiley-Interscience Publication- John Wiley &
Sons.
, pg 3-19.
66.
D.M.Sauer, (1996), “Coordination, Structure, and Vibrational Spectra of
Titanium in Silicates and Zeolites in Comparison with Related Molecules”,
J.
Phys. Chem
.,
100
, 5025-5031.
67.
K.S.Smirnov and V.D.Graaf, (1996), “On the Origin of the Band at 960 cm -1 in the Vibrational Spectra of Ti-Substituted Zeolites”,
Microporous Mater
,
7
, 133-
138.
68.
S.J.Gregg and K.S.W.Sing, (1982), “Adsorption, Surface Area and Porosity”,
2 nd
ed., Academic Press Inc.
, pg 1-5.
69.
K.Robards, P.R.Haddad and P.E.Jackson, (1997), “Principles and Practice of
Modern Chromatographic Methods”,
Academic Press
, pg 8-21.
70.
H.M.McNair, (1998), “Basic Gas Chromatography”, John Wiley & Sons, New
York.
71.
D.T.E.Hunt and A.L.Wilson, (1986), “The Chemical Analysis of Water:
General Principles and Techniques”, The Royal Society of Chemistry London , pg 449-465.
72.
G.M.Message, (1984), “Practical Aspects of Gas Chromatography/Mass
Spectrometry”,
John Wiley & Sons. Inc.
, pg 1-37.

94
73.
F.W.Karasek and R.E.Clement, (1988), “Basic Gas Chromatography-Mass
Spectrometry: Principles and Techniques”,
Elsevier
, pg 47-48.
74.
B.G.Kim, S.Y.Kang and J.J.Kim, (1997), “FTIR Study of Fluorinated Silicon
Oxide Film”, J.Phys. D: Appl. Phys.
, 30 , 1720-1724.
75.
S.Somiya, N.Yamamoto and H.Yanagida, (1988), “Science and Technology of
Zirconia”, American Ceramic Society .
76.
F.Olmi, C.Sabelli and R.T.Ferroni, (1993), “Rosenbergite, a New Mineral
From the Cetine Mine: Description and Crystal Structure”,
Eur. J. Mineral
,
5
,
1167-1174.
77.
M.Carati, (1982), “Guida Alla Mineralogia Vesuviana 2 nd Ed”,
Napoli Acc. Sc.
Fis. Mat
.
78.
V.A.Potyomkin and Y.L.Sukharev, (2003), “Formation of Liotropic Features of
Zirconium Oxyhydrate Gels”,
Chem. Phys. Letters
,
371
, 5-6.
79.
H.Yang, R.Lu, L.Wang et. al.
, (2003), “Preparation, Characterization and
Catalytic Activity of Sulfated Zirconia-Silica Nanocrystalline Catalysts”,
Mater
Letters
,
57
, 2572-2579.
80.
W.C.Zhou, C.T.Zhang and A.Z.Fu, (1988), “Modification of the Hydrofluoric
Acid Leaching Technique: Corundum-Mullite-Glassy Phase Materials”,
Journal of American Ceramic Society , 71 , 395.
81.
Y.K.Ryu, J.W.Chang and C.H.Lee, (2002), “Adsorption Isotherms of Toluene and Gasoline Vapours on DAY Zeolite”, J. Chem. Eng. Data , 47 , 363-366.
82.
M.Neurock and L.E.Manzer, (1996), “Theoretical Insights on the Mechanism of Alkene Epoxidation by H
2
O
2
with Titanium Silicalite”,
Chem. Commun
,
1133-1134.

APPENDIX A EDAX Result (Ti-NaY)
95

APPENDIX A EDAX Result (OTS-Ti-NaY)
96

APPENDIX A EDAX Result (F-Ti-NaY)
97

APPENDIX A EDAX Result (OTS-F-Ti-NaY)
98

APPENDIX A EDAX Result (CTMS-F-Ti-NaY)
99

100
WATER ADSORPTION DATA
Ti-NaY Catalysts
Time (hr) Ti-NaY OTS-Ti-NaY F-Ti-NaY OTS-F-Ti-NaY CTMS-F-Ti-NaY
0
1 3.39 2.04 1.69 1.64
2 4.84 2.98 1.84 1.89
3 5.34 3.08 1.99 1.94
1.69
1.89
1.99
4 5.68 3.18 2.24 2.24
5 5.99 3.43 2.29 2.24
6 6.04 3.98 2.49 2.24
7 6.09 3.63 3.08 2.44
2.24
2.24
2.49
2.54
8 6.09 3.53 3.08 2.44
24 6.14 4.18 3.5 2.54
25 6.14 4.33 3.54 2.69
26 6.29 4.33 3.58 2.74
27 6.24 4.33 3.58 2.44
28 6.25 4.42 3.58 2.44
29 6.24 4.45 3.68 2.39
30 6.27 4.45 3.63 2.54
2.69
2.54
2.65
2.44
2.54
2.54
2.64
2.54
Time (hr) Ti-ZrO
2
OTS-Ti-ZrO
2
F-Ti-ZrO
2
OTS-F-Ti-ZrO
2
CTMS-F-Ti-ZrO
2
0
1 1.55 1.04 0
2 2.23 1.07 0
3 2.64 1.07 0
0
0
0
0
0
0
4 2.6 1.07 0
5 2.68 1.04 0
6 2.64 1.07 0
7 2.64 1.07 0
0
0
0
0
0
0
0
0
8 2.68 1.07 0
24 2.64 1.1 0
25 2.64 1.1 0
26 2.68 1.13 0
27 2.64 1.13 0
28 2.64 1.1 0
29 2.64 1.1
30 2.64 1.1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Ti-ZrO
2
Catalysts

APPENDIX C
CHROMATOGRAM OF REACTION MIXTURE (Ti-ZrO
2
SERIES)
101

APPENDIX C
CHROMATOGRAM OF REACTION MIXTURE (Ti-NaY SERIES)
102