. win.
.5»? .
u tr! 2' l

49$

fir. :
$.rmw‘n.
a" . .3

. J: .1 x
.5: i ..
$3 I III
:. mama.
A: ..

. “raw :2.

:. .. .f
6.4.6.233 .
(«thdulil

 

 

I
‘
t
k}
_ .
t w
I.

u.‘ ’3

 

LIBRARY
Michigan State
Universnty

 

 

 

This is to certify that the
dissertation entitled

SYNTHESIS, CHARACTERIZATION, AND CATALYTIC
APPLICATION OF STRONGLY ACIDIC MESOPOROUS
ALUMINOSILICATE MATERIALS

presented by

HUI WANG

has been accepted towards fulfillment
of the requirements for the

degree In CHEMISTRY

 

 

mk/ WWW

Majo rol‘éssor’ 5 Signature

agédgi £007

Date

MSU is an affinnative-action, equal-opportunity employer

ua-c-n---.----.---u-o-o--I-u-u-n-I-u-----u-u-I-¢-o-u--n-0-.---u-n-u-s-n-----:---u-o-u-I-¢----n---u--o-

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDUE

DAIEDUE

DAJEDUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/CIRC/DateDue.indd-p.1

 

SYNTHESIS, CHARACTERIZATION, AND CATALYTIC APPLICATION OF
STRONGLY ACIDIC MESOPOROUS ALUMINOSILICATE MATERIALS

By

Hui Wang

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

2007

ABSTRACT

SYNTHESIS, CHARACTERIZATION, AND CATALYTIC APPLICATION OF
STRONGLY ACIDIC MESOPOROUS ALUMINOSILICATE MATERIALS

By
Hui Wang

Aluminosilicate materials are utilized in many industrial applications,
including adsorbents, ion-exchange agents, catalysts, and catalyst supports.‘
In catalytic applications, those aluminosilicates with large surface area and
well-defined pore structure are especially useful because such materials can
provide excellent catalytic performance and shape-selectivity that are not
available from conventional catalysts.

Mesostructured aluminosilicates templated ' from supramolecular
assembly of surfactant molecules have the above two desired properties?
However, the weak acidity and low hydrothermal stability cast a shadow over
their prospect as acidic catalysts. On the other hand, small pore sizes greatly
limited the application of zeolites, crystalline and strongly acidic
aluminosilicates.

The present work describes the synthesis of mesoporous
aluminosilicates with strong acidity. First, a family of mesostmctured
aluminosilicates was prepared from the zeolite precursor hydrolyzed in the
presence of surfactants, which interact and stabilize the zeolitic subunits. The
resultant mesostructures, MSU-Z, showed an unprecedented 74% conversion

in catalytic cumene cracking reaction compared to 11% conversion for the

conventional MOM-41 material.3 Secondary, zeolite nano—crystals as small as
30 nm with high crystallinity and yield were prepared in the presence of
organosilanes, which served as a surface modifier to control the crystal size.
The resultant nano-sized zeolites showed uniform particle size, and
constructed uniform intercrystal mesopores. Finally, organosilane modifier
polymers were used as mesoporogen in the synthesis of zeolites.‘ Zeolite
products with small and uniform intracrystal mesopores tunable in the range of
2 to 10 nm can be prepared. Compared to conventional zeolites, 6-fold and
16-fold increase in external surface and mesoporosity can be achieved. Such
materials exhibited superior catalytic performance in cumene cracking

reaction.

References

1 Corma, A. Chem. Fiev. 1997, 97(6), 2373.

2 Beck, J. S.; Vartuli, J. C.; Roth, W. J.; Leonowicz, M. E.; Kresge, C. T.;
Schmitt, K. D.; Chu, C. T. W.; Olson, D. H.; Sheppard, E. W.; et al. J. Am.
Chem. Soc. 1992, 114(27), 10834.

3 Wang, H.; Liu, Y.; Pinnavaia, T. J. J. Phys. Chem. 32006, 110(10), 4524.

4 Wang, H.; Pinnavaia, T. J. Angew. Chem. Int. Ed. 2006, 45, 7603.

ACKNOWLEDGMENTS

There are many people to thank for their contributions toward the
completion of my doctoral degree during my time at MSU. First, I want to thank
my advisor, Dr. Thomas J. Pinnavaia, for his support and guidance over the
last five years. Thanks for giving me the freedom to explore different paths and
I hope they are worthwhile. I am grateful for all the inspirational discussions we
have had over the years.

I want to express my thanks to Center for Advance Microscopy, especially
Dr. Xudong Fan, Alicia and Ewa. Thank for teaching me the basics of SEM and
TEM. They are very helpful to my research.

I am thankful to former group member Dr. Yu Liu for introducing me into the
mesostructures. You’ve been a great mentor and friend to me. Thanks to Drs.
Zhaorong Zhang and Seong-Su Kim for the help. Thanks to many other former
and current group members for their friendship. You made the lab an exciting
place to be.

Thanks to my family and friends for their support. To my parents,
parents-in-law, and sisters, thanks for the encouragements and the sacrifices
they have made.

Most importantly, I would like to express my deepest gratitude to my wife,
Xin, for all of the love and support she has given me over the years. I would not

have been able to accomplish any of these without you.

TABLE OF CONTENTS

LIST OF TABLES .......................................................................
LIST OF FIGURES .....................................................................
ABBREVIATIONS ........................................................................

1. Chapter 1: Introduction ........................................................
1 .1 Background ...............................................................

1.2 Recent advances .......................................................

1.2.1 Mesostructured aluminosilicates with zeolitic walls .......

1.2.2 Mesoporous forms of zeolites ..................................

1.3 Research objectives ...................................................

1 .4 References ...............................................................

2. Chapter 2: Highly Acidic Mesostructured Aluminosilicates
Assembled from Surfactant-medlated Zeolite Hydrolysis Products.
2.1 Abstract ....................................................................

2.2 Introduction ...............................................................

2.3 Experimental section ...................................................
2.3.1 Materials .............................................................

2.3.2 Synthesis of mesostructured aluminosilicates from

zeolite precursors .........................................................

2.3.3 Characterization ...................................................

2.3.4 Catalytic cumene cracking .......................................

2.4 Results and discussion ................................................

2.5 Conclusion and perspective ..........................................

2.6 References ...............................................................

3. Chapter 3: Organosllane-asslsted Synthesis of Zeolite
Nana-crystals with High Crystallinity and Yield .............................
3.1 Abstract ....................................................................

3.2 Introduction ...............................................................

3.3 Experimental section ...................................................
3.3.1 Materials .............................................................

3.3.2 Synthesis of nano-sized ZSM-5 zeolite ......................

3.3.2.1 Direct synthesis pathway ..............................

3.3.2.2 Post-nucleation pathway ..............................

3.3.3 Physical property characterizations ...........................

3.3.4 Catalytic cumene cracking .......................................

3.4 Results and discussion ................................................
3.4.1 Direct synthesis pathway .........................................

3.4.2 Post-nucleation pathway .........................................

vii
viii

XV

NNV-e-l-

—&

73
73
75
79
79
79
79

81
81

3.5 Conclusion and perspective .......................................... 94
3.6 References ............................................................... 113

4. Chapter 4: Synthesis, Characterization, and Catalytic

Application of Zeolites with Uniform Intracrystal Mesopores .......... 117
4.1 Abstract .................................................................... 117
4.2 Introduction ............................................................... 119
4.3 Experimental section ................................................... 123

4.3.1 Materials .............................................................. 123
4.3.2 Synthesis of mesoporogen ...................................... 123
4.3.3 Preparation of mesoporous zeolites ........................... 124
4.3.3.1 Mesoporous ZSM-5 ..................................... 124

4.3.3.2 Mesoporous faujasite .................................. 124

4.3.3.3 Mesoporous zeolite A .................................. 125

4.3.4 Zeolite characterization ........................................... 125
4.3.5 Catalytic cumene cracking ....................................... 126

4.4 Results and discussion ................................................ 127
4.4.1 Formation of mesopores ......................................... 127
4.4.2 Polymer effects on mesopore size ............................. 129
4.4.3 Effects of reaction temperature ................................. 132
4.4.4 Other types of zeolite ............................................. 132
4.4.5 Hydrothermal stability ............................................. 133
4.4.6 Catalytic activity .................................................... 134

4.5 Conclusion and perspective........................_ .................. 138
4.6 References ............................................................... 159

vi

Table 2.1

Table 2.2

Table 3.1

Table 3.2

Table 4.1

LIST OF TABLES

Summary of reaction stoichiometry for the synthesis of
MSU-Z materials from crystalline zeolites as the
aluminosilicate source .................................................

Physical properties of different mesostructured
aluminosilicate materials ..............................................

Yields of ZSM-5 (Si/AI = 50) obtained in the presence of
various organosilane surface modifiers ............................

Physical properties of ZSM-5 samples (Si/AI = 50) .............

Physical properties of MSU-MFI materials ........................

vii

56

96

97

140

Figure 1.1

Figure 1.2

Figure 1.3

Figure 1.4

Figure 1.5

Figure 1.6

Figure 1.7

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

LIST OF FIGURES

Possible mechanistic pathways for the formation of
MGM-41: (1) liquid crystal phase initiated and (2)
silicate anion initiated ...........................................

Illustration of Brosted acid site in aluminosilicate
materials ............................................................

Illustration for the preparation of nano—sized ZSM-5
zeolite using colloid-imprinted carbon (CIC) templates.

Effect of crystal size on the accessible region of a
zeolite catalyst ....................................................

TEM images of steamed faujasite crystals ................

Mesoporous zeolite single crystal templated by carbon
nanotube ...........................................................

Mechanism for meSOpores templated through the
organosilane within zeolite crystals proposed by Ryoo
et al ..................................................................

Reactor set-up for catalytic cumene cracking .............

X-ray diffraction patterns for (A), MGM-41 (Si/Al = 40)
prepared from conventional precursors, and
mesostructured aluminosilicates MSU-Z (Si/AI = 40)
prepared from surfactant-mediated hydrolysis
products of (B) ZSM-5 (trial 2) and (C) USY (trial 6) as
precursors. Each curve is offset by 10,000 cps for
clarity ................................................................

27Al MAS NMR spectra of as—synthesized (solid line)
and calcined (broken line) MSU-Z (Si/Al = 40)
prepared from hydrolyzed ZSM-5 as precursor (trial 2)

FT-IR spectra for (A), conventional MOM-41 (Si/AI =
40); (B), MSU-S (Si/AI = 50) prepared from MFl seeds
precursor; (C), MSU-Z prepared from zeolite fragment
formed by hydrolysis of ZSM-5 (Si/Al = 40) in the
presence of CTAB at a OH-IT ratio of 0.75 (trial 2); (D),
commercial ZSM-5 (Si/Al =40) ................................

viii

15

16

18

20

24

57

58

59

60

Figure 2.5

Figure 2.6

Figure 2.7

Figure 2.8

Figure 2.9

Figure 2.10

Figure 2.11

Nitrogen adsorption-desorption isotherms for (A),
MOM-41 (Si/AI = 40) synthesized from conventional
precursors; (B) MSU-Z (Si/AI = 40) made from zeolite
fragments generated from USY in the presence of
CTAB with OH'fT = 0.17 (trial 6); (C) MSU-Z (Si/Al = 40)
assembled from CTAB surfactant-mediated ZSM-5
hydrolysis products (zeolite fragments) formed with
OH‘IT = 0.75 (trial 2); and (D), mesostructured
aluminosilicate (Si/AI = 40) formed through
over-hydrolysis of ZSM-5 with OH'lT = 1.50 (trial 4) .....

BJH pore size distributions obtained from adsorption
isotherms for (A), MOM-41 (Si/AI = 40) synthesized
from conventional precursors; (B) MSU-Z (Si/Al = 40)
made from zeolite fragments generated from USY with
OH'IT = 0.17 (trial 6); (C) MSU-Z (Si/AI = 40)
assembled from surfactant-mediated ZSM-5
hydrolysis products (zeolite fragments) formed with
OH‘IT = 0.75 (trial 2); and (D), mesostructured
aluminosilicate (Si/Al = 40) formed through
over-hydrolysis of ZSM-5 with OH'IT = 1.50 (trial 4) .....

X-ray diffraction patterns of MSU-Z materials (Si/AI =
40) prepared from hydrolysis products of 2 g ZSM-5 in
the presence of (A) 2.45 9 (trial 2) and (B) 3.45 9 (trial
3) of CTAB. The diffraction peaks indexed by arrows
are attributed to the ZSM-5 residue ..........................

TEM image of MSU-Z (Si/AI = 40) prepared from
hydrolyzed ZSM-5 product (trial 2) ..........................

T-plots for (A), MSU-Z (Si/AI = 40) prepared from the
hydrolysis product of ZSM-5 as precursor (trial 2); (B)
a physical mixture of 5% ZSM-5 (Si/Al = 40) within
conventional MOM-41 (Si/AI = 40) ...........................

X-ray diffraction pattern of MSU-Z (Si/AI = 40)
prepared from incomplete hydrolyzed ZSM-5
precursor (trial 1) .................................................

N2 adsorption-desorption isotherms and BJH pore size

distributions for (A) won'nhole MSU-Z (Si/Al = 40)
prepared from the hydrolysis product of USY zeolite in

ix

61

62

65

66

Figure 2.12

Figure 3.1

Figure 3.2

Figure 3.3

Figure 3.4

Figure 3.5

Figure 3.6

the presence of 'ITeA (trial 7), and (8) large pore
hexagonal MSU-Z (Si/AI = 40) prepared from the
hydrolysis product of ZSM-5 in the presence of P123
surfactant (trial 5) .................................................

X-ray diffraction patterns for the hydrolysis products of
ZSM-5 zeolite (Si/Al = 40) in the presence of (A) 2.45
g CTAB (trial 2); (B) 2.0 g ‘ITeA (trial 9) (C) 2.4 9 P123
(trial 5) ...............................................................

Illustration for the quenching of zeolite crystal growth
using an organosilane as a capping agent ................

SEM images for ZSM-5 samples (Si/AI = 50) prepared
through the direct synthesis pathway in the absence
and presence of methyltriethoxysilane (MT ES): (A),
0% MTES; (B), 5% MT ES and (C), 8% MTES at
100°C for 48 hours ...............................................

TEM images for ZSM-5 samples (Si/AI = 50) prepared
through the direct synthesis pathway in the presence
of (A), 5% MTES; (B), 8% MT ES at 100°C for 48 hours

Nitrogen adsorption-desorption isotherms of ZSM-5
samples (Si/AI = 50) prepared through the direct
synthesis pathway in the presence of A, 0% MTES; B,
5% MTES and C, 8% MT ES at 100°C for 48 hours.
Curves B and C are offset by 50 and 150 volume units
for clarity. Inset: BJH desorption pore size distributions

X-ray diffraction patterns for ZSM-5 (Si/AI = 50)
samples prepared by (A) a conventional method in
absence of MTES, and by direct synthesis in the
presence of (B) 2 mol% MT ES, (C) 5 mol% MT ES,
and (D) 8 mol% MT ES at 100 °C for 48 hours. Curves
are offset by 4,000 counts each for clarity. Each
sample was calcined at 600 °C for 4 hours ................

N2 adsorption-desorption isotherms for (A)
conventional ZSM-5 (Si/AI = 50) in absence of
organosilane and (B) n-ZSM-5-DP (Si/Al = 50)
prepared by the direct synthesis pathway in the
presence of 6 mol% 3-aminopropyltrimethoxysilane
(APTMS) at 100 °C for 48 hours. Curve Bis offset by

67

68

98

99

100

101

102

Figure 3.7

Figure 3.8

Figure 3.9

Figure 3.10

Figure 3.11

Figure 3.12

150 volume units for clarity ....................................

Optical images for the products prepared by direct
synthesis in the presence of (A) 6 mol% APTMS, (B) 8
mol% APTMS, and (C) 10 mol% APTMS at 100 °C for
48 hours ............................................................

TEM image for the amorphous product prepared by
direct synthesis in the presence of 8 mol% APTMS at
100 °C for 48 hours ..............................................

Illustration for the formation of nano-crystals by a
direct synthesis pathway in the presence of (RO)3Si-L
surface modifier. The formation of single nanocrystals
(path A) or aggregates of intergrown nanoparticles
(path B) depends on the amount of (RO)3Si-L present
in the reaction mixture ..........................................

X-ray powder diffraction patterns for calcined ZSM-5
samples (Si/AI = 50) synthesized by the
post-nucleation pathway and the addition of (A), 4%
MT ES after 3 hours nucleation step; (B), 7% MT ES
after 3 hours nucleation step; (C), 10% MT ES after 12
hours nucleation step at 100 °C. Each curve is offset
by 4000 units for clarity .........................................

N2 adsorption/desorption isotherms for calcined
ZSM-5 samples (Si/Al = 50) synthesized by (A),
conventional method and by the post-nucleation
pathway and the addition of 10% MT ES after (B) 12
hours nucleation step; (C), 6 hours nucleation step;
(D), 3 hours nucleation step; (E), 1 hours nucleation
step at 100 °C; (F), 24 hours nucleation step at room
temperature. Each isotherm is offset by 200 volume
units for clarity .....................................................

N2 adsorption/desorption Isotherms for calcined
ZSM-5 samples (Si/AI = 50) synthesized by (A),
conventional method; and by the post-nucleation
pathway and by the addition of (B), 4% MT ES; (C), 7%
MTES; (D), 10% MT ES; (E), 15% MT ES after a 3-hour
nucleation step at 100 °C. Each isotherm is offset by
200 volume units for clarity ....................................

xi

103

104

105

106

107

108

Figure 3.13

Figure 3.14

Figure 3.15

Figure 3.16

Figure 3.17

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Figure 4.5

BJH (desorption) pore size distribution of n-ZSM-5—NP
samples (Si/AI = 50) synthesized by the
post-nucleation pathway and the addition of (A), 10%
MT ES after 3 hours nucleation step; (8), 4% MTES
after 3 hours nucleation step; (C), 10% MT ES after 12
hours nucleation step at 100 °C ..............................

TEM images of n-ZSM-5-NP (Si/AI = 50) synthesized
by the post-nucleation pathway by the addition of (A),
4% MT ES after 3 hours nucleation step; (8), 15%
MT ES after 3 hours nucleation step. Insets are the
corresponding high-resolution images ......................

Nanoparticle packing schemes. Both tetrahedral and
octahedral interparticle voids (pores) are expected for
extensively compacted spheres ..............................

Illustration for the formation of nano-crystals in a
post-nucleation pathway .......................................

Catalytic cumene cracking conversion vs. time for (A),
a commercial ZSM-5 from Zeolyst (Si/AI = 40); (B),
n-ZSM-5-DP (Si/AI = 50) prepared from 5 mol% MTES
by a direct synthesis pathway ..................................

Reaction for mesoporogen preparation .....................

X-ray diffraction patterns of (A), calcined conventional
ZSM-5 (Si/AI = 50); (B), calcined MSU-MFI (Si/AI = 50)
prepared from silylated PEI mesoporogen with an
initial NH/Si ratio of 5: 1. Curve B is offset by 5000
counts ...............................................................

”s: MASNMR spectrum of as-synthesized MSU-MFI
(Si/AI = 50) prepared from 10% silane modified PEI

porogen .............................................................

N2 adsorption/desorption isotherms of ZSM-5 (Si/Al =
50) synthesized with A, conventional method; B, 5
mol% organosilane with MP-25000(1); C, 10 mol%
organosilane with MP-25000(1); D, 10% organosilane
with MP-25000(5). Offset by 50 ...............................

TEM images of the MSU-MFI zeolite (Si/AI = 50)

xii

109

110

91

111

112

124

141

142

143

Figure 4.6

Figure 4.7

Figure 4.8

Figure 4.9

Figure 4.10

Figure 4.11

Figure 4.12

Figure 4.13

Figure 4.14

prepared in the presence of silylated polyethleneimine
with an initial molecular weight of 25,000: A) whole
particle specimen; B) thin-sectioned sample. The
light contrast areas are intracrystal mesopores ...........

Scheme for the synthesis of a zeolite with intracrystal
mesopores using a silylated polymer as the

mesoporogen ......................................................

SEM images of ZSM-5 samples (Si/Al = 50) prepared
from A, conventional method; B, 10% MP-25000(1);
and C, 5% MP-25000(1) .......................................

BJH adsorption pore size distributions for MSU-MFI
samples (Si/AI = 50) prepared from A, MP-600(5); B,
MP-1800(5); C, MP-10000(5); D, MP-25000(5) ..........

Structure of Jeffamine D surfactant ..........................

N2 adsorption/desorption isotherms of MSU-MFI
samples (Si/AI = 50) synthesized from A, MP-DZOOO;
B, MP-D4000. Inset: BJH pore size distributions from
adsorption isotherms ............................................

TEM image of thin-sectioned MSU-MFI sample (Si/AI
= 50) prepared from silylated D4000 as mesoporogen.
Inset: high-resolution image of selected area .............

X-ray diffraction patterns for (A) conventional ZSM-5
(Si/Al = 50); (B) MSU-MFI (Si/Al = 50) prepared from
silylated D2000 as mesoporogen; (C) MSU-MFI (Si/Al
= 50) prepared from silylated D4000 as mesoporogen.

N2 adsorption-desorption isotherms for MSU-MFI
samples (Si/Al = 50) prepared at 125 °C from (A)
silylated 02000 as mesoporogen and (B) from
silylated D4000 as mesoporogen. The curve B is offset
by 200 for clarity. Inset: BJH pore size distributions
obtained from adsorption isotherms ..........................

N2 adsorption/desorption isotherms of zeolite Y (Si/AI

= 2.2) synthesized with A, conventional method; 8,
with MP-25000(5) ................................................

xiii

144

145

146

147

130

148

149

150

151

152

Figure 4.15

Figure 4.16

Figure 4.17

Figure 4.18

Figure 4.19

Figure 4.20

N2 adsorption/desorption isotherm of MSU—A (Si/AI = 5)
synthesized with MP-25000(5). Inset: X-ray diffraction
pattern ..............................................................

N2 adsorption-desorption isotherms for MSU-MFI
sample (Si/AI = 50) prepared from silylated D4000. (A)
pristine MSU-MF I; (B) steamed with 20% water vapor
at 800 °C for 2 hours; (C) treated with boiling water
for 7 days. The curves 8 and C are offset by 300 and
600 for clarity. Inset: BJH pore size distributions
obtained from adsorption isotherms .........................

Catalytic cumene cracking conversion vs. time for A,
commercial USY from Zeolyst (Si/Al = 40); B,
commercial ZSM-5 from Zeolyst (Si/Al = 40); C,
MSU-MFI (Si/Al = 50) prepared from MP-25000(5); D,
MSU-MFI (Si/Al = 50) prepared from MP-D4000 .........

Illustration of catalytic cumene cracking on bulky and
mesoporous zeolite catalysts .................................

Product propylene/ethylene ratios of
methanol-to-olefins reaction catalyzed by A,
commercial bulky ZSM-5 (Si/AI = 40); B, MSU-MFI
(Si/AI = 40) prepared from MP-25000(5); and C,
MSU-MFI (Si/AI = 40) prepared from D4000. Reaction
conditions: WHSV = 0.50 h", N2 ..............................

HDPE degradation catalyzed by A, no catalyst; B,
commercial ZSM-5 (Si/AI = 280); C, commercial
ZSM-5 (Si/Al = 80); D, commercial ZSM-5 (Si/Al = 50);
E, commercial ZSM-5 (Si/AI = 23); F, MSU-MFI (Si/Al
= 40) prepared from MP-25000(5); and G, MSU-MFI
(Si/AI = 40) prepared from D4000. HOPE/Catalyst = 4:
1 ......................................................................

xiv

153

154

155

156

157

158

APTES

BEA

BET

BJH

CIC

CTAB

EtOH

FAU

FCC

FTIR

GOPTMS

HDPE

HTMS

LTA

M418

MCF

MGM-41

MFI

MSU-A

MSU-J

MSU-MFI

LIST OF ABBREVIATIONS

3-Aminopropyltriethoxysilane

Framework code for zeolite Beta
Brunauer-Emmett-Teller

Barrett-Joyner-Halenda

Colloid-imprinted carbon

Cetyltrimethylammonium Bromide, C13H33N(CH3)3Br
Ethanol

Framework code for Faujasite

Fluid Catalytic Cracking

Fourier Transform Infrared spectroscopy
Gchidoxypropyltrimethoxysilane

High-density polyethylene
Hexadecyltrimethoxysilane

Framework code for Linde type A zeolite

Mobil family of Mesostructured silicas
Mesostructured Cellular Foams

Mobil Composition of Matter 41, hexagonal mesophase
Framework code for ZSM-5 zeolite

Mesoporous zeolite A prepared from silylated polymer as
mesoporogen

Large-pore wormhole mesostructured silica assembled
using Jeffamine D series surfactants

Mesoporous ZSM-5 zeolite prepared from silylated polymer

XV

MSU-S

MSU-Y

MSU-Z

MTES

NMR

OTMS

P123

PEI

PEO

Ultra stable and strongly acidic mesostructured
aluminosilicates assembled from protozeolitic seeds
precursor

Mesoporous zeolite Y prepared from silylated polymer

Mesostructured aluminosilicates prepared from zeolitic
fragment

Methyltriethoxysilane

Nuclear Magnetic Resonance
Octyltrimethoxysilane

Pluronic 123, triblock polymer, (EO)2o(PO)7o(EO)2o
Polyethylenimine

Polyethylene oxide

PHAPTMS Phenylaminopropyltrimethoxysilane

PP

PPO

P/Po

ppm

PS

PTMS

SAPO

SBA-15

Polypropylene

Polypropylene oxide

Relative Pressure, P= pressure, Po = saturation pressure
Part per million

Polystyrene

Phenyltrimethoxysilane

lncompletely condensed silica sites, SiO2(OSi)2
lncompletely condensed silica sites, SiO(OSi)3
Completely condensed silica sites, Si(OSi)4

Silicoaluminophosphate, a family of crystalline microporous
materials with zeolite-like structures

Large pore hexagonal mesostructured silicas assembled

xvi

Seer

SEM
Si-L

STP

TEM
TEOS
TMAOH
TPAOH
'ITeA
USY

XRD

under low pH conditions with P123 surfactant

Specific surface area in m2/g obtained from the linear part of
the adsorption isotherm using BET equation

Scanning electron microscopy

Organosilane

Standard temperature and pressure
Tetrahedral sites, including Si and AI
lncompletely condensed silica sites, SiO(OSi)2R
Completely condensed silica sites, Si(OSi)3R
Transmission electron microscopy
Tetraethylorthosilicate
Tetramethylammonium hydroxide
Tetrapropylammonium hydroxide

Tallow tetramine

Ultrastable Y zeolite

X-ray diffraction

xvii

Chapter 1

Introduction

1.1 Background

Materials with well-defined porous frameworks like zeolites,“ 2 also noted
as molecular sieve, attract much attention due to their great potential in the
field of catalysis,3' 4 separation,5 and other areas.6 The typical pore diameter of
zeolite materials is less than 1.2 nm.7 Because of the limitation of the pore
diameter, only small molecules can be processed effectively. In 1992, Mobil Oil
Corporation successfully synthesized a family of silica-based mesostructured
materials, denoted M41S,8' 9 whose pore diameter ranges from 2.0 to 10.0 nm,
which exceeds the limitation of zeolite materials and allows bulky molecules to
diffuse into the pore channel. For instance, MGM-41 materials have been used
to absorb protein and other bulky biomolecules for use as a smart delivery
system.” " Compared to other materials, mesostructures showed higher
absorbing capability with desired releasing properties.”'14 The uniform
mesopore size and high surface area of mesostructures are also highly
desirable for many catalytic processes like FCC (fluid catalytic cracking), which
is one of the most important processes in the chemical industry.

MOM-41 silica was synthesized through a micelle-induced
supramolecular assembly pathway. Figure 1.1 illustrates the mechanism

proposed by Mobil scientists. Surfactant molecules, which are amphiphilic,

form micelles first. Then, an inorganic matrix precursor introduced which

interacts with the micelles to form mesostructures. Depending on the nature of

Hexagonal Array

Surfactant fl ""3 ...
Micelle MIcellar ROd ”N if It) [,4 scalcination {/1772
Silicate Ugh )I/ f/ ‘7

.f‘ --> —-—> 417 ave/f
H.954 2...)? g ’nl/F

 

MGM-41
Silicate

Figure 1.1 Possible mechanistic pathways for the formation of MOM-41: (1)

liquid crystal phase initiated and (2) silicate anion initiated.8

the interaction, the mechanism can be categorized into four most common
pathways. The Sr pathway is where the cationic surfactant (8") interacts with
an anionic inorganic reagent I' through electrostatic forces.8 Conversely, there
is an 81“ pathway, where anionic surfactants like carboxylic acids are used.”
‘6 Also, nonionic surfactants can be used either through an S"H“-X‘I+

""9 or a hydrogen-bonding S°l° pathway under neutral conditions.”22

pathway
Recently, mechanisms involving other interactions including covalent bonding
were developed.23 It should be pointed out that mesostructured silicas
synthesized through the hydrogen-bonding pathway usually have higher
cross-linking or Q“/Q3 ratio20 because no charge balance is required for
surface silicon sites. However, the ordering is generally lower for a

mesostructure made through the S°l° pathway due to the weaker H-bonding

interactions compared to electrostatic Coulomb forces. Moreover, different

mesostructures including hexagonal,9 cubic,8 wormhole,21 foam17 and IameIIar
2° structures can be assembled, depending on the reaction conditions and the
choice of surfactant molecules.24

Mesostructured aluminosilicates have been shown to have good acid
catalytic activity due to the large pore diameter and high surface area. For
instance, compared to microporous clay catalysts, mesostructured
aluminosilicate catalysts showed higher activity and selectivity for the
alkylation of diphenylamine with a-methylstyrene.25 However, mesostructures
have limited applications as refining catalysts, despite of the improvement in
mass transportation rate. The main reason behind this is that the wall of
mesostructured aluminosilicate is amorphous. Compared to zeolites,
amorphous mesostructured aluminosilicates have much weaker acidity, which
dramatically affect their performance in catalytic reactions such as FCC where
strong acidity is required. Moreover, the amorphous walls are very unstable
especially under hydrothermal conditions. The mesostructure easily collapses
and forms a low surface area degradation product. The poor hydrothermal
stability further undermines their catalytic applications because such catalysts
cannot survive the hydrothermal conditions generated during catalyst
regeneration.

The poor hydrothermal stability of mesostructured aluminosilicates also is
partially due to the thin wall of these materials. The typical wall thickness for

MGM-41 is 1~2 nm. Mokaya et al.26' 27 used post-synthesis treatment to

thicken the wall of MOM-41 and subsequently improved the hydrothermal
stability of the materials. The optimization of synthesis conditions can also
modify the wall thickness. Wang and his co-workers28 reported a two-step
synthesis route to separate the hydrolysis and condensation reactions of silica,
which in turn provided a thick-wall mesostructure with better hydrothermal
stability. In 1998, Stucky and his coworkers synthesized an MOM-41 analog,
denoted as SBA-15, using a polyethylene oxide (PEO) - polypropylene oxide
(PPO) triblock copolymer as the surfactant.17 The SBA-15 structure has a
larger pore diameter, which reaches to 10 nm, and has thicker walls compared
to MGM-41 materials. It also was reported that when PPO-PEO block
copolymers were used as mesoporogen, the wall thickness of the resultant
mesostructure was controlled by the length of the hydrophilic PEO chain,
which penetrated into the silica matrix during synthesis.29 For instance, a
FDU-1 mesostructured silica with walls thicker than 10 nm was synthesized
using (EO)39-(BO)47-(EO)39 surfactant as porogen, where 80 is a butylene
oxide unit.3o Such materials retained majority of the mesopore volume and
surface area after being boiled in water. The hydrothermal stability was further
improved when the pore wall was modified with an organosilane to increase
the hydrophobicity of the wall.

Cross-linking is another important factor that affects the structural stability
of mesostructured materials. A high degree of cross-linking is desirable in the

synthesis strategy. 29Si NMR results showed that a significant amount of Si

atoms in the MOM-41 framework are Q3 sites, which means one of the four
oxygens linked to Si is terminated by a proton. From the crosslinking
perspective, a non-electrostatic pathway is the preferred driving force for the
assembly of stable mesostructures. With non-ionic Gemini surfactant of the
type CnH2n+1NH(CH2)2NH2 as a structure-directing agent, Kim20 synthesized
MSU-G material with significantly reduced Q3/Q4 ratio than typical
mesostructures. Moreover, without electrostatic interaction, non-ionic
surfactants can be easily removed from the mesopores, as in the case of
MSU-H and HMS silicas, through a simple solvent extraction.

Another important feature of mesostructured silicates is the intrinsic
acidity.31 Oftentimes, mesostructures were prepared in a pure silica form.
Although the 29Si NMR spectrum revealed that substantial amounts of 03 sites
existed in the framework, the acidity associated with the surface silanol group
is too weak to provide desired acid catalytic activity. Strong acidity is obtained
when some of the silicon sites are isomorphically substituted by tetrahedrally
coordinated aluminum atoms. The incorporation of tetrahedral AI makes the
framework electron negative. Therefore a Bronsted acid site can be generated

when the counter ion is exchanged to proton, as shown in Figure 1.2.

le |

/N\@/ i\
I <3 i

H

Figure 1. 2 Illustration of Brosted acid site in aluminosilicate materials.

In addition to introducing framework heteroatoms by direct synthesis,32
the introduction of heteroatoms also can be achieved through a post-synthesis
method like grafting.26 However, the amount of heteroatoms that can be
incorporated into the wall of a mesostructure is quite limited. Moreover,
mesostructures lose long range ordering after the post-synthesis treatment.
Recently, Xiao et al.33 demonstrated the synthesis of mesostructured
aluminosilicate material with Si/Al close to 1. To achieve that ratio, a precursor
with preformed Si-O-Al link such as aluminosilicate ester
((BuO)2-AI-O-Si-(0Et)3) (di-sec-butoxyaluminoxytriethoxysilane) was used as
single source for both Si and Al. The resulting composition exhibited a
well-defined mesostructure and a high surface area of 491 m2/g.

However, as mentioned before, compared to commercial zeolite catalysts,
the lack of atomic ordering in mesostructured aluminosilicate materials makes
them less attractive as refining catalysts because of poor stability and, most
importantly, weak acidity. The ideal way to solve the problem would be to
synthesize aluminosilicate materials that have the large pore diameters and
high surface areas characteristic of mesostructures, but also have crystalline
walls to impart strong acidity and stability. In general, two different strategies
have been taken. On one hand, one can start with a mesostructure, and try
crystallizing the amorphous wall into a zeolite; or one can start with a zeolite,

and build mesopores within the crystal phase.

1.2 Recent advances

1.2.1 Mesostructured aluminosilicates with zeolitic walls

Shortly after the weak acidity and poor stability of mesostructured
aluminosilicates was realized, many attempts have been conducted to
crystallize the amorphous wall into a zeolite. Van Bekkum3’4 reported the
crystallization of the amorphous walls of MGM-41 materials. The
mesostructure was impregnated with tetrapropylammonium hydroxide, which
is the structure-directing agent for an MFI zeolite. The crystallization was
conducted in the solid state to minimize the mass transportation that causes
the collapse of mesostructure. Partial crystallization was achieved. Products
with improved acidity were obtained, as indicated by catalytic cumene cracking
results. However, the wall-thickness of the MGM-41 mesostructure was 1 ~ 2
nm, which is smaller than the 2.5 nm unit cell dimension of an MFI framework.
The complete crystallization of the walls caused the collapse of the
mesostructure, and formation of nano-sized zeolite crystals.35 Kaliaguine and
co-worker35 applied a similar approach to MCF mesostructured aluminosilicate,
which has a much thicker wall and a larger pore size. A product with a ZSM-5
crystallinity of 42% was formed after the crystallization. N2 isotherm for the
product confirmed the presence of mesopores and the microporosity derived
from zeolite phase. Dark-field TEM image revealed the presence of zeolite
nanocrystals around 5 nm in size. Compared to the parent mesostructure, the

crystalline product, denoted UL-ZSM-5, showed stronger acidity. However,

TEM image also revealed the loss of mesostructure in the product.

In order to stabilize the mesostructure during recrystallization, Hu et al.37
filled the mesopore of SBA-15 with carbon formed in-situ. The resultant
material showed both low-angle and wide-angle X-ray diffraction peaks,
confirming the preservation of long-range ordering. Such an ordered
mesoporous zeolite showed much higher activity in catalytic alkylation of
bulky 2-methylnaphthalene compared to conventional ZSM-5 catalyst.

Another attempted approach to achieving crystalline walls for a

mesostructured aluminosilicate involved a one-step synthesis. Mokaya38
reported using mixed templates for both simultaneously formimg a
microporous zeolite framework and a mesostructure. However, when such a
reaction mixture was heated at an elevated temperature, It failed to produce
the targeted product. In a typical reaction, cetyltrimethylammonium bromide
(CTAB), a cationic surfactant'to template mesopores, was added to a reaction
mixture for ZSM-5 synthesis. Due to the low structural stability of the surfactant
micelles at high temperature, the reaction was conducted at a temperature
lower than the one normally used for conventional ZSM-5 crystallization. The
evolution of the product with time of reaction showed that an MOM-48 type
mesostructure was formed first, without any sign of a zeolite phase. Such a
mesostructure was then transformed into a zeolite as the reaction progressed.
However, during the crystallization both the mesostructure and the mesopore

volume were lost.

In general, the attempts to crystallize the amorphous walls of a
mesostructure destroy the long-range ordering of the parent mesostructures,
which results in the significant loss of pore volume and surface area. A major
breakthrough was achieved in 2000 by the Pinnavaia group,39 who used
preformed protozeolitic nanoclusters as the precursor to assemble hexagonal
mesostructures. In the first report, a synthesis gel for faujasite was aged at
100°C. Before the zeolite product was crystallized, the reaction was quenched
by lowering the temperature. The resultant solution was then added to a
surfactant solution as the silica-alumina precursor to a mesostructure. Later,
seeds for other types of zeolite structures were also demonstrated to be
suitable precursors. Through the use of protozeolitic seeds, zeolitic
nanoclusters can be incorporated into mesostructured walls. In the case of MFI
and BEA seeds,” the presence of zeolitic nanoclusters was verified by IR
spectroscopy. The final mesostructured aluminosilicates, denoted MSU-S,
showed a distinguishable absorbance band at 550 cm". Such an absorbance
was absent in the conventional mesostructured aluminosilicates such as
MGM-41. The IR band can be assigned to the asymmetric stretching of Si-O-Si
five-member ring, which Is a subunit in the MFI and BEA zeolite frameworks.
The incorporation of other zeolite subunits was also verified by 27Al NMR that
final MSU-S materials exhibited a similar chemical shift as the corresponding
zeolite. The incorporation of zeolitic subunits greatly improved the acidity of

mesostructured aluminosilicates. Catalytic cumene cracking results showed

that a regular MGM-41 aluminosilicate catalyst had a cumene conversion of
11%. With the same Si/Al composition, a MSU-S aluminosilicates showed a
cumene conversion as high as 43%, a 4-fold increase.41 Moreover, the
introduction of Al sites became much easier. Utilizing Faujasite seeds as
precursor, up to 35% Al can be incorporated into the mesostructures. Unlike
other direct synthesis methods, the resultant MSU-S retained high
mesostructure ordering. NMR spectra revealed that the majority of Al sites are
tetrahedrally coordinated, which is very important to achieving strong acidity.
Using a high AI content MSU-S catalyst, the Pinnavaia group achieved
unprecedented cumene conversion up to 57%.

In addition, the introduction of zeolite seeds in the framework walls
greatly improved the hydrothermal stability of the resulting mesostructured
aluminosilicates. MSU-S materials retained 80% and 90% of the original
surface area and pore volume, respectively, after being steamed at 20% water
vapor at 800°C for 2 hours. The steamed samples showed the presence of
high quality mesostructures as confirmed by both XRD and N2 isotherm results.
Such strong acidity and hydrothermal stability makes this family of
mesostructured aluminosilicates good candidates as refining catalysts.42

Several studies further extended the scope of the zeolite seeds method

for producing improved aluminosilicate mesostructures.“45 Results have
showed that the zeolitic subunits of MSU-S materials survived not only under

basic conditions, but they also survived acidic environments. For example,

mesostructures such as large pore SBA-15 and MCF have been successfully
prepared under strongly acidic condition using MFI and BEA seeds
precursors. Results verified that five member ring subunits have been
successfully incorporated into the MSU-S products. All these materials
exhibited much improved hydrothermal stability and stronger acidity
compared to mesostructures prepared from conventional precursors.

In 2002, Kaliaguine and co-worker developed a method to coat the walls
of large pore mesostructures, such as SBA-1546 and MCF47 with a preformed
seeds solution. The resulting materials were subjected to elevated
temperature In a non-aqueous solvent to facilitate the crosslinking of the
zeolite subunits to the mesostruture walls. The final materials retained the
mesostructure, but the pore size, pore volume and surface area were reduced
due to the zeolite seeds coating. NMR spectra showed that a distinguishable
zeolitic phase co-existed with a regular mesostructured aluminosilicate phase.
The zeolite coated mesoporous materials exhibited strong acidity. However,
this approach is not applicable to small pore mesostructures such as MOM-41.
Moreover, severe loss of mesoporosity and surface area was inevitable.

Recently, other methods have been developed to incorporate zeolite
subunits into mesostructures. In 2002, Goto et al.48 reported the use of
base-hydrolyzed zeolite solutions as precursors for the preparation of
mesostructure/zeolite composite mixtures with improved acidity and

hydrothermal stability. To assemble a mesostructured aluminosilicates as a

ll

pure phase, lnagaki et al.49

only used the solution products of the hydrolyzed
zeolite to form the mesostructure. Because only partial hydrolysis of the zeolite
was achieved, the resulting mesostructure had a composition different from
the source zeolite. Moreover, NaOH can easily destroy the zeolite structure
completely. Thus, the resultant mesostructure showed no evidence for the
presence of zeolite subunits. By adding cationic surfactants into the hydrolysis
reaction, Wang et al.50 were able to partially or completely convert zeolites into
mesostructures. The resultant products exhibited catalytic activities that were
not observed for either conventional mesostructures or zeolites. More
importantly, this approach allowed the incorporation of subunits of certain

zeolite structures that are not easily prepared directly, and provided a good

supplement to seeds approach.

1.2.2 Mesoporous forms of zeolites

Small framework micropores limit the usefulness of zeolites for catalytic
reactions and adsorption processes when the size of the guest molecule is
larger than the pore size of the zeolite. Even when the guest molecule is
smaller than the framework pore size of the zeolite, the diffusion rate of
reactants and products into and out of the channels can be slow, thereby
limiting catalytic activity and selectivity.

Open framework structures with pore sizes a few nanometers, ie., < 10

12

nm, in dimensions are expected to function as size or shape selective catalysts
for the conversions of large molecules. For instance, using non-zeolitic
mesostructured aluminosilicates for the catalytic cracking of polymeric
macromolecules, Aguado et al.51 have shown that uniform mesopores a few
nanometers in diameter provide higher yields of liquid fuels than are obtained
using the same aluminosilicate composition with much larger and less uniform
mesopore distributions. Thus, zeolites with uniform small mesopores or
uniform large micropores can be expected to function as selective catalysts for
the cracking of large petroleum molecules and other catalytic conversions of
large molecules. However, zeolites with these desired pore size properties are
unknown.

There are two types of mesopores for zeolites, namely infer-crystal and
infra-crystal mesopores. Inter-crystal mesopores arise from the aggregation of
zeolite nanoparticles. Typical zeolite crystals are micron size, which lead to
inter-particle voids with a diameter of more than several hundred nanometers.
To prepare zeolite crystals on a nanometer scale, several methods have been

developed. Bein et al.52

prepared zeolite A by manipulating the crystallization
condition. This method required the precise control of kinetic parameters such
as reaction temperature, time, and concentration (stoichiometry). For instance,
LTA nanoparticles of 10 ~ 30 nm can be synthesized at room temperature in a

highly diluted reaction mixture. Similar methods have been applied to other

types of zeolite structures.”55 However, with kinetic control of the zeolite

crystallization process, the final zeolite products were always obtained in very
low yields and with poor crystallinity. Upon the modification of the reaction
process, Bein et al.56 were able to recycle the un—reacted precursors to
achieve higher yields. However, low crystallinity was still inevitable.

With the aid of hard templates, zeolite nanoparticles can be synthesized
under conventional conditions. Jacobsen and co-worker557' 5" first
demonstrated the concept of nano-zeolite synthesis in confined space
provided by carbon black nanoparticles. ZSM-5 nano-crystals with a diameter
of 5 ~ 50 nm can be prepared. After the removal of the carbon template and
the TPA structure-directing agent, ZSM-5 zeolite with infer-crystal mesopores
was obtained. However, due to the irregularity of carbon nanoparticle
templates, the resultant nano-zeolites were not uniform in size and shape at all,
which lead to non-uniform pore size distribution.

A much more uniform nano-sized ZSM-5 zeolite was prepared by Kim et
al.59 who used colloid-imprinted carbon (CIC) as a template (Figure 1.3).
Uniform silica nanoparticles were used to template the mesopores within the
CIC template, which were later used as nano-reactor to crystallize the zeolite
crystals. The nano-sized zeolites prepared from CIC materials reflected the
size of the initial colloidal silica templates and were very uniform in size based
on TEM observation. Based on the use of different colloidal silica precursors,
the particle sizes of the resultant ZSM-5 could be controlled within 8 ~ 90 nm.

Moreover, an unprecedented pore volume as high as 1.7 cclg could be

14

achieved. If the zeolite crystals were packed tetrahedrally, they possess
inter-crystal mesopores of 2 ~ 20 nm. However. pore size distributions
revealed that mesopore diameters far larger than expected were observed.
The results suggested that the zeolite nanoparticles were not uniform in size.
They were somehow inter-grown during crystallization or sintered during

calcination.

ZSM-5 ZSM-5
synthesis
mixture

  

 

Calcinatio/n
Colloid-Imprinted C|C/ZS|Vl-5
Carbons (CIC) ‘..$.. Composne

0.
Nano-sized ZSM-5

Figure 1.3 Illustration for the preparation of nano-sized ZSM-5 zeolite using

colloid-imprinted carbon (CIC) templates.59

More recently, Aguado et al.60 demonstrated the use of organosilanes to
modify the morphology of zeolite crystals during crystallization. The purpose of
organosilanes was to react with surface silanol groups of zeolite nanocrystals
so that the crystal growth can be quenched through the reduction of the
growing crystals. Nano-sized zeolites with very high external surface area and

greatly improved mesoporosity were prepared. In a typical experiment, a

reaction mixture for ZSM-5 zeolite was nucleated for a certain period to obtain
zeolitic nanoclusters and to avoid zeolite formation before the addition of 5%
phenylaminopropyltrimethoxysilane (PHAPTMS). Because of the bulkiness of
the organosilane molecule, they could only react with the outmost surface of
the zeolite crystals. From the TEM images, the final zeolite products consisted
of aggregated nanoparticles that are not uniform in diameter. The catalytic
cracking of polypropylene (PP) studies showed that, compared to conventional
zeolite catalysts, the silane-modified nano-zeolites were more active due to the
enhanced external surface area. A 4-fold increase of PP conversion was

achieved.

     

Diffusion
distance

Accessible

Figure 1.4 Effect of crystal size on the accessible region of a zeolite catalyst.

Non-accessible

 

The advantage of forming mesopores between nanoparticles lies
primarily in speeding-up reaction rates through improved molecular diffusion
and access to the framework pores of the crystals. For example, nanosized

zeolites have been observed to show higher catalytic performance than

conventional monolithic zeolites due to the larger external surface areas and
the more rapid diffusion of reactants and products through crystals that
typically are smaller than a few hundred nanometers in size.“53 Figure 1.4
quantitatively illustrates how the size of zeolite crystal affects accessibility, and
hence catalytic efficiency. Suppose that in a typical catalytic reaction, the
average diffusion distance for reactant molecules into zeolite channel is 20 nm.
Then, for a particle with 1 pm diameter, the accessible active sites account for
1/125000 of total active sites. If the particle size is reduced to 100 nm, 1/125 of
the total active sites become accessible. Therefore the efficiency is improved
by 1000 times. However, the mesopores generated through the aggregation of
nano-particles generally are not uniform even though the fundamental particles
are very uniform. More importantly, due to the irregular packing, the resultant
inter-crystal mesopores are much larger than 10 nm, which is much bigger
than typical molecular size to provide desirable shape-selectivity for the
products. Finally, the processing of nano—particles requires centrifugation and
sonification, which are very time-consuming. Therefore, for that perspective,
zeolites with infra-crystal mesopore are preferable.

Infra-crystal mesopores usually are generated within zeolite crystals
through a chemical leaching process, such as steaming or desilication through
NaOH leaching.64 Steaming processing is extensively used in industry to
modify the properties of zeolite catalysts. For instance, zeolite Y is usually

steamed at >600 °C. Steaming not only removes the Na+ ion trapped inside

the sodalite cages, but also removes significant amounts of framework
aluminum sites and generates mesopores. The resultant Y zeolite, denoted as
ultra-stable Y (USY), shows much higher hydrothermal stability and catalytic
performance.65 During the leaching process, silica and alumina are removed
from the framework and recrystallize or form amorphous phases, and
mesopores larger than 10 nm are generated (Figure 1.5).77 But the resulting
mesopores are not at all uniform. Moreover, the composition of the zeolites is
changed by the treatment. For instance, steaming selectively removes alumina;

while NaOH leaching is more effective for removing silica.

 

Figure 1.5 TEM images of steamed faujasite crystals.77

In another approach to the formation of intracrystal mesopores, attention
had been focused on the incorporation of carbon nanoparticles into zeolite
crystals as they crystallize. This approach is often referred as hard-templating.

The subsequent removal of the occluded carbon particles by calcination

results in intracrystal mesopores that replicate the size and shape of the hard
templates. In particular, carbon black particles66 have been used as a
templating agent to form intracrystal mesopores in zeolites. Because carbon
materials have a hydrophobic surface, which is not compatible with hydrophilic
zeolite phase, specific measures have to be taken to ensure encapsulation of
the carbon template in the zeolite crystals. For example, in order to incorporate
carbon nanoparticle reliably, a mixture of zeolite synthesis solution and carbon
template is evacuated to remove the majority of solvent. The solid residue is
then subjected to elevated temperature to crystallize zeolites under solid state
conditions. In the absence of solvent, the silica and alumina sources are less
mobile, and the final products are more likely to have the carbon nanoparticles
imbedded inside the zeolite crystals. In the presence of solvent, such a
composite structure is collapsed and phase segregation occurs. In the latter
circumstance, zeolite nanoparticles were produced.“ 58 The infra-crystal
mesopores templated through the embedding of carbon nanoparticles also are
not uniform, and they usually are very large on average due to the nature of

the mesopore templates. In a related approach, Schmidt67

used carbon
nanotubes to template infra-crystal mesopores in zeolites. TEM images
(Figure 1.6) showed the incorporation of carbon nanotube inside the ZSM-5

crystal. However, such a process is very unlikely to be commercialized due to

the high cost.

19

 

Figure 1.6 TEM image of mesoporous zeolite single crystal templated by

carbon nanotube.67

Tao and co-workers68 used relatively an inexpensive carbon aerogel,
which was formed from resorcinol and formaldehyde, as a mesopore template.
The carbon aerogel had an average mesopore diameter of 23 nm. The
mesopores were filled with a zeolite synthesis mixture to grow ZSM-5 crystals.
After the removal of the carbon template by calcination, textual mesopores
centered at 11 nm were formed with a width at half height of 3 nm. This is by
far the smallest and narrowest textural mesoporosity yet reported for a
crystalline zeolite through the hard-templating approach.

In 1999, Ryoo et al.69 first reported the synthesis of a carbon replica of a
mesostructured silica. Sucrose as well as other organics such as furfuryl
I7°

alcoho and aromatics71 were used as carbon sources. In the presence of

acid catalysts, these precursors lose water, and are further converted into
carbon at high temperature. After the removal of silica, mesostructured carbon
materials are obtained, which were widely used as templates for the synthesis
of mesostructured metal oxides such as CuO.72 In 2004, Mokaya et al.73
reported the use of mesostructured carbon CMK-3 as a template for
mesoporous zeolite synthesis. The CMK—3 carbons were prepared from a
MGM-48 template, and had a pore size of 2 ~ 4 nm. The pores of CMK-3 were
then used as nano-reactors and filled with zeolite precursors. Although the
mesopores of the template are continuous, after the crystallization of the
zeolite only nano-crystals were found and the mesostructure was completely
lost. Moreover, the observed zeolite crystal size was significantly larger than
the mesopore size. The result implied that the crystallization of zeolite
destroyed the mesostructure, hence, replication failed. Part of reason for failed
templating is the thin wall thickness of the initial MGM-48 template. In 2006, Hu
and co-work37 reported the re-crystallization of SBA-15 walls into zeolite walls.
Instead of filling the mesoporous carbon template with zeolite synthesis
mixture, the authors directly utilized the SBA-15 silica/CMK-5 carbon
composite to form the zeolite phase. The SBA-15/CMK-5 composite was
impregnated with TPAOH and crystallized at high temperature in a water
vapor-saturated environment. The product not only showed Bragg X-ray
diffraction peaks characteristic of an MFI structure at wide scattering angle, it

also exhibited the low angle diffraction peaks associated with the parent

21

SBA-15 template. The ordered mesoporous zeolite showed much higher
activity for the catalytic alkylation of bulky 2-methylnaphthalene compared to
conventional ZSM-5 catalyst.

In contrast to hard-templating, soft-templating is another approach that
has been developed recently for the synthesis of mesoporous zeolites.
Surfactant micelles have been shown to successfully template incredibly
uniform mesopores in the range of 2 to 10 nm within an amorphous
aluminosilica matrix as well as other metal oxides. However, the Van der
Waals interaction between surfactant molecules is too weak, especially at high
temperature, to balance the forces induced by zeolite crystallization. Therefore,
once the mesostructured walls crystallize, the micelles collapse and fail to
template uniform mesopores. Realizing that, Xiao et al. applied polymers as
the mesoporogen.74 The polymer has multiple quaternary ammonium sites that
not only resemble zeolite structure-directing agents, but also make the
polymer extremely water soluble. These properties allow the polymer to form a
homogeneous mixture with zeolite synthesis precursors. Although the polymer
retained its integrity throughout the zeolite synthesis, during crystallization
phase-segregation again occurred and the resultant products were
agglomerates of zeolite nanocrystals.

More recently, Ryoo and co-workers75 modified the one-pot synthesis
approach to zeolite synthesis using a specific organosilane as the soft

template (Figure 1.7). The unique organosilane had a quaternary ammonium

22

section similar to tetrapropylammonium (T PA) ions that the authors believed
could also serve as zeolite structure-directing agent. Also, the silane head
group could form Si-O-Si bond with the zeolite phase, so that phase separation
would not occur. Finally, the surfactant molecule had a structure very similar to
the commonly used mesoporogen CTAB. Therefore, it is possible that the
silylated surfactant molecules could form micelles and subsequently direct the
formation of ordered uniform mesopores within zeolite crystals. The evolution
of the reaction product with time showed that a mesostructure was obtained
initially. As the reaction continued, a new mesostructured zeolite phase
emerged upon the consumption of the initial mesostructure. Based on X-ray
diffraction results, the new structure had a much larger unit cell size that can
accommodate several zeolitic unit cells. The resultant zeolite materials
possessed very uniform mesopores comparable to those of MOM-41 materials.
Mesoporous ZSM-5 prepared from this approach exhibited enhanced catalytic
activity compared to both conventional ZSM-5 and classic mesostructured
aluminosilicate catalysts. Other than the ZSM-5 structure, the authors also
demonstrated that this soft-templating method was applicable to
aluminum-rich zeolite types such as faujasite and zeolite A. More recently, the
methodology was further extended to other crystalline microporous materials
such as SAPO.76 These works open the door for the development of novel

materials that have promising applications in catalysis.

23

GELATION
#

 

Zeolite Synthesis
Composition

   

 

 

 

 

 

CRYSTALLIZATION

 

 

Organic layer Mesopore

 

 

 

 

 

fl
CALCINATION
Figure 1.7 Mechanism for mesopores templated through the organosilane

within zeolite crystals proposed by Ryoo et al.75

24

1.3 Research objectives

There are three basic objectives that the current work tries to achieve.
The first objective is to formulate a mechanism that can incorporate more
zeolitic subunits into mesostructured aluminosilicate materials. The presence
of zeolitic subunits in MSU-S materials showed significant acidity
enhancement in comparison to convertional mesostructures. However, FTIR
data revealed that compared to fully crystallized zeolites, the concentration of
zeolitic subunits in MSU-S materials is very limited, which in turn affords
catalytic activity that is not comparable with corresponding zeolite catalysts.
Hence, a precursor with more zeolitic cross-linking is needed. The
nano-clusters obtained from the hydrolytic fragmentation of preformed zeolite
should fulfill that purpose. However, the basic hydrolysis of zeolites in NaOH
results in precursors with complete loss of zeolitic character. In order to
achieve the optimal conditions for zeolite hydrolyzation, both the type of zeolite
structure and the stabilization of the resultant nano-clusters need to be taken
into consideration.

The second objective is to develop a zeolite synthesis methodology that
offers better control of the morphology of the final products on a nanometer
scale, while at the same time providing high yields and crystallinity. The use of
carbon templates greatly limited the versatility and controllability of the particle
size. On the other hand, the formation of zeolite nano-particles is not a

thermodynamically favored process because of Ostwald ripening. In order to

25

stabilize nano-particles in the absence of carbon templates, it is essential to
deactivate the surface by planting inert species on the surface. In this way,
control of the crystal size does not rely on kinetics, and the final products can
be fully crystallized.

The third objective is to synthesize zeolites with uniform mesopores, both
inter- and intra-crystal mesopores. There are two major disadvantages
associated with the approaches described earlier for the formation of zeolites
with textural mesoporous. The first one is that, with the exception of the zeolite
formed by nanocasting in a carbon aerogel, the resulting mesopores, whether
inter- or intra- particle mesopores, typically are widely distributed in size. The
uniformity of the mesopores is dependent on the nanocrystals size or the
carbon porogens, both of which tend to be irregular in both size and shape.
Therefore, the resultant zeolites reflect the same broad distribution of
mesopores, which is not suitable for shape or size selective catalytic
conversions. Furthermore, little or no pore volume provided in the 1-2 nm large
micropore range or in the 2-10 nm small mesopore domain, which is highly
desired for shape- or size-selective catalytic conversions or separations of
large molecules. Moreover, such mesoporosity is expected to show
shape-selectivity in important catalytic industrial reactions such as petroleum

cracking and refining.

26

1.4 References
1 Corrna, A. Chem. Rev. 1997, 97(6), 2373.

2 On, D. T.; Kaliaguine, S. Series on Chemical Engineering 2004, 4, 47.
3 Davis, M. E. Micro. Meso. Mater. 1998, 21(4-6), 173.

4 Ding, Z.; Kloprogge, J. T.; Frost, R. L.; Lu, G. 0.; Zhu, H. Y. J. Porous
Mater. 2001, 8(4), 273.

5 McLeary, E. E.; Jansen, J. C.; Kapteijn, F. Micro. Meso. Mater. 2006,
90(1-3), 198.

6 Corma, A.; Garcia, H. Chem. Comm. 2004, 13, 1443.

7 Corma, A.; Diaz-Cabanas, M. J.; Jorda, J.; Martinez, 0.; Moliner, M.
Nature, 2006, 443(7113), 842.

8 Beck, J. S.; Vartuli, J. C.; Roth, W. J.; Leonowicz, M. E.; Kresge, C. T.;
Schmitt, K. D.; Chu, C. T. W.; Olson, D. H.; Sheppard, E. W.; et al. J. Am.
Chem. Soc. 1992, 114(27), 10834.

9 Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. 0.; Beck, J. 8.
Nature, 1992, 359(6397), 710.

10 Katiyar, A.; Ji, L.; Smirniotis, P. G.; Pinto, N. G. Micro. Meso. Mater. 2005,
80(1-3), 311.

11 Slowing, I. l.; Trewyn, B. 6.; Lin, V. S.-Y. J. Am. Chem. Soc. 2007, 129(28),
8845.

12 Zhu, Y.; Shi, J. Micro. Meso. Mater. 2007, 103(1-3), 243.

13 Slowing, l. I.; Trewyn, B. G.; Giri, 8.; Lin, V. S.-Y. Adv. Funct. Mater. 2007,
17(8), 1225.

14 Hoffmann, F.; Cornelius, M.; Morell, J.; Froeba, M. Angew. Chem. Int. Ed.
2006, 45(20), 3216.

15 Huo, Q.; Margolese, D. l.; Ciesla, U.; Feng, P.; Gier, T. E.; Sieger, P.; Leon,
R.; Petroff, P. M.; Schueth, F.; Stucky, G. D. Nature 1994, 368(6469), 317.

16 Gao, C.; Qiu, H.; Zeng, W.; Sakamoto, Y.; Terasaki, O.; Sakamoto, K.;

27

Chen, 0.; Che, S. Chem. Mater. 2006, 18( 16), 3904.

17

Zhao, D.; Feng, J.; Huo, 0.; Melosh, N.; Frederickson, G. H.; Chmelka, B.

F.; Stucky, G. 0. Science 1993, 279(5350), 548.

18 Liu, X.; Tian, B.; Yu, 0.; Gao, F.; Xie, 8.; Tu, 8.; Che, R.; Peng, L.; Zhao, D.
Angew. Chem. Int. Ed. 2002, 41(20), 3876.

19

Kleitz, F.; Liu, 0.; Anilkumar, G. M.; Park, I.; Solovyov, L. A.; Shmakov, A.

N.; Ryoo, R. J. Phys. Chem. B 2003, 107(51), 14296.

20

21

22

23

24

25

26

27

28

29

30

31

32

33

Kim, S-S.; Zhang, W.; Pinnavaia, T. J. Science 1998, 282(5392), 1302.
Tanev, P. T.; Pinnavaia, T. J. Science 1995, 267(5199), 865.

Kim, S.; Pauly, T. R.; Pinnavaia, T. J. Chem. Comm. 2000, 17, 1661.
Shimojima, A.; Kuroda, K. Angew. Chem. Int. Ed. 2003, 42(34), 4057.
Wan, Y.; Zhao, D. Chem. Rev. 2007, ASAP.

Liu, Y.; Kim, S.; Pinnavaia, T. J. J. Catal. 2004, 225(2), 381.

Mokaya, R.; Jones, W. Chem. Comm. 1998, 17, 1839.

Mokaya, R.; Jones, W. J. Mater. Chem. 1999, 9(2), 555.

Jin, Z.; Wang, X.; Cui, X. J. Coll. Inter. Sci. 2007, 307, 158.
Flodstrom, K.; Alfredsson, V. Micro. Meso. Mater. 2003, 59(2-3), 167.
Kruk, M.; Celer, E. B.; Jaroniec, M. Chem. Mater. 2004, 16, 698.

Liu, Y.; Pinnavaia, T. J. J. Mater. Chem. 2002, 12(11), 3179.

Chen, L.Y.; Jaenicke, S.; Chuah, G.K. Micro. Mater: 1997, 12, 323.

Yang, X.; Vantomme, A.; Lemaire, A.; Xiao, F.; Su, B. Adv. Mater. 2006,

18(16), 2117.

34

Kloetstra, R. K.; Jansen J. C.; van Bekkum, H. Chem. Commun., 1997,

2281 .

35

Verhoef, M. J.; Kooyman, P. J.; van der Waal, J. C.; Rigutto, M. S.; Peters,

28

J. A.; van Bekkum, H. Chem. Mater. 2001, 13(2), 683.

36 On, D.; Kaliaguine, S. Angew. Chem. Int. Ed. 2001, 40(17), 3248.

37 Fang, Y.; Hu, H. J. Am. Chem. Soc. 2006, 128, 10636.

38 Xia, Y.; Mokaya, R. J. Mater. Chem. 2004, 14, 863.

39 Liu, Y.; Zhang, W.; Pinnavaia, T. J. J. Am. Chem. Soc. 2000, 122, 8791.

40 Liu, Y.; Zhang, W.; Pinnavaia, T. J. Angew. Chem. Int. Ed. 2001, 40(7),
1255.

41 Liu, Y.; Pinnavaia, T. J. J. Mater. Chem. 2004, 14(7), 1099.

42 Liu, Y.; Pinnavaia, T. J. J. Mater. Chem. 2002, 12, 3179.

43 Liu, Y.; Pinnavaia, T. J. Chem. Mater. 2002, 14(1), 3.

44 Zhang, 2.; Han, Y.; Xiao, F.; Qiu, S.; Zhu, L.; Wang, R.; Yu, Y.; Zhang, Z.;
Zou, B.; Wang, Y.; Sun, H.; Zhao, D.; Wei, Y. J. Am. Chem. Soc. 2001,
123(21), 5014.

45 Zhang, Z. T.; Han, Y.; Zhu, L.; Wang, R. W.; Yu, Y.; Qiu, S. L.; Zhao, D. Y.;
Xiao, F. S. Angew. Chem. Int. Ed. 2001, 40, 1258.

46 On, D. T.; Kaliaguine, S. Angew. Chem. Int. Ed. 2002, 41(6), 1036.
47 On, D. T.; Kaliaguine, S. J. Am. Chem. Soc. 2003, 125(3), 618.

48 lnagaki, S.; Ogura, M.; Inami, T.; Sasaki, Y.; Kikuchi, E.; Matsukata, M.
Micro. Meso. Mater. 2004, 74(1-3), 163.

49 Goto, Y.; Fukushima, Y.; Ratu, P.; Imada, Y.; Kubota, Y.; Sugi, Y.; Ogura, M.;
Matsukara, M. J. Porous. Mater. 2002, 9,43.

50 Wang, S.; Dou, T.; Li, Y.; Zhang, Y.; Li, X.; Yan, Z. J. Solid State Chem.
2004, 1 77( 12), 4800.

51 Aguao, J.; Sotelo, J. L.; Serrano, D. P.; Calles, J. A.; Escola, J. M. Energy
& Fuels 1997, 11, 1225.

52 Mintova, 8.; Olson, N. H.; Valtchev, V.; Bein, T. Science 1999, 283, 958.

29

53 Majano, G.; Mintova, S.; Ovsitser, O.; Mihailova, 8.; Bein, T. Micro. Meso.
Mater. 2005, 80, 227.

54 Valtchev, V. P.; Bozhilov, K. N. J. Phys. Chem. B 2004, 108, 15587.

55 Aguado, J.; Serrano, D. P.; Escola, J. M.; Rodriguez, J. M. Micro. Meso.
Mater. 2004, 75, 41.

56 Larlus, O.; Mintova, S.; Bein, T. Micro. Meso. Mater. 2006, 96(1-3), 405.
57 Schmidt, I.; Madsen, C.; Jacobsen, C. J. H. Inorg. Chem. 2000, 39, 2279.
58 Madsen, C.; Jacobsen, C. J. H. Chem. Commun. 1999, 673.

59 Kim, S-S.; Shah, J.; Pinnavaia, T. J. Chem. Mater. 2003, 15(8), 1664.

60 Serrano, D. F.; Aguado, J.; Escola, J. M.; Rodriguez, J. M.; Peral, A. Chem.
Mater. 2006, 18, 2462.

61 Landau, M. V.; Vradman, L.; Valtchev, V.; Lezervant, J.; Liubich, E.;
Talianker, M. Ind. Eng. Chem. Res. 2003, 42, 2773

62 Zhang, R; Wang, X.; Guo, X.; Guo, H.; Zhao, L.; Hu, Y. Catal. Left. 2004,
92, 63.

63 Vogel, B.; Schneider, 0.; Klemm, E. Catal. Lett. 2002, 79, 107.

64 Groen, J. C.; Peffer, L. A. A.; Moulijn, J. A.; Perez-Ramirez, J. Chem. Eur.
J. 2005, 11(17), 4983.

65 Flanigen, E. M. Stud. Surf. Sci. Catal. 1991, 58, 13.

66 Janssen, A. H.; Schmidt, I.; Jacobsen, C. J. H.; Koster, A. J.; de Jong, K. P.
Micro. Meso. Mater. 2003, 65(1), 59.

67 Schmidt, l.; Boisen, A.; Gustavsson, E.; Stahl, K.; Pehrson, S.; Dahl, S.;
Carlsson, A.; Jacobsen, C. J. H. Chem. Mater. 2001, 13, 4416.

68 Tao, Y.; Kanoh, H.; Kaneko, K. J. Am. Chem. Soc., 2003, 125(20), 6044.
69 Ryoo, R.; Joo, 8.; Jun, S. J. Phys. Chem. B1999, 103(37), 7743.
70 Che, S.; Garcia-Bennett, A. E.; Liu, X.; Hodgkins, R. P.; Wright, P. A.;

Zhao, D.; Terasaki, 0.; Tatsumi, T. Angew. Chem. Int. Ed. 2003, 42(33), 3930.

30

71 Kim, C.; Lee, D.; Pinnavaia, T. J. Langmuir 2004, 20(13), 5157.

72 Lai, X.; Li, X.; Geng, W.; Tu, J.; Li, J.; Qiu, S. Angew. Chem. Int. Ed. 2007,
46(5). 738.

73 Yang, 2.; Xia, Y.; Mokaya, R. Adv. Mater. 2004, 16(8), 727.

74 Xiao, R, Wang, L.; Yin, 0.; Lin, K.; Di, Y.; Li, J.; Xu, R.; Su, D. S.; Schlogl,
R.; Yokoi, T.; Tatusmi, T. Angew. Chem. Int. Ed. 2006, 45, 3090.

75 Choi, M.; Cho, H. S.; Srivastava, R.; Venkatesan, C.; Choi, D.-H.; Ryoo, R.
Nature Mater. 2006, 5(9), 718.

76 Choi, M.; Srivastava, R.; Ryoo, R. Chem. Comm. 2006, 42,4380.

77 Janssen, A. H.; Koster, A. J.; de Jong, K. P. J. Phys. Chem. B 2002, 106,
11905.

31

Chapter 2
Highly acidic mesostructured aluminosilicates
assembled from surfactant-mediated zeolite

hydrolysis products

2.1 Abstract

Mesostructured aluminosilicates with strong acidity were synthesized
through a top-down approach using zeolite hydrolysis products as precursors.
The hydrolysis of zeolites was conducted in the presence of surfactants, most
preferably cetyltrimethylammonium ions, that helped to stabilize zeolitic
subunits in solution. Commercial zeolites ZSM-5 (Si/Al = 40) and USY (Si/AI =
40) were successfully transformed completely into mesostructures. The
resultant mesostructured aluminosilicates, denoted MSU-Z, showed high
surface area (> 1000 mzlg) and uniform mesopores (2.5 nm) comparable to
MOM-41 silica prepared from conventional precursors. In the case of ZSM-5
zeolite as a precursor, the successful incorporation of zeolitic subunits into the
resultant mesostructure was verified by IR spectroscopy. Compared to other
methods, this top-down approach optimized the preservation of zeolitic
subunits in solution for subsequent conversion to a mesostructure. The
resultant MSU-Z aluminosilicates showed unprecedented high activity for
catalytic cumene cracking.

Several factors affect the properties of MSU-Z materials assembled from

32

zeolite precursors. Low SilAl zeolite Y with an initial Si/Al ratio of 6: 1 can only
be partially transformed into a mesostructure even under severe hydrolysis
conditions (OH'lT = 3, 100 °C). XRD data revealed that the residual zeolite has
a lower Si/AI ratio than the starting zeolite, indicative of the preferred leaching
of silica. Moreover, even in the presence of subunit-stabilizing surfactant,
ZSM-5 zeolite can be over-hydrolyzed by reaction with excessive NaOH.
Overhydrolyzed zeolite precursors afford a mesostructure equivalent in acidity
and stability to conventional MGM-41. Several other surfactants including
fallow tetramine and P123 were also used. Compared to cationic surfactant
CTAB, these surfactants interact with zeolitic nanoclusters through H-bonding,

and provide less stabilization of the subunits than a cationic surfactant.

33

2.2 Introduction

In 1992, Mobil researchers reported the synthesis of mesostructured
MOM-41 silica through the supramolecular assembly of surfactant micelle
templates." 2 These mesostructured materials showed very high surface area,
typically above 1000 mzlg, and uniform and controllable mesopore size from 2
to 50 nm. Such materials found many interesting applications in separation} 4
adsorption} 6 electronic devices,7'9 and drug delivery.”13 The high surface
areas and large pore sizes of mesostructured aluminosilicates have been
recognized as desirable properties for the catalytic processing of large

‘4' 15 and polymers.“ ‘7 The high surface area

molecules such as enzymes
provides more surface-active sites, and the presence of uniform mesopores
allows reactant and product molecules to diffuse easily in and out of the
catalyst pore system. Moreover, the uniformity of the mesopore provide for the
possibility of shape-selectivity that is not available for other catalytic
materials.""20

The cracking of hydrocarbons to produce lower molecular weight
products is one of the most important reactions that allows the conversion of
petroleum to lighter products such as transportation fuels.21 During the
cracking reaction, a carbonium species is formed initially at strong acidic sites.
Typical cracking catalysts consist of acidic aluminosilicates as main

components. The discovery of strongly acidic zeolite catalysts greatly boosted

the yield of transportation fuels due to their strong acidity.22 However, because

34

of the pore size limitation and poor diffusion through their micropores, the
heavy fraction of petroleum is not easily processed over zeolites.23 The pore
sizes of most zeolites typimlly is limited to micropores of 0.3 ~ 1.2 nm.24
Aluminosilicates with non-uniform mesopores act as supplemental
components to zeolites for bulky molecule processing.25 Due to their high
surface area and uniform mesopores, mesostructured aluminosilicates have
been considered as promising catalysts with a desirable selectivity.26

Mesostructured aluminosilicates are not as stable hydrothermally as
zeolites. The stability limitation of mesostructured aluminosilicates has been
attributed in part to the thin and amorphous pore walls, typically 1 to 2 nm for
MCM-41 material. However, later studies have shown that even thick-wall
mesostructures such as SBA-1527 and MCF28 are not stable under
hydrothermal conditions. More importantly, mesostructured aluminosilicates
have very weak acidity, which limits their applications in acidic catalysis. The
ultimate reason for the poor hydrothermal stability and acidity is that the
mesostructured framework lacks atomic ordering in the walls.

In 2000, a novel approach to the preparation of mesostructured
derivatives with improved hydrothermal stability and acidity was reported
based on the incorporation of protozeolitic nanoclusters or “zeolite seeds" in
the framework.” 3° This was accomplished by quenching a zeolite synthesis
gel prior to the onset of zeolite crystal formation and then transforming the

quenched gel into a mesostructure. The generality of this approach and the

35

spectroscopic evidence supporting the presence of zeolitic subunits in the
mesostructured framework has been substantiated by several subsequent
studies.”34 This bottom-up approach greatly improved the catalytic activity of
resultant mesostructured aluminosilicate materials in comparison to a
conventional MOM-41 catalyst. However, their catalytic activities are still not
comparable with commercial zeolite catalysts. This could be due to the limited
amount of zeolite subunits incorporated into the mesostructures.

Several other approaches have been developed to improve the acidity of
mesostructured aluminosilicates by introducing zeolitic subunits into the pore
walls. The direct crystallization of amorphous walls of MOM-41 lead to the
complete loss of mesostructure.35 Even with very thick-wall materials like
mesostructured cellular foam silicas, partial crystallization of the amorphous
walls was achieved with the use of a non-aqueous solvent, accompanied by a
significant loss of mesopores?6 In a related approach, large pore silica
mesostructures were coated with zeolitic seeds.“ 38 The incorporation of
zeolite seeds was verified by the increase of acidity. However, through the
re-construction of the mesostructure, great reduction of pore size, pore volume,
and surface area was observed. Moreover, the coating approach is not
applicable to small pore materials such as MOM-41.

In this study, we developed a new approach to the synthesis of highly
acidic mesostructured aluminosilicates. Our top-down approach to improving

the incorporation of zeolitic subunits into the framework walls is based on the

36

hydrolysis of a pre-formed zeolite in the presence of a cationic surfactant. The
surfactant serves two purposes. The first purpose is to stabilize the zeolitic
subunits through ion-pairing during the hydrolysis so that more zeolitic
subunits can be preserved. After that, the surfactant also serves as a
structure-directing porogen for the assembly of the mesostructure. The
resultant mesostructured aluminosilicate materials, denoted MSU-Z, are

characterized by N2 isotherm, XRD, TEM and FT-IR.

37

2.3 Experimental section

2.3.1 Materials

ZSM-5 (CBV 8014), USY (CBV 780), and Y (CBV 712) zeolites with Si/AI
atomic ratios of 40 were obtained from Zeolyst. Sodium hydroxide,
cetyltrimethylammonium bromide (CTAB), and sulfuric acid (100%) were
purchased from Aldrich. Tallow tetramine (‘ITeA, C16H33(NH(03H5))3NH2) was
obtained from Tomah as a gift. P123 ((EO)2o(PO)70(EO)2o) surfactant was

obtained from BASF.

2.3.2 Synthesis of mesostructured aluminosilicates from zeolite
precursors

ZSM-5 zeolite (2.00 g, 33.3 mmol) was dispersed in a solution containing
55 mL of 0.45 M NaOH solution (OH'IT = 0.75, where T = Si + Al), to which
2.45 g (6.7 mmol) CTAB was added as mesoporogen. The mixture was stirred
at room temperature for 1 hour, and transferred into a Teflon-lined autoclave,
heated at 100 °C for 24 hours, and then cooled to room temperature. The pH
of the solution was adjusted to 9.0 through the addition of sulfuric acid and the
mixture was heated again at 100 °C for another 24 hours to form a hexagonal
mesostructure. The product was filtered, washed, and calcined at 600 °C for 4
hours to remove the organic surfactant. The final product was denoted as
MSU-Z. The protonated form of MSU-Z was obtained by duplicate

ion-exchange reaction with 0.5 M NH4N03 solution at 60 °C for 2 hours and

38

then calcining the product at 550 °C.

For the assembly of MSU-Z with a wormhole structure, ZSM-5 zeolite
(2.00 g, 33.3 mmol) was dispersed in a solution containing 55 mL of 0.45 M
NaOH solution (OH'IT = 0.75, where T = Si + Al), to which 2.0 g TTeA was
added as mesoporogen. The mixture was stirred at room temperature for 1
hour, and transferred into a Teflon-lined autoclave, heated at 100 °C for 24
hours, and then cooled to room temperature. Sulfuric acid (1.21 g, 12.3 mmol)
was added and the mixture was heated again at 60 °C for another 24 hours in
a shaking bath.

For the assembly of large pore hexagonal MSU-Z, ZSM-5 zeolite (2.00 g,
33.3 mmol) was dispersed in a solution containing 55 mL of 0.45 M NaOH
solution (OH‘lT = 0.75, where T = Si + Al), to which 2.4 9 P123 was added as
mesoporogen. The mixture was stirred at room temperature for 1 hour, and
transferred into a Teflon-lined autoclave, heated at 100 °C for 24 hours, and
then cooled to room temperature. The pH of the solution was adjusted to 4.5
with a diluted sulfuric solution. The reaction mixture was kept at 35 °C water
bath for 24 hours and then 60 °C for another 24 hours.

The reaction stoichiometries for MSU-Z syntheses are listed in Table 2.1.

2.3.3 Characterization
The physical properties of MSU-Z materials were determined using X-ray

diffraction (XRD), nitrogen adsorption/desorption isotherm, transmission

39

electron microscopy (T EM), 27AI-N MR and FT-IR.

Powder XRD patterns were measured on a Rigaku Rotaflex
Diffractometer equipped with an anode using Cu Kc radiation (A = 0.154 nm).
Scan consitions are 26 continuous reflective mode from 5° to 50° for wide
angle or 1° to 10° for low angle at 2°/min and 0.05°/step. N2 isotherms at -196
°C were obtained on a Micromeritics TriStar 3000. Samples were degassed at
150 °C and 10" Torr for a minimum of 6 hours prior to analysis. The surface
area was calculated using the BET equation. Pore size distributions were
derived from the adsorption isotherms using the BJH model. TEM images were
obtained on a JEOL 2200FS instrument operated at 200 kV. Samples were
prepared by sonicating the ground sample in EtOH for 20 min, and then
evaporating 2 drops of the suspension onto a carbon-coated holey film
supported on a 300 mesh Cu grid. 27Al MAS NMR spectra were recorded on a
Varian VXR-4OOS spectrometer with zirconia rotors 4 mm in diameter and
spun at 4 KHz. External AI(H2O)23+ was used as a chemical shift reference.
FT-IR spectra were recorded on a Mattson Galaxy Series FTIR 3000. The
adsorption band at 450 cm'1 was selected as the internal reference. This band

was present in all silica samples and was assigned to the SI-O-Si vibration.

2.3.4 Catalytic cumene cracking
Catalytic cumene cracking reactions at 300 °C were carried out in a 6 mm

id. fixed bed quartz reactor (as shown in Figure 2.1) containing 200 mg of

40

catalyst. Prior to catalytic testing, the protonated catalysts were activated at
300°C under N2 flow for a minimum of 2 hours. The cumene flow rate was 4.1
pmol/min in a 20 cm3/min carrier stream of N2. Cumene conversions were

reported after 1 hour on stream.

41

2.4 Results and discussion

Figure 2.2 provides the XRD patterns of MSU-Z aluminosilicates
assembled from ZSM-5 and USY fragments. Both samples show four
distinguishable Bragg diffraction peaks at low angle, indexed as 100, 110, 200,
and 120 respectively, indicative of the formation of the same 1-D hexagonal
mesostructure as of MOM-41. Moreover, no Bragg peaks at wide angle was
found, which excluded the presence of a crystalline zeolite phase. The 27Al
MAS NMR chemical shifts of the as-made MSU-Z prepared from ZSM-5
hydrolysis product (Figure 2.3) indicate that all of the aluminium in the
mesostructures is in tetrahedral coordination (51.2 ppm). The exclusively
tetrahedral siting of aluminum is very desirable in providing strong acidity.
Upon calcination, some octahedral aluminium is observed at ~3 ppm, but the
majority of aluminum centers remain in tetrahedral coordination.

Diluted sodium hydroxide solution has been reported to leach out
framework atoms from zeolites to generate intra-crystal mesopores.39 The
leached-out species can then be subsequently used as precursors to
assemble mesostructures. Normally, the silica is more easily removed from a I
zeolite framework under basic conditions than alumina. Therefore, under
partial hydrolysis conditions, the mesostructured aluminosilicates made from
the hydrolysis products exhibited higher Si/AI ratio than that of parent zeolite.40
Moreover, the incorporation of five-membered ring subunits in the framework

walls of the mesostructure, as well as the cracking activity of the

42

mesostructure, is no better than what can be achieved through the direct
assembly of aluminosilicate mesostructures from protozeolite seeds formed in
the presence of a zeolite template. Under the surfactant-mediated hydrolysis
conditions of the current study, however, all the zeolite starting materials are
completely transformed into mesostructures, as confirmed by the absence of
wide angle diffraction peaks in the final products. As listed in Table 2.2, no
significant change in Si/AI ratio was observed for MSU-Z materials. The results
suggested that all of the starting zeolite materials ended in the final
mesostructured aluminosilicates. The new method offers the opportunity to
easily control the Si/Al of resultant mesostructures.

As shown in Figure 2.4, the IR spectrum of MSU-Z formed from ZSM-5
fragments (curve C) exhibits an intense band near 550 cm", indicative of
five-membered ring subunits analogous to those subunits41 found for the
pristine zeolite (curve D). This band is absent in the spectrum of MOM-41
prepared from conventional aluminosilicate precursors (curve A) and
comparatively weak in the spectrum of MSU-S made from protozeolitic ZSM-5
seeds (curve B). As expected, this band is absent from the spectrum of
MSU-Z made from hydrolyzed USY zeolite, as this zeolite does not contain
five-ring subunits. On the basis of these FTIR results for MSU-Z made from
hydrolyzed ZSM-5 fragments, it is clear that the “top-down” degradation of a
zeolite in the presence of a fragment-stabilizing surfactant is superior to the

earlier “bottom-up” approach to generating zeolite subunits from conventional

43

aluminosilicate precursors.

Figure 2.5 provides the nitrogen adsorption - desorption isotherms and
framework pore size distributions for MSU-Z made from surfactant-mediated
ZSM-5 and USY zeolite fragments (curves C and B, respectively) in
comparison to conventional MOM-41 (curve A). All the samples showed typical
type IV isotherm with a nitrogen capillary condensation at P/Po ~ 0.4,
indicative of the presence of uniform cylindrical pores of 2~3 nm (Figure 2.6).
The physical properties are listed in Table 2.2. In comparison to conventional
mesostructured aluminosilicate MOM-41, MSU-Z materials exhibited similar
BET surface area (> 1000 mzlg) and pore volume (0.86 cm3/g). Furthermore,
the unit cell sizes derived from XRD patterns indicated an increase from 4.64
nm to 4.98 nm, while the pore size for MSU-Z remained the same as MGM-41.
The results suggested that MSU-Z materials prepared from zeolite fragments
have thicker walls than MOM-41, prepared from molecular precursors.

When the hydrolysis of ZSM-5 was carried out in the absence of
surfactant and the pH adjusted to 9.0 (trial 10), a low surface area
aluminosilicate gel was obtained that had no 550 cm'1 IR band and virtually no
activity for cumene cracking when exchanged with NH4" ions and calcined.
Thus, ion paring to the cationic cetyltrimethyl ammonium ion surfactant plays
an essential role in stabilizing the anionic zeolitic fragments formed upon
hydrolysis. We emphasize that it is possible to over-hydrolyze the zeolite

precursor to the point where the concentration of zeolitic fragments available

for mesostructure formation is compromised despite the presence of a cationic
surfactant (trial 4). Curve D in Figure 2.5 is the isotherm for a mesophase
formed from ZSM-5 that was deliberately over-hydrolyzed by using an
excessive concentration of sodium hydroxide. A unit cell size similar to
MOM-41 was observed.

The hypothesis of ion-pairing between inorganic silica and organic
surfactant was further investigated. Figure 2.7 provides the XRD patterns of
hydrolyzed products of ZSM-5 in the presence of different amounts of CTAB.
The broad diffraction peak near 23° is due to the Si-O bond on amorphous
silica. It can be clearly seen that as the concentration of CTAB increased, the
Bragg diffraction peaks characteristic of ZSM-5 zeolite showed up, as marked
in Figure 2.6. It is conceivable that CTAB stabilizes zeolite structures during
the hydrolyzation.

In order to test the acidity of MSU-Z mesostructures made from
surfactant-mediated zeolite hydrolysis products, the cumene cracking was
carried out as a probe reaction. Prior to catalysis the mesostructures were
exchanged with NH4N03 and calcined at 550 °C to afford the protonated form
of the mesostructure. The results are listed in Table 2.2. For comparison, the
cumene cracking also was carried out over an ammonium exchanged
conventional MGM-41 with a Si/Al ratio of 50, which showed a cumene
conversion of 11%, consistent with previous reported results.32 The MSU-Z

mesostructure prepared from USY fragments showed a substantial 3.4-fold

45

increase in cumene conversion in comparison to MGM-41. This high cumene
activity value is similar to the activity observed for MSU-S aluminosilicates
assembled from faujasite zeolite seeds. The improved catalytic activity
suggested that fragments of USY zeolite were incorporated into the
mesostructured aluminosilicate. The 73% conversion found for MSU-Z made
from the surfactant-mediated hydrolysis of ZSM-5 is 6.7-fold higher than
observed for a conventional MOM-41. This level of cumene cracking activity is
unprecedented among mesostructured aluminosilicates. The highest cumene
conversion previously observed under analogous reaction conditions (57%)
was for a MSU-S made from faujasite seeds at a Si/AI of 5.7.

The cumene cracking activity of MSU-Z made from surfactant-mediated
ZSM-5 fragments approaches the 90 - 95% conversions observed for the
pristine ZSM-5 at the same Si/Al ratio. This extremely high acidity cannot be
attributed to the presence of residual ZSM-5 crystals in the mesostructured
product, because the presence of such crystals is precluded by the absence of
wide angle Bragg peaks in the XRD powder pattern. TEM image of MSU-Z
sample exhibits a typical hexagonal ordering observed on MCM-41
mesostructures (Figure 2.8). Several composite compositions containing
mixtures of zeolite and mesostructured aluminosilicate phases have been
reported previously.42' ‘3 These mixtures show wide angle Bragg X-ray
reflections characteristic of the zeolite phase. Also, the zeolite component of

these mixtures typically exhibit lattice fringes in the bright field TEM mode and

46

dense nanocrystallite domains in dark field mode. None of these XRD or TEM
features representative of a zeolite phase were observed for MSU-Z made
from surfactant-mediated ZSM-5 fragments. Moreover, the absence of zeolitic
microporosity was verified through t-plots of the nitrogen adsorption data, as
expected for a framework that contains the secondary building units of a
zeolite, but not zeolite cavities. A t-plot is a measurement that compares the
sample with a standard surface. The deviation of the y-axis intercept from the
origin indicates the presence of micropores. For a mixture of MCM-41 with 5%
ZSM-5 (Figure 2.9, curve A), a positive deviation was observed, consistent
with the presence of microporous zeolite phase. However, the t-plot of MSU-Z
passed through origin (Figure 2.9, curve B). All the experimental results
confirmed the absence of a zeolite phase in this highly acidic MSU-Z material.
Therefore, the unprecedented high cracking activity is a result of the high
concentration of zeolitic subunits incorporated into the mesostructure, as
revealed by the FT-IR spectra.

It is noteworthy that much less NaOH is required to completely
decompose USY than ZSM-5 zeolite. Compared to ZSM-5, which has 0.55 nm
micropore size and 350 m2/g surface area, the larger pore diameter (0.74 nm)
and higher surface area (800 m2/g) of USY enable the diffusion of NaOH more
easily. Moreover, to produce USY, steaming and leaching treatments were
applied to zeolite Y to remove framework atoms (especially Al) to generate

irregular mesopores. The presence of mesopore greatly reduced the domain

47

size of zeolite so that it becomes more vulnerable to the basic hydrolysis.
Compared to ZSM-5, which requires an OHTI‘ of 0.75 to be completely
hydrolyzed, pure mesostructured aluminosilicate can be prepared from USY
with an OH'fT value of 0.17. Such a low NaOH concentration can only partially
decompose ZSM-5 structure. Therefore a mixture of zeolite and mesostructure
was resulted. Figure 2.10 provides the XRD pattern of MSU-Z prepared from
ZSM-5 hydrolyzed with insufficient NaOH. Bragg diffraction peaks
corresponding to an MFI framework appear at wide angle, in addition to the
low angle diffraction peaks of the mesostructure. The N2 isotherm data showed
a lower mesopore volume compared to MCM-41.

In addition to USY and ZSM-5, a similar methodology has also been
applied to other zeolite structures such as mordenite.44 As described above,
for different zeolite frameworks, the NaOH concentration required to achieve
complete hydrolyzation differs. In the case of zeolite Y, which has a Si/Al of 6
and same framework structure as of USY, the complete conversion into
mesostructure was not accomplished in reasonable reaction conditions. For
instance, a mixture of zeolite Y and mesostructured aluminosilicate was
obtained at an OH'IT value of 2.25. XRD pattern of the final product revealed
the presence of unconsumed zeolite Y. The reason for this observation could
be two fold. First, although USY and Y share the same structure, the presence
of mesopores within USY sample makes it more hydrolysable. In addition,

zeolite Y has a much higher aluminum content. The higher degree of isotopic

48

substitution of Si by Al helps stabilize the zeolite structure under basic
hydrolysis conditions. Moreover, a shift of d-space was observed in the wide
angle X-ray diffraction patterns for the zeolite Y residues compared to the
pristine zeolite. The d-spacing shift indicates a change in framework Si/Al ratio.
As the framework is enriched in Al, an expansion of the unit cell was observed.
We are able to determine the framework Si/Al of the residual Y zeolite using
the following equation:45
N(Al) = m (a0 — X), where m = 115.2, X = 24.191

where N(Al) equals the framework aluminum sites per unit cell ( [T1920324] per
unit cell for faujasite) and a0 is the unit cell parameter in A. After hydrolysis, a
zeolite Y with SilAI of 3 was observed, which suggested that silica was
preferably leached out from the framework during basic hydrolyzation.

Surfactants other than cationic CTAB were also capable of stabilizing
zeolite fragments and subsequently directing the formation of mesostructures.
Figure 2.11 provides the N2 isotherms and the corresponding pore size
distributions for MSU-Z materials prepared from ZSM-5 and USY fragments in
the presence of non-ionic surfactants P123 and TTeA. In both cases, high
quality mesostructured aluminosilicate materials were produced. Physical
properties are listed in Table 2.2. The resultant MSU-Z materials showed high
surface area and uniform mesopores. The resutls are in agreement with the
reported value for the large-pore hexagonal MSU-H and for the wormhole

structure assembled from conventional precursors. XRD patterns (Figure 2.12)

49

showed that no zeolite phase was observed in the hydrolysis products. It is
noteworthy that under the same hydrolysis condition (OH'IT ratio), wide angle
Bragg peaks corresponding to the ZSM-5 zeolite (Figure 2.12, curve A) were
found in the hydrolysis product using CTAB as structure-directing agent.
Whereas the cationic CTAB surfactant can ion-pair with zeolite hydrolysis
fragments in solution, the ‘ITeA and P123 surfactants can interact with zeolitic
nanoclusters only through hydrogen bonding. Apparently, the stronger
electrostatic interaction provided by CTAB stabilizes the zeolitic subunits more
efficiently than weaker H-bonding. Nevertheless, the results showed that the
top-down approach to the incorporation of zeolitic subunits into the framework
walls of a mesostructure is general and applicable to other zeolite precursors

and surfactants.

50

2.5 Conclusion and perspective

Through the incorporation of zeolitic subunits into the walls,
mesostructured aluminosilicates with strong acidity can be prepared. Two
different pathways have been developed by our group to introduce zeolitic
subunits, namely bottom-up (for MSU-S materials) and up-bottom approaches
(for MSU-Z materials). MSU-Z materials are prepared with the
surfactant-mediated hydrolysis product of preformed zeolites as precursors.
Compared to bottom-up approach, mesostructured aluminosilicates prepared
from up-bottom approach retained more zeolitic characteristics as judged by
the absorbance intensity of five-member ring in IR spectra. The resultant
MSU-Z materials are also much more acidic than conventional MGM-41
aluminosilicates and even MSU-S materials. For instance, in the catalytic
cumene cracking reaction, 2%Al-MCM-41 aluminosilicate catalyst has a
cumene conversion of 11%, at the same condition 2%AI-MSU-S materials
provide a cumene conversion as high as 43%, which can be improved to 57%
if the Al content of the mesostructure is increased to 15%. On the other hand,
2%Al-MSU-Z prepared from ZSM-5 precursor exhibits a cumene conversion of
74%, a 6-fold increase from 2%Al-MCM-41.

Other than ZSM-5, other types of zeolites such as USY can also be
used as the precursors for MSU-Z. Due to the structural difference between
ZSM-5 and USY, different hydrolysis conditions are used. Compared to ZSM-5,

USY has larger framework pore diameter, larger surface area, and more

51

textual porosity, and therefore is more vulnerable to hydrolyzation. An OH'ISi
of 0.17 is enough to completely convert USY into a mesostructure, while at
least OH’ISi of 0.75 is needed for ZSM-5. Moreover, the interaction between
surfactant molecules and zeolite subunits is very important to preserve the
zeolitic subunits during hydrolysis process. The hydrolysis product in the
absence of surfactant shows no cumene cracking activity at all. In addition,
several common types of surfactants have been successfully used to prepare
MSU-Z materials. In general, the Coulomb force between cationic surfactant
CTAB and anionic aluminosilicate cluster is much stronger that the
hydrogen-bonding interaction between neutral surfactant and aluminosilicate
cluster. Among the three surfactants, CTAB provides the best stabilization to
the zeolitic subunits.

Assembled from small and uniform precursors with zeolitic cross-linking,
MSU-S materials exhibited ultrahigh hydrothermal stability in comparison to
conventional MGM-41 materials. Compared to bottom-up approach, the
top-down approach produces precursors with substantial heterogeneity and
probably larger size. Therefore, the structure of MSU-Z materials might not be
as robust as MSU-S materials, although the former is more active in catalytic
cumene cracking reaction. Future work should be focusing on the effect of
zeolite Si/AI ratio. It is clear that framework AI stabilizes the crystal structure of
zeolite during NaOH hydrolysis.46 Therefore, the distribution of Al and the

framework Si/Al ratio would greatly affect the properties of resultant zeolite

52

fragments, and subsequently affect the hydrothermal stability of final MSU-Z

products.

53

Table 2.1 Summary of reaction stoichiometry for the synthesis of MSU-Z

materials from crystalline zeolites as the aluminosilicate source.

 

 

Trial Zeolite Si/Al NaOH/g Surfactant Surfactant/g
source"

1 ZSM-5 40 0.75 CTAB 2.45
2 ZSM-5 40 1 .0 CTAB 2.45
3 ZSM-5 40 1 .0 CTAB 3.45
4 ZSM-5 40 2.0 CTAB 2.45
5 ZSM-5 40 1 .0 F123 2.4
6 USY 40 0.228 CTAB 2.45
7 USY 40 0.228 TTeA 2.0
8 Y 6 3.0 CTAB 2.45
9 ZSM-5 40 1 .0 TTeA 2.0
10 ZSM-5 40 0.75 -- -

 

* For all the trials, the amount of zeolite is always 2.0 g.

54

Table 2.2 Physical properties of different mesostructured aluminosilicate

 

 

materials.

Trial Surface Pore Unit cell Pore Si/AI Cumene
area, size, size a, *, volume, conversioin,
mgr; nm nm cm3/g %

MCM-41 1135 2.4 4.64 0.87 50 11
2 1160 2.4 4.98 0.86 42 74
4 1114 2.2 4.75 0.73 -- 17
5 804 7.2 11.7 0.93 -- --
6 1051 2.5 4.98 0.86 43 37
7 650 4.5 5.92 0.71 -- -
10 121 22.1 -- 0.70 - < 5

 

* Hexagonal unit cell size was calculated from the d-100 spacing of XRD

patterns.

a0 = 2d/1/3.

55

9.2090 8583 03.93 .8 ago»... Louomom flu 059“.

 

 

 

 

 

 

 

 

 

 

 

     

 

 

 

 

 

 

395.5 I!
Lon—=20
ciao: macaw o.a:ooo.EoE.|. $3.5m
.23:
will. LonEaco
0.8an II fi oEEum
L
wnwpwiflw «95 5.530
u. I.” .
.sesoEeE\v E...
e H . .. . 8.3.. no as 2
co
ism: as a >
953... 29:3

5.530 00 EE m

 

56

35000

 

100

30000 ~

25000 -

 

20000 -

“695)

15000 ~

 

10000 ~

5000 ~

 

 

 

 

 

Figure 2.2‘X-ray diffraction patterns for (A), MOM-41 (Si/AI = 40) prepared from
conventional precursors, and mesostructured aluminosilicates MSU-Z (Si/AI =
40) prepared from surfactant-mediated hydrolysis products of (B) ZSM-5 (trial
2) and (C) USY (trial 6) as precursors. Each curve is offset by 10,000 cps for

clarity.

57

 

 

 

 

 

 

 

 

160 140 120 100 80 60 40 20 0 -20 -40
Chemical shift, ppm

Figure 2.3 27Al MAS NMR spectra of as-synthesized (solid line) and calcined
(broken line) MSU-Z (Si/AI = 40) prepared from hydrolyzed ZSM-5 as

precursor (trial 2).

58

0.7

 

 

 

 

 

D
0.6 -
0.5 _,
l
8 0.4
S c
e
O
8
<1: 0.3
0.2 — B
01 _ A Ail/2
0 . ,
650 600 550 500
Wavelength, cm-1

Figure 2.4 FT-IR spectra for (A), conventional MOM-41 (Si/Al = 40); (B),
MSU-S (Si/AI = 50) prepared from MFI seeds precursor; (C), MSU-Z prepared
from zeolite fragment formed by hydrolysis of ZSM-5 (Si/AI = 40) in the
presence of CTAB at a'OH-lT ratio of 0.75 (trial 2); (D), commercial ZSM-5

(Si/Al =40).

59

 

 

 

 

 

 

 

1400 «
D r
1200 - I. ,2...
0. _r”
5':
2 1000 « c 2
5
- 800: -- ,
E ,,/ ........... _._
I
° B
3 600 I”
N A
0
5 400
§
200
0 . , , .
° 0-2 0.4 0.6 0.8 1

PIPo
Figure 2.5 Nitrogen adsorption-desorption isotherms for (A), MOM-41 (Si/AI =

40) synthesized from conventional precursors; (B) MSU-Z (Si/Al = 40) made
from zeolite fragments generated from USY in the presence of CTAB with
OH'fI‘ = 0.17 (trial 6); (C) MSU-Z (Si/Al = 40) assembled from CTAB
surfactant-mediated ZSM-5 hydrolysis products (zeolite fragments) formed
with OH'lT = 0.75 (trial 2); and (D), mesostructured aluminosilicate (Si/AI = 40)

formed through over-hydrolysis of ZSM-5 with OH'IT = 1.50 (trial 4).

60

 

2.50E+00

2.00E+00 1 '-

volume

1 .50E+00 -*

, pore

1 .00E+00

dVIdD

5.00E-01 1

 

 

 

 

0.00E+00
1 2 Pore 312e, nm

Figure 2.6 BJH pore size distributions obtained from adsorption isotherms for
(A), MGM-41 (Si/Al = 40) synthesized from conventional precursors; (B)
MSU-Z (Si/AI = 40) made from zeolite fragments generated from USY with
OHTI' = 0.17 (trial 6); (C) MSU-Z (Si/AI = 40) assembled from
surfactant-mediated ZSM-5 hydrolysis products (zeolite fragments) formed
with OH'IT = 0.75 (trial 2); and (D), mesostructured aluminosilicate (Si/Al = 40)

formed through over-hydrolysis of ZSM-5 with OH'lT = 1.50 (trial 4).

6l

2000

 

1800 I

1600 ~

 

 

1400

1200 . l
1000 I l B

800

IICps)

600 -

400 -

200 .

 

 

0

 

5 15 25 29 35 45
Figure 2.7 X—ray diffraction patterns of MSU-Z materials (Si/AI = 40) prepared
from hydrolysis products of 2 g ZSM-5 in the presence of (A) 2.45 9 (trial 2)
and (B) 3.45 9 (trial 3) of CTAB. The diffraction peaks indexed by arrows are

attributed to the ZSM-5 residue.

62

 

Figure 2.8 TEM image of MSU-Z (Si/Al = 40) prepared from hydrolyzed ZSM-5

product (trial 2).

63

 

 

 

 

 

600
c 2’”.
500 . A
c

0.
'5
a 400 7
cog Q I I . ‘M
15 A B
E 300 7 A
O
8
a
g 200
_:_I e
3

100

0 -
0 0.2 10.4 0.6 0.8 1
hIckness, nm

Figure 2.9 T-plots for (A), MSU-Z (Si/Al = 40) prepared from the hydrolysis
product of ZSM-5 as precursor (trial 2); (B) a physical mixture of 5% ZSM-5
(Si/AI = 40) within conventional MGM-41 (Si/Al = 40).

64

12000 -M 2-..---

 

 

r

1 0000

8000 ~

6000

T

Ilcpsl

 

4000 ~

 

 

2000 .
X 10

 

 

 

0 5 10 15 20 25 30
28

Figure 2.10 X-ray diffraction pattern of MSU-Z (Si/Al = 40) prepared from

incomplete hydrolyzed ZSM-5 precursor (trial 1).

65

800

 

700

600 *

500 .

400

300

200 ~

 

Volume adsorbed, cm’lg STP

100 “

 

 

 

0 0.2 0.4 0.6 0.8 1
PIPo

Figure 2.11 N2 adsorption-desorption isotherms and BJH pore size
distributions for (A) wormhole MSU-Z (Si/Al = 40) prepared from the hydrolysis
product of USY zeolite in the presence of TTeA (trial 7), and (8) large pore
hexagonal MSU-Z (Si/Al = 40) prepared from the hydrolysis product of ZSM-5

in the presence of P123 surfactant (trial 5).

66

 

 

 

2500 ”..- - .- 2.-...- -_~_

 

 

 

2000 1
fl‘ 1500 ~
3 c
$5
1000 . . '
B
500 ~
A
0 y , , , '
5 15 25 35 45

20

Figure 2.12 X-ray diffraction patterns for the hydrolysis products of ZSM-5
zeolite (Si/AI = 40) in the presence of (A) 2.45 g CTAB (trial 2); (B) 2.0 g TTeA
(trial 9) (C) 2.4 9 P123 (trial 5).

67

2.6 References

1 Beck, J. S.; Vartuli, J. C.; Roth, W. J.; Leonowicz, M. E.; Kresge, C. T.;
Schmitt, K. D.; Chu, C. T. W.; Olson, D. H.; Sheppard, E. W.; et al. J. Am.
Chem. Soc. 1992, 114(27), 10834.

2 Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. 0.; Beck, J. S.
Nature, 1992, 359(6397), 710.

3 Martin, T.; Galameau, A.; Di Renzo, F.; Brunei, 0.; Fajula, F.; Heinisch, S.;
Cretier, G.; Rocca, J. Chem. Mater. 2004, 16(9), 1725.

4 Mehraban, Z.; Farzaneh, F. Micro. Meso. Mater. 2006, 88(1-3), 84.

5 Katiyar, A.; Ji, L.; Smirniotis, P. G.; Pinto, N. G. Micro. Meso. Mater. 2005,
80(1-3), 311.

6 McKimmy, E.; Dulebohn, J.; Shah, J.; Pinnavaia, T. J. Chem. Comm. 2005,
29, 3697.

7 Hatton, B. D.; Landskron, K.; Whitnall, W.; Perovic, D. D.; Ozin, G. A.
Adv. Funct. Mater. 2005, 15(5), 823.

8 Alabi, C. A.; Davis, M. E. Chem. Mater. 2006, 18(24), 5634.

9 Marschall, R.; Bannat, l.; Caro, J.; Wark, M. Micro. Meso. Mater. 2007,
99(1-2), 190.

10 Lai, C—Y.; Trewyn, B. G.; Jeftinija, D. M.; Jeftinija, K.; Xu, S.; Jeftinija, 8.;
Lin, V. S.-Y. J. Am. Chem. Soc. 2003, 125( 15), 4451.

11 Munoz, B.; Ramila, A.; Perez-Pariente, J.; Diaz, I.; Vallet-Regi, M. Chem.
Mater: 2003, 15(2), 500.

12 Slowing, l.; Trewyn, B. 6.; Lin, V. S.-Y. J. Am. Chem. Soc. 2006, 128(46),
14792.

13 Balas, F.; Manzano, M.; Horcajada, P.; Vallet-Regi, M. J. Am. Chem. Soc.
2006, 128(25), 8116.

14 Zhang, B. P.; Janicke, M. T.; Woodruff, W. H.; Bailey, J. A. J. Phys. Chem.
B 2005, 109(42), 19547.

15 Goradia, D.; Cooney, J.; Hodnett, B. K.; Magner, E. J. Mol. Catal. 8:
Enzymatic 2005, 32(5-6), 231.

16 De Stefanis, A.; Kaciulis, S.; Pandolfi, L. Micro. Meso. Mater. 2007, 99(1-2),
140.

17 Aguado, J.; Serrano, D. F.; San Miguel, G.; Escola, J. M.; Rodriguez, J. M.

68

J. Analy. Appl. Pyrolysis 2007, 78(1), 153.

18 Xu, B.; Hua, W.; Yue, Y.; Tang, Y.; Gao, Z. Catal. Left. 2005, 100(1-2), 95.
19 Liu, Y.; Kim, S. S.; Pinnavaia, T. J. J. Catal. 2004, 225(2), 381.

20 Kim, S.-S.; Liu, Y.; Pinnavaia, T. J. Micro. Meso. Mater. 2001, 44-45, 489.
21 Masakatsu Nomura, editor, Parviz M. Rahimi, editor, Omer Refa Koseoglu,
Heavy hydrocarbon resources : characterization, upgrading, and utilization,
Washington, DC : American Chemical Society : Distributed by Oxford
University Press, 2005.

22 Flanigen, E. M. Stud. Surf. Sci. Catal. 1991, 58, 13.

23 Corma, A. Chem. Rev. 1997, 97(6), 2373.

24 Meier, W. M.; Olsen, D. H. Atlas of Zeolite Structure Types, 1987,
Butterworths.

25 Stocker, M. Micro. Meso. Mater. 2005, 82, 257.
26 Davis, M. E. Nature 2002, 417(6891), 813.

27 Zhao, D.; Feng, J.; Huo, Q.; Melosh, N.; Frederickson, G. H.; Chmelka, B.
F .; Stucky, G. D. Science, 1998, 279(5350), 548.

28 Schmidt-Winkel, .; Lukens, W. W. Jr.; Zhao, D.; Yang, P.; Chmelka, B. F.;
Stucky, G. D. J. Am. Chem. Soc. 1999, 121(1), 254.

29 Liu, Y.; Zhang, W.; Pinnavaia, T. J. J. Am. Chem. Soc. 2000, 122(36),
8791.

30 Liu, Y.; Zhang, W.; Pinnavaia, T. J. Angew. Chem. Int. Ed. 2001, 40(7),
1255.

31 Liu, Y.; Pinnavaia, T. J. Chem. Mater. 2002, 14(1), 3.

32 Liu, Y.; Pinnavaia, T. J. J. Mater. Chem. 2004, 14(7), 1099.

33 Zhang, 2.; Han, Y.; Xiao, F.; Qiu, S.; Zhu, L.; Wang, R.; Yu, Y.; Zhang, Z.;
Zou, B.; Wang, Y.; Sun, H.; Zhao, D.; Wei, Y. J. Am. Chem. Soc. 2001,
123(21), 5014.

34 Zhang, Z. T.; Han, Y.; Zhu, L.; Wang, R. W.; Yu, Y.; Qiu, S. L.; Zhao, D. Y.;
Xiao, F. S. Angew. Chem. Int. Ed. 2001, 40, 1258.

35 Mokaya, R. Chem. Comm. 2001, 633.

36 On, D. T.; Kaliaguine, S. Angew. Chem. Int. Ed. 2001, 40, 3248.

69

37 On, D. T.; Kaliaguine, S. Angew. Chem. Int. Ed. 2002, 41(6), 1036.
38 On, D. T.; Kaliaguine, S. J. Am. Chem. Soc. 2003, 125(3), 618.

39 Groen, J. C.; Peffer, L. A. A.; Moulijn, J. A.; Perez-Ramirez, J. Chem. Eur.
J. 2005, 11(17), 4983.

40 lnagaki, S., Ogura, M., Inami, T., Sasaki, Y., Kikuchi, E., Matsukata, M.
Micro. Meso. Mater. 2004, 74, 163.

41 Kirschhock, C. E. A.; Ravishankar, R.; Verspeurt, F.; Grob, P. J.; Jacobs, P.
A.; Martens, J. A. J. Phys. Chem. B 1999, 103, 4965.

42 Karlsson, A.; Stocker, M.; Schmidt, R. Micro. Meso. Mater. 1999, 27, 181.

43 Goto, Y.; Fukushima, Y.; Ratu, F.; Imada, Y.; Kubota, Y.; Sugi, Y.; Ogura, M.;
Matsukara, M. J. Porous. Mater. 2002, 9,43.

44 Wang, S.; Dou, T.; Li, Y.; Zhang, Y.; Li, X.; Yan, Z. J. Solid State Chem.
2004, 177(12), 4800.

45 Szostak, R. Stud. Surf. Sci. Catal. 1991, 58, 153.

46 Groen, J. C.; Peffer, L. A. A.; Maulijn, J. A.; Perez-Ramirez, J. Chem. Eur.
J. 2005, 11,4983.

7O

Chapter 3
Organosilane-assisted Synthesis of Zeolite

Nana-crystals with High Crystallinity and Yield

3.1 Abstract

Nana-sized zeolites with controllable particle sizes were prepared from a
reaction mixture containing an organosilane as a crystal growth modifier. The
organosilanes not only mediate the crystal growth process, but also help to
stabilize the nano-sized zeolites by reducing Ostwald ripening. Compared to
other methods, this new approach allows for the use of conventional reaction
conditions, allowing products with higher yield and crystallinity to be achieved.
Two different pathways have been developed, which result in products with '
different properties. In general, the direct synthesis pathway, which combines
the organosilane and silica and alumina sources at the start of the synthesis,
produces irregular-shaped 30 ~ 150 nm particles in relatively low yields,
typically less than 20% with 8% organo-trialkyoxysilane (organo groups
include methyl, aminopropyl, octyl, and phenyl) as surface modifier.
Alternatively, the post-nucleation pathway, wherein the precursors are
nucleated prior to the addition of organosilane modifier, affords more uniform
particles, typically 70 to 100 nm in size. Moreover, the organo functionality of
the silane also plays a very important role that affects the morphology of the
final product. Genally, inert organo group helps to increase the yield and
crystallinity of final products. Nano—sized zeolite as small as 30 nm can be

synthesized in substantial yields of 20~80% through the direct synthesis

71

pathway using methyltriethoxysilane as the surface modifier. On the other
hand, yields greater than 80% and more uniform crystals can be achieved
when the post-nucleation pathway is adopted. Moreover, the particle sizes can
be more carefully controlled by regulating the amount of organosilane modifier
and the nucleation time prior to silane addition. TEM results show that the new
post-nucleation approach produces uniform and discrete nano-crystals. In the
compacted powder state, these nano-sized zeolites of ~ 100 nm pack densely
to form uniform inter-crystal mesopores of 22 ~ 60 nm, which are not provided
by other methods for generating zeolite nanoparticles. Due to the reduction in
crystal size, nano-sized zeolites show a much higher external surface area in
comparison to conventional zeolites. Compared to 65 mzlg for bulky ZSM-5,
external surface area over 200 m2/g was obtained with 4%
methyltriethoxyliane in the post-nucleation pathway. In return, the presence of
inter-crystal pores enhances the mass transportation. Cumene cracking results
show that nano—sized zeolites exhibit more than 10% higher activity compared
to conventional ZSM-5, perhaps due to improved access to both internal and

external acid sites.

72

3.2 Introduction

Zeolites are a family of crystalline aluminosilicates that have uniform

micropores that range from 0.3 nm to 1.2 nm in diameter. They are widely used

1-3 4, 5

in industry as catalysts, adsorbents, and ion-exchange reagents.°' 7
Normally, zeolites crystallize as micron size particles under hydrothermal
conditions from gel compositions. Recently, numerous studies have been
conducted to reduce the zeolite particle size from the micron scale to less than
100 nm.°'“ Nano-sized zeolite crystals hold promise for applications in sensor

1649 and hierarchical structure assembly.”22 For

technology,15 electronics,
instance, nano—sized ZSM-5 crystals have been coated on to polystyrene
microspheres that were modified to carry positive charge to interact with
negatively charged zeolite particles. After the removal of PS template, zeolitic
hollow spheres or macroporous zeolites were produced depending on the
experimental conditions. The presence of large pores (> 50 nm) allowed bulky
molecules such as proteins to be processed over these materials.23 Moreover,
a decrease in particle size results in an increase of external surface area and
pore volume. The resultant nano-sized zeolites exhibit significantly improved
catalytic activity.

Nano-sized zeolites normally are prepared from clear solutions.24 In order
to obtain such solutions, inorganic cations such as Na+ are substituted by
quaternary ammonium ions. The addition of quaternary ammonium facilitates
zeolite nucleation while the elimination of inorganic cations slows down the
crystal growth. The numerous nucleation sites in the super-saturated solution

then evolute into nano-crystals. In practice, the manipulation of kinetic

parameters is essential. Zeolite nanoparticles normally are prepared at

73

relatively low temperature, and low concentration, also the crystallization is
quenched after a short reaction period. The resultant zeolites are usually
recovered in very low yield and crystallinity through ultra-high centrifugations.
For instance, Bein et al. prepared LTA-25 and MFI-type26 zeolite nanocrystals
through kinetic control. The low yield caused the majority of the starting
materials, including the costly structure-directing agent to be wasted, inducing
environmental issues. Recently, the same group reported a multiple-step
synthetic strategy that allowed reusing the non-reacted mother liquid.27 The
yield can be boosted significantly.

Another strategy for the preparation of nano-sized zeolites is
nano-casting. Porous carbon materials were used as templates, which can be
removed later by calcinations. In 2003, Pinnavaia and co-workers28 described
a method using colloid-imprinted carbons (CIC) as templates for the
preparation of ZSM-5 nanoparticles. Uniform zeolite nano-crystals from 8 nm

I.29

to 90 nm can be synthesized. Later, Mokaya et a used the mesostructured

carbon CMK—3 as template to prepare zeolite nano-crystals within the

mesopores. In a more general method, Jacobsen” 3‘

reported the synthesis of
nano-sized zeolites in a confined space provided by the aggregation of
commercial carbon black nanoparticles. Typically, the nano-casting procedure
required the filling of the mesopores with a zeolite synthesis mixture. The
majority of solvent was then removed. Zeolite products were then crystallized
in a solid state to minimize the immigration of inorganic phases so that a
zeolite replica of the template can be prepared. However, due to the

irregularity of templates, irregular zeolite nanoparticles were formed in most

cases except for CIC templated ZSM-5. Moreover, although the fundamental

74

crystal sizes of the resulting nano-casted zeolite is very small, the aggregation
of the nanocrystals, as well as the packing of the aggregates, is non-uniform
and the resulting intercrystal pores are larger on average than the size of the
fundamental particles themselves.

Various inorganic nanoparticles with uniform size distributions have been
prepared using capping agents.”37 For instance; gold nanoparticles less than
5 nm in diameter can be synthesized through the reduction of AuCl3 solution in
the presence of thiol molecules as a capping agent.38 Due to the strong Au-S
bond, nanoparticles can be prepared and stabilized. In the zeolite synthesis,
organosilanes can serve as capping agents due to the stable Si-C bonds under
severe conditions required for zeolite synthesis. Organosilanes have been
found to modify the morphology of zeolite particles during crystallization.39
Various zeolite structures with high external surface area and pore volume
were synthesized. Such products showed much higher catalytic activity
compared to conventional zeolite catalysts. Moreover, organosilanes can also
bring desirable functionality to the resultant zeolite. Organo—functionalized
aluminosilicates have found many applications in catalysis,“°"‘1 heavy-metal
trapping,42 proton-conductive membranes,43 and drug delivery“- ‘5 In addition,
zeolites with organo-functionality also were used as building blocks for low
dielectric constant thin films.46 Compared to amorphous silica mesostructures,
crystalline zeolite thin films exhibited superior mechanical strength. And the
organo-functionality makes the thin films hydrophobic so that a low dielectric
constant can be achieved even at high humidity.

In present study, we describe the use of organosilanes to prepare

nano-sized ZSM-5 zeolites with controllable uniform particle size. This new

75

approach offers the possibility of synthesizing nano-sized zeolite products
using a conventional sol-gel process in the absence of a template. The
uniformity of the particles leads to uniform inter-crystal mesopores, which to
our best knowledge has not been observed previously. Two different pathways
have been investigated, which greatly affect the morphology and quality of the
final products. One pathway, termed the direct synthesis pathway, combines
the particle growth modifier, a silane of the type (RO)3Si-L where L is an
organo group, with the remaining reagents in a one-pot reaction mixture. In the
other pathway, termed the post-nucleation pathway, a zeolite precursor mixture

is nucleated prior to the addition of organosilane modifier.

76

3.3 Experimental section

3.3.1 Materials

Nanosized ZSM-5 zeolites were prepared from tetraethylorthosilicate
(T EOS, Aldrich), aluminum tri-sec-butoxide (Aldrich), and
tetrapropylammonium hydroxide solution (TPAOH, 1M solution, Aldrich), which
serves as a structure-directing agent for MFI zeolite. A series of
organosilanes (Si-L) were used as capping agents (or modifier), including
3-aminopropyltrimethoxysilane (APTMS, Aldrich), methyltriethoxysilane
(MTES, Gelest), phenyltrimethoxysilane (PTMS, Aldrich),
octyltrimethoxysilane (OTMS, Aldrich), and hexadecyltrimethoxysilane
(HDTMS, Aldrich). Two different methods were used to introduce the modifiers,

namely direct synthesis and post-nucleation pathway.

3.3.2 Synthesis of nano-sized ZSM-5 zeolite

3.3.2.1 Direct synthesis pathway

In the direct synthesis pathway, TEOS, aluminum tri-sec-butoxide, and a
desired amount of organosilane were mixed in a polypropylene bottle. Then
the TPAOH solution was added to the above solution with vigorous stirring to
form a clear reaction mixture with a molar composition of 1 AI2O2, : 100 SiO2 : x
Si-L : 37 TPAOH : 2000 H2O. In a typical synthesis, 0.10 g of Aluminum
isobutoxide (0.42 mmol) was dissolved in a mixture of 4.4 g of TEOS (21.2
mmol) and 0.188 g MTES (1.06 mmol). To the resultant solution, 8.0 g of
TPAOH (20 wt%) was added under stirring. After the completion of hydrolysis
at room temperature, the reaction mixture was heated to 100 °C for 48 hours to

form nanosized ZSM-5 zeolite. The final product was denoted as n-ZSM-5-DP.

77

The actual molar compositions are listed in Table 3.1.

3.3.2.2 Post-nucleation pathway

In the post-nucleation approach, silica and alumina precursors were
mixed first. To the resultant solution, TPAOH was added under stirring. After
the hydrolysis was completed, a milky solution was formed. The molar
composition of the synthesis mixture was: 1Al203: 100 SiO2: 37 TPAOH: 2000
H2O. The reaction mixture was then digested under 100 °C for 0 to 12 hours to
form protozeolitic nanoclusters or “zeolite seeds” in solution. To the above
seeds solution, a desired amount of organosilane was added under stirring.
After the hydrolysis was complete, the bottle was sealed again and heated to
100 °C for another 48 hours. The final molar composition was: 1Al203: 100
8102: y Si-L: 37 TPAOH: 2000 H2O. The product was denoted as n-ZSM-5-NP.
The molar composition was listed in Table 3.1. In a typical synthesis, 0.10 g of
Aluminum isobutoxide (0.42 mmol) was dissolved in 4.4 g of TEOS (21.2
mmol). To the resultant solution, 8.0 g of TPAOH (20 wt%) was added under
stirring. After the completion of hydrolysis at room temperature, the reaction
mixture was nucleated at 100 °C for 3 hours to form zeolitic nanoclusters. After
the mixture was cooled to room temperature, 0.377 g MTES (2.12 mmol) was
added. The resulting mixture was stirred at room temperature for 1 hour, and
then heated to100 °C for another 48 hours

All the nano—sized ZSM-5 products were recovered by repeated
centrifugation at 8,000 rpm followed by re-dispersion and washing with distilled
water until the pH of the colloidal suspension reached 8~9. Air-dried samples

were calcined at 600 °C for 4 hours to remove the organic modifier and

78

structure-directing agent.

3.3.3 Physical property characterizations

X-ray . diffraction patterns were recorded on a Rigaku Rotaflex
Diffractometer using CuKa radiation (A = 1.542 A). Scan consitions are
reflective mode from 5° to 50° 28 continuous at 2°/min and 0.05°lstep.

N2 adsorption and desorption isotherms at -196 °C were determined on a
Micromeritics ASAP 2000 sorptometer. The samples were degassed under
10'3 Torr at 250 °C for a minimum of 6 hours. The surface area was calculated
from the BET equation. Pore size distributions were derived from the
adsorption isotherms using the BJH model. The external surface area and
micropore volume were derived from t-plot. The total pore volume (mesopore +
micropore volume) was determined at P/Po of 0.95.

TEM images were taken on a Jeol 2200FS instrument operated at 200 kV.
Samples were prepared by sonicating the colloidal products in EtOH for 20 min,
and then evaporating 2 drops onto the carbon coated holey film supported on a
300 mesh Cu grid. SEM images were recorded on a JSM 6400 Electron
Microscope at an acceleration voltage of 20 W. The samples were

sputter-coated with 7 nm of gold or carbon prior to imaging.

3.3.4 Catalytic cumene cracking

Catalytic cumene cracking reactions were performed in a 6 mm id. fixed
bed quartz reactor with 200 mg catalyst at 250°C. Prior to catalytic testing, the
catalysts were ion-exchanged twice with 0.5 M NH4N03 solution followed by

calcination at 550°C for 4 hours. The cumene flow rate was 4.1 umoI/min in a

79

 

20 cm3/min carrier stream of N2. Cumene conversions were plotted against
time on stream after a 6-hour activation period at 250 °C in a pure N2

atmosphere.

80

3.4 Results and Discussions

3.4.1 Direct synthesis pathway

In comparison to TEOS, a (RO)3SI-L molecule has 3 hydrolysable Si-O
bonds and one Si-C bond, which is stable under hydrolysis conditions. The
inertness of the Si-C linkage toward hydrolysis has been demonstrated
through the synthesis of organic zeolites and organo-functionalized zeolites.
During zeolite syntheses, once an organosilane is anchored to the surface of
the crystal, crystal growth at that specific site is hindered. Under ideal
circumstances, the crystal becomes covered by the organic group and stops
growing, as illustrated in Figure 3.1. The organosilane also stabilizes the
nano-crystals by reducing Ostwald Ripening. Moreover, particle inter-growth is
alleviated. Therefore, nano-sized zeolite particles could possibly be
synthesized under more severe conditions, such as elevated temperature and

higher concentration.

Crystal growth Growth quenched

- RO S‘L
Si—OHm ._ Si——0I-i—(—l¥I .

     

Figure 3.1 Illustration for the quenching of zeolite crystal growth using an

organosilane as a capping agent.

Figure 3.2 provides SEM images of n-ZSM-S-DP samples synthesized in
the presence of methyltriethoxysilane (MTES) as the modifier through the
direct synthesis pathway. Conventional ZSM-5, prepared in the absence of

MTES, adopts a rectangular crystal form with smooth surfaces and sharp edge

81

(Figure 3.2A), which is typical for ZSM-5 zeolite. The average size of zeolite
crystals is about 1 pm. When 5 mol% MTES was added to the reaction mixture
as a surface modifier, a dramatic drop in fundamental crystal size down to 150
~ 200 nm is observed (Figure 3.2B). Some extra features, such as the dark
lines across individual particle surface were observed, may indicate the
presence of multiple domains within each crystal. A ZSM-5 zeolite with
average particle size of 30 ~ 40 nm (Figure 3.20) was obtained when 8 mol%
MTES was added to the reaction mixture. It is evident that increasing the
amount of organosilane decreases the fundamental particle size. However,
there is a limit to how much organosilane can be added. When 10 mol% MTES
was added, a clear solution was obtained after 2 days of reaction at 100 °C.
The solid residue of the solution showed no crystalline phase.

Figure 3.3 provides the high-resolution TEM images of ZSM-5
nanocrystals formed by direct synthesis in the presence of MTES. The
fundamental crystal domain sizes obtained with 5% and 8% MTES are 150
and 30 nm on average, respectively, which are consistent with the SEM
observations. Lattice fringes are observed for both samples, indicative of the
presence of a crystalline phase. The fringes for the n-ZSM-5-DP nanoparticle
prepared with 5% MTES are aligned across the entire particle aggregate,
suggesting that the aggregate is a single crystal made up of intergrown domain.
Moreover, within this 150 nm crystal aggregate there are a few voids, which
are revealed by the bright contrast in the TEM image. Such voids might also
contribute to the dark features observed through SEM (Figure 3.28). Due to
the presence of intra-crystal voids, the actual domain size of this ZSM-5—NP

sample would be well below 100 nm.

82

N2 adsorption - desorption isotherm was used to measure the physical
properties of the resultant nano—sized ZSM-5 products prepared by direct
synthesis. The results are provided in Figure 3.4. A ZSM-5 zeolite sample
synthesized through the conventional method was also included for
comparison. All the ZSM-5 samples showed a steep nitrogen uptake at P/Po <
0.02, which is characteristic of a microporous zeolite. For the conventional
ZSM-5, the adsorption curve levels out at high relative pressure. This typical
type II isotherm behavior indicated the absence of mesoporosity and a lack of
an appreciable external surface area for the sample. In comparison, the ZSM-5
products synthesized in the presence of MTES modifier showed
distinguishable hysteresis loops above P/Po 0.8, indicative of the presence of
mesopores. Interestingly, the position of hysteresis loop moved to higher P/Po
when more MTES was added to the reaction mixture, indicative of larger
average pore diameter (Figure 3.4, curve C). These hysteresis loops can be
assigned to the textural pores arising from the aggregation of nano-sized
crystals. Sample B shows an average inter-crystal pore size of 30 nm.
Although smaller inter-particle pore size is expected from a smaller particle
size. However, sample C exhibits an average pore size of 55 nm, which is
larger than the fundamental particle size itself (Figure 3.4, inset). The larger
pore size is probably due to the loose packing of irregular nanoparticles.

Figure 3.5 provides the XRD patterns of ZSM-5 samples prepared by the
direct synthesis pathway under different conditions. All the ZSM-5-NP samples
showed diffraction peaks characteristic of an MFI framework (Curve A),
thereby confirming the presence of zeolitic phases. Moreover, it is also evident

that the crystallinity of the resultant ZSM-5 samples is compromised as more

83

MTES is added to the synthesis gel, whereas the external surface area and
mesopore volume increased substantially (Table 3.2). The decrease of
diffraction intensity was proportional to the amount of MTES added.

Without the aid of templates, zeolite nanoparticles are usually prepared
by a kinetic control approach. Low yields of product are inevitable under these
reaction conditions that include low temperatures and concentrations. In the
direct synthesis method reported here, the quenching of crystal growth was
achieved by the surface silylation of the developing nano—particles, which also
minimized Ostwald ripening. Therefore, normal reaction conditions can be
utilized, which provide for higher yields of zeolites while retaining a
nano-crystal morphology. For instance, with the addition of 8 mol% MTES,
ZSM-5 with a 30 nm average particle size can be prepared at a yield of 18%.

In addition to MTES, other types of organosilanes can also be used as
the surface modifier, which also offers the opportunity to incorporating
desirable organic functionality into the zeolite nanoparticles. Figure 3.6
provides the isotherm of ZSM-5 prepared with 6 mol% APTMS as a surface
modifier. A hysteresis loop at partial pressure of 0.9 was observed, suggesting
the formation of nanoparticles. It is noted that at a silane level of 8 mol%,
MTES still generated crystalline product with a reasonable yield, while APTMS
failed to produce any crystals. Optical images of reaction products are
provided in Figure 3.7. Colloidal ZSM-5 nanoparticles were obtained when 6
mol% APTMS was added to the reaction mixture. The colloidal suspension
was stable even after several weeks, which suggests that the aggregation of
nanoparticles was hindered. At higher APTMS contents, only clear solutions

were obtained. The TEM image of the product obtained at 8 mol% APTMS

84

showed the presence of nanoparticles about 10 nm in diameter (Figure 3.8),
but no indication of crystallinity. The reason could be attributed to the strong
interaction between amino-group and silicon, which interrupted the
crystallization of zeolites. Interestingly, hydrophobic organo-groups such as
phenyl and octyl tend to boost the yield of nano-sized zeolites to up to 80%. It
is probably because hydrophobic particles aggregate in the aqueous
environment to form larger particles, therefore easier to collect by
centrifugation.

Figure 3.9 illustrates the possible mechanism for the direct synthesis
pathway. Because MTES molecules are present at the beginning, the
protozeolitic nanoclusters initially formed have a surface partially covered by
the organo group. If enough organosilane molecules are available during
crystallization, zeolite nanocrystals completely covered by organo group will be
formed as shown in Figure 3.9 (Path A). Otherwise, these partially covered
nanoclusters will merge together to form a larger particle (Path B). Under this
circumstance, the presence of organo group prevents the formation of a
homogeneous phase. During nanoparticle coalescence, the hydrophobic
organo groups tend to cluster on the surface to lower the surface energy. A
crystal aggregate with voids will be generated (Path B).

It should be pointed out that the direct synthesis pathway usually
produces nano-sized ZSM-5 products with relatively low yields compared to
the conventional zeolite synthesis. A significant amount of raw materials
including costly structure-directing agent TPAOH was wasted. This not only
creates environmental issues, but also makes it difficult to predict the

compositions of the final products. The reason most likely is related to the

85

interference of zeolite nucleation by the organosilane. Therefore, a more
desired pathway was devel0ped to boost the yield of nano-sized ZSM-S,

denoted the post-nucleation pathway, as discussed below.

3.4.2 Post-nucleation pathway

In this new pathway, a composition for the conventional ZSM-5 synthesis
was nucleated at the elevated temperature before the addition of a desired
amount of organosilane to control crystal growth. At this point, no zeolite phase
is present as confirmed by the X-ray diffraction pattern of the evaporated
reaction mixture. Because the organosilane molecules are absent during the
nucleation stage, zeolitic nanoclusters are produced without interference.
Such protozeolitic seeds have been used as precurors to assemble
mesostructured aluminosilicates with strong acidity and high hydrothermal
stability.“'51 The presence of zeolitic subunits has been confirmed by the IR
absorbance band at 550 cm", which is characteristic of five-member Si-O ring
building blocks present in MFI and BEA zeolites. After the addition of an
organosilane modifier, the preformed seeds facilitated the zeolite crystallization.
Therefore, nanocrystals with higher crystallinity and yield can be produced, as
judged by the Bragg peak intensity in the XRD patterns (Figure 3.10) and by
the micropore volumes measured by N2 isotherms.

In the post-nucleation pathway, a new parameter other than the
percentage of organosilane is introduced, which is the nucleation time for the
formation of zeolitic nanoclusters. Figure 3.11 provides the N2 isotherm plots of
n-ZSM-S-NP samples synthesized by the addition of 10 mol% MTES after

different nucleation durations. In the direct synthesis pathway, the addition of

86

10 mol% MTES at the start of reaction doesn’t give any solid product. However,
after a nucleation for 24 hours at room temperature, a ZSM-5 product is
obtained with slightly reduced crystallinity as suggested by the weak uptake of
nitrogen in the nitrogen adsorption isotherm (Figure 3.11, curve F). When the
nucleation is conducted at elevated temperature (100°C), high quality ZSM-5
products can be synthesized in more than 40% yield from 10 mol% MTES
even after short nucleation step at this temperature (1 h, Figure 3.11, curve E).
Compared to ZSM-5 synthesized under conventional condition, all these
samples show a hysteresis loop at P/Po > 0.8, indicative of high inter-particle
porosity and small particle size.

Figure 3.12 illustrates the effect of organosilane content on the
morphology of the final products. The evolution of the isotherms is very similar
to the samples made by direct synthesis. However, a 3-hour nucleation step
before the addition of the silane surface modifier greatly improves the
crystallinity and yield of the final products, as judged by XRD peak intensity
and micropore volume derived from N2 adsorption-desorption isotherms.
Whereas for direct synthesis pathway, a 10 mol% MT ES gives no solid product,
a 3-hour nucleation step produces a nano—sized ZSM-5 zeolite with high
crystallinity. Such high crystallinity is compromised as more organosilane was
added (Figure 3.12, curve E). In this case, smaller particles are produced,
which might account for the decrease in crystallinity. Figure 3.13 provides the
BJH pore size distributions of representative samples. We can see that at a
fixed MTES content (10 mol%), extending the nucleation time from 3 hours to
12 hours increases the pore size from 22 nm to 50 nm. When the same

nucleated precursor is used, reducing MT ES amount to 4 mol% increases the

87

pore size to 35 nm. The variation in the inter-particle pore size reflects the
change of the primary particle size.

Direct evidence for the formation of nanoparticles is obtained by TEM
observation (Figure 3.14). The average fundamental particle size of
n-ZSM-5-NP prepared by the addition of 4 mol% MTES after a 3-hour
nucleation is under 100 nm (Figure 3.14A). Moreover, the particle size is very
uniform. The high-resolution image shows zeolite fringes covering the whole
particle, indicative of a single crystal with high crystallinity. Particles with
average size of 60 ~ 70 nm are obtained when 15 mol% MT ES was used
(Figure 3.14B). It should be pointed out that the post-nucleation pathway not
only produced high-crystallinity zeolites, it also greatly boosted the yields of
products under the same condition. For example, a synthesis composition
nucleated for 3 hours at 100 “C gives ZSM-5 nano-crystal with over 80% yield
when less than 15 mol% MTES was added after the nucleation step.

The use of organosilane modifier during zeolite crystallization produced
nano-sized zeolites. Based on TEM observations, the final products adopted a
spherical morphology. These discrete spheres have an exceptionally uniform
particle size distribution. During aggregation, uniform inter-crystal mesopored
are formed, which have not been observed from nano-sized zeolite samples
prepared from other methods. Moreover, an average mesopore size as small
as 22 nm can be generated. The results further confirm the presence of
uniform nano-crystalline zeolite particles that are packed orderly. When
spheres with diameter of D are closest-packed as shown in Figure 3.15, two
types of inter-particle voids emerged, namely tetrahedral and octahedral pores.

The diameter (D...) of resultant inter-particle voids is a function of D as shown in

88

the following relationship,

 

 

 

 

Tetrahedral Octahedral

Figure 3.15 Nanoparticle packing schemes. Both tetrahedral and octahedral

interparticle voids (pores) are expected for extensively compacted spheres.

0,, = 0.2250 (for tetrahedral pores) or 0.414D (for octahedral pores)

Statistically, the ratio of tetrahedral to octahedral pores is 2 : 1 for extensively
compacted spheres. However, the isotherm data showed that the mesopore
size distribution is uniformly centered around an average value. For instance,
mesopores with an average pore size of 35 nm was detected on ZSM-5
prepared with 4 mol% MT ES after a 3-hour nucleation step. TEM image
showed that this sample has an average particle size of 100 nm, which
theoretically will result in uniform tetrahedral and octahedral inter-crystal
mesopores of 22 and 42 nm, respectively. The observed pore size distribution
results showed that the mathematical model gives a fair prediction. The results
suggested that the zeolite spheres were packed relatively compacted. It is

noteworthy that nano—sized ZSM-5 zeolite prepared by the direct synthesis

89

pathway exhibited much larger mesopores, which are larger than the diameter
of fundamental particles sometimes. For example, 55 nm mesopores were
observed for n-ZSM-S-DP prepared from 8 mol% MT ES in a direct synthesis
reaction.

Figure 3.16 illustrates the possible mechanism for the post-nucleation
pathway. Because organosilane molecules are eliminated from the nucleation
step, the resultant zeolitic nanoclusters are able to aggregate more densely to
form larger particles, which subsequently are covered by organo groups upon
the addition of the silane. A longer nucleation step produces larger
nanoclusters, resulting in larger zeolite particles. Meanwhile, because the
nucleation process is not interrupted by the organosilane, products with higher
crystallinity and yield are observed. Two factors determine the final products.
First, the conditions for nucleation step determine the number and size of
zeolitic nanoclusters or seeds. On the other hand, after nucleation step, the
number and size of larger zeolite particles formed are dependent on the
amount of organosilane modifier added. Therefore, the zeolite particle size is
controlled by a combination of both the nucleation step and the amount of
silane modifier added after nucleation.

Table 3.2 summarized the physical properties of ZSM-5 samples
prepared by both the direct synthesis and post-nucleation pathway under
various conditions. Compared to the conventional ZSM-5, which has large
particle size of 1 pm and external surface of 50 mzlg, nano-sized ZSM-5
materials prepared from both direct synthesis and post-nucleation pathways
exhibited a much higher external surface area. For instance, an external

surface area of 211 m2/g was observed for the 100 nm ZSM-5 product

 

prepared with 4 mol% MTES after a 3-hour nucleation step. Based on
micropore volumes, only little crystallinity was lost while a 4-fold increase on
external surface area was achieved. Moreover, all the nano-sized ZSM-5
samples showed enhanced mesoporosity up to 0.56 cm3/g with uniform pore
size distribution. Such a high mesopore volume should greatly improve the
mass transportation rate by facilitating the diffusion of molecules inside the
sample. In addition, the large external surface area will significantly increase
the number of active sites that are easily accessible to reactant molecules,
especially relatively large molecules such as cumene.

Figure 3.17 provides the cumene conversions over a nano-sized ZSM-5
catalyst prepared with 5 mol% MT ES in a direct synthesis pathway. A
conventional ZSM-5 catalyst was included for comparison. Both catalysts
showed relatively low cumene conversions initially, which increased with the
time-on-stream. This is probably due to the slow diffusion rate of cumene
molecules in the small-pore MFI zeolites. Because of the presence of large
external surface area and high mesoporosity, the n-ZSM-5-DP catalyst
exhibited a somewhat higher cracking activity despite a slightly lower

crystallinity.

9i

3.5 Conclusion and perspective

Two different pathways have been developed using organosilanes as a
surface modifier (capping agent) for the synthesis of zeolite nanocrystals. The
direct synthesis pathway, wherein the silane is added as a reagent at the initial
stages of synthesis, affords smaller crystals compared to the post-nucleation
pathway, wherein the silane is added after zeolite nucleation has been initiated.
For instance, an average particle size as small as 30 nm can be achieved with
8% MT ES as surface modifier in the direct synthesis pathway. However, due to
the irregularity and loose packing of nano-particles, the resultant inter-crystal
mesopores are quite large (even exceeding the size of fundamental particles).
Moreover, the presence of organosilane considerably affects the zeolite
nucleation process; therefore the direct synthesis pathway generally affords
nano-crystals at a yield less than 80%. On the other hand, the
organosilane-free nucleation step in the post-nucleation pathway significantly
improves the yield and crystallinity of the zeolite while the particle size of the
final product is retained at the nanometer scale. More importantly, the zeolitic
nanocluster precursors formed in the pro-nucleation step results in products
with a uniform particle size distribution (70 ~ 100 nm). Because the external
surface is covered by organo groups, the intergrowth between particles is
greatly reduced, which in turn results in discreet zeolite nanocrystals. These
well separated and uniform nano-particles can be used to assemble uniform
inter-crystal mesopores. In addition, the post-nucleation pathway allows the
use of silanes containing reactive organic groups, such as amino groups,
without sacrificing the crystallinity and the yield of zeolite nanocrystals.

Therefore, the post-nucleation pathway provides the opportunity to introduce

92

additional desirable functionalities to the final products. In any case, both
pathways produce zeolite nano—crystals with much improved inter-particle
mesoporosity and external surface area, which in turn boosts the catalytic
activity by reducing the diffusion path length.

Compared to Serrano’s method, which produces aggregates of
nanocrystalline zeolites, our post-nucleation approach produces ~ 100 nm
ZSM-5 crystals with uniform and controllable particle size. Moreover, the
resultant zeolite nanoparticles are well separated. Therefore, uniform
inter-crystal mesopores in the range 22 to 55 nm can be assembled through
close-packing of nanocrystals. It is clear that the nucleation conditions are very
critical in determining the textural properties of the final product. The uniformity
of nanocrystals is the result of uniform nucleation of protozeolitic nano-clusters
formed in the nucleation step. Therefore, future work should be focusing on the
optimizing of nucleation process. In the ZSM-5 synthesis, homogeneous
nucleation can be realized by substituting mineral ions such as Na“ with
quartemary ammonium ion TPA“, which is the structure-directing agent for MFI
structure. By increasing the TPA/Si ratio in the nucleation step, a precursor
containing smaller and more uniform zeolitic nanoclusters can be prepared,

which in turn produces smaller and uniformer zeolite nanocrystals.

93

Table 3.1 Yields of ZSM-5 (Si/Al = 50) obtained in the presence of various

organosilane surface modifiers.

 

 

Sample No. Nucleation, h Organosilane Si-L, mol% Weld, %

ZSM-5 N/A N/A 0 >90
1 N/A MT ES 2 >80
2 N/A MT ES 5 80
3 N/A MT ES 8 18
4 N/A APTMS 2 >80
5 N/A APTMS 4 >60
6 N/A APTMS 6 25
7 N/A OTMS 6 83
8 N/A PT MS 6 72
9 N/A HDTMS 6 >80
10 RT, 24 hours MT ES 10 <20
11 100°C, 1 h MTES 10 40
12 100°C,3h MTES 10 >80
13 100°C,6h MTES 10 >80
14 100°C, 12 h MTES 10 >80
15 100°C, 3 h MTES 4 >80
16 100°C, 3 h MT ES 7 >80
17 100°C,3h MTES 15 35
18 100°C, 3 h APTMS 10 >50
19 100°C, 6 h APTMS 10 >50
20 100°C, 12 h APTMS 10 >50
21 100°C, 3 h APTMS 4 >50
22 100°C, 3 h APTMS 7 >50

 

Samples 1-9 were prepared by the direct synthesis pathway wherein the silane
modifier was added at the start of the reaction. Samples 10-22 were prepared
by the post-nucleation pathway wherein the silane modifier was added to the

reaction mixture after a nucleation step.

94

Table 3.2 Physical properties of ZSM-5 samples (Si/Al = 50).

 

 

ZSM-5 389 0.04 65 0.14
1 423 0.29 138 0.13
2 389 0.35 215 0.08
3 249 0.65 162 0.04
4 540 0.26 143 0.18
5 477 0.27 123 0.16
6 549 0.30 128 0.19
7 506 0.27 177 0.15
8 497 0.45 171 0.15
9 504 0.48 177 0.15
10 179 0.55 104 0.03
11 407 0.47 144 0.12
12 414 0.55 169 0.11
13 443 0.52 155 0.13
14 433 0.49 162 0.12
15 458 0.51 212 0.11
16 466 0.46 144 0.15
17 221 0.55 146 0.04
18 156 0.40 75 0.04
19 414 0.53 180 0.11
20 451 0.56 167 0.13
21 481 0.50 152 0.15
22 456 0.52 140 0.14

 

Samples 1-9 were prepared by the direct synthesis pathway wherein the silane
modifier was added at the start of the reaction. Samples 10-22 were prepared
by the post-nucleation pathway wherein the silane modifier was added to the

reaction mixture after a nucleation step.

95

—
Sllll nm

 

Figure 3.2 SEM images for ZSM-5 samples (Si/Al = 50) prepared through the
direct synthesis pathway in the absence and presence of methyltriethoxysilane
(MTES): (A), 0% MTES; (B), 5% MTES and (C), 8% MTES at 100°C for 48

hours.

 

Figure 3.3 TEM images for ZSM-5 samples (Si/Al = 50) prepared through the
direct synthesis pathway in the presence of (A), 5% MTES; (B), 8% MT ES at

100°C for 48 hours.

97

 

 

é

 

dVIdD

 

3

 

 

i r v v m
40 60 M 1G3 120 /‘

Pore size, nm

 

“.23

 

Volume adsorbed, cclg STP
N
8
8

 

 

0.2 0.4 0.6 0.8 1
P/Po
Figure 3.4 Nitrogen adsorption-desorption isotherms of ZSM-5 samples (Si/AI
= 50) prepared through the direct synthesis pathway in the presence of A, 0%
MT ES; 8, 5% MT ES and C, 8% MTES at 100°C for 48 hours. Curves B and C

are offset by 50 and 150 volume units for clarity. Inset: BJH adsorption pore

size distributions.

98

 

 

 

 

 

 

25000
20000 —
15000“
’g A
é’v
10000"
B
#w-r‘v‘ - - - ”Nara-—
5000 — i c
D
O _Mhfir 1

 

5 15 25 35 45

2 0
Figure 3.5 X-ray diffraction patterns for calcined ZSM-5 (Si/AI = 50) samples

prepared by (A) a conventional method in absence of MT ES, and by direct
synthesis in the presence of (8)2 mol% MTES, (C) 5 mol% MTES, and (D) 8
mol% MTES at 100 °C for 48 hours. Curves are offset by 4,000 counts each for

clarity. Each sample was calcined at 600 °C for 4 hours.

99

500

 

450 ~

A
8
\

03

01

O
l

 

0)

O

O
i

N
0'1
0

8

 

Volume adsorbed, cm3/g STP

—L

01

O
I

8

so 2%

 

ii
I I I I r

0 0.2 0.4 0.6 0.8 1
P/Po
Figure 3.6 N2 adsorption-desorption isotherms for (A) conventional ZSM-5

(Si/AI = 50) in absence of organosilane and (B) n-ZSM-5-DP (Si/AI = 50)
prepared by the direct synthesis pathway in the presence of 6 mol%
3-aminopropyltrimethoxysilane (APTMS) at 100 °C for 48 hours. Curve B is

offset by 150 volume units for clarity.

100

 

Figure 3.7 Optical images for the reaction products .fonned by direct synthesis

in the presence of (A) 6 mol% APTMS, (B) 8 mol% APTMS, and (C) 10 mol%

APTMS after reaction at 100 °C for 48 hours.

101

 

 

Figure 3.8 TEM image for the amorphous product prepared by direct synthesis

in the presence of 8 mol% APTMS at 100 °C for 48 hours.

Molecular

precursors orQanosilane

Nucleation

LL L
L L .1“: LL
L‘L L

# gameiizauon\
it 1‘4:

Figure 3.9 Illustration for the formation of nano-crystals by a direct synthesis

     

pathway in the presence of (RO)3Si-L surface modifier. The formation of single
nanocrystals (path A) or aggregates of intergrown nanoparticles (path B)

depends on the amount of (RO)3$i-L present in the reaction mixture.

103

 

1 6000

14000 7

12000 r

10000 F

 

 

8000 -

l(cps)
Li.
0

6000 ~

 

 

4000 ~

2000 -

W A

0 1 .2
5 1 0 1 5 20 25 30 35 40 45 50
26

Figure 3.10 X-ray powder diffraction patterns for calcined ZSM-5 samples

 

 

 

 

 

 

(Si/Al = 50) synthesized by the post-nucleation pathway and the addition of (A),
4% MTES after 3 hours nucleation step; (B), 7% MT ES after 3 hours
nucleation step; (C), 10% MTES after 12 hours nucleation step at 100 °C.

Each curve is offset by 4000 units for clarity.

104

 

1200 -

 

§

 

Volume adsorbed, cclg STE,
8 8

8

 

 

0 I I I T
0 0.2 0.4 0.6 0.8 1

PlPo
Figure 3.11 N2 adsorption/desorption isotherms for calcined ZSM-5 samples

(Si/Al = 50) synthesized by (A), conventional method and by the
post-nucleation pathway and the addition of 10% MT ES after (B) 12 hours
nucleation step; (C), 6 hours nucleation step; (D), 3 hours nucleation step; (E),
1 hours nucleation step at 100 °C; (F), 24 hours nucleation step at room

temperature. Each isotherm is offset by 200 volume units for clarity.

105

 

1200 ~

 

 

1000 — ,
O.

l-

(I)

€800

§ 600 C
G

0

E

2

O

> 400 B

200

 

O I I I I

o 0.2 0.4 0.6 0.8 1
PlPo

Figure 3.12 N2 adsorption/desorption isotherms for calcined ZSM-5 samples

 

(Si/Al = 50) synthesized by (A), conventional method; and by the
post-nucleation pathway and by the addition of (B), 4% MTES; (C), 7% MT ES;
(D), 10% MTES; (E), 15% MT ES after a 3-hour nucleation step at 100 °C.

Each isotherm is offset by 200 volume units for clarity.

 

0.07

.....I

0.05 -

 

0.024'

0.01 ~

 

4)

 

 

 

0 20 40 60 80 100
Pore diameter, nm

Figure 3.13 BJH (adsorption) pore size distribution of n-ZSM-5-NP samples
(Si/Al = 50) synthesized by the post-nucleation pathway and the addition of (A),
10% MTES after 3 hours nucleation step; (8), 4% MT ES after 3 hours

nucleation step; (C), 10% MTES after 12 hours nucleation step at 100 °C.

107

 

 

 

Figure 3.14 TEM images of n-ZSM-5-NP (Si/Al = 50) synthesized by the
post-nucleation pathway by the addition of (A), 4% MT ES after 3 hours
nucleation step; (B), 15% MTES after 3 hours nucleation step. Insets are the

corresponding high-resolution images.

Molecular
precursors

l Nucleation

.O
.00

+ Organosilane I Crystallization

 

Figure 3.16 Illustration for the formation of nano-crystals in a post-nucleation

pathway.

 

Cumene conversion, %
8 8 8‘

8

 

 

0 I 1 .
0 100 200 300 400

 

Time on Stream, min

Figure 3.17 Catalytic cumene cracking conversion vs. time for (A), a
commercial ZSM-5 from Zeolyst (Si/Al = 40); (B), n-ZSM-5-DP (Si/AI = 50)

prepared from 5 mol% MTES by a direct synthesis pathway.

[10

3.6 References

1 Flanigen, E. M. Stud. Surf. Sci. Catal. 1991, 58, 13.
2 Carma, A. Chem. Rev. 1997, 97(6), 2373.
3 Stoecker, M. Micro. Mesa. Mater. 2005, 82(3), 257.

4 Choi, J. H.; Kim, S. D.; Noh, S. H.; Oh, S. J.; Kim, W. J. Micro. Mesa. Mater:
2006, 87(3), 163.

5 Erdem, E.; Karapinar, N.; Donat, R. J. Coll. Inter. Sci. 2004, 280(2), 309.
6 Jorgensen, T. C.; Weatherley, L. R. Water Research 2003, 37(8), 1723.

7 Caputo, D.; De Gennaro, 8.; Pansini, M.; Colella, C. Stud. Surf. Sci. Catal.
1999, 125,723.

8 Holmberg, 8. A.; Wang, H.; Norbeck, J. M.; Yan, Y. Micro. Mesa. Mater:
2003, 59, 13.

9 Valtchev, V. F.; Bozhilov, K. N. J. Phys. Chem. B 2004, 108, 15587.

10 Zhan, 8.; White, M. A.; Lumsden, M.; Mueller-Neuhaus, J.; Robertson, K.
N.; Cameron, S. T.; Gharghouri, M. Chem. Mater. 2002, 14, 3636.

11 Li, 0.; Creaser, D.; Sterte, J. Chem. Mater. 2002, 14, 1319.

12 Majano, G; Mintova, S.; Ovsitser, 0.; Mihailova, 8.; Bein, T. Micro. Mesa.
Mater. 2005, 80, 227.

13 Aguado, J.; Serrano, D. P.; Escola, J. M.; Rodriguez, J. M. Micro. Mesa.
Mater. 2004, 75, 41.

14 Song, W.; Justice, R. E. ; Jones, C. A.; Grassian, V. H.; Larsen, S. C.
Langmuir 2004, 20, 8301.

15 Wang, 2.; Larsson, M. L.; Grahn, M.; Holmgren, A.; Hedlund, J. Chem.
Commun. 2004, 2888.

16 Wang, 2.; Wang, H.; Mitra, A.; Huang, L.; Yan, Y. Adv. Mater. 2001,
13, 746.

17 Wang, Z.; Mitra, A.; Wang, H.; Huang, L.; Yan, Y. Adv. Mater. 2001,
13, 1463.

18 Li, S.; Li, 2.; Yan, Y. Adv. Mater. 2003, 15, 1528.

19 Li, Z.; Li, S.; Luo, H.; Yan, Y. Adv. Funct. Mater. 2004, 14, 1019.

III

20 Wang, X. D.; Yang, W. L.; Tang, Y.; Wang, Y. J.; Fu, S. K.; Gao, 2.
Chem. Commun. 2000, 2161.

21 Holland, 8. T.; Abrams, L.; Stein, A. J. Am. Chem. Soc. 1999, 121, 4308.
22 Valtchev, V.; Mintova, S. Micro. Mesa. Mater. 2001, 43, 41.

23 Wang, Y.; Caruso, F. AaV. Funct. Mater. 2004, 14, 1012.

24 Tosheva, L.; Valtchev, V. P. Chem. Mater. 2005, 17, 2494-2513

25 Mintova, S.; Ma, S.; Bein, T. Chem. Mater. 2001, 13,901.

26 Mintova, S.; Schoeman, 8. J.; Valtchev, V.; Sterte, J.; Ma, S.; Bein, T. Adv.
Mater. 1997, 9, 585.

27 Larlus, O.; Mintova, S.; Bein, T. Micro. Mesa. Mater. 2006, 96(1-3), 405.
28 Kim, S-S.; Shah, J.; Pinnavaia, T. J. Chem. Mater. 2003, 15(8), 1664.

29 Yang, 2.; Xia, Y.; Mokaya, R. Adv. Mater. 2004, 16(8), 727.

30 Schmidt, l.; Madsen, C.; Jacobsen, C. J. H. Inorg. Chem. 2000, 39, 2279.
31 Madsen, C.; Jacobsen, C. J. H. Chem. Commun. 1999.673.

32 Jagannathan, R.; Poddar, P.; Prabhune, A. J. Phys. Chem. C 2007,
111(19), 6933.

33 Matt, 0.; Galkowski, J.; Wang, L.; Luo, J.; Zhang, C. Langmuir 2007,
23(10), 5740.

34 Bakshi, M.; Kaur, G; Thakur, P.; Banipal, T.; Possmayer, F.; Petersen, N. O.
J. Phys. Chem. 02007, 111 (16), 5932.

35 Bakshi, M.; Possmayer, F.; Petersen, N. 0. Chem. Mater: 2007, 19(6),
1257.

36 Niesz, K.; Grass, M.; Somorjai, G A. Nana Left. 2005, 5(11), 2238.

37 Perez, H.; Pradeau, J.-P.; Albouy, P.-A.; Perez-Omil, J. Chem. Mater: 1999,
11( 12), 3460.

38 Pengo, P.; Polizzi, S.; Battagliarin, M.; Pasquato, L.; Scrimin, P. J. Mater.
Chem. 2003, 1 3( 10), 2471.

39 Serrano, D. P.; Aguado, J.; Escola, J. M.; Rodriguez, J. M.; Peral, A. Chem.
Mater. 2006, 18, 2462.

40 Mbaraka, Isa K.; Shanks, Brent H. J. Catal. 2006, 244(1), 78.

112

41 Chen, H.; Huh, 8.; Wiench, J. W.; Pruski, M.; Lin, V. S.-Y. J. Am. Chem.
Soc. 2005, 127(38), 13305.

42 McKimmy, E.; Dulebohn, J.; Shah, J.; Pinnavaia, T. J. Chem. Comm. 2005,
29, 3697.

43 Holmberg, B. A.; Hwang, S.; Davis, M. E.; Yan, Y. Micro. Mesa. Mater.
2005, 80(1-3), 347.

44 Slowing, I. I.; Trewyn, B. G; Giri, 8.; Lin, V. S.-Y. Adv. Funct. Mater. 2007,
17(8),1225.

45 Lai, C.; Trewyn, B. G; Jeftinija, D. M.; Jeftinija, K.; Xu, S.; Jeftinija, 8.; Lin,
V. S.-Y. J.Am. Chem. Soc. 2003, 125( 15), 4451.

46 Li, S.; Li, Z.; Medina, D.; Lew, 0.; Yan, Y. Chem. Mater. 2005, 17(7), 1851.

47 Liu, Y.; Zhang, W.; Pinnavaia, T. J. J. Am. Chem. Soc. 2000, 122(36),
8791.

48 Liu, Y.; Zhang, W.; Pinnavaia, T. J. Angew. Chem. Int. Ed. 2001, 40(7),
1255.

49 Liu, Y.; Pinnavaia, T. J. Chem. Mater. 2002, 14(1), 3.

50 Zhang, 2.; Han, Y.; Xiao, F.; Qiu, S.; Zhu, L.; Wang, R.; Yu, Y.; Zhang, Z.;
Zou, 8.; Wang, Y.; Sun, H.; Zhao, D.; Wei, Y. J. Am. Chem. Soc. 2001, 123(21),
5014.

51 Zhang, Z. T.; Han, Y.; Zhu, L.; Wang, R W.; Yu, Y.; Qiu, S. L.; Zhao, D. Y.;
Xiao, F. S. Angew. Chem. Int. Ed. 2001, 40, 1258.

H3

Chapter 4

Synthesis, Characterization, and Catalytic Application

of Zeolites with Uniform Intracrystal Mesopores

4.1. Abstract ’

Zeolites with intra-crystal mesopores were prepared through a sol-gel
proceSs. In this approach the nucleation of the zeolite phase is carried out in
the presence of a silylated form of a polymer such as polyethylenimine and a,
w-diamine surfactant such as Jeffamine D series
H2N(CH(CH3)CH2O)nCH2CH(CH3)NH2. As the zeolite crystals grow, the
incorporated polymer becomes phase-segregated from the zeolite matrix,
forming an intracrystal polymer network that is covalently linked to the zeolite
framework through covalent Si-O-Si linkages. The average pore size can be
tuned from 2 nm to 8 nm depending on the properties of polymer porogen. In
addition to mesoporous ZSM-5 (denoted as MSU-MF I), mesoporous forms of
other common zeolite structures of commercial importance such as Linde type
A and zeolite Y also were prepared. The presence of the mesopores resulted
in little or no decease in micropore volume. However, the external surface area
and mesopore volume increased significantly. For MSU-MFI prepared from
silylated D4000, a mesopore volume of 0.65 cm3/g was observed, which is
16-fold increase compared to a conventional ZSM-5 sample. Moreover,

stability tests showed that after various hydrothermal treatments, the majority

ll4

of mesopores were retained, indicating of the robustness of the structure. The
high external surface area and large mesopore volume greatly improved the
mass transportation rate of MSU-MFI samples, dramatically improving their
catalytic performance. In the cumene cracking reaction, MSU-MFI catalysts
showed significantly higher cumene conversion compared to commercial USY
and ZSM-5 catalysts. Moreover, the presence of mesopores facilitated
secondary reactions that changed the product distribution from a 1:1 mixture of
benzene and propylene for the conventional zeolite to benzene and C3,.
hydrocarbons for the mesoporous zeolite. This effect on product selectivity
could be very useful in many reactions such as the methanol to gasoline

process.

115

4.2. Introduction

Aluminosilicate materials have widely been used as acid catalysts in the
refining industry for decades.“3 Among all the different types of
aluminosilicates, zeolites are the most important composition because of their
super strong acidity and uniform pore size,4 which in turn provides high
petroleum conversion rates and desirable product distributions. For instance,
the 0.56 nm pore size of ZSM-5 allows a much faster diffusion rate for
para-xylene than ortho- and meta-xylene. Therefore, toluene
disproportionation with a para-selectivity over 90% can be achieved on a
modified ZSM-5 catalyst to eliminate the isomerization reaction.‘5 In addition
to product shape-selectivity}H3 the small 0.3 to 1.2 nm micropores of zeolites
also greatly reduces the diffusion rate of reactant molecules to active sites,
therefore affecting their catalytic activities. Moreover, bulky molecules can
only be processed on the external surfaces of zeolite catalysts, which
typically only accounts for less than 1% of the total surface area for a
micron-size crystal. The change of oil supply quality and the pursuit of higher
yields of transportation fuels demand the development of novel cracking
catalysts. Large pore zeolite (ITO-33) catalysts have shown the ability to
provide higher yields of diesel fuel in the cracking product.° The 1.22 nm pore
size was defined by an 18-member ring, which was constructed partially
through 3-member rings. Fluorine ions were added to the reaction mixture to
stabilize the 3-member ring. Nevertheless, synthesizing new zeolites with
pore sizes larger than 1.2 nm still remains a challenge, and several other
approaches to obtaining larger pore zeolites have been explored

extensively.“Ha

116

Mesostructured aluminosilicate materials1M5 seem to be a promising
solution to obtaining large pore catalysts due to their high surface areas and
controllable pore sizes. Moreover, due to the uniform mesopore size, desired
products In higher yield can be produced using mesostructured aluminosilicate
catalysts.""“28 However, the weak acidity associated with the amorphous walls
prevent mesostructured aluminosilicates from performing better than zeolite
catalysts in petroleum reactions, even though many improvements in acidity
have been achieved?”2 Significant improvements in acidity can be achieved
through the incorporation of zeolitic subunits into the pore walls.”39 However,
the acidity of mesostructured aluminosilicate materials still is not comparable
to zeolites.

Recently, zeolites with intra-crystal mesopores have been investigated
intensively.‘5“° Compared to inter-crystal mesopores generated through

““4 intra-crystal mesopores are more interesting

nano-particle aggregation,
because the resulting pores are intrinsically smaller and more uniform, which
are the two most favorable features needed to achieve desirable selectivity.4o
Moreover, the time-consuming steps associated with nano-particle processing
are avoided. The presence of mesopores not only increases the external
surface area that is accessible to bulky reactant molecules; it also greatly
reduces the diffusion path length through the crystal, which helps to boost the
catalytic activity.

The traditional methods for generating intra-crystal mesopores within
zeolites, such as steaming and chemical leaching, leads to uncontrollable

pores 10 nm or larger.41 Moreover, the composition, as reflected in the SilAl

ratio, changes during the treatment. Carbon-based materials‘5' ‘6 have been

117

used to template intra-crystal mesopores within zeolite crystals with limited
control of the resulting mesopores. Upon calcination to remove the carbon
template, intra-crystal mesopores are formed that reflect the size of carbon
templates. Because of the hydrophobicity and low density of the carbon
templates, phase-segregation often occurs, which results in the formation of
zeolite nano-crystals.“ In addition, the irregularity of carbon templates
inevitably results in irregular pores with large diameters, which are not suitable
for shape- or size-selective catalysis. In the best case, a mesopore size
centered at 11 nm and a width at half height of 3 nm was achieved for ZSM-5
using a preformed carbon aerogel as the carbon template.18

Supramolecular assembly has proved to successfully template uniform
mesopores within amorphous aluminosilicate materials. However, the
relatively weak interactions within the micelle and between the micelle and the
silica surface make the soft template vulnerable to deformation at the high
temperatures that are needed for zeolite crystallization.“ ‘3 As a result, no
uniform mesopores have been obtained when the mesostructured wall are
fully crystallized. Kaliaguine and co-worker‘“ partially crystallized the thick
walls of large pore mesostructured aluminosilicate material MCF with a
significant loss of pore structure. In 2006, Ryoo and co-workers‘5 used a
unique surfactant whose hydrophilic head group was functionalized with a
silane group. In the synthesis, the covalent bonding between surfactant and
zeolite phase helped stabilize the mesopore template within the crystal. The
final zeolite products exhibited uniform mesopore size as small as 2.1 nm.

In the current study, we investigate the use of silylated polymers as

molecular templates for the formation of intra-crystal zeolite mesopores. This

118

 

molecular templating approach assures the formation of small and uniform
mesopores. The presence of the silyl group in the polymer is used to secure
the mesoporogen inside the zeolite crystals. Both polyethylenimine and a, w
diamine-polypropylene oxides with different average molecular weights have
been used as mesoporogen to template intra-crystal mesopores and control
the pore size. Moreover, this new approach was applied to other types of
zeolite structures such as LTA and FAU. Hydrothermal stability of the
mesopores has been studied. Also, in order to investigate the effect of the
mesopores on catalytic behavior, catalytic cumene cracking was conducted on
mesoporous ZSM-5. Commercial ZSM-5 and USY catalysts were used for

comparison purpose.

119

4.3 Experimental Section

4.3.1 Materials

Tetraethylorthosilicate (T EOS, 98%), colloidal silica (Ludox 30%), and
aluminum tri-sec-butoxide (97%) were obtained from Aldrich as silica and
alumina sources. Aluminum metal (Spectrum Chemical) was obtained in a
form of 5-mesh powder. Tetrapropylammonium hydroxide (20 wt%, TPAOH)
and tetramethylammonium hydroxide (35 wt%, TMAOH) were obtained from
Aldrich as structure-directing agents for MFl-, FAU-, and LTA-type zeolites.
Gchidoxypropyl-trimethoxysilane (GOPTMS) was obtained from Gelest.
Polyethylenimines (PEI) (Alfa Aesar) with average molecular weights of 600,
1800, 10000, 25000 (Aldrich) and a, 00 diamine-polypropylene oxides with
molecular weights of about 2000, and 4000 (Huntsman) were used as polymer
precursors for mesoporogens. The latter polymers are commercially available

under the trade name Jeffamine D-2000, and D-4000.

4.3.2 Synthesis of Mesoporgen

The polymer and GOPTMS modifier were dissolved into ethanol, which is
used as the solvent. The resulting solution was heated to elevated temperature
to form C-N bonds between the modifier and polymer, as illustrated in Figure
4.1. The most effective molar ratio of N-H to GOPTMS was found to be 5 to 10
for PEI and 1 for Jeffamine surfactants. In a typical synthesis, 0.5 g GOPTMS
(2.12 mmol) and 2.12 g D-4000 (0.53 mmol) were dissolved in 8 9 ethanol. The
solution was then heated to 80 ~ 100 °C in a sealed container for 24 hours.
After that, the ethanol was removed by applying vacuum to the sample. The

resultant modified polymer products were denoted as MP-D2000, MP-D4000,

I20

MP-600(x), MP-1800(x), MP-10000(x), and MP-25000(x), whereas x presents

the N-I-l/silane molar ratio.

Figure 4.1 Reaction for mesoporogen preparation.

4.3.3 Preparation of Mesoporous Zeolites

4.3.3.1 Mesoporous ZSM-5

MP-25000(5) was dissolved in TPAOH first. To the resulting solution,
TEOS and aluminum tri-sec-butoxide were pro-mixed and then added under
vigorous stirring. The gel composition was as follow: 1 SiO2: 0.01 Al203: 0.37
TPAOH: 20 H20: 4 EtOH (from hydrolysis): 0.1 MP, whereas the mole of MP is
calculated on the basis of silane. The synthesis mixture was then transferred
into a Teflon-lined autoclave and heated to 150 °C for 48 hours. In the case
that MP-D2000 and MP-D4000 were used as porogen, the reaction mixture
was heated to 100 °C for 96 hours. More specifically, extra ethanol (EtOl-l/SiO2
= 4) was added to the synthesis mixture from MP-D4000 to improve the
solubility of mesoporogen. The product was washed, dried, and calcined at

600 °C for 4 hours. The final product was denoted as MSU-MFI.

4.3.3.2 Mesoporous Faujasite
A 1.503-g quantity of MP-25000(5) and 0.324 g NaOH were dissolved in
17.536 9 TMAOH (25 wt%) and 50 9 water first. To the resulting solution, 0.414

9 aluminum was added slowly under vigorous stirring. After the aluminum was

121

dissolved, 6.667 g colloidal silica (30 wt%) was added. The gel composition
was as follow: 1 SiO2: 0.23 Al203: 1.44 TMAOH: 0.243 NaOH: 113 H2O: 0.1 MP,
where the mole fraction of MP is calculated on the basis of silane. The mixture
was then transferred into a Teflon-lined autoclave and heated to 100 °C for 96
hours. The product was washed, dried, and calcined at 600 °C for 4 hours. The

final product was denoted as MSU-Y.

4.3.3.3 Mesoporous Zeolite A

A 1.503-g quantity of MP-25000(5) was dissolved in 8.507 g TMAOH
solution (25 wt%) together with 0.057 g sodium hydroxide and 19 distilled
water. Then 1.389 9 aluminum triisoproxide was added. Finally, 6.667 g
colloidal silica (Ludox SM-30, 30 wt% SIO2) was added. The resulting mixture
was put in a shaking-bath until a homogeneous sol was formed. The molar
composition was ISiO2: 0.1 Al203: 0.02 Na2O: 0.1 Silane: 0.22 (TMA)2O: 20
H2O. The reaction mixture was heated to 100 °C for 60 hours. The solid
product was collected by centrifugation, and then washed with water. After
air-drying, the product was calcined at 500 °C for 6 hours to obtain the Na“
form of MSU-LTA.

4.3.4 Zeolite characterization

X-ray diffraction patterns were recorded on a Rigaku Rotaflex
Diffractometer using CuK. radiation (A=1.542A). N2 adsorption and desorption
isotherms at -196°C were obtained on a Micromeritics ASAP 2000 sorptometer.
The samples were degassed under 10<3 Torr at 250°C overnight prior to
analysis. The surface area was calculated from the BET equation. Pore size

distributions were derived from the adsorption isotherms using the BJH model.

l22

TEM images were taken on a Jeol 2200FS instrument operated at 200 kV. To
prepare thin-sectioned specimens, powdered samples were embedded into a
therrno set resin LR White and then microtomed into 70 nm thick slices. SEM
images were recorded on a JSM 6400 Electron Microscope at an acceleration

voltage of 20 W. The samples were carbon-coated prior to imaging.

4.3.5 Catalytic cumene cracking

Catalytic cumene cracking reactions were performed in a 6 mm id. fixed
bed quartz reactor with 200 mg catalyst at 200°C. Prior to catalytic testing, the
catalysts were ion-exchanged twice with 0.5 M NH4N03 solution followed by
calcination at 550°C for 4 hours to convert the zeolite to the protonated form.
The cumene flow rate was 4.1 umoI/min in a 20 cm3/min carrier stream of N2.
Cumene conversions were plotted against time on stream after a 6-hour
activation period under N2 flow at 200°C prior to the addition of cumene. The
reaction products were analyzed using an HP 5890 GC equipped with a

30-meter SP-1 capillary column and a flame ionization detector.

123

4.4 Results and Discussions

4.4.1 Formation of mesopores

Figure 4.2 (curve B) provides the X-ray diffraction patterns of a typical
calcined mesoporous MSU-MFI zeolite synthesized using MP-25000(5) as the
porogen. For comparison, a ZSM-5 zeolite synthesized without porogen is also
included (Figure 4.2, curve A). The MSU-MFI possesses all the characteristic
peaks of an MFI framework, confirming that MSU-MFI has the same
framework structure as conventional ZSM-5 zeolite. However, the peak
intensity of MSU-MFI is slightly weaker in comparison to conventional ZSM-5.
Moreover, the diffraction peaks are broader in the case of MSU-MFI, indicative
of a smaller crystal domain size.

Figure 4.3 provides the 29Si NMR spectrum of as-made MSU-MFI with
porogen present in the mesopores. Three distinguishable resonances were
observed at 115, 105, and 70 ppm, which can be assigned to Q4, 03, and Ta
sites respectively. The deconvolution analysis shows that the, T/Q value (0.13)
in the MSU-MFI product is in agreement with the initial synthesis composition
(0.10), thus confirming the complete incorporation of the porogen into the
zeolite crystals. Moreover, no T2 sites are found in the sample, which indicates
that the organosilanes in the mesoporogen are linked to the zeolite crystal
surface through 3 Si-O bonds. Such strong covalent bonding of multiple silane
centers in a single polymer chain prevents the mesoporogen from being
ejected out of zeolite crystals during crystallization. In a related experiment, a
non-silyated polymer in place of the silylated polymer was added to a zeolite
synthesis mixture.46 No intra-crystal mesopores were found in the resulting

zeolite product. In stead, aggregates of nanocrystals were observed. After

124

 

calcination, these silicon centers of the silylated porogen presumably form
amorphous silica, which also may contribute to the decrease in the XRD
intensity (Figure 4.2, curve 8).

Figure 4.4 provides the nitrogen adsorption-desorption isotherms for
mesoporous MSU-MFI zeolites made from silylated PEI porogens. Included for
comparison is the conventional ZSM-5 synthesized without porogen under
equivalent conditions. In addition to the almost vertical N2 uptake at P/Po
below 0.05, which is characteristic of micropore filling for all samples, nitrogen
uptake at P/Po between 0.1 and 0.6 is observed for the mesoporous MSU-MFI
samples prepared from silylated mesoporogen at a Nl-l/Si of 5 (Figure 4.4,
curve D). The nitrogen uptake in this region is indicative of the presence of
small mesopores. A BJH analysis of the nitrogen adsorption curve indicated
the average mesopore size to be 3.0 nm with the width of the pore distribution
being 1.5 nm. Such features are not observed for conventional ZSM-5 (Figure
4.4, curve A). Direct evidence for the-presence of intra-crystal mesopores is
provided by the TEM images shown in Figure 4.5 for whole particle and
thin-sectioned specimens. Each particle is permeated by randomly oriented
small mesopores of nearly the same size. Higher resolution images (Figure 4.5
A) reveal lattice fringes that extend over the entire particle, indicating that each
particle is a single crystal and not an aggregate of nanocrystals.

In addition to the intracrystal mesopores, MSU-MFI also possesses
textual porosity as indicated by the hysteresis at P/Po > 0.9. Such porosity also
is found for the ZSM-5 samples prepared from fully silylated PEI polymer
(Figure 4.4, curve B and C), and can be assigned to either surface defects or

inter-particle voids. However, when fully silylated PEI was used, no intracrystal

125

mesoporosity was found for the final products.

Regarding the formation of mesopores, we hypothesize that the PEI
polymer binds to proto—zeolitic nanoclusters through SiOa units and becomes
phase-segregated from the zeolite matrix as the crystal grows (Figure 4.6).
However, when the PEI was fully silylated, the large number of SiOa' units
present after hydrolysis prevents phase segregation, and causes the polymer
to bind as a single strand to the zeolite surface or to be expelled altogether
from the crystal, resulting in either case in a loss of mesoporosity. Figure 4.7
shows the SEM image of ZSM-5 samples prepared from fully silylated PEI
porogen. It can be clearly seen that not only is the zeolite crystal size reduced
greatly, but also that the surface of each particle is full of defects (Figure 4.7 B).
It is conceivable that the fully silylated PEI polymer is mainly bonded to the
external surface of the zeolite particle and therefore limits zeolite crystal
growth to the sub-micrometer size domain, giving rise to zeolite with
inter-crystal pores as indicated by the hysterisis at P/Po = 0.9. Moreover, such
textual porosity is dependent on the amount of fully silylated PEI added. For
instance, the addition of more silylated PEI, further decreases the crystal size

(Figure 4.7 B) and further increases the textual porosity (Figure 4.4, curve C).

4.4.2 Polymer effects on mesopore size

Figure 4.8 provides the mesopore size distributions and corresponding
mesopore volumes for MSU-MFI zeolites prepared in the presence of PEI
polymers with initial molecular weights of 600, 1,800, 10,000 and 25,000. Each
porogen was prepared at a Nl-l/Si ratio of 5.0. For each derivative, the pore

size distribution is confined to values below 8 nm and the distribution of pores

126

is centered between 2.0 and 3.0 nm with a width at half-maximum < 1.5 nm.
Moreover, the resulting mesopore size is in agreement with the order of
molecular weight. These mesopore size distributions, which are in the range
anticipated for selective polymer cracking, are unprecedented among all
previously reported mesoporous forms of ZSM-5 or any other zeolite structure
type. Also, the corresponding mesopore volumes for these new ZSM-5
derivatives (0.07 — 0.11 cm3lg) compare favorably with the micropore volume
of the zeolite framework (0.12 cm3/g).

The 50% increase in average mesopore size observed upon increasing
the polyethylenimine molecular weight from 600 to 25,000 is substantially
lower than the ~3.4-fold increase expected for a 40-fold increase in polymer
molecular weight. This latter observation indicates the zeolite matrix greatly
alters the salvation and conformation of the phase-segregated polymer
embedded in the crystal. Further evidence for confinement effects on the
polymer is provided by a comparison of the hydrodynamic radius of the free
polymer in comparison to the templated pore sizes. For instance, the
hydrodynamic radius of polyethylenimine with MW = 25,000 is ~ 6.6 nm,47
whereas the average radius of the mesopores templated by the silylated form
of this polymer is ~ 1.5 nm.

Besides polyethylenimines, other N-H containing polymers such as
Jeffamine surfactants also can be used as mesoporogens for MSU-MFI
synthesis. The structure of Jeffamine D surfactants is illustrated in Figure 4.9.
Because of the low density of N-H bonds, the Jeffamine D surfactants were
fully silylated (N-H/Si = 1) to optimize the interaction between the

mesoporogen and the zeolite phase through as many Si-O-Si linkage as

127

possible. Moreover, in the case of D-4000, additional ethanol was added to the
reaction mixture to increase the solubility of the mesoporogen. Figure 4.10
provides the N2 adsorption-desorption isotherms of MSU-MFI materials
prepared in the presence of silylated Jeffamine D surfactants. Compared to
conventional ZSM-5, both MSU-MFI samples exhibit distinguishable hysteresis
loops between partial pressure of 0.5 and 0.8. The corresponding pore size
distribution plots (Figure 4.10, inset) show uniform pore diameters centered at
3.0 nm and 5.4 nm, which are slightly larger than the pores templated by
silylated polyethylenimine. The resultant mesopore diameter correlates with

the molecular weight of the original polymer that was used as mesoporogen.

"2"“ WNHz
O x
CH3 CH3

Figure 4.9 Structure of Jeffamine D surfactant
x = 33.2 for 02000, and 67.7 for D4000

The light contrast areas in the TEM image of a thin-sectioned sample
(Figure 4.11) confirms the presence of uniform intracrystal mesopores that are
distributed evenly within the crystal. In the magnified image of Figure 4.11,
(inset), lattice fringes can be clearly seen, indicative of the crystalline phase.
X-ray diffraction patterns (Figure 4.12) reveal that both MSU-MFI samples
possess all the characteristic peaks of conventional ZSM-5. No other
crystalline phase is found. It is safe to claim that MSU-MFI materials are
mesoporous aluminosilicates with zeolite walls. The presence of uniform
mesopores greatly reduces the domain size of the zeolite crystals, resulting in
weaker and broader diffraction peaks. Compared with conventional ZSM-5,

MSU-MFI samples have reduced diffraction intensity. This trend becomes

128

more evident when the mesopores get bigger.

4.4.3 Effect of reaction temperature

In addition to being dependent on the molecular weight of the
mesoporogen, the templated mesopore size is also very dependent on the
crystallization conditions. As shown in Figure 4.13, the same silylated D4000
can provide different mesopore diameters depending on the temperature. At
100 9C, the average mesopore size is 5.4 nm, whereas at 125 9C, the pore
size is expanded to 8 nm (Figure 4.13, curve B). In the case of silylated D2000
as the mesoporogen, the average mesopore size can be expanded from 3.0
nm at 100 “C to 4.5 nm it the crystallization is conducted at 125 9C (Figure 4.13,
curve A). For both mesoporogens, no uniform mesopores were formed it the
reaction temperature was higher than 150°C. It is hypothesized that at the
higher temperature, Jeffamine surfactant is not sufficiently stable to template
intracrystal mesopores. But even at lower synthesis temperatures where
uniform mesopores are obtained, no low angle X-ray diffraction peaks typical
of MSU-J mesostructures with amorphous framework walls were found.48 This
latter result rules out the possibility that a mixture of mesostructure and zeolite

was formed.

4.4.4 Other types of zeolite

Besides MFI zeolite, mesoporous forms of other zeolite structures can
also be prepared though this polymer templating approach. Zeolite Y (FAU)
and Linda type A (LTA) are two common zeolites that have found many

applications in industry. Zeolite Y has been widely used as the main

l29

composition in refining and cracking catalysts in petroleum industry. A
mesoporous form of zeolite Y based cracking catalyst could meet the demand
for high yields of transportation fuels and the challenge accompanying a
change in crude oil quality. Zeolite A is well recognized as a good
ion-exchanger and absorbent. Moreover, calcium-exchanged LTA was used to
promote blood clotting."'9 The mesoporous form of zeolite A may be expected
to perform more efficiently because the presence of mesopore allows bulky
protein molecules to diffuse and bind more easily to the zeolite surface.

Figure 4.14 provides the N2 adsorption-desorption isotherm of MSU-Y
prepared from MP-25000(5). Compared to conventional zeolite Y (Figure 4.14,
curve A), the mesoporous MSU-Y sample showed a distinguishable hysteresis
loop at partial pressure of 0.4 (Figure 4.14, curve B), indicative of the presence
of 3 nm mesopores. In addition to the enhanced mesoporosity, no loss of
microporosity was observed for MSU-Y compared to conventional Y.

Figure 4.15 provides the N2 isotherm of MSU-A. A hysteresis loop at
higher partial pressure was observed, suggestive of larger mesopore. X-ray
diffraction pattern confirmed the presence of LTA type framework (Figure 4.15,
inset). These results suggested that this approach is applicable to other zeolite

structures other than MFI.

4.4.5 Hydrothermal stability

Since zeolites have extraordinary hydrothermal stability compared to
amorphous aluminosilicates, it is conceivable that mesoporous zeolite
MSU-MFI with fully crystalline zeolite walls would exhibit much improved

stability under hydrothermal conditions. Figure 4.16 provides the N2

130

adsorption-desorption isotherms of MSU-MFI sample treated under different
hydrothermal conditions. Compared to pristine MSU-MFI, hydrothermally
treated MSU-MFI samples show expanded pore sizes. For instance, after
being steamed with 20% water vapor at 800 9C for 2 hours, the mesopore
diameter increased from 5.4 to 6 nm (Figure 4.16, curve 8); while after being
boiled in water at 100 “C for 7 days, an average mesopore size of 9 nm was
obtained (Figure 4.16, curve C). In both cases, the mesopores remained
relatively uniform, and most of the mesopore volume was retained after
hydrothermal treatment. The results suggest that the mesopores templated by
the silylated polymer are very stable under hydrothermal condition because of

the crystalline walls.

4.4.6 Catalytic activity

Table 1 summarizes the physical properties of MSU-MFI samples.
Compared to conventional ZSM-5, which only possesses a small external
surface area and essentially no mesopore volume, all MSU-MFI samples
templated by silylated Jeffamine D surfactants exhibit greatly increased
external surface area. For instance, MSU-MFI prepared from silylated D2000
mesoporogen shows an external surface of 438 m2/g, a 6.7-fold increase
compared to conventional ZSM-5. Moreover, the mesopore volume was
significantly improved at the expense of a reasonable amount of micropore
volume or crystallinity. For example, MSU-MFI templated with silylated D4000
offers a 0.65 cm3/g mesopore volume, a 16-fold increase in comparison to the
conventional zeolite. When used as a catalyst, a zeolite with such a high

mesoporosity should greatly reduce the diffusion path length, and therefore,

13]

significantly enhance the catalytic performance. It is also noteworthy that after
hydrothermal treatment, the external surface area of the zeolites decreased
significantly. It is conceived that a lot of surface roughness was generated
during mesopore templating. Under hydrothermal conditions, the surface is
restructured without destroying the mesopores. Therefore, a significant loss of
external surface area was observed with increase in mesopore size.

Cumene cracking was selected as a probe reaction to investigate the
catalytic activity of mesoporous aluminosilicates. In this study, the reaction
temperature was lowered from 300 “C to 200 9C so that commercial ZSM-5
shows a low cumene conversion. Figure 4.17 provides the activity versus
time-on-stream results. Commercial USY and ZSM-5 zeolites with a Si/Al ratio
of 40 were used as references. USY provided a very high initial cumene
conversion of 50% (Figure 4.17, curve A), which decreased quickly to 20%
after 6 hours on stream. The fast deactivation can be assigned to the larger
pore size of faujasite, which is very vulnerable to coking. On the contrary, the
commercial ZSM-5 showed steady increase of activity and reached maximum
cumene conversion at 25% after 6 hours on stream (Figure 4.17, curve B).
This increase in reactivity may be caused by a thermal stresses that cause the
zeolite crystals to crack, exposing more external surface area with time on
stream. After that, the catalyst started deactivating. The slow increase of
activity might indicate the slow diffusion rate of reactant molecules into the
catalyst. When mesoporous MSU-MFI materials were used as catalysts, not
only was higher cumene conversion achieved, but also the maximum activity
was reached almost instantly, indicative of the elimination of diffusion obstacle.

For instance, a 65% cumene conversion was achieved on MSU-MF I prepared

[32

 

from MP-25000(5) (Figure 4.17, curve C). MSU-MFI with larger uniform
mesopores had an initial cumene conversion of 50% (Figure 4.17, curve D).
This relatively lower activity of MSU-MFI prepared from MP-D4000 might be
attributed to the lower crystallinity as judged by XRD peak intensity. After 24
hours on stream, both MSU-MFI catalysts showed a cumene conversion
above 38%, which is almost a two-fold increase over conventional ZSM-5. We
noted from TGA data that all spent catalysts have no more than 2wt% carbon
residue after 24 hours on stream. Therefore, the deactivation due to coking is
not a major concern. The unprecedented higher activity of MSU-MFI materials
can be contributed to enhanced mesoporosity and much higher external
surface area. Due to the presence of uniform mesopores and a high external
surface area, more active sites can be accessed by reactant molecules more
easily. Therefore, the catalytic activity was dramatically improved.

It is noteworthy that for MSU-MFI catalyst, a benzene to propylene molar
ratio of 3:1 was observed in the product distribution. For a commercial ZSM-5
catalyst, the benzene to propylene ratio was found to be 1:1, which is the
consistent with the expected reaction stoichiometry. As illustrated in Figure
4.18, for a commercial zeolite catalyst, the majority of the accessible active
sites are located on the outer surface of crystal, where product molecules can
easily escape. However, for a mesoporous zeolite catalyst, the majority of
reaction can occur inside the mesopores due to the improved diffusion
property. Not only is the catalytic performance greatly improved, but also the
mesopores provide a constricted space. Before the product molecules can
escape from the mesopore, they are likely to undergo secondary reaction to

form heavier hydrocarbons. Since propylene is more reactive than benzene,

I33

more propylene is consumed and a benzene to propylene ratio higher than 1:1
is observed. The ability to induce secondary reaction makes these materials
good candidate catalysts for reactions such as methanol to gasoline 5° where

secondary reaction is preferred.

I34

4.5 Conclusions and perspectives

The use of silylated templates in zeolite synthesis results In small and
uniform intracrystal mesopores that are also controllable in diameter within
zeolite crystals. These mesoporous zeolites have large external surface and
mesoporosity, which in turn greatly improves their mass transportation
properties. Therefore, higher catalytic activity can be achieved. Moreover, the
presence of mesopores also increases the possibility of reactants contacting
active sites on the walls, therefore, facilitates secondary reactions. In the
cumene cracking reaction, the yield of propylene is greatly reduced on
mesoporous zeolite catalysts.

Future work should be focusing on the application of mesoporous
zeolites to other catalytic processes such as methanol to olefins conversion. In
preliminary experiments, products obtained from MSU-MFI catalysts showed
much higher propylene/ethylene ratio than the one from conventional ZSM-5
catalyst. The results are plotted in Figure 4.19. For instance, at 350 °C.
mesoporous ZSM-5 prepared from silylated PEI and D4000 showed P/E ratio
of 4.7 and 4.4 respectively, in contrast to 2.4 from conventional ZSM-5 catalyst.
The reason is because the presence of mesopores and large external surface
area greatly increases the reaction rate and helps to reach equilibrium state
faster. In this reaction, medium pore ZSM-5 favors the formation of propylene
over ethylene. Since in industry ethylene is more valuable than propylene, the
commercial MTO catalyst uses small pore SAPO-34 as catalyst, which favors
the formation of ethylene. Therefore, it will be worthwhile to synthesize
mesoporous form of SAPO-34 and use it as MT 0 catalyst to investigate the

effect of uniform mesopores on the product distribution.

135

In addition, as shown in Figure 4.20, in the polymer degradation reaction,
the addition of commercial ZSM-5 products significantly lowers the

degradation temperature of HDPE from 440 °C to 310 °C. Moreover, the

presence of mesopores in ZSM-5 catalysts further lowers the degradation

temperature to 290 °C. It is conceivable that MSU-MFI materials allow the

diffusion of bulky molecules such as HDPE inside the zeolite crystals, where
more active sites are accessible. In addition, from the results for bulky ZSM-5
catalysts, the framework Si/Al has significant influence on the degradation
temperature. First, the degradation temperature decreases along with
decreasing Si/Al, indicative of the strong correction between catalytic activity
and acidic sites. The activity of ZSM-5 starts decreasing when the SilAl ratio is
lower than 20. This is probably due to the decreasing of acidic strength when
the population of AI increases. It is well known that the framework Si/Al ratio of
zeolite catalysts also affects the product distribution in reactions such as
cracking. Therefore, the future work should be focused on the analysis of
HDPE degradation products and investigating the effect of Si/Al on product
distribution. Furthermore, mesoporogens other than the PEI and Jeffamine

should also be explored to expand the scope of the current project.

136

 

Table 4.1 Physical properties of MSU-MFI materials

 

 

Catalyst' Polymer BET External Micropore Mesopore
source Surface surface Volume, Volume,
area, mz/g area, n12/g cm’/g cm’/g

ZSM-5 N/A 389 65 0.14 0.04
MSU-MFI MP-25000(1) 467 161 0.14 0.24
MSU-MF I MP-25000(5) 451 255 0.09 0.27
MSU-MF I MP-D2000 668 438 0.10 0.44
MSU-MFI MP-D4000 613 405 0.09 0.65
MSU-MFI“ MP-D4000 517 327 0.08 0.49
MSU-MFI'“ MP-D4000 494 306 0.08 0.60

 

* Catalysts have a Si/Al ratio of 50.
** Sample was steamed at 800 “C, 20% water vapor for 2 hours.

*** Sample was boiled in water under 100 9C for 7 days.

137

 

1 2000

10000 -

8000 -

 

8

O

O
1

 

Intensity

4000 :

 

2000 r A

 

 

07* r _I‘ I I I _T—‘L‘LT

5 1 0 1 5 20 25 30 35 40 45 50

20
Figure 4.2 X-ray diffraction patterns of (A), calcined conventional ZSM-5 (Si/AI

= 50); (B), calcined MSU-MFI (Si/Al = 50) prepared from silylated PEI
mesoporogen with an initial NH/Si ratio of 5: 1. Curve 8 is offset by 5000

COURtS.

138

 

 

 

 

 

 

1 15.4 ppm
' 104.6 ppm
3: - 70.1 ppm
<0
I l I I r I
0 ~50 -100 -150
ppm

Figure 4.3 29Si MASNMR spectrum of as-synthesized MSU-MFI (Si/AI = 50)

prepared from 10% silane modified PEI porogen.

139

 

400

350-

300 —

 

 

 

 

 

o.
.—
(D
g: 250 5.
0 I;
0 :.
8
'9 200 .5
o ..
a) i:
U 1:
C5 0
a)
g 150 :
O A
>

100 '5:

50

O r I I I I
0 0.2 0.4 0.6 0.8 1

PIPo
Figure 4.4 N2 adsorption/desorption isotherms of ZSM-5 (Si/Al = 50)

synthesized in the presence of A, no MP-25000; 8, 5 mol% organosilane as
MP-25000(1); C, 10 mol% organosilane as MP-25000(1); D, 10% organosilane

as MP-25000(5). Each curve is offset by 50 cclg.

I40

 

Figure 4.5 TEM images of the MSU-MFI zeolite (Si/Al = 50) prepared in the
presence of silylated polyethleneimine with an initial molecular weight of
25,000: A) whole particle specimen; 8) thin-sectioned sample. The light

contrast areas are intracrystal mesopores.

141

.comoaoaomoE 55 we LoE>_oa “.2856 a mafia «SocomoE _waogE 5? 2:03 a _o 285%.... 65 Lo. oEosom m6 0.59".

5:25.
ago: 8:58
.32....

2.83.8
.6528 I 2:08 ...oE>.oa
5285:: 85%

ecu—95 «3.0.5

.3290 230N905 c3333:
111 1117/ 11:
9:03.

«EDGE «Am—0.5 ecu—95

 

142

.Eooomwis. o\om .0 new xtooomNH—s.

oxen: .m ”Lesion 6283:...“ o: .< Go 8:083 35 E .8885 8m u :39 mcaEmw m-2mN Go mommfi 2mm Ev 9:9”.

 

143

 

0. 1 0.09

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.09 - 0.08 - c
0-08 ‘ A 0.07 4 B
e 9 o
0.07 - ‘ o m . .
' c
0.06 «
c 0.05 .2
0.05 . 3 0 0‘ .
0.04 - 0 ' ’2
3 0.03 . g
0.03 " ‘o .
0.02 _ ‘ 0.02 ~ 3‘
0.01 « \.‘ 0-01 ‘ ’3“
0 ”’00:. 4, 0 ."°°.° ’
o 2 4 3 3 1o 0 2 4 6 8 10
0.09 0.09
0.08 - 0.08 . c
08’ a
0.07 - ’3 C 0.07 - .1» g D
Q
0.06 - 9 ‘ 0.06 - g ~
0
4 O a
0.05 .‘ 0.05 3° 3:
0.04 ~ ‘ 0.04 ~
0
0.03 - . 0.03 . s
0 e
0.02 ~ .2 0.02 ~ .
0
am- 0. 0m: ’2.
O O . O Q
o 1 7 T I. . ’ o I T I I ‘,
o 2 4 6 8 10 0 2 4 6 8 10

Pore diameter, nm
Figure 4.8 BJH adsorption pore size distributions for MSU-MFI samples (Si/Al
= 50) prepared from the following silylated PEI polymers: A, MP-600(5); B,
MP-1800(5); C, MP-10000(5); D, MP-25000(5).

144

 

600

500

400

300

200

 

100

 

 

 

15 Pore diameter, nm
0 x T I I I

 

 

0 0.2 0.4 0.6 0.8 1

P/Po
Figure 4.10 N2 adsorption/desorption isotherms of MSU-MFI samples (Si/Al =

50) synthesized in the presence of A, MP-D2000; B, MP-‘D4000. Inset: BJH

pore size distributions from adsorption isotherms.

145

 

 

Figure 4.11 TEM image of thin-sectioned MSU-MFI sample (SI/AI = 50)

prepared from silylated D4000 as mesoporogen. Inset: high-resolution Image

of selected area.

146

 

 

 

Intensity a.u.

 

 

A

 

 

 

5 10 15 20 25 30 35 40 45 50
29
Figure 4.12 X-ray diffraction patterns for (A) conventional ZSM-5 (Si/AI = 50);

(B) MSU-MFI (Si/Al = 50) prepared from silylated D2000 as mesoporogen; (C)

MSU-MFI (Si/Al = 50) prepared from silylated D4000 as mesoporogen.

147

 

700 ~

 

 

 

400

 

 

300 §

Volume adsorption, cm’lg STP

 

100

 

 

1 Pen 01.119.11.11, nm

 

o e .

0 0.2 0.4 P [Po 0.6 0.8 1

Figure 4.13 N2 adsorption-desorption isotherms for MSU-MFI samples (Si/Al =
50) prepared at 125 9C from (A) silylated D2000 as the mesoporogen and (B)
silylated D4000 as the mesoporogen. The curve B is offset by 200 for clarity.

Inset: BJH pore size distributions obtained from adsorption isotherms.

I48

 

 

 

 

 

 

 

350
B .T
300 — g-r'a"
I7, 250 —
9 A 1""
O .
O .
'6 200
0)
E
O
(D .
8 150 *j
w .
E
2
g 100
50
O I I I I I
0 0.2 0.4 0.6 0.8 1

P/Po
Figure 4.14 N2 adsorption/desorption isotherms of zeolite Y (Si/Al = 2.2)

synthesized A, in the absence of a mesoporogen; B, in the presence of

MP-25000(5) as a mesoporogen.

I49

 

500

450

400

0)
01
O

OJ
C
O

 

 

Volume adsorbed, cc/g STP
8
O

 

 

 

 

200
150
100
50
O t I I 1 1
0 0.2 0.4 0.6 0.8 1
P/Po

5)

Figure 4.15 N2 adsorption/desorption isotherm of MSU-A (Si/Al

synthesized with MP-25000(5). Inset: X-ray diffraction pattern.

150

 

 

 

 

 

 

 

 

 

 

 

an
I-
U)
«2'
n
E
0
E
.2 .-
a m f:
5 2432 -‘ ‘ ‘ A .
G ‘ I
t ....... ‘ . -
fl """ I"
“I 400
E ,
3 .
O
>
200
o I I I I
0 0.2 0.4 0.6 0.8 ‘I

PlPo
Figure 4.16 N2 adsorption-desorption isotherms for MSU-MFI sample (Si/Al =

50) prepared from silylated D4000. (A) pristine MSU-MFI; (B) steamed with
20% water vapor at 800 9C for 2 hours; (C) treated with boiling water for 7 days.
The curves 8 and C are offset by 300 and 600 for clarity. Inset: BJH pore size

distributions obtained from adsorption isotherms.

151

 

7O

60*

 

30*

Cumene Conversion 1%
o

20—

1O

 

 

 

0 i l I
O 500 1 000 1 500 2000

Time on Stream, min

Figure 4.17 Catalytic cumene cracking conversion vs. time for A, commercial
USY from Zeolyst (Si/Al = 40); 8, commercial ZSM-5 from Zeolyst (Si/AI = 40);
C, MSU-MFI (Si/Al = 50) prepared from MP-25000(5); D, MSU-MFI (Si/AI = 50)

prepared from MP-D4000.

152

308.5

  

 

99.98 9:08 maeoaomoE ucm E39 :0 9.2080 3063 2598 _o c2852: 94. 2:9“.

2203 380832 a 8:03 585

 

I53

 

4.5 ~

2.5

I

P/E ratio

1.5

m

0.5 ~

0 1 1 L 1

300 350 400 450 500 550
Temperature, s'C

 

 

 

Figure 4.19 Product propylene/ethylene ratios of methanol-to-olefins reaction
catalyzed by A, commercial bulky ZSM-5 (Si/Al = 40) with a contact time of 0.4
second; B, MSU-MFI (Si/Al = 40) prepared from MP-25000(5) with a contact
time of 0.4 second; and C, MSU-MFI (Si/AI = 40) prepared from D4000 with a

contact time of 0.2 second, Reaction conditions: WHSV = 0.50 h", N2.

154

 

 

     
  

 

 

 

 

100 r
80 —
E D
‘l/
:5 C
°\°60 ~ 1/
g E
9 V 1’»
é’ 1:
40 ~ F '5 3
G 0A
1’.
it
20 — *
0 1 L 1 1 1 7
0 100 200 300 400 500 600

Temperature, 9C
Figure 4.20 HDPE degradation catalyzed by A, no catalyst; 3, commercial
ZSM—5 (Si/Al = 280); C, commercial ZSM-S (Si/Al = 80); D, commercial ZSM-5
(Si/AI = 50); E, commercial ZSM-5 (Si/Al = 23); F,‘ MSU-MFI (Si/Al = 40)

prepared from MP-25000(5); and G. MSU-MFI (Si/AI = 40) prepared from
D4000. HDPE/Catalyst = 4: 1.

155

4.6 References

1 Flanigen, E. M. Stud. Surf. Sci. Catal. 1991, 58, 13.

2 Corma, A. Chem. Rev. 1997, 97(6), 2373.

3 Stoecker, M. Micro. Meso. Mater. 2005, 82(3), 257.

4 Corma, A.; Martinez, A. Adv. Mater. 1995, 7(2), 137.

5 Kaeding, W. W.; Chu, C.; Young, L. B.; Butter, S. A. J. Catal. 1981, 69(2),
392.

6 Dubbeldam, D.; Calero, S.; Maesen, T. L. M.; Smit, B. Angew. Chem. Int.
Ed. 2003, 42(31), 3624.

7 Benazzi, E.; Leite, L.; Marchal-George, N.; Toulhoat, H.; Raybaud, P. J.
Catal. 2003, 217(2), 376.

8 Chen, C. S. H.; Bridger, R. F. J. Catal. 1996, 161(2), 687.

9 Conna, A.; Diaz-Cabanas, M. J.; Jorda, J. L.; Martinez, 0.; Moliner, M.
Nature 2006, 443(7113), 842.

10 Mintova, 8.; Olson, N. H.; Senker, J.; Bein, T. Angew. Chem. Int. Ed. 2002,
41( 14), 2558.

11 Schmidt, l.; Madsen, C.; Jacobsen, C. J. H. Inorg. Chem. 2000, 39(11),
2279.

12 Kim, S. 3.; Shah, J.; Pinnavaia, T. J. Chem. Mater. 2003, 15, 1664.

13 Serrano, D. P.; Aguado, J.; Escola, J. M.; Rodriguez, J. M.; Peral, A. Chem.
Mater. 2006, 18, 2462.

14 Reding, G; Maurer, T.; Kraushaar-Czametzki, B. Micro. Mesa. Mater. 2003,
57, 83.

15 Janssen, A. H.; Schmidt, l.; Jacobsen, C. J. H.; Koster, A. J.; de Jong, K. P.
Micro. Meso. Mater. 2003, 65(1), 59.

16 Schmidt, l.; Boisen, A.; Gustavsson, E.; Stahl, K.; Pehrson, S.; Dahl, S.;
Carlsson, A.; Jacobsen, C. J. H. Chem. Mater. 2001, 13, 4416.

17 Jacobsen, C. J. H.; Madsen, C.; Houzvicka, J.; Schmidt, l.; Carlsson, A. J.
Am. Chem. Soc. 2000, 12, 7116.

18 Tao, Y.; Kanoh, H.; Kaneko, K. J. Am. Chem. Soc., 2003, 125(20), 6044.

19 Beck, J. S.; Vartuli, J. C.; Roth, W. J.; Leonowicz, M. E.; Kresge, C. T.;

156

 

Schmitt, K. D.; Chu, C. T. W.; Olson, D. H.; Sheppard, E. W.; et al. J. Am.
Chem. Soc. 1992, 114(27), 10834.

20 Tanev, P. T.; Pinnavaia, T. J. Science, 1996, 271(5253), 1267.
21 Kim, S. S.; Zhang, W.; Pinnavaia, T. J. Science, 1998, 282(5392), 1302.

22 Yang, P.; Zhao, D.; Margolese, D. |.; Chmelka, B. F.; Stucky, G. D. Nature,
1998, 396(6707), 152.

 

23 Schmidt-Winkel, P.; Lukens, W. W. Jr.; Zhao, D.; Yang, P.; Chmelka, B. F.;
Stucky, G. D. J. Am. Chem. Soc. 1999, 121(1), 254. m

24 Tanev, P. T.; Pinnavaia, T. J. Science, 1995, 267(5199), 865.

25 Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. 0.; Beck, J. 8.
Nature, 1992, 359(6397), 710.

26 Xu, 3.; Hua, W.; Yue, Y.; Tang, Y.; Gao, Z. Catal. Lett. 2005, 100(1-2), 95. ~

 

27 Liu, Y.; Kim, S. S.; Pinnavaia, T. J. J. Catal. 2004, 225(2), 381.

28 Kim, S.-S.; Liu, Y.; Pinnavaia, T. J. Micro. Meso. Mater. 2001, 44-45, 489.
29 Mokaya, R. Chem. Comm. 1997, 2185.

30 Mokaya, R. Angew. Chem. Int. Ed. 1999, 38, 2930.

31 Huang, L.; Guo, W.; Deng, P.; Xue, Z.; Li, 0. J. Phys. Chem. 82000, 104,
2817.

32 Mokaya, R. Chem. Comm. 2001, 633.

33 Liu, Y.; Zhang, W.; Pinnavaia, T. J. J. Am. Chem. Soc. 2000, 122(36),
8791.

34 Liu, Y.; Zhang, W.; Pinnavaia, T. J. Angew. Chem. Int. Ed. 2001, 40(7),
1255.

35 Liu, Y.; Pinnavaia, T. J. Chem. Mater. 2002, 14(1), 3.

36 Liu, Y.; Pinnavaia, T. J. J. Mater. Chem. 2004, 14(7), 1099.

37 Wang, H.; Liu, Y.; Pinnavaia, T. J. J. Phys. Chem. B2006, 110(10), 4524.
38 Zhang, 2.; Han, Y.; Xiao, F.; Qiu, S.; Zhu, L.; Wang, R.; Yu, Y.; Zhang, 2.;
Zou, B.; Wang, Y.; Sun, H.; Zhao, D.; Wei, Y. J. Am. Chem. Soc. 2001,
123(21), 5014.

39 Zhang, Z. T.; Han, Y.; Zhu, L.; Wang, R. W.; Yu, Y.; Qiu, S. L.; Zhao, D. Y.;

157

Xiao, F. S. Angew. Chem. Int. Ed. 2001, 40, 1258.

40 Aguao, J.; Sotelo, J. L.; Serrano, D. P.; Calles, J. A.; Escola, J. M. Energy
& Fuels 1997, 11, 1225.

41 Groen, J. C.; Peffer, L. A. A.; Moulijn, J. A.; Perez-Ramirez, J. Chem. Eur.
J. 2005, 11(17), 4983.

42 Karlsson, A.; Stocker, M.; Schmidt, R. Micro. Meso. Mater. 1999, 27, 181.
43 Xia, Y.; Mokaya, R. J. Mater. Chem. 2004, 14,863.
44 On, D. T.; Kaliaguine, S. Angew. Chem. Int. Ed. 2001, 40, 3248. E—

45 Choi, M.; Cho, H. S.; Srivastava, R.; Venkatesan, C.; Choi, D.-H.; Ryoo, R.
Nature Mater. 2006, 5(9), 718.

46 Xiao, R, Wang, L.; Yin, 0.; Lin, K.; Di, Y.; Li, J.; Xu, R.; Su, D. 8.; Schlogl, :
R.; Yokoi, T.; Tatusmi, T. Angew. Chem. Int. Ed. 2006, 45, 3090. s

 

47 Andersson, M. M.; Hatti-Kaul, R.; Brown, W. J. Phys. Chem. B. 2000, 104,
3660.

48 Park, l.; Wang, 2.; Pinnavaia, T. J. Chem. Mater. 2005, 17(2), 383.

49 Ostomel, T. A.; Stoimenov, P. K.; Holden, P. A.; Alam, H. B.; Stucky, G. D.
J. Thrombosis and Thrombolysis 2006, 22(1), 55.

50 Keil, F. J. Micro. Meso. Mater. 1999, 29(1-2), 49.

158

   

111111111111I