FT-IR Studies of Zeolite Materials - Characterization and Environm PDF

University of Iowa
Iowa Research Online

Theses and Dissertations
2005
FT-IR studies of zeolite materials: characterization

and environmental applications
Gonghu Li
University of Iowa
Copyright 2005 Gonghu Li

This dissertation is available at Iowa Research Online: http://ir.uiowa.edu/etd/96
Recommended Citation
Li, Gonghu. "FT-IR studies of zeolite materials: characterization and environmental applications." PhD (Doctor of Philosophy) thesis,
University of Iowa, 2005.
http://ir.uiowa.edu/etd/96.
Follow this and additional works at: http://ir.uiowa.edu/etd

Part of the Chemistry Commons
FT-IR STUDIES OF ZEOLITE MATERIALS: CHARACTERIZATION

AND ENVIRONMENTAL APPLICATIONS
by
Gonghu Li
An Abstract
Of a thesis submitted in partial fulfillment
of the requirements for the Doctor of
Philosophy degree in Chemistry
in the Graduate College of
The University of Iowa
May 2005
Thesis Supervisors: Professor Vicki H. Grassian
Associate Professor Sarah C. Larsen
ABSTRACT
Synthetic zeolites are used widely throughout the petroleum refining and chemical
process industries as selective adsorbents, catalysts and ion exchangers. The research
described in this thesis focuses on the characterization and environmental applications of
faujasite zeolites, including nanocrystalline Y zeolites that have particle sizes less than
100 nm.
In Chapter 3, photooxidation of cyclohexane and cyclohexene using molecular
oxygen in BaY is investigated as a green chemical process. The reactivity and selectivity
in photooxidation reactions are examined as a function of reactant, oxidant and
irradiation wavelength. A major portion of this thesis involves the use of nanocrystalline
zeolite catalysts in deNOx reactions. Nanocrystalline zeolites with increased external
surface areas are synthesized and characterized with a variety of techniques, including
transmission FT-IR spectroscopy, sorption of probe molecules and microscopy (Chapter
4). There are a greater number of silanol groups and extra framework aluminum (EFAL)
species, which are important in applications as adsorbents and catalysts, on the external
surface of nanocrystalline NaY zeolites relative to commercial micrometer-sized NaY.
In the absence of oxygen, the selective catalytic reduction (SCR) of NO2 with
propylene at low temperature (T 473 K) results in the complete reduction of NO2 to N2
and O2 in nanocrystalline NaY zeolite with a particle size of 8030 nm (Chapter 5). It is
revealed that silanol groups and EFAL sites on the external surface of nanocrystalline
NaY provide additional adsorption sites for propylene and NO2. EFAL species are found
to be important for SCR reactions. This is further supported by the studies using
nanocrystalline NaY with a particle size of 23 nm as a deNOx catalyst with propylene
(Chapter 6) and urea (Chapter 7) as reducing reagents. In the presence of an excess
amount of oxygen, nanocrystalline NaY appears to be a better SCR catalyst than
commercial NaY. Silanol groups and EFAL species are found to be responsible for the
better performance of nanocrystalline NaY relative to commercial NaY zeolite. Thus the
increased external surface of nanocrystalline zeolites can be utilized as a reactive surface
with unique active sites for catalysis.
Abstract Approved: ____________________________________
Thesis Supervisor
____________________________________
Title and Department
____________________________________
Date
____________________________________
Thesis Supervisor
____________________________________
Title and Department
____________________________________
Date
FT-IR STUDIES OF ZEOLITE MATERIALS: CHARACTERIZATION

AND ENVIRONMENTAL APPLICATIONS
by
Gonghu Li
A thesis submitted in partial fulfillment

of the requirements for the Doctor of
Philosophy degree in Chemistry
in the Graduate College of
May 2005
Thesis Supervisors: Professor Vicki H. Grassian
Associate Professor Sarah C. Larsen
Copyright by
GONGHU LI
2005
All Rights Reserved
Graduate College
Iowa City, Iowa
CERTIFICATE OF APPROVAL
_______________________
PH.D. THESIS
_______________
This is to certify that the Ph.D. thesis of
Gonghu Li
has been approved by the Examining Committee for the thesis requirement for
the Doctor of Philosophy degree in Chemistry at the May 2005 graduation.
Thesis Committee: ___________________________________
Vicki H. Grassian, Thesis Supervisor
___________________________________
Sarah C. Larsen, Thesis Supervisor
___________________________________
Lei Geng
___________________________________
Edward G. Gillan
___________________________________
C. Allan Guymon
ACKNOWLEDGMENTS
First of all, I would like to thank my advisors, Professor Vicki Grassian and
Professor Sarah Larsen. They have supported and helped me enormously. I would like
to thank them for showing me the way to be a great scientist and a wonderful teacher.
They always believed in me and told me not to shake my head on obstacles. They always
gave me confidence to keep moving forward when I was frustrated. Thinking of how
they spent time and energy on introducing me into the amazing world of original research
and guiding me through the difficult times, I hope I am at the level of their expectations.
I would like to thank Professor Dwight Tardy whom I took course with and
worked with as a grader and teaching assistant. I benefited a lot from his instruction and
encouragement. I am also indebted to Dr. Russell Larsen who helped me a lot with
teaching. Many thanks to my committee members: Professors Lei Geng, Edward Gillan,
Harold Goff and Allan Guymon.
I want to extend my appreciation to former and current members in Grassian
group and Larsen group. Dr. Ping Li showed me how to work with UHV chamber and
collect infrared spectra when I first came to Iowa. Dr. Weiguo Song provided me with
the magic nanocrystalline zeolites and very helpful discussion. I am very much indebted
to James Woodworth who helped me in every aspect when I worked in Larsen labs.
Thanks Jim! Many thanks to Dr. Brenda Krueger who always enjoyed helping me, to Dr.
Amy Michel and Dr. Alex Saladino for their help in experiments. I have to acknowledge
other group members for their help and encouragement: Sofia Carlos-Cuellar, Dr. Hind
Al-Abadleh, Dr. Courtney Usher, Dr. Amy Preszler Prince, Hashim Al-Hosney, Dr.
Conrad Jones, Hassan Alwy, Praveen Mogili, Elizabeth Johnson, Jonas Baltrusaitis and
Dr. Kevin Knagge.
I would like to thank people in the Chemistry department at the University of
Iowa, Janet McCune, Frank Turner, Sharon Robertson, Michele Gerot, and the staff in the
ii
Chem Store and Chemistry library. I also want to express my appreciation to the CMRF
staff and teachers in the ESL program. Their kind help greatly facilitated my study and
research.
I am grateful to Professor Weizhu An, my thesis advisor for master degree in
Chengdu Institute of Organic Chemistry, the Chinese Academy of Sciences. I would also
like to thank Proferssor Haili Zhang, Dr. Chuanming Jin and Dr. Minghua Yang, who
helped me to prepare for a career in chemistry when I was an undergraduate student at
Hubei Normal University, P. R. China.
I have to name a few of my friends whose friendship meant a lot to me during my
graduate studies: Dr. Jianjun Xu, Dr. Hui Li, Dr. Fengyi Liu, Liu Deng, Wei Li, Wei
Zhang, Yan Liu, Mingbo Qin, and Caiyun Zhao and her family for their continuous
support and encouragement. I also would like to thank the Knoll family, especially Diane
and Mike who are my parents in America, and Rikka and Jodi. Their friendship and
support are very important to my life in Iowa.
It would be impossible for me to pursue a scientific career without the support and
encouragement from my Uncle Jianbin Li and his family.
My parents gave me everything. They always have great expectations on me and
never hesitate to sacrifice for any good of my life. Thanks to my sisters and brother for
their love!
Finally, I would like to thank my wife, Yimin Gao, for her love and support!
iii
ABSTRACT
Synthetic zeolites are used widely throughout the petroleum refining and chemical
process industries as selective adsorbents, catalysts and ion exchangers. The research
described in this thesis focuses on the characterization and environmental applications of
faujasite zeolites, including nanocrystalline Y zeolites that have particle sizes less than
100 nm.
In Chapter 3, photooxidation of cyclohexane and cyclohexene using molecular
oxygen in BaY is investigated as a green chemical process. The reactivity and selectivity
in photooxidation reactions are examined as a function of reactant, oxidant and
irradiation wavelength. A major portion of this thesis involves the use of nanocrystalline
zeolite catalysts in deNOx reactions. Nanocrystalline zeolites with increased external
surface areas are synthesized and characterized with a variety of techniques, including
transmission FT-IR spectroscopy, sorption of probe molecules and microscopy (Chapter
4). There are a greater number of silanol groups and extra framework aluminum (EFAL)
species, which are important in applications as adsorbents and catalysts, on the external
surface of nanocrystalline NaY zeolites relative to commercial micrometer-sized NaY.
In the absence of oxygen, the selective catalytic reduction (SCR) of NO2 with
propylene at low temperature (T 473 K) results in the complete reduction of NO2 to N2
and O2 in nanocrystalline NaY zeolite with a particle size of 8030 nm (Chapter 5). It is
revealed that silanol groups and EFAL sites on the external surface of nanocrystalline
NaY provide additional adsorption sites for propylene and NO2. EFAL species are found
to be important for SCR reactions. This is further supported by the studies using
nanocrystalline NaY with a particle size of 23 nm as a deNOx catalyst with propylene
(Chapter 6) and urea (Chapter 7) as reducing reagents. In the presence of an excess
amount of oxygen, nanocrystalline NaY appears to be a better SCR catalyst than
commercial NaY. Silanol groups and EFAL species are found to be responsible for the
iv
better performance of nanocrystalline NaY relative to commercial NaY zeolite. Thus the
increased external surface of nanocrystalline zeolites can be utilized as a reactive surface
with unique active sites for catalysis.
TABLE OF CONTENTS
LIST OF TABLES............................................................................................................. ix
LIST OF FIGURES .............................................................................................................x
CHAPTER 1 INTRODUCTION ........................................................................................1
1.1
1.2
1.3
1.4
Zeolite Materials.....................................................................................................1
Characterization of Zeolites with Infrared Spectroscopy .......................................3
Environmental Catalysis Using Zeolites ................................................................6
Thesis Overview and Objectives ............................................................................7
CHAPTER 2 EXPERIMENTAL METHODS .................................................................10

2.1 Transmission FT-IR Spectroscopy Experiments ..................................................10
2.1.1 Sample Preparation for FT-IR Studies ........................................................10
2.1.2 Experimental Setup for Transmission FT-IR Studies .................................12
2.1.3 Protocol of In Situ FT-IR Measurements ....................................................13
2.1.4 Potassium Bromide Pellets Technique ........................................................14
2.2 Experimental Setup for Photooxidation Studies...................................................15
2.2.1 Photooxidation Studies of Hydrocarbons Followed by FT-IR
Spectroscopy ................................................................................................15
2.2.2 Batch Reactor Studies..................................................................................16
2.2.3 GC and GC-MS Analysis ............................................................................17
2.3 A Flow System for Adsorption/Desorption Studies .............................................17
2.4 Other Experimental Techniques Used in This Dissertation Research..................19
2.5 Reagents and Materials.........................................................................................20
CHAPTER 3 PHOTOOXIDATION OF CYCLOHEXANE AND
CYCLOHEXENE IN BaY ZEOLITE .........................................................21
3.1 Introduction ..........................................................................................................21
3.2 Experimental Section............................................................................................23
3.3 Results ..................................................................................................................24
3.3.1 Photooxidation of Cyclohexane Using Oxygen ..........................................24
3.3.2 Photooxidation of Cyclohexene Using Oxygen ..........................................28
3.3.3 Oxidation of Cyclohexene Using TBHP .....................................................32
3.4 Discussion.............................................................................................................35
3.4.1 Influence of Wavelength on Product Distribution.......................................35
3.4.2 Photooxidation of Cyclohexane vs. Photooxidation of Cyclohexene
in BaY ..........................................................................................................35
3.4.3 Photooxidation of Cyclohexene in BaY with Different Oxidants (O2
vs. TBHP).....................................................................................................36
3.5 Conclusions ..........................................................................................................36
3.6 Acknowledgements...............................................................................................37
vi
CHAPTER 4 CHARACTERIZATION OF NANOCRYSTALLINE NaY

ZEOLITES AND THE ADSORPTION OF SELECTED PROBE
AND SMALL MOLECULES .....................................................................38
4.1 Introduction ..........................................................................................................38
4.2 Synthesis of Nanocrystalline NaY Zeolite ...........................................................42
4.3 Characterization of Nanocrystalline NaY Zeolite ................................................43
4.3.1 Physical Characterization of Nanocrystalline NaY.....................................43
4.3.2 Characterization of Nanocrystalline Zeolite with FT-IR Spectroscopy ......45
4.3.3 Sorption of Probe Molecules in Nanocrystalline NaY Zeolite....................48
4.4 Adsorption of Other Small Molecules and Some Potential Environmental
Applications of Nanocrystalline Zeolites as Adsorbents.......................................55
4.4.1 Adsorption of Acetone in Nanocrystalline NaY .........................................55
4.4.2 Adsorption of Ammonia in Nanocrystalline NaY.......................................57
4.4.3 Adsorption of Sulfur Dioxide in Nanocrystalline NaY Zeolite...................58
4.5 Conclusions ..........................................................................................................63
CHAPTER 5 CATALYTIC REDUCTION OF NO2 IN NANOCRYSTALLINE
NaY ZEOLITE (80 nm) ...............................................................................64
5.1 Introduction ..........................................................................................................64
5.2.1 Preparation of Nanocrystalline NaY Zeolite (Particle Size 80 nm) ............65
5.2.2 FT-IR Measurements...................................................................................66
5.3 Results ..................................................................................................................67
5.3.1 An Investigation of the Nature of the Surface Sites on
Nanocrystalline NaY with A Particle Size of 80 nm ...................................67
5.3.2 Adsorption and Thermal Reduction of NO2 in Nanocrystalline NaY .........71
5.3.3 Adsorption of NO2 and Propylene (1:1) Mixture in Nanocrystalline
NaY ..............................................................................................................75
5.3.4 Low Temperature Propylene-SCR of NO2 in Nanocrystalline NaY ...........80
5.4 Discussion.............................................................................................................84
5.4.1 Formation of NO in Thermal Reduction of NO2 .........................................84
5.4.2 Formation of NO and N2O in the Adsorption of NO2 and Propylene
(1:1) Mixture at 298 K .................................................................................86
5.4.3 Reaction Stoichiometry in Propylene-SCR of NO2 ....................................87
5.4.4 A Mechanism for Propylene-SCR of NO2 at Low Temperature.................88
5.5 Conclusions ..........................................................................................................90
CHAPTER 6 SELECTIVE CATALYTIC REDUCTION OF NO2 WITH
PROPYLENE IN NANOCRYSTALLINE NaY (23 nm) ...........................91
6.1 Introduction ..........................................................................................................91
6.3 Results and Discussion .........................................................................................93
6.3.1 Propylene-SCR of NO2 in NaY Zeolite at 473 K: Gas Phase Products ......93
6.3.2 Adsorption of NO2 in NaY zeolite at 298 K................................................95
6.3.3 Adsorption of NO2 and C3H6 Mixture in NaY Zeolite at 298 K .................98
6.3.4 Adsorbed Products Following Propylene-SCR of NO2 in NaY Zeolite
at 473 K ......................................................................................................102
6.4 Conclusions ........................................................................................................106
6.5 Acknowledgements.............................................................................................106
vii
CHAPTER 7 SELECTIVE CATALYTIC REDUCTION OF NO2 WITH UREA

IN NANOCRYSTALLINE NaY (23 nm) .................................................107
7.1 Introduction ........................................................................................................107
7.2 Experimental Section..........................................................................................108
7.3 Results ................................................................................................................109
7.3.1 Adsorption of NO2 in NaY Zeolite at 298 K .............................................109
7.3.2 Urea-SCR of NO2 in NaY Zeolite at 473 K: Products in the Gas
Phase ..........................................................................................................114
7.3.3 Adsorbed Products Following Urea-SCR of NO2 in NaY Zeolite at
473 K ..........................................................................................................122
7.4 Discussion...........................................................................................................126
7.4.1 Reactions Between Gas-Phase NO2 and Adsorbed Urea in NaY at
298 K ..........................................................................................................126
7.4.2 Thermal Decomposition of Urea in Nanocrystalline NaY Zeolite at
473 K..........................................................................................................128
7.4.3 Formation of N-N Bond in Urea-SCR of NO2 over Nanocrystalline
NaY Zeolite at 473 K .................................................................................130
7.4.4 Origin of Size Effect in Urea-SCR of NO2 over NaY Zeolite ..................132
7.4.5 Urea-SCR vs. Propylene-SCR of NO2 in Nanocrystalline NaY ...............135
7.5 Conclusions ........................................................................................................136
7.6 Acknowledgements.............................................................................................137
CHAPTER 8 CONCLUDING REMARKS AND FUTURE DIRECTIONS.................138
APPENDIX A INSTRUMENTAL PARAMETERS AND MACRO PROGRAMS
FOR FT-IR MEASUREMENTS ............................................................140
APPENDIX B CALCULATION OF THE CONCENTRATIONS OF GASES
USING FT-IR SPECTROSCOPY ..........................................................145
APPENDIX C ADSORPTION AND DESORPTION STUDIES OF
CYCLOHEXANE IN BaY USING A FLOW SYSTEM.......................149
REFERENCES ................................................................................................................155
viii
LIST OF TABLES
Table 3.1 Product distribution (%) obtained from GC analysis and conversion (%)
for the photooxidation of cyclohexane in BaY.........................................................27
Table 3.2 Product distribution (%) obtained from GC and GC-MS analysis and
conversion (%) for the photooxidation of cyclohexene in BaY. ..............................30
Table 3.3 Product distribution (percent of total products) for the other products
from Table 3.2...........................................................................................................30
Table 4.1 Some frequently used probe molecules. ...........................................................48
Table 4.2 Desorption kinetic data obtained by fitting the decay curves to a double
exponential function. ................................................................................................62
Table 5.1 Product distribution and conversion (%) of gas-phase NO2 and adsorbed
NO3- in the thermal reduction of NO2 in nanocrystalline NaY at different
temperatures for 6 h. .................................................................................................74
Table 5.2 Assignment of infrared absorption bands observed following adsorption
and thermal reaction of NO2 and propylene (1:1) mixture in nanocrystalline
NaY zeolite. ..............................................................................................................78
Table 5.3 Product distribution and conversion of gas-phase NO2 and adsorbed
NO3- in propylene-SCR of NO2 at different temperatures after 6 h of reaction
time. ..........................................................................................................................82
Table 5.4 Initial rates (10-2 mol L-1 s-1) of formation or loss of gas phase species
in first 10 min of SCR reactions. ..............................................................................84
Table 6.1 Concentration (mol L-1) of gas-phase components before and after
propylene-SCR of NO2 in 7 mg NaY zeolite at 473 K for 6 h. ................................93
Table 7.1 List of vibrational frequencies of nitrogen-containing gas-phase and
adsorbed species which showed an isotopic shift upon substitution of 15N for
14
N in the adsorbed urea ((NH2)2CO) precursor. ....................................................114
Table 7.2 Rates for formation of gas-phase NH3, CO2 and N2O and loss of gasphase NO in the first 10 min during urea-SCR of NO2 and thermal
decomposition of urea at 473 K. .............................................................................121
Table 7.3 Gas-phase product distribution after urea-SCR of NO2 and thermal
reaction of urea at 473 K for 2 h in 7 mg NaY zeolite. ..........................................122
Table B.1 Calculation of the conversion factor for gases. The conversion factors
are used to calculate the gas concentration from integrated absorbance in IR
spectra. ....................................................................................................................146
Table C.1 Quantification of cyclohexane adsorbed in BaY ...........................................151
ix
LIST OF FIGURES
Figure 1.1 The faujasite framework structure of zeolite Y: the supercage structure
(left) and the 12-ring viewed along (111) (right). In the right panel, the larger
cycles represent T atoms (Si or Al) linked to each other by shared oxygen
atoms (small cycles)....................................................................................................3
Figure 1.2 Hydroxyl groups in zeolite materials: lattice termination OH groups
(left) and bridging OH groups with Bronsted acidity (right)......................................5
Figure 1.3 The formation of a hydroxyl nest and a non-framework aluminum
species during steaming. .............................................................................................5
Figure 2.1 Sample preparation in trasmission FT-IR studies: a zeolite sample
coated on half of a tungsten grid (left) and the infrared sample cell (right). ............11
Figure 2.2 Diagram of the sample holder for FT-IR studies. A stainless steel
coolant system (left) with copper power leads welded to the end of a stainless
steel pipe is put inside of a stainless steel reducing cross (right). ............................11
Figure 2.3 Experimental setup for transmission FT-IR studies. .......................................13
Figure 2.4 Collection of FT-IR spectra: (a) the gas phase, (b) clean zeolite, (c)
zeolite with surface species and (d) surface species (difference spectrum). (d)
= (c)-(b).....................................................................................................................14
Figure 2.5 Optical path for photooxidation studies of hydrocarbons and FT-IR
measurements............................................................................................................15
Figure 2.6 An example of batch reactor studies that investigates the effect of
zeolite catalyst thickness on photooxidation of cyclohexane. ..................................16
Figure 2.7 Schematic diagram of the apparatus used in the adsorption/desorption
studies. ......................................................................................................................18
Figure 2.8 Schematic of the sample cell and heating portion of the TPD reactor. ...........19
Figure 3.1 Examples of selective photooxidation of hydrocarbons with molecular
oxygen in BaY zeolite...............................................................................................21
Figure 3.2 Proposed mechanism for the photooxidation of cyclohexane with O2 in
BaY. ..........................................................................................................................22
Figure 3.3 FT-IR spectra of cyclohexane adsorbed in BaY as a function of
cyclohexane pressure at 298 K. All spectra use the clean BaY zeolite prior to
adsorption as a reference. In addition, gas-phase absorptions have been
subtracted from each of the FT-IR spectra. ..............................................................25
Figure 3.4 The uptake of cyclohexane in BaY as a function of pressure. The
coverage is calibrated using the absorption band at 1451 cm-1. ...............................26
Figure 3.5 Difference FT-IR spectra after irradiation of cyclohexane and oxygen in
BaY with > 455 nm near room temperature for 20 min, 1 h, 2 h, 6 h and 22
h (bottom to top). Cyclohexane coverage in BaY is approximately 0.5
molecules per supercage. All spectra use the clean BaY zeolite prior to
Figure 3.6 FT-IR spectra of cyclohexene adsorbed in BaY as a function of
cyclohexene pressure 298 K. All spectra use the clean BaY zeolite prior to
subtracted from each of the FT-IR spectra. The absorption band at 1636 cm-1
is used to construct the isotherm of cyclohexene in BaY (inset). .............................28
Figure 3.7 Difference FT-IR spectra after irradiation of cyclohexene and oxygen in
BaY with > 455 nm near room temperature for 20 min, 1 h, 2 h, 6 h and 22
h (bottom to top). Cyclohexene coverage in BaY is approximately 1
molecule per supercage. All spectra use the clean BaY zeolite prior to
Figure 3.8 Proposed mechanism for the photooxidation of cyclohexene with O2 in
BaY. ..........................................................................................................................31
Figure 3.9 Oxidation of cyclohexene using TBHP: (a) FT-IR spectrum of
isobutane and oxygen adsorbed in BaY and (b) difference FT-IR spectrum of
isobutane and oxygen in BaY after irradiation at > 455 nm for 4 h. Spectra
were recorded at 298 K. All spectra use the clean BaY zeolite prior to
Figure 3.10 Proposed mechanism for the oxidation of cyclohexene with TBHP in
BaY. ..........................................................................................................................33
Figure 3.11 Difference FT-IR spectra following the thermal oxidation of
cyclohexene with TBHP generated in BaY for 20 min, 1 h, 2 h, 3 h, 4 h and
44 h (bottom to top). Spectra were recorded at 298 K. All spectra use the
clean BaY zeolite prior to adsorption as a reference. In addition, gas-phase
absorptions have been subtracted from each of the FT-IR spectra...........................34
Figure 4.1 Digital pictures of zeolite films prepared from: Aldrich NaY (left,
particle size ~1 m) and nanocrystalline NaY (right, particle size 468 nm)..........39
Figure 4.2 UV-Visible spectra of the zeolite films prepared from: (a)
nanocrystalline NaY (particle size 468 nm) and (b) Aldrich NaY (particle
size ~1 m). ..............................................................................................................40
Figure 4.3 Contact mode AFM image of zeolite thin film prepared from
nanocrystalline NaY with a particle size of 468 nm...............................................40
Figure 4.4 Percent conversion in photooxidation of cyclohexane in BaY (solid
cycle) as a function of catalyst thickness. The thickness is linearly
proportional to the mass of BaY sample. The percent conversion in thermal
oxidation of cyclohexane is also shown (open cycle)...............................................41
xi
Figure 4.5 Representative SEM images of synthesized nanocrystalline NaY with a

particle size of 23 nm (scale bar = 100 nm)..............................................................44
Figure 4.6 Representative SEM images of commercial (Aldrich) NaY with a
particle size of ~1 m (scale bar = 1 m).................................................................45
Figure 4.7 FT-IR spectra of surface hydroxyl groups in: (a) nanocrystalline NaY
and (b) Aldrich NaY zeolite after heating to 623 K overnight under vacuum.
The blank grid is used as a reference. All spectra were recorded at 298 K.............46
Figure 4.8 FT-IR spectra of: (a) nanocrystalline NaY and (b) Aldrich NaY zeolite
in the range between 400 cm-1 and 1300 cm-1 showing framework vibrations.
Pure KBr was used as a reference. All spectra were recorded at 298 K. .................47
Figure 4.9 FT-IR spectra of 0.5 Torr NO2 adsorbed in: (a) nanocrystalline NaY and
(b) Aldrich NaY zeolite at 298 K. All spectra use the corresponding clean
NaY zeolite prior to adsorption as a reference. In addition, gas-phase
absorptions have been subtracted from each of the FT-IR spectra. ..........................49
Figure 4.10 FT-IR spectra of pyridine molecules adsorbed in: (a) nanocrystalline
NaY and (b) Aldrich NaY at 298 K following pyridine desorption at 473 K.
Spectra were recorded at 298 K. All spectra use the corresponding clean
zeolite prior to adsorption as a reference. .................................................................51
Figure 4.11 FT-IR spectra of propylene adsorbed in nanocrystalline NaY zeolite at
298 K as a function of propylene pressure (P = 0.1, 0.5, 1.0, 2.0 and 6.0 Torr).
The inset shows the difference spectrum (in the O-H stretching region) of
nanocrystalline NaY following the adsorption of 6.0 Torr propylene. All
spectra use the clean nanocrystalline NaY zeolite prior to adsorption as a
reference. In addition, gas-phase absorptions have been subtracted from each
of the FT-IR spectra..................................................................................................52
Figure 4.12 Difference FT-IR spectra following: (a) pyridine and (b) substituted
pyridine (DTBPy) adsorption in nanocrystalline NaY. All spectra use the
clean nanocrystalline NaY zeolite prior to adsorption as a reference. The
negative peaks show the loss of hydroxyl groups upon pyridine (or DTBPy)
adsorption in nanocrystalline NaY. ..........................................................................54
Figure 4.13 FT-IR spectra of acetone adsorbed in nanocrystalline NaY at 298 K as
a function of acetone pressure (P = 0.005, 0.011, 0.021, 0.055, 0.102 and
0.306 Torr). All spectra use the clean nanocrystalline NaY zeolite prior to
Figure 4.14 FT-IR spectra of ammonia adsorbed in nanocrystalline NaY at 298 K
as a function of ammonia pressure (P = 0.05, 0.1, 0.2, 0.5 and 1 Torr). The
inset displays the difference FT-IR spectrum in the O-H/N-H region
following adsorption of 1 Torr NH3 in nanocrystalline NaY at 298 K. The
FT-IR spectrum of 1 Torr NH3 and 1 Torr H2O mixture adsorbed in
nanocrystalline NaY at 298 K is also shown (in green). All spectra use the
clean nanocrystalline NaY zeolite prior to adsorption as a reference. In
addition, gas-phase absorptions have been subtracted from each of the FT-IR
spectra. ......................................................................................................................57
xii
Figure 4.15 FT-IR spectra of SO2 adsorbed in nanocrystalline NaY at 298 K as a

function of SO2 pressure (P = 0.05, 0.1, 0.2, 0.5, 1 Torr). All spectra use the
clean nanocrystalline NaY zeolite prior to adsorption as a reference. In
addition, gas-phase absorptions have been subtracted from each of the FT-IR
spectra. ......................................................................................................................59
Figure 4.16 FT-IR spectra of 1 Torr SO2 and 1 Torr H2O mixture adsorbed in
nanocrystalline NaY at 298 K as a function of pumping time (t = 0, 1, 3, 10,
20, 120 min). All spectra use the clean nanocrystalline NaY zeolite prior to
adsorption as a reference...........................................................................................60
Figure 4.17 Decay curves of surface adsorbed sulfur species formed in the absence
(black circle) and presence (green circle) of water during evacuation of
nanocrystalline NaY at room temperature. The integrated areas were
obtained in the spectral range from 1240 cm-1 to 1365 cm-1. The decay curves
were further fitted with a double exponential function (see text for further
details). Spectra were recorded every 1 min. ...........................................................61
Figure 5.1 Nanocrystalline NaY zeolite with a particle size of 8030 nm: a
representative SEM image (left) and the particle size distribution (right). ..............66
Figure 5.2 FT-IR spectra of surface hydroxyl groups in nanocrystalline NaY
zeolite after heating to: (a) 573 K and (b) 673 K overnight under vacuum.
Spectra were recorded at 298 K. The bland grid is used as a reference. .................67
Figure 5.3 FT-IR spectra of propylene adsorbed in nanocrystalline NaY zeolite at
298 K as a function of propylene pressure (P = 0.050, 0.493, 1.019, 2.028 and
4.061 Torr). All spectra use the clean nanocrystalline NaY zeolite prior to
Figure 5.4 FT-IR spectra of pyridine molecules adsorbed in nanocrystalline NaY
zeolite at 298 K following pyridine desorption at: (a) 298 K, (b) 473 K and
(c) 573 K. Spectra were recorded at 298 K. All spectra use the clean
nanocrystalline NaY zeolite prior to adsorption as a reference. The inset
shows the spectral region extending from 3635 to 3760 cm-1. .................................70
Figure 5.5 FT-IR spectra of NO2 adsorbed in nanocrystalline NaY zeolite at 298 K
as a function of NO2 pressure (P = 0.055, 0.107, 0.498 and 1.033 Torr).
Spectra were recorded at 298 K. All spectra use the clean nanocrystalline
Figure 5.6 Time course concentration of gas-phase NO2, NO and N2O during the
thermal reduction of NO2 in nanocrystalline NaY at 373 K. ....................................73
Figure 5.7 FT-IR spectra of NO2 adsorbed in nanocrystalline NaY zeolite after
adsorption at: (a) 298 K followed by evacuating at (b) 298 K, (c) 373 K, (d)
473 K and (e) 573 K for 30 min. Spectra were recorded at 298 K. All spectra
use the clean nanocrystalline NaY zeolite prior to adsorption as a reference.
In addition, gas-phase absorptions have been subtracted from each of the FTIR spectra. .................................................................................................................75
xiii
Figure 5.8 FT-IR spectra of NO2 and propylene (1:1) mixture adsorbed in
nanocrystalline NaY zeolite at 298 K recorded as a function of increasing
total pressure (P = 0.100, 0.200, 0.623, 1.005 Torr). All spectra use the clean
nanocrystalline NaY zeolite prior to adsorption as a reference. In addition,
gas-phase absorptions have been subtracted from each of the FT-IR spectra. .........76
Figure 5.9 FT-IR spectra of adsorbed species following adsorption of an NO2 and
propylene (1:1) mixture at: (a) 298 K and heating to (b) 373 K and (c) 473 K.
Spectra were recorded at 298 K. All spectra use the clean nanocrystalline
Figure 5.10 FT-IR spectra of the gas phase following propylene-SCR of NO2 in
nanocrystalline NaY at 473 K for (a) 0 min, (b) 10 min, (c) 30 min, (d) 1 h
and (e) 4 h. ................................................................................................................80
Figure 5.11 Time course concentration of gaseous N-containing species during
propylene-SCR of NO2 in nanocrystalline NaY at 473 K. The inset shows the
decay and evolution of propylene and carbon monoxide, respectively. ...................81
Figure 6.1 FT-IR spectra of 0.5 Torr NO2 adsorbed in nanocrystalline NaY and
commercial NaY zeolite in the presence and absence of 10 mol adsorbed
water at 298 K. All spectra use the corresponding clean NaY zeolite prior to
adsorption as a background. In addition, gas-phase absorptions have been
Figure 6.2 FT-IR spectra of nanocrystalline NaY and commercial NaY zeolite in
the presence and absence of adsorbed water at 298 K following the adsorption
of NO2 and C3H6 (1:1) mixture. All spectra use the corresponding clean NaY
zeolite prior to adsorption as a background. In addition, gas-phase
Figure 6.3 The integrated absorbance of surface NO3- on EFAL sites (), NO+ ()
in nanocrystalline NaY zeolite and NO+ () in commercial NaY zeolite as a
function of time. The changes of the integrated absorbance are shown in the
process of NO2 adsorption, evacuation of the IR cell and exposure of
propylene to the surface species in the zeolites. Spectra used for band
integration were recorded every 5 seconds. The integrated absorbance of
surface NO3- on EFAL sites in nanocrystalline NaY zeolite is 20 times larger
than the original absorbance. ..................................................................................101
Figure 6.4 FT-IR spectra of adsorbed species following propylene-SCR of NO2 at
473 K for 6 h in nanocrystalline NaY and commercial NaY zeolite in the
presence and absence of adsorbed water. The spectra were recorded at 298 K.
All spectra use the corresponding clean NaY zeolite prior to adsorption as a
background. In addition, gas-phase absorptions have been subtracted from
each of the FT-IR spectra........................................................................................103
Figure 7.1 FT-IR spectrum of commercial NaY with adsorbed urea at 298 K. The
zeolite sample with adsorbed urea was dried in air and evacuated at 298 K
overnight. The FT-IR spectrum of a clean commercial NaY sample is also
plotted (dotted line). The blank grid is used as a reference. All spectra were
recorded at 298 K....................................................................................................110
xiv
Figure 7.2 Difference FT-IR spectra of nanocrystalline NaY with adsorbed urea
and commercial NaY zeolite with adsorbed urea following adsorption of 0.5
Torr NO2 at 298 K. The spectra use the corresponding NaY zeolite with
adsorbed urea prior to NO2 adsorption as a background. In addition, gasphase absorptions have been subtracted from each of the FT-IR spectra. ..............111
Figure 7.3 The FT-IR spectra of the gas phase upon adsorption of NO2 in: (a)
commercial NaY zeolite with adsorbed 15N-urea, (b) nanocrystalline NaY
zeolite with adsorbed 15N-urea, (c) commercial NaY zeolite with adsorbed
unlabeled urea, and (d) nanocrystalline NaY zeolite with adsorbed unlabeled
urea. All spectra were recorded at 298 K after adsorption equilibrium. ................113
Figure 7.4 FT-IR spectra of the gas phase as a function of reaction time in 15N-urea
SCR of NO2 over nanocrystalline NaY zeolite. The spectra shown were
recorded at 473 K and at t (reaction time) = 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,
12, 13, 14, 15, 16, 17, 18, 19, 21, 23, 25, 27, 30 min, respectively. The blank
grid is used as a reference. ......................................................................................115
Figure 7.5 FT-IR spectra of the gas phase as a function of reaction time in ureaSCR of NO2 over commercial NaY zeolite. The spectra shown were recorded
at 473 K and at t (reaction time) = 0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 18,
20, 25, 30, 35, 40, 50, 60, 70, 80, 90, 120 min, respectively. The blank grid is
used as a reference. .................................................................................................117
Figure 7.6 Evolution of gas-phase (a) NH3, (b) CO2 and (c) H2O during urea-SCR
of NO2 in nanocrystalline NaY zeolite (), urea-SCR of NO2 in commercial
NaY zeolite () and thermal decomposition of urea in commercial NaY
zeolite () at 473 K. ................................................................................................118
Figure 7.7 Evolution of gas-phase (a) NO and (b) N2O during urea-SCR of NO2 in
nanocrystalline NaY zeolite () and in commercial NaY zeolite () at 473 K.......120
Figure 7.8 Difference FT-IR spectra of zeolite samples following: (a) 15N-urea
SCR of NO2 in nanocrystalline NaY for 30 min, (b) urea-SCR of NO2 in
nanocrystalline NaY for 2 h, (c) urea-SCR of NO2 in commercial NaY for 2 h
and (d) thermal decomposition of urea in commercial NaY for 2 h at 473
(from top to bottom). The spectra were recorded at 298 K. All spectra shown
use the corresponding NaY zeolite with adsorbed urea prior to NO2
adsorption as a background. In addition, gas-phase absorptions have been
subtracted from each of the FT-IR spectra shown. .................................................123
Figure 7.9 Thermal decomposition of urea on silanol (SiOH) and EFAL (AlOx)
sites in nanocrystalline NaY zeolite........................................................................129
Figure 7.10 15N-urea SCR of NO2 on silanol (SiOH) and EFAL (AlOx) sites in
nanocrystalline NaY zeolite. ...................................................................................132
Figure 7.11 A nanocrystalline zeolite particle with a particle size of 23 nm. The
internal surface of nanocrystalline zeolite provides sites for NOx- storage (as
NO2- and NO3-) and the minority of the SCR reactions, the external surface
provides sites for additional NOx storage (as NO3-) and the majority of the
SCR reactions. ........................................................................................................135
Figure B.1 FT-IR spectra of gas-phase NO2, NO and N2O. ...........................................146
xv
Figure B.2 FT-IR spectra of gas-phase C3H6, CO2 and CO............................................147

Figure B.3 FT-IR spectra of gas-phase NH3, SO2 and H2O............................................147
Figure B.4 FT-IR spectrum of gas phase in a sample that contains N2O and CO. .........148
Figure C.1 FID signal following six pulses of cyclohexane in the absence (dotted
line) and presence (solid green line) of BaY zeolite...............................................150
Figure C.2 TPD profiles of cyclohexane in BaY at different heating rates (He flow
rate 125 mL min-1, cyclohexane coverage 0.21).....................................................152
Figure C.3 Determination of activation energy of desorption (Ed) for cyclohexane
in BaY. The data are linearly fitted to calculate Ed (see text for details)...............153
Figure C.4 TPD profiles of cyclohexane in BaY at different Helium flow rates
(heating rate 15 K min-1, cyclohexane coverage 0.21). ..........................................154
Figure C.5 TPD profiles of cyclohexane in BaY at different cyclohexane coverage
(He flow rate 125 mL min-1, heating rate 15 K min-1). .................................154
xvi
CHAPTER 1
INTRODUCTION
Zeolites are crystalline aluminosilicates with three-dimensional framework
structures that form pores with uniform sizes of molecular dimensions [1]. In 1756, Axel
F. Cronstedt named hydrated aluminosilicate minerals zeolites after the Greek zein (to
boil) and lithos (stone). For the next 200 years, research efforts on zeolites were sparse in
part due to the limited availability of material. Richard M. Barrers early pioneering
work in the 1930s began the era of synthetic zeolites [2]. In the late 1950s faujasite
(FAU)-type zeolite was used in oil refining, which marked the first commercial success
of synthetic zeolite materials. Research activities on zeolites have escalated since the
1960s.
Today synthetic zeolites are used widely throughout the petroleum refining and
chemical process industries as selective adsorbents, catalysts and ion exchangers [2-5].
Driven by the increasing concerns related to environmental problems, many recent
studies used zeolites as catalysts in pollution abatement [6]. This chapter provides a brief
introduction to zeolite materials including nanocrystalline zeolites, characterization of
zeolites with infrared spectroscopy and environmental catalysis using zeolites.
1.1 Zeolite Materials
Structurally, zeolites are inorganic materials based on an infinitely extending
four-connected framework of SiO4 and AlO4 tetrahedra that are linked to each other by
shared oxygen atoms [2]. Each AlO4 tetrahedron in the framework bears a net negative
charge that is balanced by an extra framework cation. The structural formula of a zeolite
is based on the crystallographic unit cell, the smallest unit of structure, represented by
Mx/n [(AlO2)x(SiO2)y]wH2O
where n is the valence of cation M, w is the number of water molecules per unit cell, x
and y are the total number of tetrahedral atoms per unit cell. The y/x ratio (Si/Al ratio)
usually ranges from 1 to 5 or 10 to 100 for high silica zeolites [2]. The framework
structure contains channels or interconnected voids that are occupied by the cations and
water molecules. The cations are mobile and ordinarily undergo ion exchange. The
water may be removed reversibly, generally by application of heat [2].
The special structures of zeolite materials contribute to their unique properties and
applications. For example, the microporous nature of zeolites with uniform pores of
molecular dimensions allows certain hydrocarbon molecules to enter the crystals while
other molecules are rejected based on the molecular size. This makes the phenomenon of
shape selectivity possible in catalytic reactions [7]. The ion exchange properties of
zeolite materials result from the exchangeable cations and lead to many different
applications, such as water softening. The high surface area and high thermal stability of
zeolites make them desirable materials for a wide range of applications including
adsorbents, detergents and catalysts [1].
Zeolite Y has an FAU-type structure and is widely used in petroleum refining [8,
9]. The basic structural units for Y zeolites are sodalite cages, which are arranged so as
to form supercages that are cage-like voids interconnected to each other by 12-ring
openings. Figure 1.1 shows the supercage structure of zeolite Y (left) and the 12-ring
pore opening (right) [10]. The diameter of the supercages and the size of the 12-ring
opening of Y zeolites are ~12 and ~7.4 , respectively. The combination of large void
volume (ca. 50%), 12-ring pore openings and 3-dimensional channel system make zeolite
Y ideal for many catalytic applications [8, 11].
Figure 1.1 The faujasite framework structure of zeolite Y: the supercage structure (left)
and the 12-ring viewed along (111) (right). In the right panel, the larger cycles represent
T atoms (Si or Al) linked to each other by shared oxygen atoms (small cycles).
Nanocrystalline zeolites are zeolites with discrete, uniform crystals with

dimensions of less than 100 nm that have unique properties relative to conventional
micrometer-sized zeolite crystals [12, 13]. Nanocrystalline zeolites have significantly
higher external surface areas and reduced diffusion path lengths relative to micrometersized zeolites. The external surface area of a nanocrystalline zeolite with a particle size
of less than 50 nm is comparable to its internal surface area. For example,
nanocrystalline zeolite Y with a particle size of 23 nm has an external surface area of 178
m2 g-1, which accounts for ~30% of its total surface area. The relatively high external
surface areas of nanocrystalline zeolites make them potential materials for bifunctional
catalysis using both the external and internal surfaces and are desirable for their
applications as adsorbents and catalysts [12-14].
1.2 Characterization of Zeolites with Infrared
Spectroscopy
Various characterization techniques have been employed to obtain information
about the structural, chemical and catalytic characteristics of zeolite materials [15]. For
example, powder X-Ray diffraction (XRD) is frequently used to assess the crystallinity
and verify the identity of zeolites. Electron microscopy is used to determine particle
morphology and particle size. N2 adsorption isotherms are used to measure surface area
and solid state nuclear magnetic resonance (NMR) is used to investigate structural
properties of the zeolite framework. Temperature programmed desorption (TPD) is used
to gain insight into the adsorption/desorption of adsorbates in zeolites.
In this thesis, the characterization of zeolites with transmission Fourier transform
infrared spectroscopy (FT-IR) is described. FT-IR spectroscopy is used to probe the
structure of zeolites and monitor reactions in zeolite pores. Specifically, structural
information can be obtained from the vibrational frequencies of the zeolite lattice
observed in the range between 200 and 1500 cm-1 [15]. In the O-H stretching region,
infrared spectra of zeolites provide a wealth of information on hydroxyl groups attached
to zeolite structures. The hydroxyl groups are important for the chemistry of zeolite
materials [16, 17]. At least five types of hydroxyl groups are present in zeolite Y,
including (i) lattice termination silanol groups (~3745 cm-1), (ii) hydroxyl groups
occurring at defect sites, i.e. hydroxyl nests (~3720 cm-1), (iii) OH groups attached to
cations which compensate the negative charge of the framework (~3695 cm-1), (iv) OH
groups attached to extra framework aluminum (EFAL) species (~3655 cm-1), and (v) for
zeolite Y in the H-form, the bridging OH groups with Bronsted acidity (~3630 and 3560
cm-1) [16].
Lattice termination silanol groups are mainly located on the external surface
(Figure 1.2 left). Thus, terminal silanol groups are especially important for
nanocrystalline zeolites that have relatively large external surface areas. The relative
intensity of the absorption band at ~3745 cm-1 depends on the primary size of the zeolite
crystals [15]. For zeolites in the H-form, the hydroxyl groups bridging a Si and an Al
atom possess strong Bronsted acid properties (Figure 1.2 right).
H
H O
Si
O O
Si
O O
Si
O O
Si
O O
Al
O O
Si
O O
Figure 1.2 Hydroxyl groups in zeolite materials: lattice termination OH groups (left) and
bridging OH groups with Bronsted acidity (right).
Hydroxyl nests can be produced in the process of steaming (heating to high

temperatures in the presence of water vapor), as described by the equation shown in
Figure 1.3 [18, 19]. The removal of framework aluminum also occurs during steaming,
forming non-framework aluminum species such as Al(OH)3.
Si
Si
Si
O H+
O Al O
O
Si
+ 3H2O
H
Si
O
Si
O
H
Si + Al(OH)3
O
Si
Figure 1.3 The formation of a hydroxyl nest and a non-framework aluminum species
during steaming.
With longer reaction time and at high temperatures, the non-framework aluminum
species can be transformed into a complex variety of species with Lewis acidity, which
are mainly classified as condensed and noncondensed EFAL species [3, 20]. The
condensed species as Al2O3 phases form mostly on the external surface of the zeolites
[21, 22], while the noncondensed species are normally associated with aluminum and
oxyaluminum cations (AlO+, Al(OH)2+, Al3+) [18].
Functional groups of Y zeolites including surface acid sites as well as other active
sites of catalytic importance can be further detected by the adsorption of probe molecules
coupled with infrared analysis. For example, cation sites generated in the ion exchange
process can be probed by NO or NO2 [23, 24]. The type (Bronsted or Lewis acidity),
concentration and accessibility of surface acid sites can be determined by the adsorption
of ammonia, pyridine or substituted pyridine [25]. In addition to the functional groups of
zeolites, the chemical and physical interactions of molecules with the zeolite framework
should always be taken into consideration regarding the activity and selectivity of zeolite
catalysts [26].
1.3 Environmental Catalysis Using Zeolites
Zeolite catalysts have been used in industry for many years [4, 9]. Large
quantities of the FAU-type zeolite catalysts are used in oil refineries to manufacture
gasoline from crude oil in the fluidized catalytic cracking (FCC) process. In this thesis,
the applications of Y zeolites, including nanocrystalline Y zeolites, in environmental
catalysis are investigated. Environmental catalysis encompasses the study of catalysis
and catalytic reactions that impact the environment [27, 28]. Recent studies in
environmental catalysis emphasize understanding global catalytic processes in natural
systems and applying catalysis in the development of green chemical processes and in
environmental remediation [27].
Organic synthesis is essential to the production of many chemical compounds
including pharmaceuticals. The selective production of one molecular species out of
many other thermodynamically feasible product molecules is highly favored in green
chemical processes [29, 30]. Photooxidation of hydrocarbons using zeolite catalysts
combines several key aspects to be an environmentally benign catalyst-based chemical
process [31, 32]. For example, visible light instead of UV irradiation is used to initiate
the selective oxidation and control further selective steps in the reaction. Zeolite pores
are used as reaction vessels that possess inherent selectivity in terms of their molecular
dimensions. In addition, these reactions are all done in the gas phase thereby eliminating
the use of organic solvents.
The treatment of gaseous emissions from potentially noxious compounds is
among the most studied topics in environmental catalysis. This involves the abatement of
VOCs (volatile organic compounds), NOx (NO+NO2), SOx (SO2+SO3) and greenhouse
gases [6]. Efficient emission control systems are necessary to meet the increasingly
stringent environmental regulations. One of the most urgent problems is the removal of
NOx emitted from stationary and mobile sources. NOx is typically produced during hightemperature combustion and contributes to the formation of smog and ground-level ozone
[33]. In the past decade, the selective catalytic reduction (SCR) of NOx using zeolitebased catalysts has been intensively studied [33]. Currently, SCR of NOx using NH3 as
the reducing agent is used worldwide for controlling NOx emissions from stationary
sources [34]. However, NH3 is not a practical reducing agent for NOx emissions from
mobile sources due to its toxicity and difficulties in its storage, transportation and
handling [33]. Hydrocarbons (such as propylene) and ammonia precursors (such as urea)
have been investigated as alternative reducing agents in deNOx reactions using zeolite
catalysts [33].
1.4 Thesis Overview and Objectives
The experimental methods used in this thesis, including transmission FT-IR
spectroscopy, gas chromatography (GC), GC-MS analysis and a flow system for
adsorption/desorption studies, are described in Chapter 2. In Chapter 3, selective
photooxidation of hydrocarbons using molecular oxygen in zeolite catalysts is
investigated as an environmentally benign alternative to conventional liquid phase
oxidation of industrially important chemicals [31, 35-37]. In situ transmission FT-IR
spectroscopy and ex situ GC and GC-MS analysis are employed to study photooxidation
of cyclohexane and cyclohexene in BaY zeolite. The selectivity to the formation of

partially oxidized products in photooxidation is examined as a function of reactant,
oxidant and irradiation wavelength [38].
A major objective of this thesis research is to explore the chemically and
physically distinct properties and environmental applications of nanocrystalline zeolites.
In Chapter 4, nanocrystalline zeolite materials are characterized with a variety of
techniques, including microscopy, FT-IR spectroscopy and sorption of probe molecules
[14]. Nanocrystalline NaY zeolite is also examined as an adsorbent for small molecules,
such as acetone and sulfur dioxide (Chapter 4). In Chapter 5, thermal reduction and
propylene-SCR of NO2 at low temperatures (T 473 K) using nanocrystalline NaY with
a particle size of 8030 nm are investigated in the absence of oxygen [39].
In Chapters 6 and 7, selective catalytic reduction of NO2 is studied in the presence
of oxygen using propylene and urea as the reducing agents, respectively. Nanocrystalline
NaY zeolite with a particle size of 23 nm and a relatively large external surface area is
used as the deNOx catalyst. Kinetic analysis and isotopic studies are conducted to help to
elucidate the reaction mechanisms [40, 41]. The performance of commercial
micrometer-sized NaY in SCR reactions is also studied in Chapters 6 and 7 for the
purpose of comparison. The effect of adsorbed water on the activity and selectivity in
propylene-SCR of NO2 is also investigated (Chapter 6). In SCR reactions, the sizegenerated differences in activity and selectivity are usually explained on the basis of
intracrystalline diffusion such that a faster diffusion of reactants and/or products through
the small crystals prevents side reactions from occurring [42, 43]. In this thesis, the
enhancement of activity and selectivity in SCR reactions using nanocrystalline zeolite
catalysts are explored based on the physically and chemically distinct properties of the
external surface.
The overall conclusions on characterization and environmental applications of
zeolites are summarized in Chapter 8. Appendix A lists details regarding the
instrumental parameters of FT-IR measurements and macro programs for automatic

scanning. In Appendix B, the calculation of gas concentration from the infrared spectra
of the gases is described. Appendix C provides a brief introduction to the adsorption and
desorption studies of cyclohexane in BaY using a flow system.
10
CHAPTER 2
EXPERIMENTAL METHODS
In this thesis, the characterization and environmental applications of zeolites are
investigated using a variety of different techniques. This chapter provides details about
the experimental methods used in this research. Transmission FT-IR spectroscopy is
used to a large extent in this research to characterize the zeolite materials and to follow
catalytic reactions. The home-built apparatus used for these FT-IR studies is described in
this chapter. In addition to transmission FT-IR spectroscopy, several other techniques
including GC, GC-MS, BET measurements and microscopy are also briefly introduced.
2.1 Transmission FT-IR Spectroscopy Experiments
2.1.1 Sample Preparation for FT-IR Studies
A desired amount of the zeolite mixed with water was coated onto a 3 cm 2 cm
tungsten grid held in place by nickel jaws (Figure 2.1 left). The nickel jaws are attached
to copper leads so that the sample can be resistively heated. A thermocouple wire is spotwelded to the tungsten grid to measure the temperature of the sample. The zeolite sample
is usually dried at room temperature before being placed in a stainless steel cube with a
volume of 330 mL (Figure 2.1 right). The stainless steel cube is outfitted with two BaF2
windows for infrared measurements (vide infra) and is connected to a vacuum/gas
handling system.
11
Copper power leads
Sample holder
Nickel jaws
To gas
handling
system
BaF2 window
Zeolite sample
Tungsten grid
Stainless
steel cube
Figure 2.1 Sample preparation in trasmission FT-IR studies: a zeolite sample coated on
half of a tungsten grid (left) and the infrared sample cell (right).
Coolant out
3/8 stainless
steel tubing
Coolant in
1 1/2 OD
stainless
steel pipe
2 3/4 conflat
flange
Dividing plate
Copper
power leads
1 OD
stainless
steel pipe
Reducing
cross 2 3/4
to 1 1/3
Thermocouple
wires
Insulated flexible
copper braid
Figure 2.2 Diagram of the sample holder for FT-IR studies. A stainless steel coolant
system (left) with copper power leads welded to the end of a stainless steel pipe is put
inside of a stainless steel reducing cross (right).
12
The nickel jaw assembly is attached to a sample holder through the copper leads.
The structure of the sample holder is shown in Figure 2.2. The entire reducing cross
(Figure 2.2 right) is then place the infrared cell such that the tungsten grid with zeolite
sample can be held in place inside the stainless steel cube. The zeolite sample can be
cooled by the stainless coolant system, or heated to an elevated temperature by running a
current across the copper leads. Unless otherwise noted, the zeolite sample inside the IR
cell is usually pretreated by gradually heating under vacuum to 623 K or higher
temperatures overnight to remove adsorbed water.
2.1.2 Experimental Setup for Transmission FT-IR
Studies
The stainless steel IR cell is placed inside the sample compartment of a Mattson
Galaxy 6000 infrared spectrometer equipped with a narrowband MCT detector. A PC
with commercial software (WinFirst) is used to record the FT-IR spectra collected by the
spectrometer. Figure 2.3 shows a schematic representation of the experimental setup for
transmission FT-IR studies. As can be seen from Figure 2.3, the IR cell is put in the dry
air purge area and is connected to a vacuum/gas handling system. Gases (or vapors of
liquids) can be introduced into the IR cell through a premix chamber (volume 820 mL).
An absolute pressure transducer is used to monitor the pressure in the premix chamber.
The IR cell and the gas handling system can be differentially pumped under a vacuum of
10-3 Torr.
13
MCT detector
Dry air purge area

BaF2 window
IR
Reflecting
mirror
Gases
Pressure
transducer
Premix chamber
Liquids
Turbo/Mechanical pump
Figure 2.3 Experimental setup for transmission FT-IR studies.
2.1.3 Protocol of In Situ FT-IR Measurements

In Figure 2.3, the IR cell is held in place by a linear translator inside the sample
compartment of the infrared spectrometer. The linear translator allows each half of the
tungsten grid to be translated into the infrared beam. This permits the in situ detection of
gas-phase reactants, intermediates and products (Figure 2.4 (a)), and the zeolite sample
(Figure 2.4 (b)) or the zeolite sample with adsorbates (Figure 2.4 (c)) under identical
reaction conditions. A difference spectrum containing information of adsorbed species
(Figure 2.4 (d)) can be obtained by subtracting the spectrum of clean zeolite (Figure 2.4
(b)) from the spectrum of zeolite with adsorbed species (Figure 2.4 (c)). Each spectrum
was obtained by averaging 64 scans at an instrument resolution of 4 cm-1. Each
absorbance spectrum represents a single beam scan referenced to the appropriate single
beam scan of the clean zeolite or the blank grid, unless otherwise noted. Time course
14
experiments can be done by automatically recording infrared spectra of the zeolite sample
or gas phase at a fixed time interval. Details regarding the instrumental parameters of
FT-IR measurements and macro programs for automatic scanning are listed in Appendix
A.
(b) Clean zeolite

A
b
s
o
r
b
a
n
c
e
A
b
s
o
r
b
a
n
c
e
4000
3500
3000
2500
2000
1500
1000
(c) Zeolite with

surface species
4000
3500
3000
2500
2000
(d) Surface species

A
b
s
o
r
b
a
n
c
e
1500
Wavenumber (cm-1 )
Wavenumber (cm-1 )
1000
4000
3500
3000
2500
1500
1000
IR
(a) Gas phase

A
b
s
o
r
b
a
n
c
e
4000
2000
Wavenumber (cm-1 )
Linear translator
3500
3000
2500
2000
1500
1000
Wavenumber (cm-1 )
Figure 2.4 Collection of FT-IR spectra: (a) the gas phase, (b) clean zeolite, (c) zeolite
with surface species and (d) surface species (difference spectrum). (d) = (c)-(b).
2.1.4 Potassium Bromide Pellets Technique

KBr pressed-pellet technique is used to obtain structural information of the zeolite
lattice. A finely divided zeolite sample is intimately mixed with KBr powder and then is
pressed in a die to from a transparent pellet. The transmission FT-IR spectrum of the
pellet in the range between 400 and 1500 cm-1 is collected at room temperature on an
Infinity Gold FT-IR spectrometer equipped with a DTGS Mid-IR detector. The FT-IR
spectrum of a pure KBr pellet is used as a reference.
15
2.2 Experimental Setup for Photooxidation Studies

2.2.1 Photooxidation Studies of Hydrocarbons
Followed by FT-IR Spectroscopy
A schematic of the optical setup used in studies of the photooxidation of
hydrocarbons in zeolites is shown in Figure 2.5. In these photooxidation studies of
hydrocarbons, a 500-W mercury arc lamp (Oriel Corp) with a water filter is used as the
light source. A broadband long pass filter is placed in front of the lamp. The broadband
light is then turned by a 1-inch quartz prism onto the zeolite sample. The quartz prism is
mounted inside the sample compartment of the FT-IR spectrometer so that the dry air
purge is not broken during irradiation. The infrared sample cell is connected to the
vacuum/gas handling system described in Figure 2.3. The blank grid (gas phase) or
zeolite sample can be moved into the infrared beam by the linear translator to record in
situ infrared spectra at different reaction stages.
Dry air purge area
MCT detector
Al foil
Quartz prism
IR
Long pass
filter
Linear
translator
Mercury
lamp
Figure 2.5 Optical path for photooxidation studies of hydrocarbons and FT-IR
measurements.
16
2.2.2 Batch Reactor Studies

Figure 2.6 describes the batch reactions regarding how the thickness of zeolite
catalyst influences the photooxidation of cyclohexane using molecular oxygen. Batch
reactor experiments are designed as a convenient approach to investigate some factors
that may influence the activity, selectivity and kinetics in photooxidation of
hydrocarbons. In batch reactions, zeolite samples of different weight are first mixed with
water in a Branson 2210 Ultrasonic Cleaner. The zeolite hydrosols are then transferred
into quartz vials of same sizes. Zeolite samples deposit on the bottom of the vials after
sitting still overnight. The water in the vials is evaporated at 353 K under atmospheric
pressure to form zeolite thin films on the botton of the vials. The zeolite samples are
prepared carefully so that the thickness of zeolite films is approximately proportional to
the weight of the zeolite samples.
25 mg
50 mg
100 mg 150 mg 200 mg 300 mg
Dry at 353 K
Dehydrate at 623 K
Zeolite hydrosol
Load reactants at r.t.
GC analysis
Product extracted
> 400 nm
using CH3CN
Cover with Aluminum foil

for thermal oxidation
Figure 2.6 An example of batch reactor studies that investigates the effect of zeolite
catalyst thickness on photooxidation of cyclohexane.
17
Reactant gases such as cyclohexane and oxygen are loaded in the vials after the
zeolite samples are degassed under vacuum at 623 K for 18 h. Photooxidation is then
carried out by irradiating the zeolite samples at > 400 nm. A control experiment is
done by covering one of the vials with aluminum foil. This is to determine whether the
thermal oxidation is significant in photooxidation of hydrocarbons since the temperature
of zeolite samples slightly increases during the irradiation.
2.2.3 GC and GC-MS Analysis
In these photooxidation studies of hydrocarbons done in the IR cell and in the
batch reactor setup, ex situ GC and GC-MS analysis are conducted to obtain information
on products trapped in zeolite catalysts. Products are extracted from the zeolite samples
in acetonitrile for 90 min. The samples are then centrifuged for 2 min at 10,000 g and the
supernatants are analyzed by GC or GC-MS. The GC (Varian 3600) is equipped with an
FID detector and a 5% phenyl/95% methylpolysiloxane capillary column. In some
experiments GC-MS analysis is conducted, using a GC-MS (HP G1800 C) with a HP5MS column (cross linked 5% PH ME siloxane). When available, standards of the
products are injected separately to determine retention times and GC response factors.
2.3 A Flow System for Adsorption/Desorption Studies
Both quantitative and qualitative information regarding the adsorption/desorption
of hydrocarbons and other adsorbates in zeolites can be obtained with transmission FT-IR
spectroscopy. However, the FT-IR setup is designed in a way such that adsorption and
reactions occur in a closed system. In desorption experiments, FT-IR spectroscopy is
unable to monitor the desorbed species when the zeolite sample is pumped under
vacuum.
A flow system is designed and constructed to further understand the
adsorption/desorption and sometimes diffusion in zeolite materials. Figure 2.7 describes
the experimental setup for adsorption/desorption studies. The flow system consists of
18
feeding gases, a six-way valve for pulse injection, a quartz tube with heating elements
and a flame ionized detector (FID).
NH3
He
He
Sampling loop
MFC
(0.25 mL)
MFC
MFC
He
Vapor generator
Heating element
FID
Figure 2.7 Schematic diagram of the apparatus used in the adsorption/desorption studies.
A zeolite thin film is prepared by depositing zeolite sample on a 5.5 cm 0.7 cm

quartz plate according to the procedure described in Section 2.2.2. The quartz plate with
the zeolite film is placed in the quartz tube and degassed at 573 K for 1 h under helium
prior to any adsorption/desorption experiments. At the same time, a gas (such as NH3) or
the vapor of a liquid (such as cyclohexane) carried by helium flows continuously through
a 0.25 mL sampling loop that is connected to the six-way valve. A third gas line of
helium is used in case that the adsorbate in the sampling loop needs to be diluted in
concentration. By switching the six-way valve, a pulse of the adsorbate is directed
through the quartz tube and adsorbs in the zeolite sample. The FID detector is used to
monitor the adsorbate that is not adsorbed in the zeolite sample.
19
With a temperature controller (Omega i/16 Temperature & Process Controller),

temperature programmed desorption can be carried out using the same flow system. An
enlarged view is shown in Figure 2.8.
Thermal
couple
Heating element
FID
Zeolite sample
He
Figure 2.8 Schematic of the sample cell and heating portion of the TPD reactor.
A thermal couple is placed closely to the zeolite sample to monitor the

temperature. In TPD experiments, a zeolite sample is initially dosed with one pulse
injection of the adsorbate and purged under helium at room temperature until no
adsorbate comes out of the quartz tube. Desorption experiments are then carried out at
variable heating rates or helium flow rates.
2.4 Other Experimental Techniques Used in This
Dissertation Research
The specific surface areas of zeolite materials are measured on an automated
Quantachrome Nova 1200 multipoint BET apparatus using approximately 0.2 g of
sample for each measurement. Prior to the N2 adsorption, each sample is degassed under
vacuum at 393 K for 1 h. The external surface area and total surface area are measured
before and after removing the template from synthesized zeolite sample, respectively, by
calcination at 773 K under oxygen flow for 16 h.
Scanning electron microscopy (SEM, Hitachi S-4000) is used to determine
particle morphology and particle sizes of the zeolite materials. X-ray microanalysis is
conducted using energy dispersive spectrometry (EDS) to obtain information on
20
elemental composition of individual particles. Atomic force microscopy (AFM, Digital

Instruments Nanoscope III) is used to evaluate the roughness of zeolite films prepared
according to the procedure described in Section 2.2.2. The absorbance of a zeolite film
can be measured on an ultraviolet-visible spectrometer (Perkin-Elmer Lambda 20).
2.5 Reagents and Materials
Research-grade purity oxygen and carbon dioxide are purchased from Air
Products. Research-grade purity nitrogen dioxide, propylene, nitric oxide, nitrous oxide,
carbon monoxide, anhydrous ammonia and sulfur dioxide are purchased from Matheson.
Isobutane (99%) is purchased from Aldrich. Ultra high purity helium and hydrogen are
purchased from Airgas. All gases are used without further purification.
Commercial NaY zeolite, tetramethylammonium hydroxide (TMAOH) solution
(20% wt), 2,6-Di-tert-butylpyridine (DTBPy, 97%), cyclohexane (99%), cyclohexanol
(99%), cyclohexanone (99.8%), cyclohexene (99%), 1,2-epoxycyclohexane (98%), 2cyclohexen-1-ol (95%), 2-cyclohexen-1-one (95%) and acetonitrile (99%) are purchased
from Aldrich. Aluminum isopropyloxide, tetraethyl oxyorthosilane (TEOS) and sodium
hydroxide are purchased from Alfa Aesar. Potassium bromide is purchased from J. T.
Baker. Urea (>99.5%) is purchased from Acros Organics. N-15 and C-13 labeled urea
are purchased from Cambridge Isotopes. Acetone (HPLC grade) and pyridine (99.9%)
are purchased from Fisher Scientific. DTBPy, cyclohexane, cyclohexanol,
cyclohexanone, cyclohexene, 1,2-epoxycyclohexane, 2-cyclohexen-1-ol, 2-cyclohexen-1one, acetone and pyridine are further dehydrated over molecular sieves and purified
before use by several freeze-pump-thaw cycles.
21
CHAPTER 3
PHOTOOXIDATION OF CYCLOHEXANE AND
CYCLOHEXENE IN BaY ZEOLITE
3.1 Introduction
Catalyzed partial oxidation of small hydrocarbons using molecular oxygen has
been a very important process used to obtain industrially important chemicals. The
products serve as building blocks for plastics and synthetic fibers, or as industrial
intermediates in the manufacture of fine chemicals [44-47]. Typically, direct oxidations
of small hydrocarbons by O2 give very poor selectivity due to the free radical nature of
the process, the high exothermicity of the reactions and/or overoxidation [48]. The
accumulation of byproducts also limits conversion to a few percent in most practical
processes. The inefficiency associated with low conversion has motivated the search for
solid catalysts that are active for the oxidation of hydrocarbons [49-55].
CH2OOH
CH3
+ O2
h
BaY
+ O2
h
BaY
+ O2
h
BaY
CHO
+ H2O
CH2OOH
OOH
CHO + H2O
(a)
(b)
O
+ H2O
(c)
Figure 3.1 Examples of selective photooxidation of hydrocarbons with molecular oxygen

in BaY zeolite.
22
Frei and coworkers [32, 44, 55-64] and Larsen, Grassian and coworkers [37, 47,
65-69] have explored the use of cation-exchanged zeolites in the room temperature
selective photo- and thermal oxidation of alkyl-substituted benzenes (Figure 3.1 (a)),
olefins (Figure 3.1 (b)) and alkanes (Figure 3.1 (c)). The high selectivity of these
reactions can be attributed to several factors. First, using low reaction temperatures
inhibits some free radical chemistry. Second, the use of visible light rather than UV light
provides a low-energy pathway and third, the zeolite cavity results in a caging effect.
In a previous study, Larsen, Grassian and coworkers investigated the kinetics of
the photo- and thermal reactions of cyclohexane in BaY [69]. The proposed reaction
mechanism for this reaction is shown in Figure 3.2 [55].
OOH
+
+ O2
h
BaY
O2-
(a)
+ H2O
OOH
(b)
OH
(c)
Figure 3.2 Proposed mechanism for the photooxidation of cyclohexane with O2 in BaY.
In the proposed mechanism, the oxidation is initiated by excitation of a

hydrocarbon-oxygen complex occluded in the zeolite to form a low energy charge
transfer state that is stabilized by the electric field at the cation sites in the zeolite [32,
23
63]. Cyclohexyl hydroperoxide is produced as an intermediate following proton

abstraction (Figure 3.2 (a)). Using visible light or mild thermal conditions, oxidation of
cyclohexane in BaY with molecular oxygen produced cyclohexanone (Figure 3.2 (b)) and
cyclohexanol (Figure 3.2 (c)). Based on kinetic isotope effect studies, the ratedetermining step was assigned to the proton abstraction step (Figure 3.2 (a)). Since the
kinetic isotope effects were the same for the thermal and photochemical processes, the
mechanism was proposed to be the same for both reactions. The activation energy was
measured for the thermal process and it was consistent with the mechanism shown in
Figure 3.2.
In this study, the photooxidation of cyclohexane and cyclohexene in BaY zeolite
using in situ FT-IR and ex situ GC and GC-MS was examined. A direct comparison of
the photooxidation of cyclohexane versus cyclohexene in BaY using molecular oxygen
and tertiary butyl hydroperoxide (TBHP) was made in order to elucidate the factors that
influence product selectivity in these reactions. These factors include the excitation
wavelength, the reactant and the oxidant.
3.2 Experimental Section
BaY was prepared from NaY purchased from Aldrich by standard ion-exchange
procedures. NaY was added to an aqueous 0.5 M BaCl2 solution and heated to 363 K for
24 h. The resulting solution was filtered and the zeolite was washed with deionized water
until the solution was free of Cl- ions as determined by adding AgNO3 (aq) to the filtrate
solution until no precipitate was observed. The elemental composition of Al, Si and Ba
was determined by inductively coupled plasma/atomic emission spectroscopy (ICP/AES)
using a Perkin Elmer Plasma 400. The Si/Al and Ba/Al ratios for BaY were 2.4 and 0.33,
respectively. BaY was activated by heating overnight to 573 K under vacuum.
The infrared sample cell and experimental setup for photooxidation studies are
described in Figures 2.1 and 2.5, respectively. Approximate 50 mg BaY zeolite was
24
coated onto the 3 cm 2 cm tungsten grid (Figure 2.1). In oxidation experiments using
molecular oxygen, cyclohexane or cyclohexene was loaded into the zeolite by adsorption
at different hydrocarbon gas pressures that corresponded to different hydrocarbon
loadings as described in detail in Section 3.3. The excess hydrocarbon was then pumped
out, and oxygen was added to the IR cell at a pressure of 750 Torr. In oxidation
experiments using TBHP, TBHP was obtained by photooxidation of isobutane using
molecular oxygen at 273 K [61]. The excess isobutane and oxygen were then pumped
out, and cyclohexane or cyclohexene was added to the IR cell at room temperature for
oxidation in the dark. Products were extracted from the zeolite in acetonitrile for 90 min
and were analyzed by GC or GC-MS, as described in Section 2.2.3.
3.3 Results
3.3.1 Photooxidation of Cyclohexane Using Oxygen
Figure 3.3 shows the IR spectra of cyclohexane adsorbed in BaY as a function of
cyclohexane pressure. The band at 1451 cm-1 is assigned to (CH2) mode of adsorbed
cyclohexane and increases in intensity as the pressure of cyclohexane increases. The
amount of adsorbed cyclohexane was calibrated using volumetric methods. The
difference in expansion pressure of cyclohexane with and without BaY present in the IR
cell was determined at the highest pressures measured. From the calibrated volumes of
the infrared cell and gas handling system, p, the pressure difference, was converted to
the number of cyclohexane molecules absorbed in a sample of BaY. From the weight of
the BaY sample, the number of supercages for that sample was determined as well. The
integrated absorbance of the band at 1451 cm-1 at high pressure was equated to the
number of cyclohexane molecules per supercage determined from volumetric
measurements.
25
0.5
A
b
s
o
r
b
a
n
c
e
Pressure
3500
1451
3000
2500
2000
Wavenumber (cm-1 )
1500
1000
Figure 3.3 FT-IR spectra of cyclohexane adsorbed in BaY as a function of cyclohexane

pressure at 298 K. All spectra use the clean BaY zeolite prior to adsorption as a
reference. In addition, gas-phase absorptions have been subtracted from each of the FTIR spectra.
Assuming that the band at 1451 cm-1 is linearly proportional to the number of
cyclohexane molecules adsorbed in the zeolite, the number of cyclohexane molecules
adsorbed per supercage can be determined at lower pressures as well. The uptake of
cyclohexane in the zeolite as a function of pressure obtained by the volumetric method is
shown in Figure 3.4. It can be seen in Figure 3.4 that a coverage of 1 molecule
cyclohexane per supercage is obtained at an equilibrium pressure of 200 mTorr. At a
pressure above 1500 mTorr, the coverage reaches a maximum of approximately 3.5
cyclohexane molecules per supercage. An alternative approach to study the adsorption of
cyclohexane in BaY is desribed in Appendix C.
26
Coverage (molecules/supercage)
3.5
3
2.5
2
1.5
1
0.5
0
500
1000 1500 2000 2500 3000 3500 4000

Expansion Pressure (mTorr)
Figure 3.4 The uptake of cyclohexane in BaY as a function of pressure. The coverage is
calibrated using the absorption band at 1451 cm-1.
The photooxidation of cyclohexane in BaY with O2 was investigated using FT-IR

spectroscopy. Figure 3.5 shows the difference FT-IR spectra obtained after irradiation of
cyclohexane (100 mTorr, 0.5 cyclohexane per supercage) and O2 in BaY at > 455 nm.
After broadband irradiation with > 455 nm, absorption bands at 1672 cm-1 and 1367
cm-1 are apparent in the spectrum. These indicate that cyclohexanone and cyclohexyl
hydroperoxide, respectively, were produced [52]. An absorption band at ~1644 cm-1 can
be assigned to the bending mode of adsorbed water that was produced in the reaction
shown in Figure 3.2 (b). The percent conversion of cyclohexane was calculated by
measuring the decrease in the integrated area of the cyclohexane absorption at 1451 cm-1
(not observed in difference spectra) as the reaction occurs. The percent conversion after
irradiation at > 455 nm for 18 h was determined to be 7%. Irradiation using a shorter
wavelength filter, > 325 nm, gave a greater percent conversion of 48%. GC analysis of
the extracted products was then conducted to determine the product distributions that are
presented in Table 3.1.
27
1672
0.5
A
b
s
o
r
b
a
n
c
e
2200
1367
1644
Irradiation time
2000
1800
1600
1400
1200
W avenumber (cm-1 )
Figure 3.5 Difference FT-IR spectra after irradiation of cyclohexane and oxygen in BaY
with > 455 nm near room temperature for 20 min, 1 h, 2 h, 6 h and 22 h (bottom to top).
Cyclohexane coverage in BaY is approximately 0.5 molecules per supercage. All spectra
use the clean BaY zeolite prior to adsorption as a reference. In addition, gas-phase
absorptions have been subtracted from each of the FT-IR spectra.
Table 3.1 Product distribution (%) obtained from GC analysis and conversion
(%) for the photooxidation of cyclohexane in BaY.
OH
OOH
BaY
Conversion
(%) a
O2, 325nm, 100 mTorr, 18 h
81
11
48
O2, 455nm, 100 mTorr, 18 h
86
14
Conversion (%) was calculated by integrating the absorption at 1451 cm-1 in the FTIR spectrum (estimated error = 3%).
28
The results indicate that ~80% of the products formed in photooxidation of

cyclohexane in BaY are cyclohexanone independent of the excitation wavelength. The
other products formed are cyclohexanol [54, 62] (not observed in the FT-IR spectra) and
cyclohexyl hydroperoxide. Cyclohexyl hydroperoxide decomposes thermally to form
cyclohexanone and water and may be the reason why it is not observed in GC analysis
but is in FT-IR spectroscopy (Figure 3.5).
3.3.2 Photooxidation of Cyclohexene Using Oxygen
The FT-IR spectra of cyclohexene in BaY as a function of cyclohexene pressure
Coverage (molecules/supercage)
are shown in Figure 3.6.
0.5
A
b
s
o
r
b
a
n
c
e
Pressure
3500
3.5
3
2.5
1440
1450
1.5
1
0.5
1636
0
0
500 1000 1500 2000 2500 3000 3500 4000
Expansion Pressure (mTorr)
3000
2500
2000
1500
1000
-1
W avenumber (cm )
Figure 3.6 FT-IR spectra of cyclohexene adsorbed in BaY as a function of cyclohexene

pressure 298 K. All spectra use the clean BaY zeolite prior to adsorption as a reference.
In addition, gas-phase absorptions have been subtracted from each of the FT-IR spectra.
The absorption band at 1636 cm-1 is used to construct the isotherm of cyclohexene in
BaY (inset).
29
In Figure 3.6, the absorption band at 1636 cm-1 is assigned to the C=C stretching
mode and is used to construct the uptake curve of cyclohexene in BaY as a function of
expansion pressure (see inset of Figure 3.6). A coverage of 1 molecule cyclohexene per
supercage is seen to occur around 350 mTorr. Similar to that of cyclohexane, the
coverage leveled off at approximately 3 cyclohexene molecules per supercage at
expansion pressures greater than ~1000 mTorr.
The difference FT-IR spectra obtained after irradiation of cyclohexene (350
mTorr, 1 cyclohexene/supercage) and O2 in BaY at > 455 nm are shown in Figure 3.7.
1698
0.2
A
b
s
o
r
b
a
n
c
e
1640
1631
Irradiation time
1440
2200
2000
1800
1600
Wavenumber (cm-1 )
1400
1200
Figure 3.7 Difference FT-IR spectra after irradiation of cyclohexene and oxygen in BaY
with > 455 nm near room temperature for 20 min, 1 h, 2 h, 6 h and 22 h (bottom to top).
Cyclohexene coverage in BaY is approximately 1 molecule per supercage. All spectra
30
In Figure 3.7, several product bands attributed to C=O and C=C containing
products are apparent in the 1600-1700 cm-1 region. The negative feature at 1440 cm-1 is
due to loss of the reactant, cyclohexene. Similar difference spectra were observed when
the sample is irradiated with photons of shorter wavelength, > 325 nm. The product
distributions obtained from the GC and GC-MS analysis are listed in Tables 3.2 and 3.3.
Table 3.2 Product distribution (%) obtained from GC and GC-MS analysis and
conversion (%) for the photooxidation of cyclohexene in BaY.
O
BaY
CHO
CHO
O2, 325nm, 23.5 h

O2, 455nm, 23.5 h
TBHP, thermal, 16 h
OH
Other
products
Conversion
(%) a
35
23
35
64
33
10
19
38
57
10
11
46
25
Conversion (%) was calculated by integrating the absorption at 1636 cm-1 in the FT-IR spectrum
(estimated error = 3%). The pressure of cyclohexene was 350 mTorr. b. TBHP: tertiary butyl
hydroperoxide (see text for further details).
Table 3.3 Product distribution (percent of total products) for the other products from
Table 3.2.
OH
OH
BaY
CHO
OH
Condensation
product
O2, 325nm, 23.5 h
13
11
O2, 455nm, 23.5 h
11
11
TBHP, thermal, 16 h
31
In photooxidation of cyclohexene using O2, the major product formed was 2cyclohexen-1-one (35%). Smaller amounts of 2-cyclohexen-1-ol (7%) and hexanedial
(23%) were also formed. Various other products were also formed and are listed in
Table 3.3. The product distribution is not altered to any great extent with excitation
wavelength. Possible reaction paths for the formation of three major products in the
photooxidation of cyclohexene with O2 in BaY are shown in Figure 3.8.
OOH
+
+ O2
h
BaY
O2-
(a)
+ H2O
OOH
(b)
OH
(c)
O
O2-
(d)
O
O
Figure 3.8 Proposed mechanism for the photooxidation of cyclohexene with O2 in BaY.
In Figure 3.8, the first three reactions are analogous to the cyclohexane chemistry
shown in Figure 3.2. The reaction in Figure 3.8 (d) accounts for the formation of
hexanedial through a dioxetane intermediate. Similar chemistry has been observed in
related alkene photooxidation reactions in Y zeolites [37, 47, 60]. The other products
32
listed in Table 3.3 are formed from secondary reactions of the products in Table 3.2. For
example, 2,3-epoxycyclohexanone and 2,3-epoxycyclohexanol are formed from the
epoxidation of 2-cyclohexen-1-one and 2-cyclohexen-1-ol, respectively.
3.3.3 Oxidation of Cyclohexene Using TBHP
TBHP is a major industrial oxidant for the transformation of small olefins to

epoxides [70-73]. In this study, TBHP was generated in situ by irradiation ( > 400 nm)
of isobutane and molecular oxygen in BaY at 273 K [61]. The difference IR spectra of
isobutane and oxygen in BaY before and after irradiation are shown in Figure 3.9.
1377
0.4
A
b
s
o
r
b
a
n
c
e
1693
(b) After irradiation with O2
1475
1368
(a) Isobutane in BaY
2200
2000
1800
1600
1400
1200
-1
Wavenumber (cm )
Figure 3.9 Oxidation of cyclohexene using TBHP: (a) FT-IR spectrum of isobutane and
oxygen adsorbed in BaY and (b) difference FT-IR spectrum of isobutane and oxygen in
BaY after irradiation at > 455 nm for 4 h. Spectra were recorded at 298 K. All spectra
33
The formation of TBHP (1377 cm-1) and its thermal reaction product, acetone
(1693 cm-1), are observed. Frei and coworkers observed similar FT-IR spectra of
isobutane and TBHP in BaY [61]. The proposed mechanism for the formation of TBHP
is shown in Figure 3.10 (a). A competing, thermal reaction results in the formation of
acetone and methanol from TBHP as shown in Figure 3.10 (b) [61].
(CH3)3CH + O2
h
BaY
(CH3)3COOH
(a)
O
(CH3)3COOH
(b)
CH3CCH3 + CH3OH
O
+ (CH3)3COOH
+ (CH3)3COH
(c)
Figure 3.10 Proposed mechanism for the oxidation of cyclohexene with TBHP in BaY.
After removing excess isobutane and oxygen, cyclohexene was allowed to react
thermally with TBHP in BaY. TBHP was then used to thermally oxidize cyclohexene (in
the dark). The difference FT-IR spectra obtained following the thermal oxidation of
cyclohexene with TBHP in BaY as a function of time are shown in Figure 3.11. The
band at 1676 cm-1 indicates the formation of a C=O containing product. No reaction was
observed when cyclohexane was added to BaY in the presence of TBHP.
34
1676
0.5
A
b
s
o
r
b
a
n
c
e
Time
1438
1256
1370
2200
2000
1800
1600
1400
1200
-1
Wavenumber(cm )
Figure 3.11 Difference FT-IR spectra following the thermal oxidation of cyclohexene
with TBHP generated in BaY for 20 min, 1 h, 2 h, 3 h, 4 h and 44 h (bottom to top).
Spectra were recorded at 298 K. All spectra use the clean BaY zeolite prior to adsorption
as a reference. In addition, gas-phase absorptions have been subtracted from each of the
FT-IR spectra.
The product distributions determined by GC and GC-MS analysis are listed in

Tables 3.2 and 3.3. The main product formed from the oxidation of cyclohexene with
TBHP is 1,2-epoxycyclohexane (46%) that is not seen in the FT-IR spectra because it has
weak absorptions in the region where losses are seen for the reactant molecules. Smaller
amounts of 2-cyclohexen-1-one (8%), 2-cyclohexen-1-ol (10%) and hexanedial (11%)
are also formed as well as the other products listed in Table 3.3.
35
3.4 Discussion
3.4.1 Influence of Wavelength on Product Distribution
For cyclohexane, GC product analysis indicates that photooxidation with O2 at

different wavelengths results in similar product distributions (Table 3.1). The
cyclohexane conversion measured by FT-IR spectroscopy increases from 7% at > 455
nm to 48% at > 325 nm. Thus better conversion of cyclohexane can be achieved at
shorter wavelengths with no loss in selectivity. The wavelength dependence of the
photooxidation of cyclohexene with O2 is less dramatic (Tables 3.2 and 3.3). A slight
increase in conversion is observed with wavelength: 64% at 325 nm relative to 57% at
455 nm. The product distribution does not change substantially. More hexanedial (23%)
is formed at 325 nm relative to 455 nm (19%). This is consistent with the prevalence of
dioxetane chemistry at shorter irradiation wavelengths [37].
3.4.2 Photooxidation of Cyclohexane vs. Photooxidation
of Cyclohexene in BaY
The photooxidation of cyclohexane in BaY with molecular oxygen yields

cyclohexanone in approximately 80% product selectivity with smaller amounts of
cyclohexanol (~10%) and cyclohexyl hydroperoxide (~0-10%). Photooxidation of
cyclohexene in BaY with molecular oxygen yields only ~35% of the analogous major
product, 2-cyclohexen-1-one and 10% 2-cyclohexen-1-ol. Numerous other oxidation
products, such as hexanedial (~20%) are also formed.
These results indicate that photooxidation of cyclohexane in BaY is more
selective than photooxidation of cyclohexene in BaY. The observed difference in
selectivity can be attributed to the presence of two reactive centers (-H and C=C bond)
in cyclohexene leading to multiple reaction paths as shown in Figure 3.8. The percent
conversion of cyclohexene (57%) after irradiation with > 455nm is much greater that
the percent conversion of cyclohexane (7%) under similar reaction conditions. The
36
relative activity of H-abstraction by peroxide radicals from cyclohexene relative to

cyclohexane is 18 to 0.014 suggesting that cyclohexene should be much more reactive
than cyclohexane [74] in these reactions.
3.4.3 Photooxidation of Cyclohexene in BaY with
Different Oxidants (O2 vs. TBHP)
The main product formed in oxidation of cyclohexene with TBHP is 1,2-epoxy

cyclohexane which accounted for 46% of the extracted products (Table 3.2). Notably, no
1,2-epoxy cyclohexane is formed in photooxidation of cyclohexene using molecular
oxygen as the oxidant. The reaction pathway for the formation of 1,2-epoxy cyclohexane
is shown in Figure 3.10 (c). Tert-butanol is also formed stoichiometrically but is not
detected by GC analysis. Smaller amounts of 2-cyclohexen-1-one, 2-cyclohexen-1-ol
and hexanedial are also formed after oxidation of cyclohexene with TBHP, as well as
several other oxidation products listed in Tables 3.2 and 3.3. No reaction between
cyclohexane and TBHP occurs under comparable reaction conditions. Presumably this is
due to the fact that cyclohexane does not have a reactive double bond.
3.5 Conclusions
In the photooxidation of cyclohexane and cyclohexene in BaY using molecular

oxygen, the main products formed were cyclohexanone and 2-cyclohexen-1-one,
respectively. Photooxidation of cyclohexane was found to be more selective than
phootooxidation of cyclohexene in BaY with O2 under comparable reaction conditions.
The decrease in selectivity of cyclohexene relative to cyclohexane was attributed to the
presence of two active reaction centers (H and C=C bond) in cyclohexene. The C=C
functionality provided a route for dioxetane chemistry in cyclohexene that was not
possible with cyclohexane as a reactant. Thermal oxidation of cyclohexene in BaY with
TBHP changed the product distribution dramatically and resultd in the formation of 1,2-
37
epoxy cyclohexane as the major product. No products were formed when cyclohexane
was allowed to react with TBHP in BaY.
3.6 Acknowledgements
The authors want to thank Dr. Ping Li, Dr. Alexander Saladino and Jennifer Mann
for their valuable assistance with experiments. The authors gratefully acknowledge the
Environmental Protection Agency for the support of this work (R829600). The results of
this work are presented in a publication under the authorship of: Gonghu Li, Minna Xu,
Sarah C. Larsen, Vicki H. Grassian, J. Mol. Catal. A 2003, 194, 169.
38
CHAPTER 4
CHARACTERIZATION OF NANOCRYSTALLINE NaY
ZEOLITES AND THE ADSORPTION OF SELECTED
PROBE AND SMALL MOLECULES
4.1 Introduction
Zeolites are considered nanomaterials because of their pore sizes that range from
approximately 0.4 to 1 nm. Recently, there has been a great deal of interest in another
aspect of zeolites related to nanoscience, the primary crystal size, which can potentially
be exploited for use in nanotechnology. Nanocrystalline zeolites are zeolites with
discrete, uniform crystals with dimensions of less than 100 nm that have unique
properties relative to conventional micrometer-sized zeolite crystals. Nanocrystalline
zeolites have higher external surface areas and reduced diffusion path lengths relative to
conventional micrometer-sized zeolites [12, 13]. The external surface of nanocrystalline
zeolites can potentially be utilized for adsorption and catalytic reactions. It was shown
that the adsorption capacity of nanocrystalline silicalite increased approximately 50%
relative to micrometer-sized silicalite [12]. Nanocrystalline ZSM-5 as a catalyst exhibits
increased selectivity and toluene conversion into cresol and decreased coke formation
relative to conventional ZSM-5 materials [75]. In addition, a bifunctional material could
be prepared from nanocrystalline zeolites, such that the external and internal surfaces
have different reactivities. For example, the external surface active sites could catalyze
the degradation reaction of a large molecule that could then diffuse into the zeolite pores
for further reactions.
Nanometer-sized zeolites can be assembled into thin films and other porous
nanoarchitectures for use in various applications [76-80]. Initially, nanocrystalline NaY
with a particle size of 468 nm was synthesized according to the method described in
Section 5.2.1. Then a hydrosol of the nanocrystalline NaY was prepared according to the
39
method described in Section 2.2.2 and was pipeted onto a quartz slide and dried in
ambient air. A film of commercial NaY purchased from Aldrich (particle size ~1 m)
was prepared using the same method. Digital images of the films are shown in Figure
4.1. In each case, the film was prepared using approximately the same mass of zeolite Y.
Figure 4.1 Digital pictures of zeolite films prepared from: Aldrich NaY (left, particle size
~1 m) and nanocrystalline NaY (right, particle size 468 nm).
The film prepared from nanocrystalline NaY is much more uniform than the film
prepared from Aldrich NaY (Figure 4.1). The increased transparency of the films can be
observed visually. The Y printed on the paper behind the film can be clearly seen
through the nanocrystalline NaY film (right) but is much more difficult to see through the
Aldrich NaY film (left). To obtain more quantitative information, the percent
transmittance (%T) of the films was measured using UV-Visible spectroscopy. The UVVisible spectra shown in Figure 4.2 indicate that the nanocrystalline NaY film had a %T
of 70-80% in the 300-800 nm range compared to a %T of 30-40 for the Aldrich NaY
film.
Images of the nanocrystalline NaY film were also obtained using atomic force
microscopy and are shown in Figure 4.3. The AFM images show that the film surface is
continuous and smooth on the hundreds of nanometer scale. Attempts to obtain AFM
images of Aldrich NaY film were unsuccessful due to the extreme roughness of the film.
40
1
(a)
0.8
I/I0
0.6
0.4
(b)
0.2
0
300
400
500
600
W avelength (nm)
700
800
Figure 4.2 UV-Visible spectra of the zeolite films prepared from: (a) nanocrystalline
NaY (particle size 468 nm) and (b) Aldrich NaY (particle size ~1 m).
Figure 4.3 Contact mode AFM image of zeolite thin film prepared from nanocrystalline
NaY with a particle size of 468 nm.
41
The above comparison demonstrates that nanocrystalline zeolites are superior

materials relative to Aldrich zeolites in such applications as self assembly to produce
zeolite films. Batch reactor experiments were designed to examine how the thickness of
the thin film of micrometer-sized BaY catalyst influences the activity in photooxidation
of cyclohexane using molecular oxygen (see Section 2.2.2). Figure 4.4 demonstrates that
the percent conversion in the photooxidation of cyclohexane varies with the thickness of
BaY films. Significantly higher conversion of cyclohexane was achieved using a thinner
film of BaY catalyst. For BaY catalysts with a thickness greater than 0.5 mm, thermal
reaction is the major contribution to the oxidation of cyclohexane since irradiation cannot
penetrate the BaY film. This indicates that nanocrystalline zeolites may be better
catalysts in photooxidation of hydrocarbons relative to micrometer-sized zeolites.
100
1.5
60
1
40
Thermal oxidation
0.5
20
Thickness (mm)
Percent Conversion
80
50
100
150
200
250
0
300
M ass of BaY (mg)
Figure 4.4 Percent conversion in photooxidation of cyclohexane in BaY (solid cycle) as

a function of catalyst thickness. The thickness is linearly proportional to the mass of BaY
sample. The percent conversion in thermal oxidation of cyclohexane is also shown (open
cycle).
42
In this study, the physically and chemically distinct properties of nanocrystalline

NaY are explored, including enhanced external surface area, unique surface
functionalities and improved catalytic activity. Nanocrystalline zeolite Y with a particle
size of 23 nm was synthesized and characterized since the external surface area for
zeolite crystals with particle sizes smaller than 30 nm is greater than 100 m2/g, the same
order of magnitude as the internal surface area for a typical zeolite.
The synthesis of nanocrystalline zeolite Y has been reported by several groups
[79, 81-84]. The synthetic method used here is a modification of the method of Creaser
and co-workers [82]. The physicochemical properties of the synthesized nanocrystalline
NaY samples are characterized with a variety of techniques, including microscopy, FT-IR
spectroscopy and sorption of probe molecules. Aldrich NaY zeolite is also characterized
for a direct comparison. In addition, the application of nanocrystalline zeolite Y as a
potential adsorbent for chosen small molecules is explored.
4.2 Synthesis of Nanocrystalline NaY Zeolite
Nanocrystalline Zeolite Y was synthesized using clear solutions according to the

method reported by Creaser and co-workers [82]. Aluminum isopropoxide, half of the
tetramethylammonium hydroxide (TMAOH) solution, and water were mixed and stirred
until the mixture became a clear solution. Next, tetraethoxyorthosilane (TEOS), the rest
of the TMAOH solution, and water were added to the clear solution. The mixture was
stirred overnight to ensure complete hydrolysis of the aluminum and silicon sources. The
final clear synthesis gel had the following composition: 0.07Na : 2.4TMAOH : 1.0Al :
2.0Si :132H2O : 3.0i-PrOH : 8.0EtOH; the latter two alcohols came from the hydrolysis
of the aluminum isopropoxide and TEOS, respectively. To optimize the formation of
small NaY crystals, the sodium content was intentionally set too low relative to
aluminum content in the synthesis gel.
43
The clear solution was transferred into a 500 mL flask equipped with an air
cooled condenser and was heated to 368 K in an oil bath for 84 h with stirring.
Nanocrystalline Y powders were recovered from the milky, colloidal suspension of NaY
after two cycles of centrifugation, washed with deionized water, and dried at 393 K in air.
Nanocrystalline NaY samples were calcined at 773 K under oxygen flow for 16 h to
remove organic templates.
4.3 Characterization of Nanocrystalline NaY Zeolite
4.3.1 Physical Characterization of Nanocrystalline NaY
A variety of techniques were used to characterize the synthesized nanocrystalline

NaY. XRD patterns for synthesized nanocrystalline NaY agree with those for the
faujasite structure of NaY. Elemental analysis indicates that the Si/Al ratios were 2.0 and
1.8 for Aldrich NaY and nanocrystalline NaY, respectively [14]. The external and total
surface areas of nanocrystalline NaY were obtained from the nitrogen adsorption
isotherms using BET method. The surface area of the as-synthesized nanocrystalline
NaY, in which the internal surface is blocked by template molecules, represents the
external surface area of the zeolite sample. The total surface area obtained from the
calcined samples contains contributions from both the internal and the external surfaces.
The external and total surface areas for the nanocrystalline NaY sample were 178 and
584 m2/g, respectively [14]. For Aldrich NaY sample, only the total surface area of 477
m2/g was measured since the material was only available in the calcined form.
Using the external surface area measured from the BET method, the particle size
can be accurately estimated assuming the crystals are cubic [12, 13]. The particle size
can be estimated using the relationship
x = 4061 / Sext
44
where Sext (in m2/g) is the external surface area of NaY and x is the particle size in nm
(assuming cubic crystals). Using this relationship, the particle size of synthesized
nanocrystalline NaY was estimated from the external surface area to be 23 nm [14]. An
analogous relationship was derived for nanocrystalline silicalite, and the experimentally
determined external surface areas and particle sizes determined from SEM agreed
remarkably well with the results of the calculation assuming cubic silicalite crystals [12,
13].
Figures 4.5 and 4.6 show representative SEM images of the synthesized
nancorystalline NaY and Aldrich NaY, respectively. Small NaY crystals are envisioned
as cubic particles in order to estimate the particle size. Close inspection of the SEM
image of nanocrystalline NaY shown in Figure 4.5 indicates that the crystals are discrete
with narrow size distributions and the particle size is in good agreement with that
calculated from the external surface (23 nm).
Figure 4.5 Representative SEM images of synthesized nanocrystalline NaY with a

particle size of 23 nm (scale bar = 100 nm)
45
The particle size of Aldrich NaY (Figure 4.6) is around 1 m with a broader size
distribution relative to nanocrystalline NaY. The external surface area of Aldrich NaY
was estimated from the particle size to be ~4 m2/g
Figure 4.6 Representative SEM images of commercial (Aldrich) NaY with a particle size
of ~1 m (scale bar = 1 m).
4.3.2 Characterization of Nanocrystalline Zeolite with

FT-IR Spectroscopy
Figure 4.7 shows the FT-IR spectra of nanocrystalline NaY zeolite and Aldrich
sample after being heated at 623 K overnight under vacuum. In the spectrum of
nanocrystalline NaY zeolite, three absorptions are observed in the spectral region
between 3765 and 3630 cm-1 (Figure 4.7 (a)). The most intense and highest frequency
band at 3744 cm-1 is assigned to terminal silanol groups that are on the external surface of
46
the zeolite crystals [16]. This feature is of much weaker intensity in Aldrich NaY sample
as expected since the external surface area is ~40 times smaller than the nanocrystalline
NaY sample. The absorption band at 3695 cm-1 is assigned to hydroxyl groups attached
to Na+ [85]. An absorption band at 3656 cm-1, associated with hydroxyl groups attached
to extra framework alumina (EFAL) species [17], is only observed in the FT-IR spectrum
of nanocrystalline NaY zeolite (Figure 4.7).
0.02
3744
A
b
s
o
r
b
a
n
c
e
3695
3656
(a)
(b)
3760
3740
3720
3700
Wavenumber
3680
3660
3640
(cm-1 )
Figure 4.7 FT-IR spectra of surface hydroxyl groups in: (a) nanocrystalline NaY and (b)
Aldrich NaY zeolite after heating to 623 K overnight under vacuum. The blank grid is
used as a reference. All spectra were recorded at 298 K.
The FT-IR spectra of nanocrystalline NaY and Aldrich NaY in the range between
400 and 1300 cm-1 were obtained using KBr pellet technique. The absorption features
shown in Figure 4.8 are resulted from stretching and bending modes of the T-O units (T =
Si or Al) in zeolite framework [15]. For example, the bands at 1135 and 725 cm-1 are
47
assigned to the asymmetric and symmetric stretching modes of internal tetrahedra,

respectively. The bands at 1020 and 792 cm-1 are associated with the asymmetric and
symmetric stretching modes of external linkages [16]. These absorption features in the
FT-IR spectrum of nanocrystalline NaY (Figure 4.8 (a)) are much more broadened and
show more band overlap relative to those of Aldrich NaY, indicating that nanocrystalline
NaY has a more heterogeneous surface than Aldrich NaY.
450
0.2
0.1
466
A
b
s
o
r
b
a
n
c
e
500
1020
1135
792
725
566
(a)
(b)
1200
1050
900
800
700
W avenumber (cm -1)
600
500
400
Figure 4.8 FT-IR spectra of: (a) nanocrystalline NaY and (b) Aldrich NaY zeolite in the
range between 400 cm-1 and 1300 cm-1 showing framework vibrations. Pure KBr was
used as a reference. All spectra were recorded at 298 K.
The double ring opening vibration at 566 cm-1 in the FT-IR spectrum of
nanocrystalline NaY (Figure 4.8 (a)) is characteristic of faujasite zeolites [86]. The T-O
bending vibrations of internal tetrahedra in nanocrystalline NaY can be identified by
48
absorption bands around 500, 466 and 450 cm-1 [16]. These absorption bands
characterizing T-O bending vibrations can be shifted to lower frequencies with deceasing
Si/Al ratio in the internal linkages due to the different length of the Al-O (1.73 ) and SiO (1.62 ) bonds [15]. In the spectra shown in Figure 4.8, the absorption features for TO bending modes have lower frequencies in the spectrum of nanocrystalline NaY relative
to Aldrich NaY. As mentioned earlier, nanocrystalline NaY (1.8) and Aldrich NaY (2.0)
have similar Si/Al ratio. Most likely, the percentage of Si-O bonds in the internal linkage
of nanocrystalline NaY is lower than in Aldrich NaY since a significant amount of Si-O
bonds are on the external linkage of nanocrystalline NaY. This further confirms that
nanocrystalline NaY has a significantly greater external surface than Aldrich NaY.
4.3.3 Sorption of Probe Molecules in Nanocrystalline
NaY Zeolite
Probe molecules are frequently used to probe surface properties of zeolite

materials, i.e. acid character, as shown in Table 4.1 [15].
Table 4.1 Some frequently used probe molecules.

Probe Molecules
Type of sites probed
Ammonia
Type and concentration of acidic sites
Pyridine
Type and concentration of acidic sites
Substituted pyridine
Type and concentration of acidic sites, accessibility of sites
Alkanes, alkenes
Acid strength, accessibility of sites
Nitrogen dioxide
Metal cations, type and concentration of acidic sites
Carbon monoxide
Metal cations, type and concentration of acidic sites
In this study, surface features of nanocrystalline NaY were investigated with

probe molecules including nitrogen dioxide, pyridine, propylene and substituted pyridine.
Before the adsorption of probe molecules, nanocrystalline NaY was heated at 623 K
49
overnight under vacuum. Then the zeolite was cooled back to room temperature and
equilibrated with gases (or vapors of liquids) prior to a spectrum being recorded.
4.3.3.1 Adsorption of NO2 in Nanocrystalline NaY
The NO2 adsorption experiment was conducted at room temperature by

equilibrating the zeolite with 0.5 Torr NO2 for 30 min and then pumping out the gas
phase for several minutes before acquiring the FT-IR spectra. Figure 4.9 shows the FTIR spectra of NO2 adsorbed in nanocrystalline NaY and Aldrich NaY. Spectral features
observed can be identified as NO+ (2000-2180 cm-1 spectral region) and NO3- (1300-1600
cm-1 spectral region) [23, 24, 87-89].
0.2
1400
1327
A
b
s
o
r
b
a
n
c
e
2135
2024
1565
(a)
(b)
2200
2000
1800
1600
Wavenumber (cm-1 )
1400
1200
Figure 4.9 FT-IR spectra of 0.5 Torr NO2 adsorbed in: (a) nanocrystalline NaY and (b)
Aldrich NaY zeolite at 298 K. All spectra use the corresponding clean NaY zeolite prior
to adsorption as a reference. In addition, gas-phase absorptions have been subtracted
from each of the FT-IR spectra.
50
The NO+ and NO3- species are proposed to form through a cooperative effect
whereby two adsorbed NO2 molecules in close proximity autoionize according to the
reaction 4.1.
2NO2 NO+ + NO3-
(4.1)
In accord with previous assignments in the literature, the band at ~1400 cm-1 is
assigned to NO3- adsorbed on Na+ cation site in NaY, and the bands around 2024 and
2135 cm-1 are due to NO+ adsorbed onto different cationic positions in the zeolite
framework [89, 90]. The bands at ~1565 and 1327 cm-1 are assigned to NO3- adsorbed on
EFAL sites [88, 89].
Clear differences are observed in the spectra of NO2 adsorbed in Aldrich NaY as
compared to NO2 adsorbed in nanocrystalline NaY. Two prominent peaks are observed
at 1565 and 1327 cm-1 in the FT-IR spectrum of NO2 adsorbed in nanocrystalline NaY
(Figure 4.9 (a)). These two peaks are absent in the FT-IR spectrum of NO2 adsorbed in
Aldrich NaY (Figure 4.9 (b)). This is consistent with the FT-IR spectra of the hydroxyl
group region shown in Figure 4.7, which indicates that EFAL sites are present in larger
quantities in nanocrystalline NaY relative to Aldrich NaY.
4.3.3.2 Adsorption of Pyridine in Nanocrystalline NaY
In pyridine adsorption experiments, adsorption was carried out by equilibrating

the zeolite powder with 1 Torr pressure of pyridine vapor at 473 K. The sample was then
evacuated for 30 min at 473 K and cooled back to room temperature prior to recording a
spectrum. Figure 4.10 displays the FT-IR spectra of pyridine molecules adsorbed in the
nanocrystalline NaY and Aldrich NaY zeolite following desorption at 473 K. Pyridine
molecules can adsorb and react at different sites in NaY zeolites [25, 91]. Pyridine
molecules can interact with Bronsted acid sites forming pyridinium ions, which are
characterized by bands at 1545 and 1400 cm-1 in the FT-IR spectrum of pyridine
51
adsorbed in nanocrystalline NaY (Figure 4.10). The absorptions at 1621 and 1454 cm-1
are assigned to pyridine molecules adsorbed on Lewis acid sites. At the same time,
pyridine molecules can physically adsorb via hydrogen bonding with surface hydroxyl
groups, giving absorption bands at 1595, 1574, and 1443 cm-1. All three interactions
contribute to the band at 1489 cm-1.
0.05
1443
A
b
s
o
r
b
a
n
c
e
1454
1545
1621
1400
1489
1595
1574
(a)
(b)
1650
1600
1550
W avenumber
1500
1450
1400
(cm-1)
Figure 4.10 FT-IR spectra of pyridine molecules adsorbed in: (a) nanocrystalline NaY
and (b) Aldrich NaY at 298 K following pyridine desorption at 473 K. Spectra were
recorded at 298 K. All spectra use the corresponding clean zeolite prior to adsorption as
a reference.
Notable are the differences between the FT-IR spectra of pyridine adsorbed in
nanocrystalline NaY (Figure 4.10 (a)) and Aldrich NaY (Figure 4.10 (b)). The peaks at
1545 and 1400 cm-1 attributed to pyridine molecules on Bronsted acid sites are absent in
the spectrum of pyridine adsorbed in Aldrich NaY. In addition, the absorption bands at
52
1621 and 1454 cm-1 assigned to pyridine adsorbed on Lewis acid sites are also very weak
or absent in the spectrum of pyridine adsorbed in Aldrich NaY.
4.3.3.3 Adsorption of Propylene in Nanocrystalline NaY
The adsorption of propylene in nanocrystalline NaY results in an absorption band

at 1635 cm-1 assigned to the C=C stretching mode and several bands between 1350-1500
cm-1 associated with the CH3 deformation mode and the CH2 scissor mode (Figure 4.11).
Absorptions due to C-H stretching modes between 2800 and 3100 cm-1 were also
observed (not shown).
0.02
A
b
s
o
r
b
a
n
c
e
1634
A
b
s
o
r
b
a
n
c
e
1455
0.002
3718 3695
3655
3744
3745
3710
3675
3640
Wavenumber (cm-1 )
1379
Pressure
1640
1600
1560 1520 1480 1440

Wavenumber (cm-1 )
1400
1360
Figure 4.11 FT-IR spectra of propylene adsorbed in nanocrystalline NaY zeolite at 298
K as a function of propylene pressure (P = 0.1, 0.5, 1.0, 2.0 and 6.0 Torr). The inset
shows the difference spectrum (in the O-H stretching region) of nanocrystalline NaY
following the adsorption of 6.0 Torr propylene. All spectra use the clean nanocrystalline
NaY zeolite prior to adsorption as a reference. In addition, gas-phase absorptions have
been subtracted from each of the FT-IR spectra.
53
In Figure 4.11 (inset), the loss of hydroxyl groups with infrared absorptions at
3744 cm-1 (silanol groups), 3718 cm-1 (hydroxyl nests) [16], 3695 cm-1 (hydroxyl groups
on Na+ sites) and 3655 cm-1 (hydroxyl groups on EFAL sites) are apparent following the
adsorption of propylene. This indicates that these surface hydroxyl groups are important
adsorption sites for propylene.
All of the absorption features in the spectra shown in Figure 4.11 disappeared
after evacuation of gas-phase propylene at room temperature for several minutes. With
excess propylene in the IR cell, no pronounced change of the band at 1635 cm-1 was
observed at room temperature for 6 h, indicating that propylene did not undergo
polymerization in the nanocrystalline NaY zeolite. It can therefore be concluded that
strong Bronsted acid sites that initiate propylene polymerization are not present in the
nanocrystalline NaY zeolite pretreated at 623 K [92]. The FT-IR spectra shown in Figure
4.7 are consistent with this conclusion as strong Bronsted acid sites are typically bridging
OH groups with vibrational frequencies of 3630 and 3560 cm-1 [17, 91]. These bands are
not observed in the infrared spectrum of nanocrystalline NaY shown in Figure 4.7.
4.3.3.4 Adsorption of Substituted Pyridine in
Nanocrystalline NaY Zeolite
The above analysis suggests that relatively weak Bronsted acid sites as well as
Lewis acid sites are present in nanocrystalline NaY. These acidic sites are most likely
associated with EFAL sites in nanocrystalline NaY since the loss of hydroxyl groups on
EFAL sites (3656 cm-1) is observed upon pyridine adsorption (Figure 4.12 (a)). The loss
of non-acidic silanol groups (3744 cm-1) is also observed in the difference spectrum of
nanocrystalline NaY following pyridine adsorption. It has been suggested that most
EFAL species are located on the external surface of zeolite Y [20, 21].
54
A
b
s
o
r
b
a
n
c
e
(a)
3744
(b)
-0.62
-0.51
3656
0.01
3760
3720
3680
W avenumber (cm-1 )
3640
Figure 4.12 Difference FT-IR spectra following: (a) pyridine and (b) substituted pyridine
(DTBPy) adsorption in nanocrystalline NaY. All spectra use the clean nanocrystalline
NaY zeolite prior to adsorption as a reference. The negative peaks show the loss of
hydroxyl groups upon pyridine (or DTBPy) adsorption in nanocrystalline NaY.
In order to determine the location of the EFAL species in nanocrystalline NaY

zeolite, adsorption of 2,6-Di-tert-butylpyridine (DTBPy) was conducted. While pyridine
can adsorb into the zeolite pores, DTBPy can only interact with the hydroxyl groups on
the external surface (and in the pore mouth region) of the zeolites since its kinetic
diameter is larger than the pore opening of Y zeolites [93]. Thus DTBPy can be used as a
probe molecule to characterize acid sites located on the external surface of the zeolites
[93]. The difference FT-IR spectrum in the hydroxyl group region after DTBPy
adsorption on nanocrystalline NaY zeolite is shown in Figure 4.12 (b). The integrated
absorbance of the negative peak at 3656 cm-1 is approximately proportional to the amount
of EFAL species in nanocrystalline NaY zeolite. After DTBPy and pyridine adsorption
on nanocrystalline NaY zeolite, the integrated absorbance of the negative peak at 3656
55
cm-1 are -0.51 and -0.62, respectively. Therefore, it can be concluded that more than 80%
of the EFAL species are located on the external surface of nanocrystalline NaY zeolite.
To summarize, several key differences between nanocrystalline NaY and Aldrich
NaY zeolite are highlighted using the FT-IR results. A greater number of silanol groups
and EFAL sites are detected in nanocrystalline NaY relative to Aldrich NaY sample.
Relatively weak Bronsted acid sites and Lewis acid sites are present in nanocrystalline
NaY, but not in Aldrich NaY zeolite. These acidic sites are associated with EFAL sites
located on the external surface of the nanocrystalline NaY.
4.4 Adsorption of Other Small Molecules and Some
Potential Environmental Applications of
Nanocrystalline Zeolites as Adsorbents
Zeolites have emerged as important materials for environmental applications,

such as emission abatement of nitrogen oxides, environmentally benign synthesis, and
adsorption of VOCs and other pollutants [36]. Nanocrystalline zeolites have several
advantages relative to conventional micrometer-sized zeolites that may enhance their
usefulness for environmental applications. The increased surface area of nanocrystalline
zeolites relative to commercial zeolites is one advantage that can be exploited for
applications of zeolites as catalysts [28, 36] and adsorbents of VOCs and other pollutants.
The catalytic application of nanocrystalline NaY in environmental remediation will be
discussed in Chapters 5, 6 and 7. In this chapter, nanocrystalline NaY with a particle size
of 23 nm is investigated as an adsorbent for several small molecules such as acetone,
ammonia and sulfur dioxide.
4.4.1 Adsorption of Acetone in Nanocrystalline NaY
Figure 4.13 shows the FT-IR spectra of acetone adsorbed in nanocrystalline NaY.
Several adsorption features can be identified at 1237 cm-1 (asymmetric C-C stretching
mode), 1366 cm-1 (symmetric CH3 bending mode) and 1421 cm-1 (asymmetric CH3
56
bending mode). Two overlapping bands at 1688 and 1709 cm-1 are assigned to C=O
stretching mode of acetone molecules adsorbed on Lewis acid sites and silanol groups,
respectively [94]. The loss of silanol groups and hydroxyl groups attached to EFAL sites
was observed upon acetone adsorption in nanocrystalline NaY (not shown). This
indicates that the external surface of nanocrystalline NaY is important for acetone
adsorption since silanol groups and EFAL sites are located on the external surface (vide
supra).
1366
0.1
A
b
s
o
r
b
a
n
c
e
1709
1421
1688
Pressure
1237
1760 1680 1600 1520 1440 1360 1280 1200

Wavenumber (cm-1 )
Figure 4.13 FT-IR spectra of acetone adsorbed in nanocrystalline NaY at 298 K as a

function of acetone pressure (P = 0.005, 0.011, 0.021, 0.055, 0.102 and 0.306 Torr). All
spectra use the clean nanocrystalline NaY zeolite prior to adsorption as a reference. In
addition, gas-phase absorptions have been subtracted from each of the FT-IR spectra.
57
4.4.2 Adsorption of Ammonia in Nanocrystalline NaY
In the absence of water, NH3 molecularly adsorbed in nanocrystalline NaY can be

identified by the absorption bands at 1308 cm-1 and 1617 cm-1 attributed to NH3 bending
mode of ammonia (Figure 4.14) [95]. NH3 molecules most likely adsorbed via hydrogen
bonding with silanol groups of nanocrystalline NaY since the loss of silanol groups (3747
cm-1) upon NH3 adsorption was observed (Figure 4.14 inset).
A
b
s
o
r
b
a
n
c
e
A
b
s
o
r
b
a
n
c
e
3309
0.005
3374
3747
3243
1433
1308
3655
3750 3600 3450 3300 3150

W avenumber (cm -1)
1641
1442
1617
1590
N H3 + H 2O
N H3
Pressure
0.02
2000 1900 1800 1700 1600 1500 1400 1300 1200
Wavenumber (cm-1 )
Figure 4.14 FT-IR spectra of ammonia adsorbed in nanocrystalline NaY at 298 K as a

function of ammonia pressure (P = 0.05, 0.1, 0.2, 0.5 and 1 Torr). The inset displays the
difference FT-IR spectrum in the O-H/N-H region following adsorption of 1 Torr NH3 in
nanocrystalline NaY at 298 K. The FT-IR spectrum of 1 Torr NH3 and 1 Torr H2O
mixture adsorbed in nanocrystalline NaY at 298 K is also shown (in green). All spectra
use the clean nanocrystalline NaY zeolite prior to adsorption as a reference. In addition,
gas-phase absorptions have been subtracted from each of the FT-IR spectra.
The loss of hydroxyl groups attached to EFAL sites (3655 cm-1) is also observed
(not shown), suggesting that NH3 adsorption in part occurred on EFAL sites. This is
58
further supported by several absorption features between 1442 and 1590 cm-1
characterizing NH3 adsorbed on Lewis acid sites that are associated with EFAL species in
nanocrystalline NaY. The formation of ammonium ions in nanocrystalline NaY can be
identified by the absorptions due to bending vibration at 1433 cm-1 (Figure 4.14) and
stretching vibrations at 3374, 3309 and 3243 cm-1 (Figure 4.14 inset). The formation of
ammonium ions can be resulted from the adsorption of NH3 on Bronsted acid sites
(hydroxyl groups attached to EFAL species).
In the spectra shown in Figure 4.14, the absorption band at 1433 cm-1
characterizing ammonium ions in the presence of water (1641 cm-1) is much more intense
relative to the absorption in the absence of water. The formation of ammonium ions upon
adsorption of NH3 and H2O mixture must involve both the interaction between NH3 and
surface Bronsted acid sites and the reaction between NH3 and H2O (reaction 4.2).
NH3 + H2O NH4+ + OH-
(4.2)
The formation of molecularly adsorbed NH3 (1308 cm-1) was not substantially
influenced by the presence of water. This is probably due to the existence of Lewis acid
sites and/or silanol groups in nanocrystalline NaY that are important sites for NH3
adsorption.
4.4.3 Adsorption of Sulfur Dioxide in Nanocrystalline
NaY Zeolite
SOx (SO2+SO3) are among the main pollutants in gaseous emissions concern [6].
Adsorption of SO2 has been investigated in catalytic control of SOx emissions. SOx is
also a component of engine exhaust that can have an inhibitive effect on the performance
of zeolites as deNOx catalysts [6, 96]. Figure 4.15 shows the FT-IR spectra of SO2
adsorbed in nanocrystalline NaY as a function of SO2 pressure. Two absorption bands at
1320 and 1155 cm-1 are assigned to the asymmetric and symmetric S-O stretching
59
vibrations, respectively, of weakly chemisorbed SO2 in nanocrystalline NaY (Figure

4.15) [97-99]. At relatively high pressure, a band around 1334 cm-1, featuring physically
adsorbed SO2, appeared in the spectra shown in Figure 4.15 [98].
1320
0.05
A
b
s
o
r
b
a
n
c
e
1500
1334
Pressure
1155
1450
1400
1350
1300
1250
1200
1150
1100
W avenumber (cm-1 )
Figure 4.15 FT-IR spectra of SO2 adsorbed in nanocrystalline NaY at 298 K as a

function of SO2 pressure (P = 0.05, 0.1, 0.2, 0.5, 1 Torr). All spectra use the clean
nanocrystalline NaY zeolite prior to adsorption as a reference. In addition, gas-phase
In the presence of H2O, SO2 weakly adsorbed in nanocrystalline NaY can be

identified with the absorption bands at 1325 and 1153 cm-1 in the FT-IR spectra shown in
Figure 4.16. A relatively weak absorption around 1180 cm-1 and a broad band around
1310 cm-1 are apparent. The surface species attributing to the1310 cm-1 band remained
on the nanocrystalline NaY after being evacuated at room temperature for 2 h, while
weakly adsorbed SO2 (1325 cm-1) decayed gradually. The absorption bands around 1310
60
and 1080 cm-1 are assigned to the asymmetric and symmetric SO2 stretching vibrations,
respectively, of surface HSO3- in nanocrystalline NaY [100].
1325
0.05
A
b
s
o
r
b
a
n
c
e
1310
1365
1640
1153
1080
Time
1800
1700
1600
1500 1400 1300

Wavenumber (cm-1 )
1200
1100
1000
Figure 4.16 FT-IR spectra of 1 Torr SO2 and 1 Torr H2O mixture adsorbed in
nanocrystalline NaY at 298 K as a function of pumping time (t = 0, 1, 3, 10, 20, 120
min). All spectra use the clean nanocrystalline NaY zeolite prior to adsorption as a
reference.
The above observation suggests that reaction 4.3 occurred upon adsorption of SO2
and H2O mixture in nanocrystalline NaY.
SO2 + H2O H2SO3
(4.3)
In the spectra shown in Figure 4.16, an absorption centered at 1365 cm-1 is due to
the gas-phase SO2 before evacuating the sample cell. The loss of silanol groups and
hydroxyl groups on EFAL sites were also observed in SO2 adsorption and during the
adsorption of SO2 and H2O mixture (not shown). It has been suggested that Na+ cations
and hydroxyl groups in NaY zeolite are the centers for SO2 adsorption [99]. The
61
formation of HSO3- on Na+ sites upon chemical adsorption of SO2 is characterized by an

absorption at 1240 cm-1 [101], which is not observed in this study (Figure 4.15 and
Figure 4.16). Thus, the adsorption of SO2 or SO2 and H2O mixture in nanocrystalline
NaY most likely occurred on the hydroxyl groups (silanol groups and/or hydroxyl groups
attached to EFAL sites).
Figure 4.17 describes the decay curves of surface species derived from SO2.
Normalized Integrated Absorbance

(1365-1240 cm-1)
SO2
SO2+H2O
0.8
0.6
0.4
0.2
100
200
300
Pumping Time (min)
400
500
Figure 4.17 Decay curves of surface adsorbed sulfur species formed in the absence
(black circle) and presence (green circle) of water during evacuation of nanocrystalline
NaY at room temperature. The integrated areas were obtained in the spectral range from
1240 cm-1 to 1365 cm-1. The decay curves were further fitted with a double exponential
function (see text for further details). Spectra were recorded every 1 min.
Prior to the time course experiments, nanocrystalline NaY was equilibrated with 1
Torr SO2 (or 1 Torr SO2 and 1 Torr H2O) at room temperature. Then the NaY sample
was subject to evacuating under vacuum at room temperature. The FT-IR spectra of
surface sulfur species were recorded every 1 min. The normalized integrated area of the
62
absorption between 1240 and 1365 cm-1 was then plotted as a function of pumping time.
Assuming first-order desorption behavior for all of the sulfur species, the decay curves
can be fitted with a double exponential function
y = A1XP(-k1t) + A2XP(-k2t)
where y and t represent the integrated absorbance in the spectral range from 1240 cm-1 to
1365 cm-1 and pumping time in the unit of min, respectively. A1, A2 and k1, k2 denote the
pre-exponential constants and desorption rate constants in the unit of min-1, respectively.
The curve fitting results are listed in Table 4.2.
Table 4.2 Desorption kinetic data obtained by fitting the

decay curves to a double exponential function.
Parameters
SO2
SO2 + H2O
A1
0.48
0.56
k1
0.111
0.375
A2
0.42
0.37
k2
0.026
0.001
The results in Table 4.2 suggest that two types of sulfur species were formed in
SO2 adsorption in nanocrystalline NaY. One of them desorbs at a greater rate (k1 =
0.111) than the other one (k2 = 0.026). Based on the FT-IR results (Figure 4.15), these
two species are identified as physically adsorbed SO2 (1334 cm-1) and weakly
chemisorbed SO2 (1320 cm-1), respectively. Similarly, two surface species formed in the
adsorption of SO2 and H2O mixture can be identified as physically adsorbed SO2 (k1 =
0.375, 1325 cm-1) and chemically adsorbed HSO3- (k2 = 0.001, 1310 cm-1), respectively
(Figure 4.16). Most likely, physically adsorbed SO2 species were attached to silanol
groups via hydrogen bonding, while chemical adsorption of SO2 occurred on EFAL sites.
63
4.5 Conclusions
Synthesized nanocrystalline NaY has a much greater external surface area than
commercial micrometer-sized NaY sample. Characterization of nanocrystalline NaY
with FT-IR spectroscopy and sorption of probe molecules indicate that a greater number
of silanol groups and EFAL sites are present in nanocrystalline NaY. The EFAL species
are associated with Bronsted and Lewis acid sites and are mainly located on the external
surface of the nanocrystalline NaY.
The large external surface area and the functional groups (silanol groups and
EFAL species) on the external surface are important in the application of nanocrystalline
NaY as an adsorbent for small molecules such as acetone, ammonia and sulfur dioxide.
Further studies will focus on the physically and chemically distinct properties of
nanocrystalline NaY in environmental catalysis.
The authors thank Hassan Alwy for providing the nanocrystalline NaY (particle
size 468 nm) and Dr. Weiguo Song for providing the nanocrystalline NaY (particle size
23 nm) and the SEM image. The authors gratefully acknowledge the Environmental
Protection Agency for the support of this work (R829600). The results of this work are
presented in a publication under the authorship of: Weiguo Song, Gonghu Li, Vicki H.
Grassian, Sarah C. Larsen, Environ. Sci. Technol. 2005, 39, 1214.
64
CHAPTER 5
CATALYTIC REDUCTION OF NO2 IN
NANOCRYSTALLINE NaY ZEOLITE (80 nm)
5.1 Introduction
The emission control of nitrogen oxides (NOx) from engine exhaust is of great
importance due to their contribution to processes harmful to the environment [6, 33, 102].
Selective catalytic reduction (SCR) of NOx with NH3 or hydrocarbons (HCs) has
emerged as an effective technology using oxide and zeolite based catalysts [103-105].
Zeolites are considered particularly promising SCR catalysts because of their cation
exchange capacity and acid-base properties.
Several recent studies have investigated alkali and alkaline earth-substituted Y,
FAU zeolites in the adsorption and HC-SCR of NOx [88, 89, 106-111]. NOx can be
stored in zeolite Y in the form of stable nitrate and nitrite species. The adsorption of NO2
(or NO+O2) in zeolite Y has been suggested to proceed via the disproportionation of
N2O4 or the reaction of NO2 with the residual water in zeolite. As for the mechanism of
HC-SCR, previous studies have suggested isocyanates to be key intermediates in this
reaction [112, 113]. Organic nitro/nitrito compounds are first generated in the reaction
between adsorbed hydrocarbons and surface nitrate/nitrite species. These compounds are
then converted to isocyanates. Isocyanates will react with NOx from the gas phase
producing dinitrogen. Although the general reaction scheme can be understood from
these steps, many details still remain elusive, such as the mechanism for the formation of
the nitrogen-nitrogen bond in complete reduction. Ammonium ions or free radicals have
been proposed as key intermediates in the formation of dinitrogen [106, 114].
Most HC-SCR studies in zeolites have been carried out at temperatures above 673
K. At these elevated temperatures, many side reactions can take place [33]. In this study,
thermal reduction and propylene-SCR of NO2 were done at lower temperatures (T 473
65
K) and monitored with in situ FT-IR spectroscopy. Infrared spectroscopy has been
widely used to investigate the adsorption and the selective catalytic reduction of NOx
[115-119]. In particular, information concerning the kinetics and mechanism of these
reactions may be discerned from this spectroscopic technique. In addition, synthesized
nanocrystalline NaY zeolite with a particle size of 8030 nm was used as the catalyst
instead of commercially available zeolite catalysts. As discussed in a recent publication,
nanocrystalline zeolites may be particularly useful in environmental applications due to
the small particle size and large internal and external surface areas [14, 35].
5.2.1 Preparation of Nanocrystalline NaY Zeolite
(Particle Size 80 nm)
Nanocrystalline NaY zeolite was synthesized according to methods described in

the literature [82]. Two solutions (A and B) were prepared. Solution A was obtained by
dissolving NaOH and tetramethylammonium hydroxide (TMAOH) in distilled water and
then adding aluminum isopropylate to the alkali solution. Solution B was prepared by
adding tetraethyl orthosilicate (TEOS) to the remainder of the TMAOH solution. Both
solutions were filtered through a 0.2 m filter paper, then mixed together and stirred
overnight at room temperature. The clear solution mixture obtained was then heated in a
Teflon-lined stainless steel autoclave for 3 days at 403 K. The resulting solution was
centrifuged for 30 min at 3400 rmp. The product was washed with distilled water and
dried in air.
The synthesized zeolite was characterized using powder X-ray diffraction
(Siemans D5000) to assess the crystallinity and to verify the identity of the zeolite.
Scanning electron microscopy was used to determine particle morphology and particle
size. The average particle size was determined to be 8030 nm from SEM (Figure 5.1).
The internal and external surface areas of the synthesized zeolite were 474 and 46 m2/g,
66
respectively. Based on the method described in Section 4.3.1, the particle size of the
zeolite was calculated using the external surface area to be approximately 88 nm, in
reasonable agreement with the SEM measurement.
27
Al and 29Si magic angle spinning
(MAS) NMR (300 MHz wide bore, Tecmag) were performed on the sample. NMR
signals from tetrahedral Al and Si atoms were identified. The Si/Al ratio was determined
to be 1.7 from 29Si MAS NMR.
0.3
Number Density
0.25
0.2
0.15
0.1
0.05
40 53
67 80
93 107 120 133 147 160 187 200 227 253 267
Size (nm)
Figure 5.1 Nanocrystalline NaY zeolite with a particle size of 8030 nm: a
representative SEM image (left) and the particle size distribution (right).
5.2.2 FT-IR Measurements
Approximately 8 mg of nanocrystalline NaY zeolite mixed with some water were

coated onto the tungsten grid described in Figure 2.1. The NaY sample was gradually
heated under vacuum at 573 K or higher temperature to remove adsorbed water.
Reactant gases were loaded into the zeolite through the gas handling system (Figure 2.3).
Typically the zeolite was equilibrated with each gas prior to a spectrum being recorded.
Pyridine adsorption experiments were conducted according to the method described in
67
Section 4.3.3.2. In SCR reactions, NO2 and propylene (1:1 ratio) were mixed in the
premix chamber before being introduced into the IR cell at room temperature (Figure
2.3). The extinction coefficient of individual gases was calibrated using the characteristic
IR absorption band and measuring the pressure using an absolute pressure transducer, as
described in Appendix B.
5.3 Results
5.3.1 An Investigation of the Nature of the Surface Sites
on Nanocrystalline NaY with A Particle Size of 80 nm
The FT-IR spectra after heating the NaY zeolite overnight under vacuum at 573 K
and 673 K are shown in Figure 5.2.
3651
3694
0.05
3605
3723
A
b
s
o
r
b
a
n
c
e
3745
(a)
(b)
3800
3750
3700
3650
3600
3550
3500
3450
3400
Wavenumber (cm-1 )
Figure 5.2 FT-IR spectra of surface hydroxyl groups in nanocrystalline NaY zeolite after
heating to: (a) 573 K and (b) 673 K overnight under vacuum. Spectra were recorded at
298 K. The bland grid is used as a reference.
68
The spectrum recorded after heating to 673 K shown in Figure 5.2 (b) is fit to five
bands with Lorentzian line-shapes. Five absorption bands due to hydroxyl groups were
identified in the infrared spectra. Although there is an overall decrease in intensity in the
bands in this region after heating from 573 K to 673 K, the five absorption bands are still
evident at 673 K. The relatively intense high frequency absorption band at 3745 cm-1 is
assigned to terminal silanol groups that are on the external surface of the particles. This
feature is of much weaker intensity in a commercial sample of NaY zeolite that has a
much larger particle size [14]. Thus, this result confirms that the synthesized NaY zeolite
consists of small particles. The next highest frequency absorption band around 3723 cm-1
is associated with hydroxyl groups occurring at defect sites (also called hydroxyl nests)
[16]. The absorption band at 3694 cm-1 has been assigned to surface hydroxyl groups
near Na+ sites, while the bands at 3651 and 3605 cm-1 are associated with hydroxyl
groups attached to extra framework aluminum (EFAL) species [17, 85] and adsorbed
water, respectively.
Several probe molecules were used to detect the presence of additional surface
adsorption sites for the synthesized nanocrystalline zeolite Y used in this study. The
adsorption of propylene in NaY results in an absorption band at 1635 cm-1 (Figure 5.3)
assigned to the C=C stretching mode and several bands between 1350-1500 cm-1
associated with the CH3 deformation mode and the CH2 scissor mode. Absorptions due
to C-H stretching modes between 2800 and 3100 cm-1 were also observed (not shown).
All of the absorption features disappeared after evacuation of gas-phase propylene at
room temperature for several minutes. With excess propylene in the IR cell, no
pronounced change of the band at 1635 cm-1 was observed at room temperature for 6 h,
indicating that propylene did not undergo polymerization in the nanocrystalline NaY
zeolite. It can therefore be concluded that strong Bronsted acid sites that initiate
propylene polymerization are not present in the nanocrystalline NaY zeolite pretreated at
573 K [92]. The FT-IR spectra shown in Figure 5.2 are consistent with this conclusion as
69
strong Bronsted acid sites are typically bridging OH groups with vibrational frequencies
of 3630 and 3560 cm-1 [17, 91]. These bands are not observed in the infrared spectra of
nanocrystalline NaY shown in Figure 5.2.
0.05
1455
1635
A
b
s
o
r
b
a
n
c
e
1700
1442
1433
Pressure increasing
1650
1600
1550
1500
1418
1450
1380
1400
1350
Wavenumber (cm-1 )
Figure 5.3 FT-IR spectra of propylene adsorbed in nanocrystalline NaY zeolite at 298 K
as a function of propylene pressure (P = 0.050, 0.493, 1.019, 2.028 and 4.061 Torr). All
spectra use the clean nanocrystalline NaY zeolite prior to adsorption as a reference. In
addition, gas-phase absorptions have been subtracted from each of the FT-IR spectra.
Pyridine adsorption, however, revealed the existence of weaker Bronsted acid

sites as well as Lewis acid sites in the NaY zeolite. Figure 5.4 displays the FT-IR spectra
of pyridine molecules adsorbed in the nanocrystalline NaY zeolite following desorption
at three different temperatures (298, 373 and 473 K). It is well known that weakly
adsorbed pyridine with characteristic vibrational frequencies of 1617, 1592, 1574 and
1441 cm-1 desorbs near 470 K [25, 91]. In contrast, pyridinium ions with characteristic
vibrational frequencies of 1632 and 1545 cm-1 generated by reaction with Bronsted acid
sites desorb at much higher temperatures. Similarly, pyridine molecules that adsorbed at
70
Lewis acid sites are also more strongly adsorbed. Pyridine molecules adsorbed at Lewis
acid sites can be identified from characteristic absorptions at 1621 and 1454 cm-1. All
three types of adsorbed pyridine molecules contribute to the absorption at 1489 cm-1. It
can be seen from the infrared spectrum at 298 K (Figure 5.4 (a)) that all three types of
adsorbed pyridine molecules are observed. After heating to the higher temperatures, only
pyridine molecules adsorbed on Bronsted and Lewis acid sites are apparent in the
infrared spectra. The Bronsted sites may be associated with the surface hydroxyl groups
attached to EFAL species.
(a)
0.002
1441
0.02
A
b
s
o
r
b
a
n
c
e
(b)
(c)
1592
3717
3747
3695
1489
3653
1454
1617
1621
1574
1632
1545
(a)
(b)
(c)
1650
1600
1550
1500
Wavenumber (cm-1 )
1450
Figure 5.4 FT-IR spectra of pyridine molecules adsorbed in nanocrystalline NaY zeolite
at 298 K following pyridine desorption at: (a) 298 K, (b) 473 K and (c) 573 K. Spectra
were recorded at 298 K. All spectra use the clean nanocrystalline NaY zeolite prior to
adsorption as a reference. The inset shows the spectral region extending from 3635 to
3760 cm-1.
71
5.3.2 Adsorption and Thermal Reduction of NO2 in

Nanocrystalline NaY
The adsorption and reaction of NO2 on surfaces have been extensively studied due
to its environmental importance. Difference infrared spectra recorded of the NaY zeolite
as a function of NO2 pressure are shown in Figure 5.5. It is evident from the spectra that
exposure to NO2 results in the formation of several surface species whose frequencies fall
into two spectral regions: 1200-1660 cm-1 due to NO3- and NO2-, and 1900-2300 cm-1 due
to NO+.
1407
0.2
A
b
s
o
r
b
a
n
c
e
1327
3649
3694
3560
3716
Pressure
increasing
3746
2187
2124
2450
2024
1554
1581
1627
1764
1658
3600
3200
2800
2400
Wavenumber
2000
1230
1410 1386
1600
1200
(cm-1 )
Figure 5.5 FT-IR spectra of NO2 adsorbed in nanocrystalline NaY zeolite at 298 K as a
function of NO2 pressure (P = 0.055, 0.107, 0.498 and 1.033 Torr). Spectra were
recorded at 298 K. All spectra use the clean nanocrystalline NaY zeolite prior to
adsorption as a reference. In addition, gas-phase absorptions have been subtracted from
each of the FT-IR spectra.
72
Previous studies on nitrogen oxides adsorption make the assignment of the spectra
relatively straightforward [23, 24, 120-123]. At low pressures, a doublet with absorbance
maxima at 1410 and 1386 cm-1 can be attributed to the 3 mode splitting of surface nitrate
(NO3-) adsorbed on Na+ sites. At higher loading, the doublet converges into a singleton
with a frequency of 1407 cm-1. This broadening of the nitrate bands is most likely due to
intermolecular interactions [88, 89]. The absorption at 1230 cm-1 is associated with
adsorbed nitrite (NO2-). Absorption bands at 1627, 1581 and 1554 cm-1 are associated
with nitrate absorptions on the EFAL bonded in different coordinations (bridging,
bidendate and monodentate, respectively) [102]. The absorption at 1327 cm-1 is also
associated with nitrate adsorption on EFAL sites.
The absorption bands within the spectral region between 1900 and 2300 cm-1 are
assigned to nitrosonium ions (NO+) [89, 90]. At low pressures, two bands at 2024 and
2124 cm-1 are due to the formation of NO+ species adsorbed onto different cationic
position in the zeolite framework. At higher pressure, an absorption feature at 2187 cm-1
develops. This feature can be assigned to the N-O stretching mode from [NO+][NO2] or
[NO+][N2O4] adducts adsorbed onto Lewis base sites [88, 89]. Two other bands at 2450
cm-1 and 1764 cm-1 can be assigned to an overtone and combination bands of adsorbed
nitrates [119].
In the hydroxyl group region, loss of three bands at 3694, 3716 and 3746 cm-1
was observed because of either the displacement or interaction of hydroxyl groups with
adsorbed NOx- species (Figure 5.5). Perturbed hydroxyl groups on EFAL species gave
rise to two absorptions at 3649 and 3560 cm-1 [88]. At low NO2 pressure, a negative
peak around 1658 cm-1 was visible, indicating the consumption of residual H2O in the
zeolite upon NO2 adsorption.
Thermal reduction of NO2 at temperatures from 298 K to 573 K was investigated.
FT-IR spectra of the gas phase were taken at two minutes time intervals. The
concentration of the gases in the infrared cell can be determined from the integrated
73
absorbances of the bands at 1616 cm-1 (NO2), 1875 cm-1 (NO) and 2224 cm-1 (N2O) using
calibrated extinction coefficients. The time course concentration of gas-phase products in
thermal reduction of NO2 at 373 K is shown in Figure 5.6. Immediately after heating
began, the concentration of NO2 in the gas phase jumped to a maximum and then
decreased quickly. Clearly a large amount of NO2 was released from the NaY zeolite
upon heating, possibly via the reaction of highly active surface nitrate [124]. NO and
N2O were the only other gas-phase products detected by the FT-IR spectroscopy.
50
Concentration (10-6 mol L -1)
NO 2
NO
40
N 2O
30
20
10
0
0
50
100
150
200
250
300
350
Time (min)
Figure 5.6 Time course concentration of gas-phase NO2, NO and N2O during the thermal
reduction of NO2 in nanocrystalline NaY at 373 K.
The product distribution and conversion of both gaseous NO2 and adsorbed NO3at three different temperatures are listed in Table 5.1. In thermal reduction, the total
loading of NO2 was controlled to be 505 mol L-1. The initial concentration of gasphase NO2 after adsorption equilibrium was 142 mol L-1. In thermal reduction of NO2,
NO was the major product. N2O accounted for less than 2% of the products in the
temperature range studied. More than 70% of gas-phase NO2 decomposed after reaction
74
at 373 K for 6 h. The efficiency reached 100% when the temperature was above 523 K.
The concentration of N2 and O2 could not be monitored in this study. However, based on
the nitrogen mass balance, it appears that little N2 and O2 were formed in the thermal
reduction of NO2.
Table 5.1 Product distribution and conversion (%) of gas-phase

NO2 and adsorbed NO3- in the thermal reduction of NO2 in
nanocrystalline NaY at different temperatures for 6 h.
Reaction
Temperature
Gas Products (mol L-1)

NO
N2O
% Conversion a
NO2
NO3-
298 K
10.1
79
11
373 K
23.6
0.2
71
64
473 K
51.3
0.8
92
90
Conversion of NO2 refers to change of NO2 in gas phase; conversion of

NO3- describes the change associated with the integrated absorbance in
the entire 1240-1660 cm-1 spectral region.
In order to measure the thermal stability of the various surface species in the
zeolite, NO2 was first adsorbed in nanocrystalline NaY zeolite at 298 K and the sample
was then heated to different temperatures. Figure 5.7 shows the FT-IR spectra of surface
species adsorbed in nanocrystalline NaY zeolite after evacuating the zeolite at different
temperatures. Surface adsorbed NO+ was removed completely by evacuating at 298 K
for 30 min, while less than 10% of the original NO3- species still remained adsorbed in
the zeolite after heating to 573 K. The quantitative study by Sedlmair and co-workers has
shown the relative stability of the NOx surface species as following: Na+-nitrates > Alnitrates > nitrites [88]. In our work, nitrates also appear to be the more stable species
relative to nitrites and nitrosonium ions.
75
1386
1410
0.1
A
b
s
o
r
b
a
n
c
e
1327
1554
1581
1627
2187
2024
1764
(a)
(b)
(c)
(d)
(e)
2200
2000
1800
W avenumber
1600
1400
1200
(cm-1 )
Figure 5.7 FT-IR spectra of NO2 adsorbed in nanocrystalline NaY zeolite after
adsorption at: (a) 298 K followed by evacuating at (b) 298 K, (c) 373 K, (d) 473 K and
(e) 573 K for 30 min. Spectra were recorded at 298 K. All spectra use the clean
nanocrystalline NaY zeolite prior to adsorption as a reference. In addition, gas-phase
5.3.3 Adsorption of NO2 and Propylene (1:1) Mixture

in Nanocrystalline NaY
In the propylene-SCR reaction, NO2 and propylene (1:1 mole ratio) were mixed in
prior to being introduced into the infrared cell. The NaY zeolite was equilibrated with an
NO2 and propylene (1:1) mixture for 30 minutes. The FT-IR spectra of the resulting
mixture adsorbed in NaY zeolite at 298 K are shown in Figure 5.8. Thermal stability of
the surface compounds was investigated to assist in making spectral assignments. In the
experiments regarding the thermal stability of surface compounds, the NaY zeolite was
heated to different temperatures for 30 min after evacuating the gas phase, and cooled
back to room temperature before recording each of the spectra shown in Figure 5.9.
76
1406
1387
1354
0.2
A
b
s
o
r
b
a
n
c
e
Pressure increasing
1647
1557
1528
1508
3562
2985
1282
2451
2317
3650
1320
1943
1969
3747
1723
1763
3695
3600
3000
2400
2000 1900 1800 1700 1600 1500 1400 1300 1200

Wavenumber (cm -1 )
Figure 5.8 FT-IR spectra of NO2 and propylene (1:1) mixture adsorbed in
nanocrystalline NaY zeolite at 298 K recorded as a function of increasing total pressure
(P = 0.100, 0.200, 0.623, 1.005 Torr). All spectra use the clean nanocrystalline NaY
zeolite prior to adsorption as a reference. In addition, gas-phase absorptions have been
subtracted from each of the FT-IR spectra.
In comparison to the adsorption of NO2 alone, similar absorption features were

observed in the NO3- region following the adsorption of NO2 and propylene mixture
(Figure 5.8). In addition, several new absorption bands were also observed. It is
important to note that the spectra contain a wealth of information and there is some
considerable overlap of some of the bands in several spectral regions. For example, both
Na+-nitrates and surface adsorbed organic nitro compounds can contribute to the intensity
of the observed band at 1387 cm-1. The absorption at 1647 cm-1 is assigned to organic
nitrito species, but it may also, in part, be due to the formation of carbonyl groups in
primary amides or the C=N stretching mode of oximes [125, 126]. Therefore, careful
77
analysis of the FT-IR spectra is important in identifying the adsorbates present.

Assignments of the absorption bands shown in the FT-IR spectra in Figures 5.8 and 5.9
are given in Table 5.2. Most assignments are based on the assignments made in several
previous studies reported in the literature [127-131].
1404
1387
0.2
A
b
s
o
r
b
a
n
c
e
1354
1557
1647
1528
1320
1282
3563
2985
3649
1694
2451
1723
2317
1943
3747
2272
2239
(a)
(b)
(c)
2355 2187
3600
3000
2400
1598
1483
2000 1900 1800 1700 1600 1500 1400 1300 1200

Wavenumber (cm -1 )
Figure 5.9 FT-IR spectra of adsorbed species following adsorption of an NO2 and
propylene (1:1) mixture at: (a) 298 K and heating to (b) 373 K and (c) 473 K. Spectra
were recorded at 298 K. All spectra use the clean nanocrystalline NaY zeolite prior to
adsorption as a reference. In addition, gas-phase absorptions have been subtracted from
each of the FT-IR spectra.
Adsorption of the NO2 and propylene mixture results in the formation of several
oxygenated and nitrogen-containing organic compounds. As seen in Figures 5.8 and 5.9,
adsorbed oxygenated hydrocarbons (CxHyOz) with absorptions at 2985, 1723, 1694,
1528, 1508 and 1354 cm-1 appear immediately in the spectrum recorded at 298 K.
78
Carbonyl group containing compounds may come about from the direct oxidation of
adsorbed propylene by adsorbed NOx- [119, 124, 132]. Two bands at 1528 and 1508 cm-1
develop quickly after the adsorption of NO2 and propylene mixture. These have been
previously assigned to carbenium ions formed upon adsorption of olefins on acidic
zeolites [133]. Since there are no strong Bronsted acid sites in the nanocrystalline NaY
zeolite used in our study, these two bands are most likely due to the formation of
monodentate carbonate ions and formate ions, respectively [134]. These organic
compounds may be intermediates or possibly final products in the SCR reaction.
Table 5.2 Assignment of infrared absorption bands observed following adsorption and
thermal reaction of NO2 and propylene (1:1) mixture in nanocrystalline NaY zeolite.
Bands (cm-1)
Assignments
Possible Species
References
(OH)
Oximes
[115, 125]
2985
(=CH2)
RCH=CH2
[127]
2317
(NCO)
SiNCO
[128, 129]
2272
(NCO)
HNCO
[106, 129]
2239
(NCO)
RNCO
[116, 126, 135-138]
2187
(CN)
RCN
[126, 137, 138]
1969
(NO)
Adsorbed NO
[121]
1943
(NO)
Adsorbed N2O3
[121]
1723, 1694
(CO)
Carbonyl species
[106, 125]
(ONO)
RONO (or RCONH2)
[106, 117, 118, 125]
1598,1483
Carbonaceous species
[131]
1557,1387
(NO2)
RNO2
[106, 117, 118, 125]
3300-3150
1647
2-
1528
(CO3 )
Carbonate ion
[127]
1508
(COO-)
Formate ion
[127]
1354
(CH3)
RCH3
[106]
1282
(CO)
RONO
[130]
79
Surface adsorbed NO (1969 cm-1) and N2O3 (1943 cm-1) all appear upon
adsorption of the NO2 and propylene mixture. These species are stable up to hours at
room temperature [135]. However, they disappeared quickly upon heating to 373 K.
Another surface species that appears to have similar stability is identified as NCO species
on Si4+ sites. NCO is identified by the absorption at 2317 cm-1 and has been identified in
previous studies using isotope labels [106]. Two surface species characterized by fairly
intense absorption bands at 1557 cm-1 and 1647 cm-1 appear to be more stable than NO,
N2O3 and NCO as these bands persist in the spectra even after heating to 373 K. Yeom et
al. investigated the adsorption and reactions of nitromethane in zeolite Y [106]. Based on
their isotopic studies, the two intense absorption bands are assigned to organic nitro
(1557 cm-1) and nitrito (1647 cm-1) compounds, respectively. Another possible stable
intermediate that some consider to be important in the SCR reactions is oxime. Oxime
may be responsible for the broad absorbance between 3150 and 3300 cm-1 [115].
Most of the absorption bands decreased in intensity after the nanocrystalline NaY
zeolite was heated at elevated temperature. Ketones or aldehydes are produced at 373 K
as indicated by the band at 1694 cm-1 but then convert into other products by heating to
473 K. Surface adsorbed carbon dioxide (2355 cm-1) and isocyanic acid (2272 cm-1)
develop in the zeolite upon heating to 373 K and 473 K (Figure 5.9). At elevated
temperatures, two other absorption peaks appear in the spectra at 2239 and 2187 cm-1.
These two features have been assigned to organic isocyanate and nitrile in several
previous studies using alumina-supported materials as SCR catalysts [116, 126, 135-138].
The assignments of isocyanate and nitrile have been supported by other experiments
involving NO+CO reactions [136]. Since there is EFAL detected in the nanocrystalline
NaY zeolite used in our study, these two absorptions are assigned to organic isocyanate
and nitrile adsorbed on EFAL sites. They most likely come from the thermal reaction of
organic nitro/nitrito compounds or oxime. Unlike the band at 1557 cm-1, the absorption
at 1647 cm-1 persists even after heating to 473 K. This may indicate the formation of
80
primary amide, which can be generated via the hydrolysis of nitriles and/or thermal
reaction of organic nitro/nitrito species. The assignment for the bands at 1598 and 1483
cm-1 is difficult since there are numerous literature assignments for bands at these two
frequencies and are most likely due to the presence of unsaturated carbonaceous residues
or coke [131]. Organic isocyanates (2239 cm-1) and coke are the only compounds
adsorbed in NaY zeolite after evacuating the sample overnight at 673 K.
5.3.4 Low Temperature Propylene-SCR of NO2 in
Nanocrystalline NaY
Propylene-SCR of NO2 at different temperatures was investigated next. The FTIR spectra of the gas phase as a function of time at 473 K are shown in Figure 5.10.
CO2
2348
A
b
s
o
r
b
a
n
c
e
0.002
CO
N2O 2143
2224
NO
1875
NO2
1616
C3H6
1657
C3H6
1459
(e)
(d)
(c)
(b)
(a)
2400
2200
2000
1800
Wavenumber (cm-1)
1600
1400
Figure 5.10 FT-IR spectra of the gas phase following propylene-SCR of NO2 in
nanocrystalline NaY at 473 K for (a) 0 min, (b) 10 min, (c) 30 min, (d) 1 h and (e) 4 h.
81
Figure 5.11 shows the time course concentration of the various gaseous species in
propylene-SCR of NO2 at 473 K.
NO 2
NO
N 2O
15
10
30
Concentration (10-6 mol L-1)
Concentration (10-6 mol L-1 )
20
25
C 3H 6
CO
20
15
10
50
100
150
200
250
300
350
Time (min)
0
0
50
100
150
200
250
300
350
Time (min)
Figure 5.11 Time course concentration of gaseous N-containing species during

propylene-SCR of NO2 in nanocrystalline NaY at 473 K. The inset shows the decay and
evolution of propylene and carbon monoxide, respectively.
Immediately after the heating process begins, gaseous NO2 decreases and is
completely gone after 25 min. The concentrations of NO and N2O initially increase until
a maximum is reached at approximately t = 10 min, then gradually decrease to zero.
Propylene (1459 and 1657 cm-1), CO (2143 cm-1) and CO2 (2348 cm-1) were the only
carbon-containing molecules detected in the gas phase. Although not directly measured,
N2 and O2 most likely form as well based on the estimated nitrogen mass balance. Some
water is apparent in the gas-phase spectra as well. A similar time course experiment was
done at 373 K.
82
Table 5.3 summarizes the gas-phase product distribution and percent conversion
of gas-phase NO2 and adsorbed nitrate (vide infra) in propylene-SCR of NO2 at three
different temperatures (298 K, 373 K and 473 K). The total loading of NO2 was 505
mol L-1. In the absence of oxygen, the initial concentration of gas-phase NO2 after
adsorption equilibrium was 164 mol L-1. Based on the nitrogen balance, the complete
reduction of NO2 into N2 and O2 was found at 473 K after 6 h of reaction. Although a
small amount of N2O existed after 6 h at 373 K, it was expected that it too would
disappear with longer reaction time. CO and CO2 were generated as the major carbon
containing products in the gas phase.
Table 5.3 Product distribution and conversion of gas-phase NO2 and adsorbed
NO3- in propylene-SCR of NO2 at different temperatures after 6 h of reaction time.
Reaction
Temperature
Product Distribution (mol L-1)

NO
N2O
CO
% Conversion a
CO2
NO3-
NO2
298 K
12.5
3.0
82
373 K
0.4
24.7
1.6
100
473 K
28.1
8.6
100
29
473 K
40.9
3.7
6.4
13.9
100
68
25.1
4.6
1.3
40.7
100
87
473 K, O2
Conversion of NO2 refers to the change of NO2 in gas phase; conversion of NO3describes the change associated with the integrated absorbance in the entire 1230-1600
cm-1 spectral region (Because of overlapping absorption bands due to other adsorbed
species, this estimate could be in error by as much as 20%). b Propylene-SCR of NO2 was
done using a partially deactivated zeolite catalyst (see text for further details).
a
It should be pointed out that the conversion of NO3- in Table 5.3 stands for the
conversion of species responsible for the absorption bands between 1230 and 1600 cm-1.
Even though some of the absorptions in the region are due to other adsorbates, it can still
83
be an approximate measure of how much NO3- species reacted since NO3- accounts for
most of the absorbance in that region (estimated to be ~90%).
Complete conversion of NO2 did not occur when coke accumulated in the NaY
zeolite (referred to as partially deactivated NaY zeolite in Table 5.3). Propylene-SCR
of NO2 at 473 K in a partially deactivated NaY zeolite converted approximately 90% of
loaded NO2 into NO and N2O. The evolution of NO2, NO and N2O with time was similar
to that of thermal reduction, except for the higher concentration of N2O. Compared to
SCR reaction in active NaY zeolite, less propylene was consumed, but the CO2 to CO
ratio and conversion of NO3- were much higher. More undesired products, i.e. partially
oxidized organics, were generated in the zeolite as well. These products include surface
adsorbed carbonyl compounds (1723 and 1694 cm-1).
Although an excess amount of oxygen is present under typical lean burn
conditions, only propylene and NO2 were included in the starting reaction mixture in this
study for several reasons. First, an intermediacy of NO2 in SCR of NO has been
suggested [105]. Furthermore, the oxidation of propylene by oxygen competes with the
SCR reaction. This is confirmed by the propylene-SCR of NO2 in nanocrystalline NaY
zeolite at 473 K in the presence of oxygen, propylene was found to be consumed within 2
h at 473 K. The concentration of NO reached a maximum and then decreased slowly.
Table 5.4 summarizes the initial rates of formation or loss of gas phase species in
catalytic reduction of NO2. The initial rates were obtained by linear fitting the
concentration of individual gaseous species in the first 10 min of the reaction. NO for
thermal reduction, NO2, propylene and CO for SCR reaction were chosen because their
concentration changed monotonically with reaction time. In thermal reduction of NO2,
the initial rate of NO formation increased with temperature. Propylene-SCR of NO2 did
not occur to any great extent in nanocrystalline NaY zeolite at room temperature. In
propylene-SCR of NO2, the initial rates at 373 K were approximately the same as those at
473 K. This reveals a free radical behavior for the early stage of SCR reactions as the
84
initial rates are independent of temperature once a certain threshold temperature is

reached in which the free radicals are formed. The high CO to CO2 ratio in Table 5.3
also suggests a free radical pathway [106].
Table 5.4 Initial rates (10-2 mol L-1 s-1) of formation or loss
of gas phase species in first 10 min of SCR reactions.
Reaction
Temperature
Thermal
Propylene-SCR
NO2
(loss)
NO
C3H6
(loss)
CO
298 K
0.1
0.4
373 K
1.0
2.1
0.7
2.9
473 K
2.8
2.0
0.8
2.6
473 K
0.9
0.3
0.4
Propylene-SCR of NO2 was done using a partially deactivated

zeolite catalyst (see text for further details).
5.4 Discussion
5.4.1 Formation of NO in Thermal Reduction of NO2
It was found that the adsorption of NO2 on ion exchanged Y zeolite occurred
preferentially via surface nitrates on exchangeable metal cation sites and partly on
cationic EFAL [88]. A second pathway has been proposed involving the reaction
between strongly adsorbed water and NO2 [89, 108]. Other studies indicate the
possibility that nitrate is formed via two or more different mechanisms that populate
different sites [106]. The disproportionation of NO2 has been suggested to explain the
formation of surface species and NO over CuZSM-5 [124]. Underwood and co-workers
proposed a mechanism involving the formation of nitrite first followed by nitrate [123].
85
Based on the experimental observations, the formation of surface nitrate and

nitrite in NaY zeolite can be best explained by reactions 5.1 and 5.2.
2NO2 + H2O HNO3 + HONO
(5.1)
2NO2 NO+ + NO3-
(5.2)
Reaction 5.1 is the minor path since only a trace amount of water was left in the
zeolite after heating to 673 K under vacuum. However, adsorption of NO2 happens
preferentially via reaction 5.1 in the presence of water, as indicated by the formation of
nitrite (1230 cm-1) in the spectra shown in Figure 5.5.
In this study, decomposition of NO2 should involve the participation of NO2 from
the gas phase. An equilibrium might exist between surface NOx- species and gaseous
NO2 (reaction 5.3). At elevated temperature, NO2 was released from the NaY zeolite and
reacted with surface cationic sites (reaction 5.4a) or nitrite (reaction 5.4b). Mn+
represents Na+ or EFAL sites.
Mn+-NO3- NO2 + Mn+-O-
(at elevated temperature)
(5.3)
Mn+-O2- + 2NO2 Mn+-NO3- + NO
(5.4a)
Mn+-NO2- + NO2 Mn+-NO3- + NO
(5.4b)
The small amount of N2O produced in thermal reduction of NO2 might come from
the disproportionation of NO [24, 88, 139]. Significant thermal reduction of NO to N2O
was observed only at temperature above 573 K.
Notably, the concentration of NO reached a maximum at t = 150 min and was
constant thereafter (Figure 5.6). Conversion of surface NO3- species increased from 64%
at 373 K to 90% at 473 K (Table 5.1). This means that some of the NO3- species are
86
more strongly adsorbed and less reactive. Two different reactivity classes of surface
nitrates in zeolite Y have been observed in other studies [106].
5.4.2 Formation of NO and N2O in the Adsorption of
NO2 and Propylene (1:1) Mixture at 298 K
The concentration of NO and N2O upon adsorption of the NO2 and propylene
mixture (Figure 5.11) was consistently higher than that after the adsorption of NO2 alone
(Figure 5.6). Surface adsorbed NO and N2O3 (1969 and 1943 cm-1, respectively) were
observed upon exposure of the mixture to NaY zeolite (Figure 5.8), but not in the
adsorption of NO2 alone (Figure 5.5).
Although NO+ was not observed in the adsorption of NO2 and propylene mixture,
it could play an important role due to its low thermal stability and high reactivity.
Gerlach and co-workers studied the reactivity of NO+ towards propylene over acidic
Mordenite [115]. Their study demonstrated that oxime was produced in the reaction
between NO+ and propylene at 393 K and converted into nitrile at higher temperature. In
this study, the disappearance of NO+ by propylene adsorption was observed at room
temperature. First the NaY zeolite was equilibrated with NO2 at room temperature. Then
propylene was introduced to NO2 pre-adsorbed NaY zeolite. NO+ vanished immediately
upon the introduction of propylene. At the same time, surface adsorbed NO and N2O3 are
observed. Other new surface species observed in adsorption of NO2 and propylene
mixture (Figure 5.8) appear within 1 min after adsorption. The intensity of 1407 cm-1
band characterizing surface Na+-nitrates didnt change significantly. This implies that at
298 K surface Na+-nitrate is inactive toward reaction with adsorbed propylene [106].
Thus it is suggested from the data that NO+ and NO3- on EFAL produced upon
adsorption of NO2 (reaction 5.2) readily react with adsorbed propylene to form NO and
organic nitro/nitrito compounds (reaction 5.5 and 5.6).
NO+ + CH3-CH=CH2 [CH3-CH=CH2]+ + NO
(5.5)
87
Mn+-NO3- + [CH3-CH=CH2]+ C3H5NO2 + Mn+-OH
(5.6)
The high reactivity of the -H in propylene and the stabilization of allylic species
by cations in zeolite Y could facilitate these two reactions [140, 141]. Reaction 5.6 can
lead to the generation of acidic hydroxyl groups. This is consistent with the appearance
of a broad band at 3562 cm-1 associated with hydroxyl groups on EFAL (Figure 5.8).
Oxime was then produced in the reaction between NO+ and adsorbed propylene in
presence of the weak Bronsted acidic sites. It is also possible that organic nitro/nitrito
compounds came from the direct reaction between adsorbed propylene and certain nitrate
species, most likely those attached to EFAL sites populated on the external surface of the
nanocrystalline NaY zeolite used in this study.
5.4.3 Reaction Stoichiometry in Propylene-SCR of NO2
In the absence of oxygen, the balanced reaction for complete reduction of NO2 by
propylene can be written either as
(9/2)NO2 + C3H6 = 3CO2 + (9/4)N2 + 3H2O
(5.7)
3NO2 + C3H6 = 3CO + (3/2)N2 + 3H2O
(5.8)
assuming CO2 or CO as the only carbon-containing products, respectively.

As shown in Table 5.3, a high CO to CO2 ratio was observed at both 373 K and
473 K. Therefore, it can be concluded that the majority of the propylene-SCR of NO2 in
nanocrystalline NaY zeolite occurs via reaction 5.8 at low temperatures. The
concentration of CO2 was much higher at 473 K than 373 K. This can be easily
understood since higher temperature promotes decomposition of oxygenated species and
the oxidation of CO to CO2. The initial rates for NO2 loss and CO formation at both 373
K and 473 K are approximately 3 folds that found for propylene (Table 5.4). Thus, the
stoichiometry shown in reaction 5.8 is in agreement with the measured rate data.
88
Water was also observed in the gas phase in propylene-SCR of NO2 at both 373 K
and 473 K (Figure 5.10). However, surface adsorbed water was not present to any great
extent in the zeolite after SCR reaction for 6 hours. It is most likely that surface adsorbed
water was consumed in the hydrolysis of surface species, such as organic nitrile.
It is difficult to write the balanced reaction for propylene-SCR of NO2 in the
presence of excess oxygen because of the multiple reaction pathways involved and the
complexities of these reactions. Complete oxidation of propylene by oxygen to CO2 and
water does occur in nanocrystalline NaY zeolite at 473 K. This can be seen by the high
CO2 to CO ratio (Table 5.3) and the existence of surface adsorbed water after SCR
reaction for 6 hours. As mentioned previously, NO reduction occurred very slowly once
propylene was completely consumed. The low SCR reaction rate is possibly due to a
combination of no propylene and the presence of surface adsorbed water. Although
oxygen does decrease the zeolite deactivation by oxidizing carbonaceous deposits [105,
142], overall it is detrimental to the propylene-SCR of NO2.
5.4.4 A Mechanism for Propylene-SCR of NO2 at Low
Temperature
Several studies have focused on the mechanism for HC-SCR of NOx. Most of
them have suggested that isocyanate is an important intermediate. The most
controversial and least understood aspect of the selective catalytic reduction of NO2 is the
formation of the nitrogen-nitrogen bond which is necessary for the production of N2. In
the SCR reaction over FeZSM-5, isotopic studies have shown that the nitrogen in N2
comes from both the gas phase NOx and surface nitrogen-containing species [143].
Radical mechanisms involving azoxy or dinitroso species have been proposed to explain
the formation of dinitrogen in zeolites containing transition metals [114, 140]. Others
have considered -NH complexes as the key intermediate [126]. Recently, Yeom and coworkers studied the reactions of acetaldehyde, acetic acid and nitromethane with NO2 on
89
BaNa-Y zeolite [106]. Their study led to the conclusion that the formation of ammonium
nitrite (NH4NO2) is a crucial step for deNOx catalysis. They also suggested a parallel
reaction channel involving free radicals to explain reaction pathways which result in a
high CO to CO2 ratio.
Based on the observed kinetics and the discussion of the literature above, a
mechanism involving free radicals appeared to be the most plausible pathway for
propylene-SCR of NO2 in nanocrystalline NaY zeolite. Thermal treatment is necessary
for C-H bond dissociation in propylene [144]. At elevated temperatures, the SCR
reaction is initiated by the abstraction of H from adsorbed propylene to yield the allyl
radical, according to reaction 5.9 [145, 146].
NO2 + CH3-CH=CH2 CH2-CH=CH2 + HNO2
(5.9)
The allylic radical can react with HNO2 to give oxygenated species and NO
(reaction 5.10).
HNO2 + CH2-CH=CH2 C3H6O + NO
(5.10)
Upon thermal treatment, surface nitrogen containing compounds (organic

nitro/nitrito species and oximes) adsorbed on EFAL sites are converted to nitrile or
isocyanate. Acrylamide species can be produced via the hydrolysis of organic nitrile or
isocyanate. The reaction between acrylamide and HNO2 (reaction 5.11) provides one
possible pathway for the dinitrogen formation [147, 148].
RCONH2 + HNO2 RCOOH + N2 + H2O
(5.11)
Another possible channel for N-N bond formation is the reaction between acrylamide and
N2O3 in the presence of water.
In propylene-SCR of NO2, isocyanate radical (2317 cm-1) and isocyanic acid
(2272 cm-1) were generated as stable species containing one atom each of nitrogen,
carbon and oxygen. Isocyanic acid adsorbed in nanocrystalline NaY zeolite very strongly
90
and can be removed by heating to 573 K under vacuum. It was considered as the
precursor of ammonia, which is essential for the formation of dinitrogen [106, 149]. In
our study, a minor pathway for nitrogen-nitrogen bond formation might involve the
reactions of isocyanate radical and isocyanic acid as evidenced by the presence of these
species in the FT-IR spectra [137].
5.5 Conclusions
Catalytic reduction of NO2 in nanocrystalline NaY zeolite was investigated using

in situ transmission FT-IR spectroscopy. NO2 was stored in the NaY zeolite as nitrate
and nitrite. Because of the relatively high external surface area of the nanocrystalline
zeolite, there was considerable adsorption on the external EFAL sites. Thermal reduction
of NO2 showed high selectivity to NO formation in the temperature range studied.
In the absence of oxygen, the selective catalytic reduction of NO2 with propylene
at low temperature (T 473 K) resulted in complete reduction of NO2 to N2 and O2 in
nanocrystalline NaY zeolite. Isocyanate and nitrile species formed on external EFAL
sites are found to be important intermediates. A pathway involving radical reactions
appeared to be the most plausible mechanism for the complete reduction of NO2. In the
proposed mechanism, complete reduction of NO2 was initiated by H abstraction from
propylene. The resulting radical, HNO2, then continues on to form N2.
The authors thank Hassan Alwy for providing the nanocrystalline NaY zeolite
sample used in this work and gratefully acknowledge the Environmental Protection
Agency for the support of this work (R829600). The results of this work are presented in
a publication under the authorship of: Gonghu Li, Sarah C. Larsen, Vicki H. Grassian, J.
Mol. Catal. A 2005, 227, 25.
91
CHAPTER 6
SELECTIVE CATALYTIC REDUCTION OF NO2 WITH
PROPYLENE IN NANOCRYSTALLINE NaY (23 nm)
6.1 Introduction
As discussed in Chapter 4, nanocrystalline zeolites may be particularly useful in

environmental applications due to the small particle size and relatively large internal and
external surface areas [14]. Since NOx is mainly stored as adsorbed nitrate and nitrite in
zeolite Y, the high total surface areas of nanocrystalline Y zeolites could offer additional
advantages for NOx storage. Furthermore, the small particle size of nanocrystalline
zeolites can dramatically alter the diffusional properties of these materials. The
dependence of the SCR reaction rate, NOx conversion and catalyst durability on zeolite
particle size has been previously reported by others [150-154]. Ogura and co-workers
reported the dependence of catalytic activity on H-ZSM-5 particle size in deNOx
reactions [150]. Tabata and co-workers studied the intracrystalline diffusion of NO in
ZSM-5 zeolites with different particle sizes [153]. Higher conversion of NOx and
propane in SCR reactions was obtained over zeolites with a smaller particle size. Shichi
and co-workers discovered that the reaction rate and conversion of NOx in SCR reaction
over MFI-type zeolite were influenced by intracrystalline diffusion of hydrocarbons [151,
152]. In addition, it has been reported that the durability of CoHZSM-5 increased with
decreasing particle size [154]. With small particle size and large external surface area,
nanocrystalline zeolites can provide additional adsorption/reaction sites with diverse
functionality and diffusional advantages.
It has been shown conclusively in Chapter 4 that there are a greater number of
silanol groups and EFAL sites detected in nanocrystalline NaY compared to commercial
micrometer-sized NaY. Bronsted acid sites and Lewis acid sites are present in large
quantities in nanocrystalline NaY, but not in commercial NaY zeolite. These acidic sites
92
are associated with EFAL sites located on the external surface of the nanocrystalline
NaY. In this chapter, in situ transmission FT-IR spectroscopy is used to investigate
propylene-SCR of NO2 over nanocrystalline NaY zeolite at low temperature (T = 473 K).
The zeolite particle size used in this study is 23 nm. This represents the smallest
nanocrystalline NaY zeolite material reported in the literatures for use in any SCR
reaction. The performance of nanocrystalline NaY zeolite in SCR reactions is compared
to commercial NaY zeolite with a larger particle size in order to explore the potential
advantage of the small particle size of nanocrystalline NaY zeolite and to determine if
differences in the chemical as well as physical properties of nano-sized zeolite are
important in SCR reactions.
The synthesis and characterization of nanocrystalline NaY with a particle size of

23 nm is described in chapter 4. Commercial NaY was purchased from Aldrich. In this
study, the NaY sample was heated under vacuum at 623 K or higher temperature
overnight to remove adsorbed water. In SCR reactions, NO2 and propylene (1:1 mole
ratio) were first mixed in a chamber before being introduced into the IR cell at room
temperature (Figure 2.3). Thirty minutes were allowed for adsorption equilibrium before
an excess amount of oxygen was added into the IR cell. Time course experiments were
conducted by automatically recording infrared spectra of the NaY zeolite or gas phase
every 5 seconds. In experiments designed to determine the effect of adsorbed water on
the SCR reaction, NaY zeolite was exposed to 0.2 Torr of water vapor before loading
reactant gases.
93
6.3 Results and Discussion

6.3.1 Propylene-SCR of NO2 in NaY Zeolite at 473 K:
Gas Phase Products
The gas-phase product distributions were measured by IR before and after

propylene-SCR of NO2 in NaY (nanocrystalline and commercial) at 473 K and the results
are summarized in Table 6.1.
Table 6.1 Concentration (mol L-1) of gas-phase components before and after
propylene-SCR of NO2 in 7 mg NaY zeolite at 473 K for 6 h.
Before SCR a
Catalyst
NO
Nanocrystalline NaY
N2O
After SCR
C3H6
NO
N2O
C3H6
CO2
11
18
174
Nanocrystalline NaY + H2O
18
37
11
13
33
48
Commercial NaY
16
31
31
22
52
Commercial NaY + H2O
13
37
28
21
60
The loadings of both NO2 and C3H6 are 705 mol L-1; the concentration of oxygen in the
IR cell is 55020 mol L-1.
Gas-phase NO2 was not observed after adsorbing the reactant gases for 30 min at
room temperature in nanocrystalline NaY or in commercial NaY zeolite. However, NO,
N2O, and C3H6 were evident in the gas phase. More nitrogen products (NO and N2O) and
more propylene were observed in the gas phase on commercial NaY compared to
nanocrystalline NaY zeolite suggesting that more NO2 and propylene were adsorbed or
reacted in the nanocrystalline NaY zeolite. It can be seen from the data in Table 6.1 that
water markedly decreased the adsorption capacity of propylene in nanocrystalline NaY
zeolite. The surface sites in nanocrystalline NaY zeolite may be poisoned by water due
to hydrogen bonding with silanol groups on the external surface, which are important
94
adsorption sites for NOx and propylene [39]. The effect of adsorbed water on the
adsorption capacity of NO2 and propylene in the commercial sample is much smaller
compared to the nanocrystalline NaY zeolite.
Next, propylene-SCR of NO2 in nanocrystalline and commercial NaY zeolite was
carried out at 473 K in the presence of an excess amount of oxygen. After 6 h of SCR, no
gas-phase NO2 or NO was observed indicating that conversion of gas-phase NOx
(combined gas-phase NO2 and NO concentrations) was 100% in nanocrystalline NaY
zeolite in the absence of adsorbed water. N2O was the major N-containing gas-phase
product detected with FT-IR spectroscopy. In SCR reactions over the commercial NaY
zeolite, 44% of total loaded NO2 was converted into NO and the selectivity toward N2O
formation was 14% compared to 52% for nanocrystalline NaY zeolite. The relatively
high concentration of propylene and low concentration of CO2 in the gas phase after SCR
reactions also indicate that propylene-SCR of NO2 using commercial sample did not
occur to the same extent as in nanocrystalline NaY zeolite. Based on the nitrogen mass
balance, the selectivity toward N2 formation was estimated to be approximately 20% in
SCR reactions using nanocrystalline and commercial NaY zeolite as the catalyst. CO2
was the major carbon-containing gas-phase product, while CO was detected as the other
gas-phase carbon-containing product with a concentration lower than 15 mol L-1.
Adsorbed water markedly decreased the activity and selectivity of the
nanocrystallline NaY catalyst. In SCR reaction over nanocrystalline NaY zeolite with
adsorbed water, the selectivity toward NO and N2O formation was ~14% and ~36%,
respectively, compared to 0% and ~52% in the absence of adsorbed water. Complete
oxidation of propylene into CO2 was severely inhibited by adsorbed water in
nanocrystalline NaY zeolite. This is indicated by the relatively high concentration of
propylene and low concentration of CO2 in the gas phase after SCR reaction in
nanocrystalline NaY zeolite in the presence of adsorbed water. Adsorbed water did not
appear to influence the reactivity of the commercial NaY sample to any great extent.
95
In order to understand these differences between the propylene-SCR of NO2 in

nanocrystalline NaY zeolite and the commercial NaY zeolite, FT-IR spectroscopy was
further used to (i) identify differences in NO2 adsorption in nanocrystalline NaY zeolite
and commercial NaY at 298 K and (ii) identify differences in NO2 and C3H6 mixture
adsorption in nanocrystalline NaY zeolite and commercial NaY at 298 K.
6.3.2 Adsorption of NO2 in NaY zeolite at 298 K
FT-IR analysis of NO2 adsorption was conducted in the absence and presence of
adsorbed water. Figure 6.1 shows the difference FT-IR spectra of nanocrystalline and
commerical NaY zeolite in the presence of 0.5 Torr NO2 at 298 K. Gas-phase
absorptions have been subtracted from the spectra shown in Figure 6.1 to highlight the
spectra of the adsorbed species.
The assignment of the adsorbed species in the zeolite can be based on several
earlier studies [23, 39, 88, 89, 121]. It is well known that nitrate (NO3-) and nitrosonium
ion (NO+) were produced via the NO2 disproportionation (reaction 6.1) in zeolite Y in the
absence of adsorbed water.
2NO2 NO3- + NO+
(6.1)
In the spectra shown in Figure 6.1, the broad absorption bands near 1409 and
1389 cm-1 can be assigned to the 3 mode of nitrate ions attached to Na+ sites [88, 89].
The absorption bands at 1612, 1579 and 1547 cm-1 are associated with the 3 mode of
adsorbed nitrate bonded in different coordinations (bridging, bidendate and monodentate,
respectively) on the EFAL sties in nanocrystalline NaY zeolite [39, 102]. The absorption
at 1327 cm-1 is also associated with nitrate adsorption on EFAL sites [39]. The
absorption bands at 2131 and 2010 cm-1 are assigned to nitrosonium ions adsorbed onto
different cationic positions in the zeolite framework [88, 89]. Another absorption feature
at 2165 cm-1 can be attributed to the N-O stretching mode from [NO+][NO2] or
96
[NO+][N2O4] adducts adsorbed on Lewis base sites [88, 89]. An absorption feature at
1212 cm-1 readily observed only in the FT-IR spectra of NO2 adsorbed in commercial
NaY sample indicates the formation of nitrosyl ion (NO-) [121], which adsorbs on surface
cationic sites.
0.2
1579
A
b
s
o
r
b
a
n
c
e
2131
1612
2010
2165
1389
1409
1327
1272
1547
1638
5
Nano-NaY
Nano-NaY with H2 O
1212
Aldrich-NaY
Aldrich-NaY with H2 O
2200
2000
1800
1600
W avenumber (cm-1 )
1400
1200
Figure 6.1 FT-IR spectra of 0.5 Torr NO2 adsorbed in nanocrystalline NaY and
commercial NaY zeolite in the presence and absence of 10 mol adsorbed water at 298
K. All spectra use the corresponding clean NaY zeolite prior to adsorption as a
background. In addition, gas-phase absorptions have been subtracted from each of the
FT-IR spectra.
A comparison of the spectrum of NO2 adsorbed in nanocrystalline and

commercial NaY zeolite shows several differences (see Figure 6.1). These include: (1)
nitrate absorptions due to nitrate adsorbed on EFAL sites for nanocrystalline NaY zeolite
97
but not for the commercial NaY sample; (2) the NO+ absorption bands between 1900 and
2300 cm-1 in the spectrum of NO2 adsorbed in commercial NaY zeolite are 5 times more
intense compared to those in nanocrystalline NaY zeolite; (3) broader absorption bands in
spectrum of the nanocrystalline NaY relative to the commercial NaY sample.
Nitrate adsorbed on EFAL sites is only observed in nanocrystalline NaY zeolite.
This can be understood from the characterization studies discussed in Chapter 4 and that
the nanocrsytalline NaY zeolite has a much greater external surface area compared to
commercial NaY zeolite. Since EFAL are associated with external surface, there is a
much greater number of these sites per unit weight for nanocrystalline NaY zeolite
compared to commercial NaY zeolite. The difference in the intensity of NO+ bands
between nanocrystalline and commercial NaY zeolite may also be understood in terms of
EFAL sites. In nanocrystalline NaY zeolite, NO2 adsorption in part occurs on cationic
sites of EFAL species (Mn+-O2-) without the production of NO+, according to reaction 6.2
[88].
Mn+-O2- + NO2 Mn+-NO3-
(6.2)
Furthermore, the FT-IR spectra of SCR surface species in nanocrystalline NaY

show additional broadening of the absorption bands and a greater degree of spectral
overlap and interference relative to those in the commercial NaY sample. This suggests
that the surface adsorption sites in nanocrystalline NaY zeolite are more heterogeneous
than in commercial NaY zeolite. In commercial NaY zeolite, the internal surface
provides most of the adsorption sites and Na+ is the only cationic site. In nanocrystalline
NaY zeolite, this is not the case as the external surface accounts for ~30% of the total
surface. The silanol groups and EFAL species on the external surface of nanocrystalline
NaY zeolite provide further surface sites [14].
FT-IR spectra of NO2 adsorption in the presence of adsorbed water were recorded
and several differences were noted. Nitrosonium ion did not form in NaY zeolites which
98
contained adsorbed water. Instead, an absorption peak at 1272 cm-1 associated with
adsorbed nitrite (NO2-) was identified in the FT-IR spectra of surface NOx species formed
upon exposing of 0.5 Torr pressure of NO2 to NaY zeolites (Figure 6.1). In the presence
of surface adsorbed water (as indicated by the H2O bending mode absorption at 1638 cm1
), NO2 reacts preferentially according to reaction 6.3 [39, 108].

(6.3)
The nitric and nitrous acids deprotonate and adsorb on surface cationic sites to
form surface nitrate and nitrite.
6.3.3 Adsorption of NO2 and C3H6 Mixture in NaY
Zeolite at 298 K
To further understand the differences in the potential of nanocrystalline NaY

zeolite in deNOx catalysis compared to the commercial NaY sample, the co-adsorption of
NO2 and C3H6 was investigated. NO2 and C3H6 were first mixed (1:1 mole ratio) prior to
being introduced into the IR cell at 298 K. Thirty minutes was allowed for adsorption
equilibrium. The FT-IR spectra of resulting surface species in nanocrystalline NaY
zeolite and commercial zeolite sample are shown in Figure 6.2.
Although there are considerable spectral overlap and interference, several surface
species can be identified based on previous reports in the literature. In the spectra shown
in Figure 6.2, adsorbed oxygenated hydrocarbons (CxHyOz) can be identified by several
absorptions near 1698, 1614, 1525, 1510, 1354, and 1317 cm-1. For example, carbonyl
containing compounds ((C=O) at 1698 cm-1) may come about from the direct oxidation
of adsorbed propylene by NO3- [119, 124, 132]. Two bands at 1525 and 1510 cm-1 are
most likely due to the formation of monodentate carbonate ions and formate ions,
respectively [134]. The absorption at 1614 cm-1 can be assigned to carboxylic acids. The
absorption peaks at 1354 and 1317 cm-1 can be assigned to surface CHx species [106]. A
99
number of nitrogen containing surface species can also be identified in the spectra shown
in Figure 6.2. The broad band with a maximum at 1406 cm-1 characterizes nitrate species
adsorbed on Na+ sites as discussed in the previous section. An absorption feature near
1382 cm-1 most likely is due to the formation of Na+-nitrate, organic nitro compound
and/or surface CHx species [39].
0.2
1554
1614
A
b
s
o
r
b
a
n
c
e
1382 1317
1406 1354
1698
1525
1283
1510
1635
Nano-NaY
1278
Nano-NaY
with H2 O
1640
1562
Aldrich-NaY
Aldrich-NaY
with H 2O
1800
1700
1600
1500
1400
-1
W avenumber (cm )
1300
1200
Figure 6.2 FT-IR spectra of nanocrystalline NaY and commercial NaY zeolite in the
presence and absence of adsorbed water at 298 K following the adsorption of NO2 and
C3H6 (1:1) mixture. All spectra use the corresponding clean NaY zeolite prior to
adsorption as a background. In addition, gas-phase absorptions have been subtracted
In the spectrum of nanocrystalline NaY zeolite after NO2 and C3H6 adsorption
(Figure 6.2), two relatively intense absorption bands at 1554 cm-1 and 1635 cm-1 are also
100
apparent. Yeom and co-workers investigated the adsorption and reactions of

nitromethane in zeolite Y [106]. With the aid of isotopic labeling, nitromethane and
methyl nitrite were identified with absorption peaks at 1556 and 1653 cm-1, respectively.
Therefore, the absorption at 1554 cm-1 can be assigned to organic nitro species [39]. The
broad peak at 1635 cm-1, however, cannot be definitely assigned to a single compound
based on the infrared data presented here. It may include the contribution from (ONO)
mode of organic nitrito compounds, (C=C) mode of adsorbed propylene and/or bending
mode of adsorbed water. The relatively intense band at 1283 cm-1 should be associated
with (C-O) mode of organic nitrito compounds [39, 130].
In the spectra of surface species adsorbed in commercial NaY zeolite, organic
nitro compounds can be identified by the absorption band at 1562 cm-1. Similarly, the
broad band at 1640 cm-1 may include the contribution from (ONO) mode of organic
nitrito compounds, (C=C) mode of adsorbed propylene and/or bending mode of
adsorbed water. The absorption band at 1278 cm-1 should also be associated with organic
nitrito compounds. A difference of 5-8 cm-1 is observed between the absorption bands of
the nitro/nitrito compounds adsorbed in nanocrystalline NaY zeolite and the commercial
NaY sample. In nanocrystalline NaY zeolite, organic nitro/nitrito compounds most likely
formed over EFAL species on the external surface [39].
Slight differences are observed between the FT-IR spectra of SCR surface species
in NaY zeolites in the absence and presence of adsorbed water (Figure 6.2). For
example, it is known that water suppresses the formation and/or promoted the desorption
of carbonate ion (1525 cm-1) and formate ions (1510 cm-1) [155]. This can be seen in the
spectra shown in Figure 6.2 for commercial NaY zeolite.
The reactivity of surface NOx species towards propylene at room temperature was
also investigated with FT-IR spectroscopy. First, the NaY zeolite sample was
equilibrated with 0.5 Torr NO2, followed by evacuating the IR cell under 10-3 Torr
vacuum at room temperature. Then the surface NOx species remaining in the zeolite
101
sample were exposed to 0.5 Torr C3H6 at room temperature. The reactivity of surface
NOx species towards propylene was quantitatively characterized by integrating the
individual absorption band in the FT-IR spectra recorded in the NO2 adsorptionevacuation-C3H6 exposure process, as shown in Figure 6.3.
120
NO 2 adsorption
Exposure to C3 H 6
Nano EFAL-NO3-
100
Integrated Absorbance
r.t. evacuation
20
80
60
Aldrich-NaY NO+
40
20
Nano-NaY NO+
0
0
10
15
20
25
Time (min)
30
35
40
Figure 6.3 The integrated absorbance of surface NO3- on EFAL sites (), NO+ () in
nanocrystalline NaY zeolite and NO+ () in commercial NaY zeolite as a function of
time. The changes of the integrated absorbance are shown in the process of NO2
adsorption, evacuation of the IR cell and exposure of propylene to the surface species in
the zeolites. Spectra used for band integration were recorded every 5 seconds. The
integrated absorbance of surface NO3- on EFAL sites in nanocrystalline NaY zeolite is 20
times larger than the original absorbance.
Figure 6.3 indicates that NO+ in both nanocrystalline and commercial NaY zeolite
is very reactive toward C3H6 at room temperature. 44% of NO+ in nanocrystalline NaY
102
zeolite and 42% of NO+ in the commercial NaY sample were gone after being evacuated
at room temperature for 13 min. All of the nitrosonium ions disappeared immediately in
contact with C3H6 at room temperature. The high reactivity of NO+ in NaY zeolite
toward molecules such as H2O and C3H6 has been reported previously [39, 108].
NO3- adsorbed on EFAL sites in nanocrystalline NaY zeolite also displays high
reactivity toward propylene at room temperature. As shown in Figure 6.3, 10% of the
NO3- on EFAL sites in nanocrystalline NaY zeolite was pumped away from the zeolite
during the evacuation process. Another 16% loss of the NO3- on EFAL sites was
observed upon exposure of propylene to nanocrystalline NaY zeoilte at room
temperature. Considering the fact that the formation of several SCR surface species
(such as organic nitro and nitrito compounds) in this process can contribute to the
absorption bands in the same spectral region as NO3- on EFAL sites in the FT-IR spectra,
the actual loss of NO3- on EFAL sites upon exposure of C3H6 is most likely more than
16%.
In both nanocrystalline and commercial NaY zeolite samples, the intensity of
NO3- on Na+ sites barely changed in contact with C3H6, indicating the low activity of
NO3- on Na+ sites toward C3H6 at room temperature. The reactivity of NO3- on EFAL
sites toward C3H6 is consistent with the fact that organic nitro and nitrito compounds
formed preferentially on EFAL sites in nanocrystalline NaY zeolite in the adsorption of
NO2 and C3H6, as discussed previously (Figure 6.2).
6.3.4 Adsorbed Products Following Propylene-SCR of
NO2 in NaY Zeolite at 473 K
At elevated temperature, organic nitro and nitrito compounds can be converted

into organic isocyanate [39]. In this study, the formation of isocyanic acid and organic
isocyanate on EFAL sites is observed when nanocrystalline NaY zeolite was used as the
103
SCR catalyst. The FT-IR spectra of surface species after SCR reactions in both
nanocrystalline and commercial NaY zeolite are shown in Figure 6.4.
0.1
A
b
s
o
r
b
a
n
c
e
1723
2276
2352
1380
1414
1640
1272
1593 1477
2239
5
Nano-NaY
Nano-NaY with H2 O
1562
1354
1525
2244
Aldrich-NaY
Aldrich-NaY with H2 O
2400
2200
2000
1800
1600
1400
1200
Wavenumber (cm-1 )
Figure 6.4 FT-IR spectra of adsorbed species following propylene-SCR of NO2 at 473 K
for 6 h in nanocrystalline NaY and commercial NaY zeolite in the presence and absence
of adsorbed water. The spectra were recorded at 298 K. All spectra use the
corresponding clean NaY zeolite prior to adsorption as a background. In addition, gasphase absorptions have been subtracted from each of the FT-IR spectra.
After propylene-SCR of NO2 at 473 K for 6 h in nanocrystalline NaY zeolite, two

surface species can be identified as characterized by the absorption bands at 2276 and
2239 cm-1, respectively. The absorption band at 2276 cm-1 has been assigned to
isocyanic acid [39, 106]. The other band at 2239 cm-1 has been assigned to organic
104
isocyanate in several previous studies using alumina-supported materials as SCR

catalysts [116, 126, 135-138]. Since a substantial amount of EFAL species is detected in
nanocrystalline NaY zeolite used in this study, the band at 2239 cm-1 should be
associated with organic isocyanate on EFAL sites. Isocyanic acid and organic isocyante
most likely came from the thermal reactions of organic nitro/nitrito compounds. When
commercial NaY zeolite was used as the SCR catalyst, organic isocyante was produced
as characterized by an absorption band at 2244 cm-1 in the spectra shown in Figure 6.4.
Other absorption features shown in Figure 6.4 can be assigned based on the
discussion in the previous section and relevant reports in the literature. Nitrate remained
in NaY zeolite after SCR reactions can be identified by two absorption bands at 1414 and
1380 cm-1 [39, 88, 89]. Carbon atom-containing surface species can be identified by the
absorption bands at 1723, 1593 and 1477 cm-1 in nanocrystalline NaY zeolite and 1723,
1525 and 1354 cm-1 in commercial sample [39, 131]. Since acetic acid adsorbed in NaY
zeolite gave rise to absorption bands at 1722 and 1271 cm-1, the absorption features at
1723 and 1272 cm-1 should be assigned to carboxylic acid.
However, the absorption band at 1272 cm-1 may partly attribute to the formation
of nitrite, considering the fact that adsorbed water (bending mode at 1640 cm-1) was
observed in both nanocrystalline and commercial NaY samples. As mentioned
previously, nitrite formed upon adsorption of NO2 in NaY zeolites in the presence of
adsorbed water (Figure 6.1). An absorption band at 1562 cm-1 characterizing organic
nitrito compounds, which is considered as one of the important SCR intermediates, is
evident in the spectra of commercial NaY zeolite after SCR reactions at 473 K for 6 h.
Surface adsorbed CO2 can be identified by the absorption band at 2352 cm-1 in the
spectra shown in Figure 6.4.
It has been reported isocyanic acid can be generated in the reactions between
organic nitro compound and gas-phase NOx [106]. Isotopic labeling studies have shown
that one nitrogen atom in N2 originates from the gas-phase NOx and the other nitrogen
105
atom originates from a surface nitrogen-containing species in SCR reactions [143]. In

nanocrystalline NaY zeolite, the reactions between gas-phase NOx and surface NCO
compounds or their hydrolysis products (such as ammonium ions) on EFAL sites are
important for the N-N bond formation to produce N2 and N2O. Compared to surface sites
in commercial NaY zeolite, EFAL species on the external surface of nanocrystalline NaY
zeolite yield chemically distinct adsorption and reaction sites in SCR reactions.
It should be noticed that a greater amount of adsorbed water is observed in the
FT-IR spectra of commercial NaY zeolite after SCR reactions relative to those of
nanocrystalline NaY zeolite (Figure 6.4). The accumulation of surface adsorbed water in
commercial NaY zeolite during the SCR reactions (Figure 6.4) might indicate that
hydrolysis of isocyanate did not occur in the commercial NaY sample to the same extent
as it does in nanocrystalline NaY zeolite. Supporting this is the fact that more NO but
less N2O formation was observed after propylene-SCR of NO2 in commercial NaY
zeolite compared to that happened in nanocrystalline NaY zeolite (Table 6.1). In
addition, SCR reactions in nanocrystalline NaY zeolite are more sensitive to adsorbed
water than in commercial sample, as shown by the comparison in Table 6.1. Most likely,
water generated in SCR reactions over nanocrystalline NaY zeolite reacted with surface
SCR intermediates (such as organic isocyanate), and did not inhibit the further adsorption
of gas-phase NOx and C3H6. Pre-adsorbed water in nanocrystalline NaY zeolite before
starting SCR reactions, however, strongly inhibited the adsorption of NO2 and C3H6 via
poisoning adsorption sites on external surface and further SCR reactions. In commercial
NaY zeolite, external surface did not play an important role in SCR reactions at 473 K.
Consequently, pre-adsorbed water did not have a significant effect on propylene-SCR of
NO2.
106
6.4 Conclusions
Compared to commercial NaY zeolite with a larger particle size, nanocrystalline

NaY zeolite demonstrated markedly higher adsorption capacity and activity toward N-N
bond formation in propylene-SCR of NO2 at low temperature (T = 473 K). FT-IR
analysis suggests that a significant amount of SCR reactions in nanocrystalline NaY
zeolite occurred on the external surface. The study on the reactivity of surface NOx
species in nanocrystalline NaY zeolite indicated that EFAL sites in nanocrystalline NaY
zeolite were responsible for the higher NOx (NO2 + NO) conversion and higher
selectivity toward N2O formation in SCR reactions. However, nanocrystalline NaY
zeolite is more sensitive to adsorbed water in propylene-SCR of NO2 at 473 K relative to
commercial NaY zeolite. Most likely, adsorbed water poisons important surface
adsorption sites by hydrogen bonding to silanol groups on the external surface of
nanocrystalline NaY zeolite.
The authors thank Dr. Weiguo Song for providing the nanocrystalline NaY zeolite
sample and gratefully acknowledge the Environmental Protection Agency for the support
of this work (R829600). The results of this work are presented in a publication under the
authorship of: Gonghu Li, Sarah C. Larsen, Vicki H. Grassian, Catal. Lett. 2005 (in
press).
107
CHAPTER 7
SELECTIVE CATALYTIC REDUCTION OF NO2 WITH
UREA IN NANOCRYSTALLINE NaY (23 nm)
7.1 Introduction
Selective catalytic reduction (SCR) using ammonia or hydrocarbons is a

promising technology for post-combustion treatment of NOx (NO and NO2) [33]. NH3SCR has been developed and used worldwide for the control of NOx emissions in fuel
combustion from stationary sources due to its efficiency, selectivity and economics [156].
However, NH3 is not a practical reducing agent for NOx emissions from mobile sources
due to its toxicity and difficulties in its storage, transportation and handling [33]. A great
deal of interest has focused on using urea as a safer source of ammonia in automotive
applications [157-161]. Currently, a solution of 32.5% urea in water is the preferred
choice among different precursors for ammonia [162-165]. It is generally accepted that
urea thermally decomposes in two steps (reaction 7.1 and 7.2) to form ammonia and
carbon dioxide [166-168].
(NH2)2CO NH3 + HNCO
(7.1)
HNCO + H2O NH3 + CO2
(7.2)
Transition metal-containing zeolites have been extensively studied as SCR

catalysts [6, 169, 170]. Several recent studies have also used alkali and/or alkaline earth
substituted Y zeolites as deNOx catalysts [88, 106, 109]. In the absence of transition
metals, zeolite Y potentially offers novel SCR pathways different from those that occur in
transition metal containing catalysts. As discussed in Chapter 4, nanocrystalline Y
zeolites may be particularly useful in environmental applications due to the small particle
sizes and large internal and external surface areas [14, 39, 40]. The influence of
108
intracrystalline diffusion on SCR reaction rates has been previously investigated for
transition metal-exchanged zeolites with different particle sizes [150-154]. In Chapter 6,
FT-IR spectroscopy was used to study propylene-SCR of NO2 over nanocrystalline NaY
zeolite with a particle size of 23 nm. At 473 K, significantly higher NOx conversion and
higher selectivity toward N-N bond formation was observed in nanocrystalline NaY
zeolite compared to commercial NaY zeolite with a larger particle size. It was also
observed that nanocrystalline NaY zeolite was more sensitive to the presence of adsorbed
water than commercial NaY zeolite in propylene-SCR of NO2 at 473 K. The silanol
groups and extra framework aluminum (EFAL) species with Bronsted acidity located on
the external surface of nanocrystalline NaY zeolite were found to account for the
observed differences in activity.
In the study reported here, in situ transmission FT-IR spectroscopy is used to
investigate urea-SCR of NO2 over nanocrystalline NaY zeolite at relatively low
temperature (T = 473 K). This work is motivated by the fact that the silanol groups and
Bronsted acid sites on the external surface of nanocrystalline NaY may be effective
surface sites for the decomposition of urea to NH3 and the subsequent SCR of NO2 [156,
162]. The performance of nanocrystalline NaY zeolite in urea-SCR reactions is
compared to a commercial NaY zeolite with a larger particle size in order to explore the
potential advantages of the small particle size of nanocrystalline NaY zeolite and to
determine if differences in the chemical as well as physical properties of nanocrystalline
NaY are important in urea-SCR reactions.
4 mg urea in aqueous solution (~0.3 M) and 7 mg NaY zeolite were mixed and
sonicated for 30 min at room temperature. The resulting hydrosol was coated onto the
tungsten grid described in Figure 2.1. The NaY zeolite sample with adsorbed urea was
dried in air and evacuated at room temperature overnight to remove weakly adsorbed
109
water. Reactant gases (NO2 and O2) were loaded into the NaY zeolite through the gas
handling system (Figure 2.3). In SCR reactions, NO2 was first introduced into the IR cell
at room temperature. Thirty minutes were allowed for adsorption equilibrium before an
excess amount of oxygen was added into the IR cell. Time course experiments were
conducted by automatically recording infrared spectra of the gas phase every 1 min.
7.3 Results
7.3.1 Adsorption of NO2 in NaY Zeolite at 298 K
As discussed above, an aqueous urea solution and the NaY zeolite powder were
mixed at room temperature. Prior to adsorbing NO2, the NaY zeolite sample was dried in
air and evacuated at room temperature overnight to remove weakly adsorbed water. The
FT-IR spectrum of commercial NaY zeolite with adsorbed urea is shown in Figure 7.1.
The FT-IR spectrum of a clean commercial NaY sample is also plotted in Figure 7.1
(dotted line) for comparison.
Adsorbed urea in NaY zeolite can be identified by several absorption bands at
1456, 1584, 1610, 1650 and 1712 cm-1 in the FT-IR spectrum shown in Figure 7.1. The
absorption at 1456 cm-1 is assigned to the asymmetric N-C-N stretching mode of
adsorbed urea [168]. The absorptions at 1584 and 1610 cm-1 are both due to the NH2
bending mode of adsorbed urea [168]. The bands at 1650 cm-1 [168] and 1712 cm-1 (both
C=O stretching mode) can be assigned to urea adsorbed at different sites or different
coordination environments in NaY zeolite. Some additional bands in the spectral region
3100-3800 cm-1 due to O-H/N-H stretching are also seen (not shown). The general
features of the FT-IR spectrum of nanocrystalline NaY zeolite with adsorbed urea (not
shown) are similar to those of commercial NaY zeolite, except that there is an increase in
the spectral broadening of the peaks. This broadening has been previously attributed to
an increase in site heterogeneity for nanocrystalline zeolite samples (Chapter 4).
110
1610
0.2
1650
1584
1456
1712
A
b
s
o
r
b
a
n
c
e
Commercial NaY
with adsorbed urea
Commercial NaY
2000
1800
1600
1400
Wavenumber (cm-1 )
1200
Figure 7.1 FT-IR spectrum of commercial NaY with adsorbed urea at 298 K. The zeolite
sample with adsorbed urea was dried in air and evacuated at 298 K overnight. The FT-IR
spectrum of a clean commercial NaY sample is also plotted (dotted line). The blank grid
is used as a reference. All spectra were recorded at 298 K.
The NaY sample with adsorbed urea was then exposed to 0.5 Torr pressure of
NO2 at room temperature and was equilibrated for 30 min. The changes in the FT-IR
spectra of NaY sample upon exposure to NO2 are clearly seen in the difference spectra
shown in Figure 7.2. Each difference spectrum was obtained by subtracting the spectrum
of NaY zeolite before adsorbing NO2 from the corresponding spectrum after the
adsorption equilibrium. As shown in the difference FT-IR spectrum of commercial NaY
zeolite, several absorption bands appeared at 1272, 1372, 1514 and 1770 cm-1. The
absorption bands at 1272 and 1372 cm-1 are attributed to the formation of nitrite and
nitrate on Na+ sites, respectively [39, 40, 88, 89, 108]. The appearance of two absorption
peaks at 1514 cm-1 (CN + NH) and 1770 cm-1 (CO) indicate the formation of a surface
species with a linear imido structure (e.g. (C=O)-NH-(C=O)-) [166]. The simultaneous
111
loss in intensity of the bands around 1456 and 1650 cm-1 was observed, indicating that
adsorbed urea reacted with NO2 (or surface species generated in NO2 adsorption).
1372
0.1
A
b
s
o
r
b
a
n
c
e
1336
1272
Commercial NaY
2276
1514
1770
Nanocrystalline NaY
1650
2400
2200
2000
1800
1600
W avenumber (cm-1 )
1456
1400
1200
Figure 7.2 Difference FT-IR spectra of nanocrystalline NaY with adsorbed urea and
commercial NaY zeolite with adsorbed urea following adsorption of 0.5 Torr NO2 at 298
K. The spectra use the corresponding NaY zeolite with adsorbed urea prior to NO2
adsorption as a background. In addition, gas-phase absorptions have been subtracted
The difference FT-IR spectrum of nanocrystalline NaY with adsorbed urea after
being exposed to NO2 is shown in Figure 7.2 (bottom). Similar to the commercial NaY
zeolite, the formation of nitrite (1272 cm-1) and disappearance of adsorbed urea (1456
and 1650 cm-1) were observed for nanocrystalline NaY zeolite. However, several
differences were observed between the spectra of nanocrystalline NaY zeolite and
commercial NaY zeolite. For example, the formation of the surface species with the
linear imido structure (1514 and 1770 cm-1) occurs to a much less extent in
112
nanocrystalline NaY (Figure 7.2). Furthermore, NO2 adsorption in nanocrystalline NaY

zeolite resulted in the formation of nitrate adsorbed on external (vide infra) EFAL sites
(1336 cm-1) [39, 40] as well as nitrate on internal Na+ sites (1372 cm-1). Finally, the
formation of isocyanic acid (HNCO) adsorbed on EFAL sites is observed in
nanocrystalline NaY zeolite, as indicated by a broad absorption band around 2276 cm-1
[39, 40, 106]. In NO2 adsorption at room temperature, the formation of surface adsorbed
isocyanic acid is not appreciable in commercial NaY zeolite (Figure 7.2). Previous
studies of pyridine adsorption coupled with infrared analysis indicate that EFAL sites are
located on the external surface of nanocrystalline NaY zeolite and are not present to any
great extent in the commercial NaY sample due to its low external surface area (Chapter
4) [14, 39, 40].
Unlabeled urea and 15N labeled urea were used in SCR reactions over both
nanocrystalline NaY and commercial NaY zeolite. Compared to the FT-IR spectrum of
NaY zeolite with unlabeled urea, several differences were observed when 15N-urea
solution was mixed with the zeolite sample. For example, the absorption band for
adsorbed H15NCO shifts from 2276 cm-1 to 2258 cm-1 when nanocrystalline NaY with
adsorbed 15N-urea was exposed to NO2 at room temperature (not shown). A similar
isotopic shift was also observed for gas-phase HNCO. The FT-IR spectra of the gas
phase after NO2 adsorption in commercial and nanocrystalline NaY zeolite with
unlabeled urea and 15N-urea are shown in Figure 7.3. Gas-phase NO2 (1616 cm-1), NO
(1875 cm-1), HNCO (2268 cm-1) and CO2 (2349 cm-1) were observed when NaY zeolite
with adsorbed urea was exposed to NO2 at room temperature [39, 106, 171]. For gasphase HNCO, an isotopic shift of 8 cm-1 was observed on substituting 15N for 14N [106].
113
NO 2
1616
0.005
A
b
s
o
r
b
a
n
c
e
CO 2
2349
H 15 NCO
2260
NO
1875
(a)
(b)
(c)
(d)
HNCO
2268
2400
2300
2200
2100 2000 1900 1800

Wavenumber (cm-1 )
1700
1600
Figure 7.3 The FT-IR spectra of the gas phase upon adsorption of NO2 in: (a)
commercial NaY zeolite with adsorbed 15N-urea, (b) nanocrystalline NaY zeolite with
adsorbed 15N-urea, (c) commercial NaY zeolite with adsorbed unlabeled urea, and (d)
nanocrystalline NaY zeolite with adsorbed unlabeled urea. All spectra were recorded at
298 K after adsorption equilibrium.
The complete list of isotopic shifts for nitrogen-containing gas-phase and

adsorbed species is given in Table 7.1 and will be discussed in detail in later sections. In
the spectra shown in Figure 7.3, the absorption band for gas-phase NO2 over
nanocrystalline NaY zeolite is less intense than that over commercial NaY, indicating
that more NO2 was adsorbed and/or reacted with nanocrystalline NaY zeolite relative to
commercial NaY. This is consistent with our previous observation that nanocrystalline
NaY zeolite has a greater adsorption capacity for NO2 relative to commercial NaY zeolite
[14, 40].
114
Table 7.1 List of vibrational frequencies of nitrogen-containing gas-phase and
adsorbed species which showed an isotopic shift upon substitution of 15N for 14N
in the adsorbed urea ((NH2)2CO) precursor.
Phase
Gas Phase
Adsorbed in
Nanocrystalline NaY
Assignments
(15N)
(14N)/(15N)
HNCO
NCO
2268
2260
1.004
N2O
NN
2224
2201
23
1.010
N2O
NO
1285
1270
15
1.012
NH3
NH3
965
960
1.005
HNCO
NCO
~2276
~2258
~18
1.008
OCN-
NC
2169
2153
16
1.007
(NH2)2CO
CNH
1527
1522
1.003
(NH2)2CO
CNH
1271
1266
1.004
HNCO
NCO
~2282
~2250
~32
1.014
OCN
NC
2175
2159
16
1.007
(NH2CO)2NH
CO
1759
1751
1.005
(HNCO)3
CO
1735
1727
1.005
Adsorbed in
Commercial NaY
(14N)
= (14N) - (15N)
7.3.2 Urea-SCR of NO2 in NaY Zeolite at 473 K:

Products in the Gas Phase
An excess amount of O2 was introduced into the IR cell after the NaY zeolite
sample was equilibrated with NO2 for 30 min at room temperature. Urea-SCR of NO2
was then carried out at 473 K. Figure 7.4 shows the FT-IR spectra of gas-phase species
formed during the first 30 minutes of 15N-urea SCR over nanocrystalline NaY zeolite at
473 K.
In the beginning of the SCR reaction (t = 0), gas-phase NO2, NO, H15NCO, CO2
and a small amount of H2O were observed in the FT-IR spectrum. At t = 1 min, more
H2O and H15NCO appeared in the gas phase as shown by the spectra (Figure 7.4). Gasphase H15NCO decayed quickly after t = 2 min. Several additional changes in the gas
115
phase composition can be seen in these time course experiments, including the rapid
production of 15NH3 (927, 960, 1623 and 3331 cm-1) and CO2 at t = 2 min.
0.01
CO2
15
NH3
15
NNO
15
15
NH3
3500
3000
NO2
2500
2000 -1 1500
Wavenumber (cm )
ac
t
NO
1000
Re
H15NCO
io
tim
NH3
A
b
s
o
r
b
a
n
c
e
Figure 7.4 FT-IR spectra of the gas phase as a function of reaction time in 15N-urea SCR
of NO2 over nanocrystalline NaY zeolite. The spectra shown were recorded at 473 K and
at t (reaction time) = 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 21,
23, 25, 27, 30 min, respectively. The blank grid is used as a reference.
Gas-phase 15N-urea appeared and is identified by several characteristic

absorptions including three broad bands between 3100 and 3520 cm-1 and several
absorptions between 1290 and 1720 cm-1 in the FT-IR spectra shown in Figure 7.4.
Furthermore, gas-phase NO2 disappeared quickly relative to the concentration of NO.
15
NNO (2201 and 1270 cm-1) appeared in these spectra as the only isomer of N2O (Table
116
7.1) [172]. Other isomers of N2O such as N15NO and 15N2O have different absorption
frequencies (2178/1280 cm-1 for N15NO and 2155/1265 cm-1 for 15N2O) and were not
detected in our experiments [172]. The formation of 15NNO in the gas phase indicates
that N-N bond formation occurred at room temperature between a 15N-atom from urea
and a 14N-atom from NO2 and that the N=O bond in NO2 remains intact during the
formation of 15NNO. There was no significant change in the gas phase after 30 minutes
of SCR reactions over nanocrystalline NaY zeolite.
Compared to the SCR reaction in nanocrystalline NaY zeolite, the urea-SCR of
NO2 in commercial NaY zeolite was much slower. The FT-IR spectra of the gas phase
species present during the 2 hours of urea-SCR over commercial NaY zeolite are shown
in Figure 7.5. Similar to the spectra over the nanocrystalline NaY, NO2, NO, HNCO,
CO2 and a small amount of H2O were among the detectable gas-phase species present
before the SCR reaction started (t = 0). NO2 disappeared quickly, but NO persisted in the
gas phase during the two-hour reaction period. More gas-phase HNCO and H2O
appeared immediately after the heating began. The concentration of HNCO reached a
maximum at t = 5 min and HNCO remained detectable in the IR cell for at least 60 min
during the SCR reaction process (Figure 7.5). Compared to nanocrystalline NaY, a larger
amount of gas-phase urea appeared over commercial NaY zeolite. There was an
induction period (~3 min) for the production of NH3 and CO2 which did not appear in the
gas phase with significant concentrations until t = 4 min (Figure 7.5). N2O appeared as
one of final products in gas phase [39, 40].
It should be noted that a gas-phase species with an absorption around 2185 cm-1
was present in the FT-IR spectra from t = 4 min to 10 min. Since another absorption
band at 1255 cm-1 was also observed, this species is assigned to formonitrile oxide
(HCNO) [173]. Further supporting the assignment is the fact that HCNO (2190 and 1255
cm-1) was produced in the presence of a large amount of HNCO during the thermal
decomposition of urea at 473 K in commercial NaY zeolite (not shown).
117
CO2
0.01
HNCO
NH3
Urea
Urea
NO
3500
3000
2500
2000
-1
Wavenumber (cm )
NO2
1500
1000
Re
ac
tio
nt
im
e
N2O
A
b
s
o
r
b
a
n
c
e
Figure 7.5 FT-IR spectra of the gas phase as a function of reaction time in urea-SCR of
NO2 over commercial NaY zeolite. The spectra shown were recorded at 473 K and at t
(reaction time) = 0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 18, 20, 25, 30, 35, 40, 50, 60, 70,
80, 90, 120 min, respectively. The blank grid is used as a reference.
A control experiment was also done by heating the commercial NaY zeolite with
adsorbed urea at 473 K in the presence of oxygen. The time course concentrations of the
gases in the infrared cell can be determined from the integrated absorbance of the
corresponding bands using calibrated extinction coefficients, as described in Appendix B
[39, 40]. The time course concentrations of gas-phase NH3, CO2 and H2O in urea-SCR of
NO2 are shown in Figure 7.6 (a), 7.6 (b) and 7.6 (c), respectively. The time course
118
concentrations for the urea thermal decomposition over commercial NaY zeolite are also
included in Figure 7.6.
150
(a) NH 3
100
Nanocrystalline
Commercial
Urea Thermal
Concentration (10-6 mol L-1)
50
0
150
(b) CO 2
100
50
0
40
(c) H 2 O
30
20
10
0
20
40
60
80
Reaction Time (min)
100
120
Figure 7.6 Evolution of gas-phase (a) NH3, (b) CO2 and (c) H2O during urea-SCR of
NO2 in nanocrystalline NaY zeolite (), urea-SCR of NO2 in commercial NaY zeolite ()
and thermal decomposition of urea in commercial NaY zeolite () at 473 K.
In urea-SCR of NO2 using nanocrystalline NaY zeolite as the catalyst, the

concentration of NH3 in the gas phase initially increased until a maximum was reached at
119
approximately t = 15 min (Figure 7.6 (a)). The concentration of NH3 then gradually
decreased and the remained at a constant level at approximately 55 mol L-1. The
concentration of CO2 also increased quickly during the first 15 min after the urea-SCR of
NO2 in nanocrystalline started (Figure 7.6 (b)). The concentration of gas-phase H2O
rapidly increased to a maximum at t = 2 min, then decreased sharply until a minimum
was reached at approximately t = 15 min and started increasing again thereafter (Figure
7.6 (c)). Most likely, strongly adsorbed water that remained in nanocrystalline NaY
zeolite was released initially upon heating to 473 K [39], followed by the consumption
and production of H2O in SCR reactions [166].
In urea-SCR of NO2 over commercial NaY zeolite, the changes in concentrations
of NH3, CO2 and H2O are quite different relative to the corresponding ones in
nanocrystalline NaY zeolite (Figure 7.6). For example, the concentration of NH3 during
urea-SCR over commercial NaY zeolite increased slowly, remained at a concentration
lower than 30 mol L-1 and decreased after t = 60 min (Figure 7.6 (a)). The final
concentration of CO2 over commercial NaY zeolite was about half of the final
concentration over nanocrystalline NaY zeolite (Figure 7.6 (b)). The concentration of
H2O in the gas phase during urea-SCR over commercial NaY zeolite increased initially
and changed slightly after t = 10 min (Figure 7.6 (c)). In the thermal decomposition of
urea over commercial NaY zeolite, the concentration of NH3 increased until a maximum
was reached at t = 40 min and only slightly changed thereafter (Figure 7.6 (a)). The time
course concentrations of CO2 and H2O in the thermal decomposition of urea are just
slightly lower than those in urea-SCR of NO2 over commercial NaY zeolite (Figure 7.6
(b) and 7.6 (c)).
The time course concentrations of NO and N2O in the gas phase are shown in
Figure 7.7 (a) and 7.7 (b), respectively. In urea-SCR of NO2 over nanocrystalline NaY
zeolite, the concentration of NO decreased monotonically and gas-phase NO completely
disappeared around t = 40 min (Figure 7.7 (a)). When commercial NaY zeolite was used
120
as the SCR catalyst, the concentration of NO initially increased until a maximum and
decreased slowly thereafter. The time course concentration of N2O demonstrates that
N2O was produced more quickly and in much higher concentration in nanocrystalline
NaY zeolite than in commercial NaY zeolite (Figure 7.7 (b)).
25
(a) NO
20
Nanocrystalline
Commercial
Concentration (10-6 mol L -1)
15
10
5
0
15
(b) N 2 O
12
9
6
3
0
0
20
40
60
80
Reaction Time (min)
100
120
Figure 7.7 Evolution of gas-phase (a) NO and (b) N2O during urea-SCR of NO2 in
nanocrystalline NaY zeolite () and in commercial NaY zeolite () at 473 K.
The initial rates for formation of gas-phase NH3, CO2 and N2O and loss of gasphase NO can be obtained by linear fitting of the concentrations of individual gas-phase
species in the beginning of the reactions, as listed in Table 7.2. In urea-SCR of NO2 over
nanocrystalline NaY zeolite, the initial rate of NH3 formation is slightly higher than the
121
initial formation rate of CO2. Time course concentrations of NH3 (Figure 7.6 (a)) and
CO2 (Figure 7.6 (b)) in the first 15 min of SCR reactions also indicate that the amount of
NH3 released into the gas phase is almost the same as the amount of CO2. In the thermal
reaction of urea over commercial NaY zeolite, the ratio of NH3 formation rate to CO2
formation rate is close to 2:1, which matches the fact that 1 mole of urea thermally
decomposes into 2 moles of NH3 and 1 mole of CO2 (reactions 7.1 and 7.2).
Furthermore, the initial rates for NH3 and CO2 formation are much higher in
nanocrystallline NaY than in commercial NaY. Table 7.2 also includes the initial rates of
N2O formation and NO loss. It can be concluded from the data listed in Table 7.2 that
urea-SCR of NO2 occurred more quickly in nanocrystalline NaY than in commercial NaY
zeolite.
Table 7.2 Rates for formation of gas-phase NH3, CO2 and N2O and
loss of gas-phase NO in the first 10 min during urea-SCR of NO2 and
thermal decomposition of urea at 473 K.
Catalyst
Nanocrystalline NaY
Commercial NaY
Commercial NaY
Initial Rate (mol L-1 min-1)

NH3
CO2
NO
N2O
10.3
8.9
-0.4
1.1
2.4
1.5
-0.2 a
0.1
3.0
1.6
The initial rate of NO loss during the urea-SCR of NO2 over commercial
NaY zeolite was measured from t = 9 min to t = 18 min; b Thermal
decomposition of urea in the presence of O2.
Table 7.3 summarizes the product distribution in the gas phase after urea-SCR of
NO2 and thermal reaction of urea. In nanocrystalline NaY zeolite, the concentrations of
NH3 and CO2 after SCR reaction are much higher than in commercial NaY zeolite. In
addition, urea-SCR of NO2 over nanocrystalline NaY zeolite resulted in 100% NOx (NO2
and NO) conversion and higher selectivity toward N2O formation relative to commercial
122
NaY zeolite. Based on mass balance for nitrogen, more than 80% of NO2 was converted
into N2 in urea-SCR over nanocrystalline NaY zeolite (Table 7.3). Although about 17%
of NO2 was converted into N2O in urea-SCR reactions over nanocrystalline NaY zeolite,
it was observed that N2O could be reduced into N2 in nanocrystalline NaY zeolite at
higher temperatures, which are more practical for typical engine operation.
Table 7.3 Gas-phase product distribution after urea-SCR of NO2 and

thermal reaction of urea at 473 K for 2 h in 7 mg NaY zeolite.
Catalyst
Concentration (mol L-1) a

NH3
CO2
NO
N2O
Nanocrystalline NaY
56
138
12
Commercial NaY
17
74
Commercial NaY b
53
69
The concentration of gas-phase NH3 was measured at 473 K, the

concentrations of gas-pase CO2, NO and N2O were measured at 298 K. The
loadings of NO2, urea and O2 are 705 mol L-1, 17020 mol L-1 and
55020 mol L-1, respectively. b Thermal reaction of urea in commercial
NaY in the absence of NO2.
a
7.3.3 Adsorbed Products Following Urea-SCR of NO2

in NaY Zeolite at 473 K
Besides kinetics studies, infrared analysis of surface adsorbed intermediates or

products can assist in understanding the mechanism for urea-SCR of NO2 [39, 40].
Difference FT-IR spectra of nanocrystalline NaY and commercial NaY zeolite after ureaSCR of NO2 are shown in Figure 7.8. The difference FT-IR spectrum of nanocrystalline
NaY zeolite after 15N-urea SCR of NO2 and commercial NaY zeolite after thermal
reaction of urea are also shown. The zeolite samples were cooled back to room
temperature prior to recording the spectra. It was observed that most gas-phase NH3,
123
H2O and urea disappeared from the gas phase after cooling the zeolites back to room
temperature due to adsorption in the zeolite samples.
3742
0.5
2258
3695
2153
5
(a)
(b)
2276 2169
A
b
s
o
r
b
a
n
c
e
1368
1522
1266
3514
3623
1712
3424
3489
1527
3380
1271
1759
1735
2175
1376
(c)
1455
1495
(d)
3555
1651
3436
1712
0.1
3800
3600
1286
3400
2200 2000 1800

Wavenumber (cm-1 )
1600
1400
1200
Figure 7.8 Difference FT-IR spectra of zeolite samples following: (a) 15N-urea SCR of
NO2 in nanocrystalline NaY for 30 min, (b) urea-SCR of NO2 in nanocrystalline NaY for
2 h, (c) urea-SCR of NO2 in commercial NaY for 2 h and (d) thermal decomposition of
urea in commercial NaY for 2 h at 473 (from top to bottom). The spectra were recorded
at 298 K. All spectra shown use the corresponding NaY zeolite with adsorbed urea prior
to NO2 adsorption as a background. In addition, gas-phase absorptions have been
subtracted from each of the FT-IR spectra shown.
In urea-SCR of NO2 over nanocrystalline NaY zeolite, the formation of adsorbed

HNCO and OCN- can be identified by the absorption bands at 2276 cm-1 [39, 40, 106]
124
and 2169 cm-1 [106, 174], respectively, in the FT-IR spectra shown in Figure 7.8 (b).
The loss of adsorbed urea, characterized by the negative absorptions at 1271, 1527, 1712
cm-1 (CO), 3424 and 3514 cm-1 (both NH), can be seen as well in the spectrum. The
absorption band at 1527 cm-1 is assigned to the CNH bend-stretch mode where the
nitrogen and hydrogen move in opposite directions relative to the carbon, while the
absorption band at 1271 cm-1 is assigned to the CNH open-stretch mode where N and H
atoms move in the same direction relative to the carbon [175]. In addition, the recovery
of silanol groups (3742 cm-1) and OH groups attached to Na+ (3695 cm-1) are evident in
the spectrum of nanocrystalline NaY zeolite shown in Figure 7.8 (b) [14, 39, 40]. This
indicates that adsorption and SCR reactions occurred to a significant extent on the
external surface since silanol groups are mainly present on the external surface of the
nanocrystalline NaY zeolite [14, 40].
The difference FT-IR spectrum of nanocrystalline after 15N-urea SCR of NO2 for
30 min (Figure 7.8 (a)) is similar to that after urea-SCR for 2 h (Figure 7.8 (b)). The
formation of adsorbed H15NCO (2258 cm-1) and OC15N- (2153 cm-1) as well as the loss of
adsorbed 15N-urea (1522 and 1266 cm-1) are apparent in the spectrum. The isotopic shifts
for these nitrogen-containing species are apparent in the spectra and are listed in Table
7.1. In the spectrum shown in Figure 7.8 (a), a dotted baseline is drawn in the range
between 1200 and 1800 cm-1, revealing that a small amount of nitrate remained in
nanocrystalline NaY zeolite after 30 min of SCR reactions at 473 K.
When commercial NaY zeolite was used as the SCR catalyst, the loss of adsorbed
urea was observed, as indicated by the negative absorption bands at 1286 cm-1 (CNH),
1651, 1712 cm-1 (both C=O), 3436 and 3555 cm-1 (both NH) in the spectrum of
commercial NaY zeolite after urea-SCR reactions (Figure 7.8 (c)). The other CNH
vibration which should fall between 1400 and 1600 cm-1 may be less pronounced in the
spectrum due to the formation of two absorption bands at 1455 and 1495 cm-1 (Figure 7.8
(c)). The two absorption bands at 1455 and 1495 cm-1 are assigned to the bending mode
125
of adsorbed ammonium ion and stretching mode of surface species containing imine
groups (C=NH), respectively [166]. In the spectrum shown in Figure 7.8 (c), the
corrected baseline (dotted line) in the range between 1200 and 1800 cm-1 clearly
demonstrates the formation of adsorbed biuret (1759 cm-1) [166] and cyanuric acid (1735
cm-1) [149] in commercial NaY zeolite after SCR reactions. Furthermore, surface OCN(2175 cm-1) [174] and nitrate (1376 cm-1) were observed after urea-SCR of NO2 for 2 h in
commercial NaY zeolite. In the OH/NH stretching region (Figure 7.8 (c)), the
absorptions in the spectrum of commercial NaY zeolite indicate the formation of OH
group on Na+ (3695 cm-1), surface adsorbed H2O (3623 cm-1) and NH-containing species
(3220, 3380, 3489 cm-1) as well as the loss of adsorbed urea (3436 and 3555 cm-1). The
FT-IR spectrum of commercial NaY zeolite after thermal reaction of urea is almost the
same as the spectrum after urea-SCR of NO2, except that the formation of nitrate, biuret
or cyanuric acid are not observed in the thermal decomposition of urea (Figure 7.8 (d)).
The comparison between the difference FT-IR spectra of the zeolite samples after
urea-SCR of NO2 for 2 h indicates that higher conversion of nitrate was achieved in
nanocrystalline NaY zeolite than in commercial NaY zeolite. This is consistent with the
fact that urea-SCR of NO2 is much slower in commercial NaY than in nanocrystalline
NaY, as mentioned previously (Table 7.2). Furthermore, undesired surface products
(biuret and cyanuric acid) were observed in commercial NaY zeolite after urea-SCR of
NO2 but not in nanocrystalline NaY zeolite.
The assignments of absorption bands in FT-IR spectra shown in Figure 7.8 are
further supported by isotopic studies (Table 7.1). The absorption frequencies for HNCO
and OCN- in nanocrystalline NaY zeolite are different from those in commercial NaY
zeolite. In nanocrystalline NaY, HNCO (2276 cm-1) and OCN- (2169 cm-1) most likely
adsorbed on EFAL sites [39, 40]. In commercial NaY zeolite, OCN- is characterized by
the relatively intense absorption band at 2175 cm-1 in the FT-IR spectrum shown in
Figure 7.8 (c) and most likely attached to Na+ sites [106]. In nanocrystalline NaY zeolite,
126
the isotopic shift for the CNH vibration mode of adsorbed urea (5 cm-1) is exactly the
same as that for the NH3 deformation mode of gas-phase NH3. This suggests that urea
adsorbed in nanocrystalline NaY zeolite in a way such that at least one NH2 group was
almost unperturbed [168]. In commercial NaY zeolite, the isotopic shift for the C=O
stretching mode of adsorbed biuret and cyanuric acid (8 cm-1) is the same as that for the
NCO stretching mode of gas-phase HNCO. This may be due to the configuration
exchange (C=O)-NH- -C(OH)=N- in adsorbed biuret and cyanuric acid in
commercial NaY zeolite.
7.4 Discussion
7.4.1 Reactions Between Gas-Phase NO2 and Adsorbed
Urea in NaY at 298 K
The adsorption of NO2 in zeolite Y has been described in detail in the literature
[39, 40, 88, 89, 106, 108]. In this study, NO2 adsorption in NaY zeolite occurs
preferentially according to reaction 7.3 or 7.4 since there was a considerable amount of
water remaining in the zeolite after evacuating the zeolite samples at room temperature
[40, 108].
(7.3)
3NO2 + H2O 2HNO3 + NO
(7.4)
The nitric acid and nitrous acid further deprotonate and adsorb on surface cationic
sites to form surface nitrate and nitrite. This explains the formation of surface nitrate,
nitrite (Figure 7.2) and gas-phase NO upon NO2 adsorption in NaY zeolite (Figure 7.3).
The formation of nitrate on EFAL sites in nanocrystalline NaY zeolite was also observed
(Figure 7.2). In nanocrystalline NaY zeolite, NO2 adsorption in part occurs on cationic
sites of EFAL species (Mn+-O2-), according to reaction 7.5 [39, 40, 88].
127
Mn+-O- + NO2 Mn+-NO3-
(7.5)
The formation of NO, HNCO and a small amount of CO2 and H2O was observed
when NO2 was adsorbed in NaY zeolite with adsorbed urea at room temperature (Figures
7.3, 7.4 and 7.5). Isotopic labeling shows that the N-atom in HNCO originates from urea
(Figure 7.4). The loss of adsorbed urea was observed as well upon the adsorption of NO2
(Figure 7.3). Since the decomposition of urea into HNCO and NH3 is not significant at
room temperature [166], HNCO most likely formed in the reactions between urea and
HNO3 (reaction 7.6) [168]. The reaction between urea and HONO can lead to the
formation of CO2 and H2O at room temperature.
(NH2)2CO + HNO3 HNCO + NH4NO3
(7.6)
(NH2)2CO + 2HONO 2N2 + CO2 + 3H2O
(7.7)
NH4NO3 can thermally decompose into N2O + 2H2O, providing a minor route for
N-N bond formation in urea-SCR of NO2 in NaY zeolite [176]. Reaction 7.7 is used in
destroying excess nitrous acid in diazotization [147].
When commercial NaY zeolite was exposed to NO2 at room temperature, surface
species with a linear imido structure such as biuret appeared, most likely generated in the
reactions between HNCO and adsorbed urea (Figure 7.2). It is reported that amide
carbonyls of bi- and tri-urets have characteristic absorption bands around 1750-1770 cm-1
[166]. Biuret can be produced via reaction 7.8 [147].
NH2-CO-NH2 + HNCO NH2-CO-NH-CO-NH2
(7.8)
The above analysis suggests that EFAL species in nanocrystalline NaY zeolite
provide additional surface sites for NO2 adsorption and its further reactions at room
temperature. Previous studies show that EFAL sites on the external surface of
nanocrystalline NaY zeolite are important for the formation of reactive nitrate and for the
following SCR reactions [40]. In this study, EFAL sites in nanocrystalline NaY zeolite
128
also provide important surface adsorption sites for HNCO and inhibit the formation of
undesired products such as biuret or triuret.
7.4.2 Thermal Decomposition of Urea in
Nanocrystalline NaY Zeolite at 473 K
The FT-IR spectra of the gas phase shown in Figure 7.4 demonstrate that the
thermal decomposition of urea occurred at 473 K in nanocrystalline NaY zeolite. As
described previously, gas-phase HNCO was produced initially and decayed rapidly after t
= 2 min (Figure 7.4). The initial change in H2O concentration (Figure 7.6 (c)) is similar
to that of HNCO. Not until t = 2 min did the concentration of NH3 in gas phase start to
increase sharply (Figure 7.6 (a)). According to reaction 7.1, the first step in thermal
decomposition of urea produces equal amounts of HNCO and NH3. This suggests that
most of the NH3 produced in thermal decomposition of urea remained in the
nanocrystalline NaY zeolite. In addition, the concentrations of NH3 and CO2 in gas phase
increased sharply after t = 2 min (Figure 7.6 (a) and 7.6 (b)). The kinetic data in Table
7.2 show that the initial formation rates of NH3 and CO2 in gas phase are 10.3 and 8.9
mol L-1 min-1, respectively. This correlates well with the stoichiometry shown by
reaction 7.2, in which 1 mole of HNCO hydrolyzes to form 1 mole of NH3 and 1 mole of
CO2. Most likely, NH3 formed in hydrolysis of HNCO in the zeolite immediately goes
into the gas-phase at 473 K.
Furthermore, both the loss of adsorbed urea and regeneration of surface silanol
groups were observed in nanocrystalline NaY zeolite after urea-SCR of NO2 (Figure 7.8).
The absorption frequencies for the urea bands (1527 and 1271 cm-1) in the FT-IR
spectrum of nanocrystalline NaY zeolite are different from those (1286 cm-1) in the
spectrum of commercial NaY zeolite which contains few silanol groups (Table 7.1) [14,
40]. It can be concluded that in nanocrystalline NaY zeolite the adsorption and the
thermal decomposition of urea involved the silanol groups on the external surface.
129
Finally, HNCO produced in thermal decomposition of urea adsorbed and hydrolyzed on

EFAL sites in nanocrystalline NaY zeolite. This is supported by the fact that HNCO
adsorbed on EFAL sites was observed before (Figure 7.2) and after urea-SCR of NO2
(Figure 7.8) in nanocrystalline NaY zeolite.
Based on the above discussion, a mechanism can be proposed for the thermal
decomposition of urea in nanocrystalline NaY zeolite (Figure 7.9). Initially, urea
adsorbed in nanocrystalline NaY zeolite by hydrogen bonding to silanol groups on the
external surface. At elevated temperature (T = 473 K), urea decomposed into HNCO and
NH3. While NH3 remained on the silanol groups of nanocyrstalline NaY zeolite via
hydrogen bonding, HNCO adsorbed on EFAL sites and continued to hydrolyze and
H
H2O
CO2, NH3
AlOx
SiOH
SiOH
AlOx
AlOx
C
SiOH
OCNH
produce NH3 and CO2.
Figure 7.9 Thermal decomposition of urea on silanol (SiOH) and EFAL (AlOx) sites in
It is generally accepted that NH3 is the actual reducing agent in urea-SCR of NOx
[6, 156, 162, 166-168]. In Figure 7.9, urea adsorbed in nanocrystalline NaY zeolite by
hydrogen bonding to silanol groups via one N-atom, leaving the other NH2 almost
unperturbed [168]. This is consistent with the isotopic studies showing that the isotopic
shift for the CNH vibration mode is the same as the isotopic shift for the NH3
deformation mode of gas-phase NH3 (Table 7.1). In addition, we suggest that NH3
molecules produced in thermal decomposition of urea remained in nanocrystalline NaY
130
zeolite by hydrogen boding with silanol groups on the external surface. When NH3 was
adsorbed in nanocrystalline NaY zeolite at room temperature, the loss of silanol groups
was observed. NH3 adsorbed on EFAL sites in nanocrystalline NaY zeolite was
observed, too. In urea-SCR of NO2 over nanocrystalline NaY, however, EFAL sites are
more likely associated with the adsorption/reactions of nitrate and HNCO as discussed
previously.
Desorption of urea occurred when the nanocrystalline NaY zeolite with adsorbed
urea was heated at 473 K, as shown by the presence of urea in the gas phase (Figure 4)
[166, 167]. The spectra shown in Figure 7.4 also indicate that most gas-phase urea stayed
as stable species during the SCR reactions at 473 K.
7.4.3 Formation of N-N Bond in Urea-SCR of NO2 over
Nanocrystalline NaY Zeolite at 473 K
When nanocrystalline NaY zeolite was used in urea-SCR of NO2, more than 80%
of the total loaded NO2 was converted into N2, based on nitrogen mass balance. In this
work, N2 cannot be monitored directly by infrared spectroscopy. Isotopic studies show
that only 15NNO among four isomers of N2O was produced when 15N-urea was used in
the SCR reactions (Figure 7.4). Previous studies in literature suggest that N-N bond
formation occurs in the reaction between NOx and NH3 produced in thermal
decomposition of urea [156, 166-168]. Based on isotopic labeling studies, it can be
concluded that the formation of dinitrogen in urea-SCR of NO2 over NaY zeolite
involves both a N-atom from NO2 and a N-atom from urea. As for the production of
N2O, the bond formation occurs when a N-atom from urea binds with NO from NO2
through the N-atom.
In selective catalytic reduction of NOx using hydrocarbons, isotopic studies have
shown that the N-N bond formation occurs between NOx from the gas phase and a
surface N-containing adsorbate [106, 143]. In urea-SCR of NO2 over nanocrystalline
131
NaY zeolite, the N-N bond formation is complex since reaction might involve (i) surface
adsorbed NH3 and/or gas-phase NH3, and (ii) gas-phase NOx and/or surface NOx- (NO3and NO2-).
Previous studies on thermal reduction of NO2 in nanocrystalline NaY zeolite
demonstrated that NO was initially generated in thermal reactions of nitrate and nitrite at
elevated temperatures according to reactions 7.9-7.11 [39, 40]. Mn+ represents Na+ or
EFAL sites.
Mn+-NO3- Mn+-O- + NO2
(7.9)
Mn+ + 2NO2 Mn+-NO3- + NO
(7.10)
Mn+-NO2- + NO2 Mn+-NO3- + NO
(7.11)
Although the concentration of NO decreased monotonically in urea-SCR of NO2

over nanocrystalline NaY zeolite (Figure 7.7 (a)), further analysis suggests that the
formation of NO via reactions 7.9-7.11 could have occurred. For example, the initial rate
of N2O formation was much greater than the initial rate of NO loss in urea-SCR of NO2
over nanocrystalline NaY zeolite (Table 7.2). The actual initial rate for N-N bond
formation should be even greater since more than 80% of NO2 was converted into N2 in
urea-SCR of NO2 over nanocrystalline NaY zeolite. In urea-SCR of NO2 over
commercial NaY zeolite, the concentration of NO initially increased and then decreased
after a maximum at t = 9 min (Figure 7.7 (a)), indicating that both production and
consumption of NO occurred at elevated temperature. Similar time course concentration
of NO has been observed in propylene-SCR of NO2 in nanocrystalline NaY zeolite [39].
It has been suggested that OH groups and Bronsted acid sites on the catalyst
surface can coordinate with and further activate NH3 [156, 162]. Figure 7.10 describes a
possible pathway for N-N bond formation in urea-SCR of NO2 over nanocrystallline NaY
zeolite. In the mechanism shown by Figure 7.10, NO is released during the thermal
132
reactions of nitrate and nitrite attached to Na+ or EFAL sites and reacts with hydrogen
bonded NH3 to produce N2 (or N2O) and H2O.
NO, O2
15NN, 15NNO,
H2O
AlOx
AlOx
15N
SiOH
SiOH
O2
AlOx
SiOH
15N
NO3-
H
H
Figure 7.10 15N-urea SCR of NO2 on silanol (SiOH) and EFAL (AlOx) sites in
Reaction 7.12 is the standard SCR reaction using NH3 [166].

4NH3 + 4NO + O2 4N2 + 6H2O
(7.12)
From the above discussion, a mechanism in which N-N bond formation in ureaSCR over nanocrystalline NaY zeolite occurs on the external surface and involves surface
adsorbed NH3 and NO from the gas phase is proposed. Minor pathways for N-N bond
formation in nanocrystalline NaY zeolite include the thermal decomposition of NH4NO3
produced in the reaction between HNO3 and urea (reaction 7.6). As it has been
previously suggested that urea initially decomposes into CO and an NH2 radical which
consequently reacts with NOx to form N2 [33]. However, this pathway is unlikely in
urea-SCR over nanocrystalline NaY zeolite since CO was not observed in this study.
7.4.4 Origin of Size Effect in Urea-SCR of NO2 over
NaY Zeolite
Reducing the zeolite particle size can increase the external surface area and
dramatically alter the diffusional properties of the zeolite materials [151-153]. Infrared
characterization revealed that silanol groups and EFAL species on the external surface
133
are present in greater concentration in nanocrystalline NaY zeolite than in commercial

NaY zeolite due to the difference in external surface areas (see chapter 4). The external
surface areas of nanocrystalline NaY and commercial NaY zeolite are 178 and ~4 m2/g,
respectively [14]. Silanol groups and EFAL species on the external surface were found
to be responsible for the improved performance of nanocrystalline NaY zeolite in
propylene-SCR of NO2. In the present study, urea-SCR of NO2 using nanocrystalline
NaY zeolite as the catalyst led to a significantly greater SCR reaction rate and less
formation of undesired products than urea-SCR in commercial NaY.
The SCR reactions in nanocrystalline NaY zeolite occurred more quickly than in
commercial NaY zeolite, as indicated by the initial rates for NO loss and N2O production
listed in Table 7.2. Consequently, higher NOx conversion was achieved in
nanocrystalline NaY than in commercial NaY zeolite after urea-SCR for 2 h (Table 7.3).
A significant amount of nitrate remained in commercial NaY zeolite after urea-SCR of
NO2 for 2 h, but nitrate was barely observed in nanocrystalline NaY zeolite after 30 min
(Figure 7.8 (a)). The enhanced SCR reaction rate in urea-SCR over nanocrystalline NaY
zeolite can be explained by the existence of both silanol groups and EFAL sites. As
described by Figure 7.10, urea-SCR of NO2 in nanocrystalline NaY zeolite involves both
the production of NO from nitrate and the N-N bond formation between NO and NH3
adsorbed on silanol groups. First, the silanol groups and EFAL species possess
diffusional advantages since they are both located on the external surface of the
nanocrystalline NaY zeolite. Furthermore, they are considered as chemically distinct
surface sites. A major conclusion of this study and previous studies is that nitrate
adsorbed on EFAL sites on the external surface of nanocrystalline NaY zeolite is more
reactive than nitrate in the internal pores adsorbed to Na+ [40]. The activation of NH3 by
silanol groups is also important for N-N bond formation [156].
As shown in Figure 7.9, EFAL species on nanocrystalline NaY zeolite also
provides important sites for the adsorption and fast hydrolysis of HNCO, which is
134
essential for avoiding isocyanic acid emission [6] and inhibiting the formation of
undesired products. The kinetic data listed in Table 7.2 indicate that the hydrolysis of
HNCO in nanocrystalline NaY zeolite occurred at a much faster rate compared to that in
commercial NaY zeolite. This is consistent with the observation that gas-phase HNCO
persists for a much longer period of time (>60 min, Figure 7.5) in the presence of
commercial NaY zeolite compared to nanocrystalline NaY zeolite (<20 min, Figure 7.4).
The time course measurements for H2O also indicate that hydrolysis happens much faster
in nanocrystalline NaY zeolite than in commercial NaY zeolite (Figure 7.6 (c)).
To conclude, silanol groups and EFAL species are physically and chemically
distinct surface sites, which are present on the external surface of nanocrystalline NaY
zeolite. These sites are responsible for the improved performance of nanocrystalline NaY
relative to commercial NaY since the majority of urea-SCR reactions occur on the
external surface. The role of the internal surface is to store the NOx prior to SCR
reactions on the external surface. A pictorial representation of this multifunctional
capability of the nanocrystalline zeolite catalyst is shown in Figure 7.11. The internal
surface of nanocrystalline zeolite provides sites for NOx- storage (as NO2- and NO3-) and
the minority of the SCR reactions, the external surface provides sites for additional NOx
storage (as NO3-) and the majority of the SCR reactions. Thus nanocrystalline alkali
zeolites can be classified as new materials for SCR catalysts with potentially significantly
better performance.
135
NOx
N2, N2O
NOxStorage
Figure 7.11 A nanocrystalline zeolite particle with a particle size of 23 nm. The internal
surface of nanocrystalline zeolite provides sites for NOx- storage (as NO2- and NO3-) and
the minority of the SCR reactions, the external surface provides sites for additional NOx
storage (as NO3-) and the majority of the SCR reactions.
7.4.5 Urea-SCR vs. Propylene-SCR of NO2 in

Nanocrystalline NaY
In Chapter 6, nanocrystalline NaY zeolite was examined in deNOx reactions at

473 K using propylene as the reducing agent. It was found that nanocrystalline NaY
zeolite was very sensitive to pre-adsorbed water in propylene-SCR of NO2. Compared to
urea-SCR, propylene-SCR appears to be a less realistic deNOx technology under typical
conditions of humid, oxygen rich exhaust streams [34, 162].
Although the effect of adsorbed water on urea-SCR is not studied in this work,
water should not be detrimental to deNOx with urea over nanocrystalline NaY zeolite.
First, water is necessary for the hydrolysis of HNCO in urea-SCR reactions (reaction
7.2). Second, it is suggested from the above discussion that at elevated temperature H2O
did not inhibit the adsorption of NH3 in nanocrystalline NaY zeolite. In urea-SCR of
NO2 over nanocrystalline NaY zeolite, the concentration of H2O in gas phase kept
increasing after the urea-SCR reactions were finished around t = 40 min (Figure 7.6 (c)).
136
Figure 7.6 (a) indicates that the concentration of NH3 in gas phase further decreased after
t = 40 min. Since the oxidation of NH3 by O2 was not significant under the experimental
conditions described in this work [170], NH3 most likely adsorbed on available silanol
groups and/or EFAL sites in nanocrystalline NaY zeolite in the presence of water [14,
40].
Compared to urea-SCR, a significant amount of nitrate and carbonaceous deposit
remained in nanocrystalline NaY zeolite after propylene-SCR of NO2 at 473 K for 6 h
(Figure 6.4). The carbonaceous deposit in nanocrystalline NaY zeolite was found to
hinder the adsorption of propylene and NO2 and further SCR reactions. In urea-SCR of
NO2 over nanocrystalline NaY zeolite, surface nitrate was not observed after 2 h of SCR
reactions. Surface HNCO and OCN- were the only undesired products formed in ureaSCR over nanocrystalline NaY (Figure 7.8). Furthermore, more than 80% of NO2 was
converted into N2 in urea-SCR relative to ~20% selectivity to N2 formation in propyleneSCR of NO2 over nanocrystalline NaY zeolite (see Chapter 6). As mentioned previously,
NOx emissions emerge at high velocity under typical lean burn conditions [34]. At low
temperatures, urea-SCR over nanocrystalline NaY zeolite appears to be an even more
promising deNOx technology than propylene-SCR.
7.5 Conclusions
Nanocrystalline NaY zeolite exhibits enhanced deNOx at low temperature (T = 473

K) compared to commercial NaY zeolite, as shown by the present study on the selective
catalytic reduction of NO2 with urea. Silanol groups and EFAL species on the external
surface of nanocrystalline NaY were found to be responsible for the higher SCR reaction
rate and decreased formation of undesired products relative to commercial NaY zeolite.
Isotopic labeling coupled with infrared analysis indicates that N-N bond formation
involves both a N-atom from NO2 and a N-atom from urea. In nanocrystalline NaY
zeolite, kinetic studies indicated that NH3 formed in thermal decomposition of urea was
137
activated by hydrogen bonding with silanol groups and NO was initially produced in the
thermal reaction of nitrate on EFAL sites. Consequently N-N bond formation occurred
between NH3 adsorbed on silanol groups and NO from the gas phase.
Nanocrystalline alkali zeolites can be visualized as new catalytic materials that have
NOx storage capacity in the internal pores and high reactivity on the external surface.
This provides multifunctional capabilities beyond that found for zeolites with a larger
crystal size and smaller external surface area. Although for shape selective catalysis, the
reactivity of the external surface may be undesirable, in reactions such as deNOx or
hydrocarbon cracking, nanocrystalline zeolite catalysts may have greatly enhanced
reactivity due to reactions on the external surface. This is the first example in the
literature demonstrating that the increased external surface area of nanocrystalline
zeolites can be utilized as a reactive surface with unique active sites for catalysis.
The authors thank Dr. Weiguo Song for providing the nanocrystalline NaY zeolite
sample. The authors gratefully acknowledge the Environmental Protection Agency for
the support of this work (R829600). The results of this work are presented in a
publication under the authorship of: Gonghu Li, Conrad A. Jones, Vicki H. Grassian,
Sarah C. Larsen, J. Catal. (submitted).
138
CHAPTER 8
CONCLUDING REMARKS AND FUTURE DIRECTIONS
Zeolite materials can be used in environmental applications as adsorbents and
catalysts. Photooxidation of hydrocarbons using molecular oxygen in ion-exchanged
zeolite Y is potentially an environmentally benign alternative to conventional liquid
phase oxidation process. In photooxidation of hydrocarbons, the unique framework
structure of zeolite catalyst and exchanged ions facilitated the formation of a charge
transfer complex upon irradiation. Following reactions proceeded selectively in the
supercages of BaY to manufacture partially oxidized products. The reactivity and
selectivity in photooxidation reactions were dependent on reactant, oxidant and
irradiation wavelength.
Synthesized nanocrystalline zeolites have relatively high external surface areas
that are comparable to the total surface areas of typical zeolite materials. A greater
number of silanol groups and extra framework aluminum (EFAL) species are present on
the external surface of nanocrystalline NaY zeolites relative to commercial micrometersized NaY. The EFAL species are associated with the Bronsted and Lewis acidity and
are only present in large quantity in nanocrystalline zeolites. The relatively large external
surface area and the functional groups (silanol groups and EFAL species) on the external
surface are important in applications of nanocrystalline NaY as adsorbents for chemicals
of environmental importance.
In my opinion, the most promising results with the greatest potential are those
using nanocrystalline zeolite materials in environmental catalysis. As shown here, in the
absence of oxygen, the selective catalytic reduction (SCR) of NO2 with propylene at low
temperature (T 473 K) resulted in complete reduction of NO2 to N2 and O2 in
nanocrystalline NaY zeolite with a particle size of 8030 nm. Silanol groups and EFAL
sites on the external surface of nanocrystalline NaY provided additional adsorption sites
139
for propylene and NO2. EFAL species are important for SCR reactions since several
SCR intermediates formed on EFAL sites.
In the presence of an excess amount of oxygen, 100% NOx (NO2+NO) conversion
was realized in propylene-SCR of NO2 using nanocrystalline NaY with a particle size of
23 nm as catalyst. Compared to commercial NaY zeolite with a larger particle size and
negligible external surface area, nanocrystalline NaY possesses markedly higher
adsorption capacity and activity toward N-N bond formation in propylene-SCR of NO2.
A significant amount of SCR reactions in nanocrystalline NaY involved the silanol
groups and EFAL species on the external surface. However, nanocrystalline NaY zeolite
is more sensitive to adsorbed water in propylene-SCR of NO2 since adsorbed water
poisons important surface adsorption sites by hydrogen bonding to silanol groups on the
external surface.
In urea-SCR of NO2, silanol groups and EFAL species on the external surface are
also responsible for the higher SCR reaction rate and decreased formation of undesired
products in nanocrystalline NaY relative to commercial NaY zeolite. In nanocrystalline
NaY zeolite, urea adsorbed on silanol groups and thermally decomposed into NH3 and
HNCO. EFAL species were essential for the adsorption and effective hydrolysis of
HNCO. N-N bond formation involved NH3 that was adsorbed and activated on silanol
groups.
Thus nanocrystalline zeolites are new materials for environmental applications as
adsorbents and catalysts. Compared to the internal surface, the external surface of
nanocrystalline zeolites possesses diffusional advantage and chemically distinct
properties, and provides a tunable platform for various applications. Future directions in
the use of these materials may include selective photooxidation of hydrocarbons,
bifunctional catalysis, adsorption and degradation of toxic chemical compounds.
140
APPENDIX A
INSTRUMENTAL PARAMETERS AND MACRO
PROGRAMS FOR FT-IR MEASUREMENTS
Instrumental parameters for FT-IR measurements on the Mattson Galaxy 6000
infrared spectrometer were specified and saved in the Method Setup of the Control Panel
window in the WinFirst program. Macro programs were written and modified using the
macro toolbox in the WinFirst program.
A.1 Instrumental Setup for FT-IR Measurements
This scan method is used to collect single beam files (.sbm) by averaging 64
sample scans. The file names are assigned in a prompt window.
*****************************************************
-------------------------------------------------------------------------------Sample Scans: 64
Background Scans: 8
IR Data Type: Sample Single Beam
Resolution: 4.000
Signal Gain: 1
Auto Gain: Off
-------------------------------------------------------------------------------Forward Velocity: 40.0 kHz
Reverse Velocity: 40.0 kHz
Scan Direction: Forward
Frequency Start: 600.0000
Frequency End: 4000.0000
Spectral Range: Mid-IR
Laser Sampling: 2
Low-Pass Filter: Auto
ADC Sensitivity: Off
ZPD Polarity: Positive
ZPD Threshold: 170
FFT on Spectrometer: Off
Zero Fill: 1X
FFT Symmetry: Double
Apodization: Triangle
Phase Correction: Mertz
Phase Apodization: Triangle
Phase Type: Real
--------------------------------------------------------------------------------
141
Data Collection Style: Noninteractive
Data Collection Mode: Standard
Display Sample: On
Write Sample: On
Display Background: On
Write Background: On
Save Raw Data: Off
Signal Clipping Check: Off
Audible Scan Alarm: Off
Use Noise Threshold: Off
Noise Ratio: 100
Reference Threshold Time: 1000
Auto Name: On
Auto Increment: On
Starting Value: 1
Scan Filename: Jan24
Scan Data Directory: c:\gonghu\5Jan\
Prompt For Title: Off
Use Control String: On
Title Control String: %n %d
Prompt Level: Silent
Verbosity Level: Silent
Tune Autoscale: On
Tune Points: 300
-------------------------------------------------------------------------------Bench File: 60
Bench Name: Galaxy 6020
Com Settings: COM2:38400,Even,8,1
-------------------------------------------------------------------------------*****************************************************
A.2 Macro Program for Collecting FT-IR Spectra
This macro program is used to instruct the spectrometer to continuously collect

150 scans at a rate of 1 scan per 2 minutes. The collected single beam files are converted
into absorbance spectra (.abs) and saved in a specified directory. In addition, the
absorbance between 1200 and 1400 cm-1 in every absorbance spectrum is integrated. The
integrated absorbance can be saved in a WinFirst Report file (.txt).
*****************************************************
For x=1 to 150 step 1
begin
142
sleep 120
Bench:Scan
ratio c:\gonghu\5Jan\Jan24001.sbm
ras2abs
integrate 1200 1400
save
end
*****************************************************
The instrumental parameters used with the macro program are listed below. The
file names are automatically assigned (Jan24100.sbm -- Jan24249.sbm).
*****************************************************
-------------------------------------------------------------------------------Sample Scans: 64
Background Scans: 8
IR Data Type: Sample Single Beam
Resolution: 4.000
Signal Gain: 1
Auto Gain: Off
-------------------------------------------------------------------------------Forward Velocity: 40.0 kHz
Reverse Velocity: 40.0 kHz
Scan Direction: Forward
Frequency Start: 600.0000
Frequency End: 4000.0000
Spectral Range: Mid-IR
Laser Sampling: 2
Low-Pass Filter: Auto
ADC Sensitivity: Off
ZPD Polarity: Positive
ZPD Threshold: 170
FFT on Spectrometer: Off
Zero Fill: 1X
FFT Symmetry: Double
Apodization: Triangle
Phase Correction: Mertz
Phase Apodization: Triangle
Phase Type: Real
-------------------------------------------------------------------------------Data Collection Style: Noninteractive
Data Collection Mode: Standard
Display Sample: On
Write Sample: On
Display Background: On
Write Background: On
143
Save Raw Data: Off
Signal Clipping Check: Off
Audible Scan Alarm: Off
Use Noise Threshold: Off
Noise Ratio: 100
Reference Threshold Time: 1000
Auto Name: On
Auto Increment: On
Starting Value: 100
Scan Filename: Jan24
Scan Data Directory: c:\gonghu\5Jan\
Prompt For Title: Off
Use Control String: On
Title Control String: %n %d
Prompt Level: Silent
Verbosity Level: Silent
Tune Autoscale: On
Tune Points: 300
-------------------------------------------------------------------------------Bench File: 60
Bench Name: Galaxy 6020
Com Settings: COM2:38400,Even,8,1
-------------------------------------------------------------------------------*****************************************************
A.3 Macro Program for Integrating Specific Bands in
Existing Spectra
This macro program is used to load a series of single beam files and convert them
into absorbance spectra. Integrated areas of specific bands are then obtained following
subtracting a reference spectrum (Jan24002.abs) from each of the absorbance spectra.
The integrated absorbance can be saved in a WinFirst Report file that can be opened in
Excel for further data processing.
*****************************************************
S = findFirstFile "c:\gonghu\5Jan\Jan24*.sbm"
A = stringLength S
if (A>0)
{
While (A>0)
{
144
B=load S
ratio "C:\gonghu\5Jan\Jan24001.sbm"
ras2abs
subtract "C:\gonghu\5Jan\Jan24002.abs" 1
integrate 1790 1750
integrate 2268 2168
integrate 3760 3690
S=findnextfile
A=stringlength S
}
}
*****************************************************
145
APPENDIX B
CALCULATION OF THE CONCENTRATIONS OF
GASES USING FT-IR SPECTROSCOPY
This section describes how the concentration of individual gases using FT-IR
spectroscopy was determined. Usually, about 4 Torr pressure of gas (2 Torr for H2O) in
the premix chamber expanded to the IR cell without zeolite sample in it (Figure 2.3). The
equilibrium pressure of the gas in the IR cell was recorded with an absolute pressure
transducer. Assuming ideal gas behavior, concentration of the gas can be calculated from
the measured pressure at room temperature based on perfect gas equation.
Then an FT-IR spectrum of the gas was collected. Characteristic absorption
band(s) in the infrared spectrum can be integrated, as listed in Table B.1. It is observed
that the integrated absorbance (A) in the specified spectral region is linearly proportional
to the gas concentration in the IR cell. Thus a conversion factor (in mol L-1 A-1) can be
obtained for each gas to calculate concentration from integrated absorbance of the
characteristic band(s) in any other FT-IR spectra. The FT-IR spectra of gases are shown
in Figures B.1-B.3.
Different gases may have absorption bands in the same spectral region. For
example, a band between 1670 and 1540 cm-1 is present in both the FT-IR spectra of NO2
and NO (Figure B.1). So the calculation of concentration is more complex if the gas
phase contains both NO2 and NO. In this case, the concentration of NO is calculated
using the integrated absorbance in the spectra region between 1980 and 1750 cm-1. Next
the integrated absorbance between 1670 and 1540 cm-1 for NO can be estimated since it
is proportional (0.430/0.513) to that between 1980 and 1750 cm-1. The absorbance
between 1670 and 1540 cm-1 for NO2 is obtained by subtracting the absorbance attributed
to NO from the integrated absorbance in the same spectral region. This way the
concentration of NO2 can be calculated using the FT-IR spectrum.
146
Table B.1 Calculation of the conversion factor for gases. The conversion factors are
used to calculate the gas concentration from integrated absorbance in IR spectra.
Formula
Pressure
Concentration
-1
Spectral region
-1 a
Integrated
Conversion
(mTorr)
(mol L )
(cm )
absorbance (A)
(mol L-1 A-1)
NO2
2729
1.4710-4
1670-1540
4.74
0.3110-4
NO
2936
1.5810-4
1980-1750
0.513
3.0810-4
N2O
2191
1.1810-4
2271-2124
1.71
0.6910-4
C3H6
2799
1.5110-4
1540-1335
0.974
1.5510-4
CO2
3002
1.6210-4
2400-2230
2.06
0.7810-4
CO
3058
1.6510-4
2272-1995
0.388
4.2510-4
NH3
2450
1.3210-4
979-910
0.518
2.5310-4
H2O
800
0.4310-4
3955-3790
0.196
2.2010-4
SO2
2810
1.5110-4
1430-1280
6.39
0.2410-4
Other spectral regions that can be used to calculate the concentration: 1662-1542 cm-1 (A =
0.430) for NO; 1342-1217 cm-1 (A = 0.599) for N2O; 1695-1595 cm-1 (A = 0.564) and 19001775 cm-1 (A = 0.138) for C3H6; 3955-3515 cm-1 (A = 0.636) for H2O.
0.01
A
b
s
o
r
b
a
n
c
e
NO 2 ( 0.25)
NO ( 3)
N 2O
4000
3500
3000
2500
Wavenumber
2000
(cm-1 )
Figure B.1 FT-IR spectra of gas-phase NO2, NO and N2O.
1500
1000
147
0.02
A
b
s
C 3H 6
o
r
b
a
n
CO 2
c
e
CO ( 10)
4000
3500
3000
2500
2000
1500
wavenumber (cm-1 )
Figure B.2 FT-IR spectra of gas-phase C3H6, CO2 and CO.
0.05
A
b
s
o
r
b
a
n
c
e
NH 3 ( 5)
SO 2
H 2 O ( 10)
4000
3500
3000
2500
2000
1500
Wavenumber (cm-1 )
Figure B.3 FT-IR spectra of gas-phase NH3, SO2 and H2O.
1000
148
Another example that may cause difficulty in determining concentrations is

shown in Figure B.4. In this case, the gas phase contains both N2O (2224 cm-1) and CO
(2143 cm-1), whose characteristic absorptions are close in frequency to each other such
that spectral overlap is a problem. In this dissertation research, an approximation is made
to solve this problem since the concentration of N2O or CO is usually low (< 510-5 mol
L-1). It is assumed that the integrated absorbance between 2143 and 1995 cm-1 (AI in
Figure B.4) is half of that between 2265 and 1995 cm-1 for CO. Thus the concentration
of CO can be calculated using 2AI as the integrated absorbance and the conversion factor
from Table B.1. Then the absorbance between 2265 and 1995 cm-1 (AII) is integrated.
Finally the concentration of N2O can be obtained using the integrated absorbance AII-2AI.
A
b
s
o
r
b
a
n
c
e
N 2O
2224
0.0005
CO
2143
AI
A II
2240
2200
2160
2120
2080
2040
2000
W avenumber (cm-1 )
Figure B.4 FT-IR spectrum of gas phase in a sample that contains N2O and CO.
149
APPENDIX C
ADSORPTION AND DESORPTION STUDIES OF
CYCLOHEXANE IN BaY USING A FLOW SYSTEM
This section describes a quantitative study of cyclohexane adsorption and
desorption in BaY zeolite using a newly constructed adsorption/desorption apparatus
(Figure 2.7). The preparation of BaY sample is described in Chapter 3.
C.1 Adsorption of Cyclohexane in BaY
C.1.1 Pulse Injections in the Absence of BaY
1. Continuously flow He through the vapor generator and the sampling loop; purge the
quartz tube with He at room temperature;
2. Switch the six-way valve for 30 seconds to direct a pulse of cyclohexane through the
quartz tube;
3. Repeat the pulse injection every 10 minutes until the FID signal for each pulse reaches
a constant level (Figure C.1, dotted line).
C.1.2 Pulse Injections in the Presence of BaY
1. Prepare a thin film of BaY according to the method described in Section 2.2.2, put the
BaY sample in the quartz tube and heat to 573 K under helium for 1 hour;
2. Cool the BaY sample back to room temperature under helium, continuously flow He
through the vapor generator and the sampling loop;
3. Switch the six-way valve for 30 seconds to direct a pulse of cyclohexane through the
quartz tube with the BaY sample;
4. Repeat the pulse injection every 10 minutes until the FID signal for each pulse reaches
a constant level, i.e. the BaY sample is saturated (Figure C.1, solid green line);
5. Stop the pulse injection and weigh the BaY sample.
150
1
Blank
W ith BaY
FID Signal (a.u.)
0.8
0.6
0.4
0.2
0
0
10
20
30
Time (min)
40
50
60
Figure C.1 FID signal following six pulses of cyclohexane in the absence (dotted line)
and presence (solid green line) of BaY zeolite.
C.1.3 Calculation
As shown in Figure C.1, the BaY sample is saturated with 3 pulses of

cyclohexane. The integrated areas of FID signals for pulse injections are listed in Table
C.1. The average area of FID signal for a single pulse in the absence of BaY is
determined to be 31.7105 counts. Assuming ideal gas behavior, the amount of
cyclohexane in a single pulse (n) can be calculated from the partial pressure of
cyclohexane (p = 11344 Pa at 295 K) and the volume of the sampling loop (V = 0.25
mL). The temperature of the sampling loop was kept at 373 K.
n = pV/RT = (113440.2510-6)/(8.314373) = 9.2510-7 mole
151
Table C.1 Quantification of cyclohexane adsorbed in BaY
Blank
Integrated areas (10 ) of FID signal

for six pulse injections
Difference in integrated areas (105) a

-7
Molecules adsorbed (10 mole)

Mass of BaY sample (mg)
-7
Supercages in BaY (10 mole)

Molecules per supercage
Sample 1
Sample 2
31.9
1.2
2.0
32.0
4.2
2.6
32.0
22.7
4.0
31.7
29.8
19.4
31.2
28.7
31.4
31.3
27.6
31.4
67.0
98.8
19.6
28.9
2.5
3.1
10.6
13.2
1.8
2.2
Difference between the integrated areas of FID signal in the absence and in the
presence of BaY sample; b the quantity of molecules adsorbed in BaY is
calculated from difference in integrated areas (see text for details).
Then the difference between the total area of FID signal for pulse injections in the
absence and in the presence of BaY sample can be obtained and converted into the
amount of cyclohexane (n) adsorbed in BaY. For sample 1 in Table C.1,
n = (67.01059.2510-7)/31.7105 = 19.610-7 mole
The quantity of supercages in BaY can be calculated from the mass of BaY and
the molecular weight of supercage (2349.5 g mol-1). Consequently the maximum uptake
of cyclohexane in BaY can be determined. Under the experimental conditions, an uptake
of ~2 cyclohexane molecules per supercage is obtained for BaY. This is consistent with a
previous study showing that the conversion of cyclohexane dramatically decreased at
loadings greater than 2 molecules per supercage in thermal oxidation using BaY catalyst
at 338 K [69].
152
C.2 Temperature Programmed Desorption of

Cyclohexane in BaY
Temperature programmed desorption (TPD) of cyclohexane in BaY is carried out

in the experimental setup described in Figures 2.7 and 2.8. In TPD studies, a thin film of
BaY sample is degassed and dosed with desired amount of adsorbate (cyclohexane). The
coverage of cyclohexane () can be calculated based on the adsorption results (Section
C.1). The BaY sample is then purged under helium at room temperature until no
cyclohexane comes out of the quartz tube. Desorption experiments are then carried out at
variable heating rates and helium flow rates.
0.6
3 K/min
15 K/min
30 K/min
60 K/min
Desorption Rate (a.u.)
0.5
0.4
0.3
0.2
0.1
0
300
350
400
450
500
550
600
650
700
Temperature (K)
Figure C.2 TPD profiles of cyclohexane in BaY at different heating rates (He flow rate
125 mL min-1, cyclohexane coverage 0.21).
Figure C.2 shows the desorption profiles of cyclohexane (coverage 0.2) in BaY at
different heating rates. With increasing heating rate, the maximum desorption
temperature (Tm) of cyclohexane decreases, but the desorption rate (rm) at Tm increases.
153
Based on Redhead analysis, the activation energy of desorption (Ed) can be

obtained using the following equation:
d[ln(/Tm2)]/d(1/Tm) = -Ed/k
where is heating rate and k is Boltzman constant. The activation energy of desorption
for cyclohexane in BaY is determined to be (-4266.6)(-k) = 35.3 kJ mol-1 (Figure C.3).
Similarly, The activation energy of desorption for cyclohexanone in BaY is determined to
be 98.2 kJ mol-1. This indicates that cyclohexanone, the major product in photooxidation
of cyclohexane (see Chapter 3), adsorbs more strongly than cyclohexane in BaY.
-12.5
ln(heat rate/Tm2)
-13
-13.5
-14
-14.5
-15
y = -4.9141 - 4266.6x R= 0.98953

-15.5
0.0017
0.0018
0.0019
0.002
0.0021
0.0022
0.0023
0.0024
-1
1/Tm (K )
Figure C.3 Determination of activation energy of desorption (Ed) for cyclohexane in

BaY. The data are linearly fitted to calculate Ed (see text for details).
Desorption of cyclohexane in BaY can be carried out at different helium flow rate
(Figure C.4) or with different cyclohexane coverage (Figure C.5). TPD results at
different flow rate can help to understand the diffusion and re-adsorption of cyclohexane
in BaY. Diffusion and re-adsorption are important in studies using zeolite materials,
especially nanocrystalline zeolites, as adsorbents and catalysts.
154
0.12
45 ml/min
85 ml/min
125 ml/min
165 ml/min
0.1
0.08
0.06
0.04
0.02
0
400
450
500
Temperature (K)
550
600
Figure C.4 TPD profiles of cyclohexane in BaY at different Helium flow rates (heating
rate 15 K min-1, cyclohexane coverage 0.21).
0.3
1.00
0.74
0.44
0.21
0.25
0.2
0.15
0.1
0.05
0
300
350
400
450
500
550
Temperature (K)
Figure C.5 TPD profiles of cyclohexane in BaY at different cyclohexane coverage (He
flow rate 125 mL min-1, heating rate 15 K min-1).
155
REFERENCES
[1] T. Maesen and B. Marcus, Studies in Surface Science and Catalysis (Vol. 137), Eds.
H. van Bekkum, E.M. Flanigen, P.A. Jacobs and J.C. Jensen. Elsevier Science B.V.,
Amsterdam. 2001, p 1.
[2] E.M Flanigen, Studies in Surface Science and Catalysis (Vol. 137), Eds. H. van
Bekkum, E.M. Flanigen, P.A. Jacobs and J.C. Jensen. Elsevier Science B.V.,
[3] C.O. Arean, Comments Inorg. Chem. 2000, 22, 241.
[4] M.E. Davis, Micro. Meso. Mater. 1998, 21, 173.
[5] L. Moscou, Studies in Surface Science and Catalysis (Vol. 58), Eds. H. van Bekkum,
E.M. Flanigen and J.C. Jensen. Elsevier Science B.V., Amsterdam. 1991, p 1.
[6] G. Delahay and B. Coq, Catalytic Science Series 2002, 3, 345.
[7] P.J. Kunkeler, R.S. Downing and H. van Bekkum, Studies in Surface Science and
Catalysis (Vol. 137), Eds. H. van Bekkum, E.M. Flanigen, P.A. Jacobs and J.C.
Jensen. Elsevier Science B.V., Amsterdam. 2001, p 987.
[8] L.B McCusker and C. Baerlocher, Studies in Surface Science and Catalysis (Vol.
137), Eds. H. van Bekkum, E.M. Flanigen, P.A. Jacobs and J.C. Jensen. Elsevier
Science B.V., Amsterdam. 2001, p 37.
[9] T.F. Degnan, Jr., Topics in Catalysis 2000, 13, 349.
[10] http://www.iza-structure.org.
[11] V. Ramamurthy, J. Photochem. Photobio. C 2000, 1 145.
[12] W. Song, R.E. Justice, C.A. Jones, V.H. Grassian and S.C. Larsen, Langmuir 2004,
20, 4696.
[13] W. Song, R.E. Justice, C.A. Jones, V.H. Grassian and S.C. Larsen, Langmuir 2004,
20, 8301.
[14] W. Song, G. Li, V.H. Grassian and S.C. Larsen, Environ. Sci. Tech. 2005, 39, 1214.
[15] A. Jentys and J.A. Lercher, Studies in Surface Science and Catalysis (Vol. 137),
Eds. H. van Bekkum, E.M. Flanigen, P.A. Jacobs and J.C. Jensen. Elsevier Science
B.V., Amsterdam. 2001, p 345.
[16] H.G. Karge, Micro. Meso. Mater. 1998, 22, 547.
[17] S. Khabtou, T. Chevreau and J.C. Lavalley, Micro. Mater. 1994, 3, 133.
[18] R. Szostak, Studies in Surface Science and Catalysis (Vol. 137), Eds. H. van
Bekkum, E.M. Flanigen, P.A. Jacobs and J.C. Jensen. Elsevier Science B.V.,
156
[19] T. Kawai and K. Tsutsumi, J. Coll. Inter. Sci. 1999, 212, 310.
[20] D.L. Bhering, A. Ramirez-Solis and C.A. Mota, J. Phys. Chem. B 2003, 107, 4342.
[21] M.J. Remy, D. Stanica, G. Poncelet, E.P. Feijen, P.J. Grobet, J.A. Martens and P.A.
Jacobs, J. Phys. Chem. 1996, 100, 12440.
[22] R.D. Shannon, K.H. Gardner, R.H. Staley, R. Bergeret, P. Gallezot and A. Auroux,
J. Phys. Chem. 1985, 89, 4778.
[23] C. Chao and J.H. Lunsford, J. Am. Chem. Soc. 1971, 93, 6794.
[24] C. Chao and J.H. Lunsford, J. Am. Chem. Soc. 1971, 93, 71.
[25] M.R. Basila, T.R. Kantner and K.H. Rhee, J. Phys. Chem. 1964, 68, 3197.
[26] J.A. Martens and P.A. Jacobs, Studies in Surface Science and Catalysis (Vol. 137),
Eds. H. van Bekkum, E.M. Flanigen, P.A. Jacobs and J.C. Jensen. Elsevier Science
B.V., Amsterdam. 2001, p 633.
[27] Environmental Catalysis, Ed. V.H. Grassian, CRC Publishing, Boca Raton, FL,
2005.
[28] Environmental Catalysis, Ed. J.N. Armor, American Chemical Society, Washington
DC, 1994.
[29] G.A. Somorjai and Y.G. Borodko, Catal. Lett. 2001, 76, 1.
[30] G.A. Somorjai and K. McCrea, Appl. Catal. A 2001, 222, 3.
[31] V.H. Grassian and S.C. Larsen, The Handbook of Photochemistry and
Photobiolody, Ed. H.S. Nalwa, American Scientific Publishers, CA, 2003, p 451.
[32] F. Blatter and H. Frei, J. Am. Chem. Soc. 1993, 115, 7501.
[33] S. Bhattacharyya and R.K. Das, Int. J. Energy Res. 1999, 23, 351.
[34] J.N. Armor, Catal. Today 1995, 26, 99.
[35] S.C. Larsen and V.H. Grassian, Encyclopedia of Nanoscience and Nanotechnology,
Eds. J.A. Schwarz, C.I. Contescu, K. Putyera, Marcell Dekker Publishing, Co., NY,
2004, p1137.
[36] S.C. Larsen, Environmental Catalysis, Ed. V.H. Grassian, CRC Publishing, Boca
Raton, FL, 2005.
[37] Y. Xiang, S.C. Larsen and V.H. Grassian, J. Am. Chem. Soc. 1999, 121, 5063.
[38] G. Li, M. Xu, S.C. Larsen and V.H. Grassian, J. Mol. Catal. A 2003, 194, 169.
[39] G. Li, S.C. Larsen and V.H. Grassian, J. Mol. Catal. A 2005, 227, 25.
[40] G. Li, S.C. Larsen and V.H. Grassian, Catal. Lett. (in press).
157
[41] G. Li, C.A. Jones, S.C. Larsen and V.H. Grassian, J. Catal. (submitted).
[42] M.A. Arribas and A. Martinez, Catal. Today 2001, 65, 117.
[43] S. Al-Khattaf and H. de Lasa, Ind. Eng. Chem. Res. 1999, 38, 1350.
[44] F. Blatter, H. Sun, S. Vasenkov and H. Frei, Catal. Today 1998, 41, 297.
[45] B.C. Gates, Catalytic Chemistry, New York: Wiley. 1992.
[46] G.W. Parshall and S.D. Ittel, Homogeneous Catalysis. New York: Wiley. 1992.
[47] A.G. Panov, R.G. Larsen, N.I. Totah, S.C. Larsen and V.H. Grassian, J. Phys.
Chem. B 2000, 104, 5706.
[48] K.A. Suresh, M.M. Sharma and T. Sridhar, Ind. Eng. Chem. Res. 2000, 39, 3958.
[49] I.W.C.E. Arends, R.A. Sheldon, M. Wallau and U. Schuchardt, Angew. Chem. Int.
Ed. 1997, 36, 1144.
[50] R. Raja, G. Sankar and J.M. Thomas, J. Am. Chem. Soc. 1999, 121, 11926.
[51] J.D. Chen and R.A. Sheldon, J. Catal. 1995, 153, 1.
[52] D.L. Vanoppen, D.E. Devos, M.J. Genet, P.G. Rouxhet and P.A. Jacobs, Angew.
Chem. Int. Ed. 1995, 34, 560.
[53] G.X. Lu, H.X. Gao, J.H. Suo and S.B. Li, J. Chem. Soc. Chem. Comm. 1994, 21,
2423.
[54] E.L. Pires, J.C. Magalhaes and U. Schuchardt, Appl. Catal. A 2000, 203, 231.
[55] H. Sun, F. Blatter and H. Frei, J. Am. Chem. Soc. 1996, 118, 6873.
[56] F. Blatter and H. Frei, J. Am. Chem. Soc. 1994, 116, 1812.
[57] H. Sun, F. Blatter and H. Frei, J. Am. Chem. Soc. 1994, 116, 7951.
[58] F. Blatter, F. Moreau and H. Frei, J. Phys. Chem. B 1994, 98, 13403.
[59] F. Blatter, H. Sun and H. Frei, Catal. Lett. 1995, 35, 1.
[60] H. Frei, F. Blatter and H. Sun, Chemtech 1996, 26, 24.
[61] F. Blatter, H. Sun and H. Frei, Chem. Eur. J. 1996, 2, 385.
[62] H. Sun, F. Blatter and H. Frei, Catal. Lett. 1997, 44, 247.
[63] S. Vasenkov and H. Frei, J. Phys. Chem. B 1997, 101, 4539.
[64] S. Vasenkov and H. Frei, J. Phys. Chem. B 1998, 102, 8177.
[65] K.B. Myli, S.C. Larsen and V.H. Grassian, Catal. Lett. 1997, 48, 199.
158
[66] P. Li, Y. Xiang, V.H. Grassian and S.C. Larsen, J. Phys. Chem. B 1999, 103, 5058.
[67] P. Carl and S.C. Larsen, J. Catal. 1999, 182, 208.
[68] A.G. Panov, K.B. Myli, Y. Xiang, V.H. Grassian and S.C. Larsen, Green Chemical
Syntheses and Processes, Eds. P. Anastas, L.G. Heine, and T.C. Williamson, ACS:
Washington DC, 2000, p 206.
[69] R.G. Larsen, A.C. Saladino, T.A. Hunt, J.E. Mann, M. Xu, V.H. Grassian and S.C.
Larsen, J. Catal. 2001, 204, 440.
[70] M.S. Niassary, F. Farzaneh, M. Ghandi and L. Turkian, J. Mol. Catal. A 2000, 157,
183.
[71] Q.H. Xia, X. Chen and T.. Tatsumi, J. Mol. Catal. A 2001, 176, 179.
[72] M. Fujiwara, H. Wessell, P. Hyung-Suh and H.W. Roesky, Tetrahedron 2002, 58,
239.
[73] J.P.M. Niederer and W.F. Holderich, Appl. Catal. A 2002, 229, 51.
[74] D. Nonhebel and J. Walton, Free Radical Chemistry: Structure and Mechanism.
Cambridge: Cambridge University Press, 1974.
[75] B. Vogel, C. Schneider and E Klemm, Catal. Lett. 2002, 79, 107.
[76] T. Bein, Chem. Mater. 1996, 8, 1636.
[77] J.L. Meinershagen and T. Bein, J. Am. Chem. Soc. 1999, 121, 448.
[78] S. Mintova and T. Bein, Micro. Meso. Mater. 2001, 50, 159.
[79] N.B. Castagnola and P.K. Dutta, J. Phys. Chem B 2001, 105, 1537.
[80] N.B. Castagnola and P.K. Dutta, J. Phys. Chem B 1998, 102, 1696.
[81] S. Mintova, N.H. Olsen and T. Bein, Angew. Chem. Int. Ed. 1999, 38, 3201.
[82] Q. Li, D. Creaser and J. Sterte, Chem. Mater. 2002, 14, 1319.
[83] B.A. Holmberg, H. Wang, J.M Norbech and Y. Yan, Micro. Meso. Mater. 2003, 59,
13.
[84] G. Zhu, S. Qiu, J. Yu, Y. Sakamoto, F. Xiao, R. Xu and O. Terasaki, Chem. Mater.
1998, 10, 1483.
[85] P.O. Fritz and J.H. Lunsford, J. Catal. 1989, 118, 85.
[86] B. Zhan, M.A. White, M. Lumsden, J. Mueller-Neuhaus, K.N. Robertson, T.S.
Cameron and M. Gharghouri, Chem. Mater. 2002, 14, 3636.
[87] C. L. Angell and P. C. Schaffer, J. Phys. Chem. 1965, 69, 3463.
159
[88] C. Sedlmair, B. Gil, K. Seshan, A. Jentys and J. Lercher, Phys. Chem. Chem. Phys.
2003, 5, 1897.
[89] J. Szanyi, J. Kwak, R.A. Moline and C.H. F. Peden, Phys. Chem. Chem. Phys. 2003,
5, 4045.
[90] K. Hadjiivanov, J. Saussey, J.L. Freysz and J.C. Lavalley, Catal. Lett. 1998, 52,
103.
[91] F. Thibault-Starzyk, B. Gil, S. Aiello, T. Chevreau and J. Gilson, Micro. Meso.
Mater. 2004, 67, 107.
[92] Y. Xiang, S.C. Larsen and V.H. Grassian, J. Am. Chem. Soc. 1999, 121, 5063.
[93] S. Zheng, H.R. Heydenrych, A. Jentys and J.A. Lercher, J. Phys. Chem. B 2002,
106, 9552.
[94] A. Panov and J.J. Fripiat, Langmuir 1998, 14, 3788.
[95] D.R. Flentge, J.H. Lunsford, P.A. Jacobs and J.B. Uytterhoeven, J. Phys. Chem.
1975, 179, 354.
[96] J. Garcia-Martinez, D. Cazorla-Amoros and A. Linares-Solano, Langmuir 2002, 18,
9778.
[97] J.A. Wang, L.F. Chen, R. Limas-Ballesteros, A. Montoya and J.M Dominguez, J.
Mol. Catal. A 2003, 194, 181.
[98] A. Dutta, R. G.RCavell, R.W. Tower and Z.M. George, J. Phys. Chem. 1985, 89,
443.
[99] A. M. Shor and A.I. Rubaylo, J. Mol. Structure 1997, 410, 133.
[100] Y. Kuo, B. Cheng and Y. Lee, Chem. Phys. Lett. 1991, 177, 195.
[101] M. Laniecki, M. Ziolek and H. G. Karge, J. Phys. Chem. 1987, 91, 4.
[102] V.H. Grassian, Int. Rev. Phys. Chem. 2001, 20, 467.
[103] R. Burch, J.P. Breen and F.C. Meunier, Appl. Catal. B 2002, 39, 283.
[104] Y. Traa, B. Burger and J. Weitkamp, Micro. Meso. Mater. 1999, 30, 3.
[105] M. Shelef, Chem. Rev. 1995, 95, 209.
[106] Y. Yeom, B. Wen, W.M. H. Sachtler and E. Weitz, J. Phys. Chem. B 2004, 108,
5386.
[107] B. Wen, Y. Yeom, E. Weitz and W.M.H. Sachtler, Appl. Catal. B 2004, 48, 125.
[108] J. Szanyi, J. Kwak and C.H.F. Peden, J. Phys. Chem. B 2004, 108, 3746.
[109] S.J. Schmieg, B.K. Cho and S.H. Oh, Appl. Catal. B 2004, 49, 113.
160
[110] A. Sultana, R. Loenders, O. Monticelli, C. Kirschhock, P.A. Jacobs and J.A.
Martens, Angew. Chem. Int. Ed. 2000, 39, 2934.
[111] O. Monticelli, R. Loenders, P.A. Jacobs and J.A. Martens, Appl. Catal. B 1999, 21,
215.
[112] R. Brosius and J.A. Martens, Top. Catal. 2004, 28, 119.
[113] V. A. Sadykov, V.V. Lunin, V.A. Matyshak, E.A. Paukshtis, A. Rozovskii, N.N.
Bulgakov and J. R.H. Ross, Kinet. Catal. 2003, 44, 379.
[114] R.H. Smits and Y. Iwasawa, Appl. Catal. B 1995, 6, L201.
[115] T. Gerlach, F. Schutze and M. Baerns, J. Catal. 1999, 185, 131.
[116] Y. Yu, H. He, Q. Feng, H. Gao and X. Yang, Appl. Catal. B 2004, 49, 159.
[117] M. Haneda, Y. Kintaichi, M. Inaba and H. Hamada, Catal. Today 1998, 2, 127.
[118] T. Tanaka, T. Okuhara and M. Misono, Appl. Catal. B 1994, 4, L1.
[119] A. Satsuma, T. Enjoji, K. Shimizu, K. Sato, H. Yoshida and T. Hattori, J. Chem.
Soc., Faraday Trans. 1998, 94, 301.
[120] A.L. Goodman, E.T. Bernard and V.H. Grassian, J. Phys. Chem. A 2001, 105,
6443.
[121] K.I. Hadjiivanov, Catal. Rev. Sci. Eng. 2000, 42, 71.
[122] A.L. Goodman, T.M. Miller and V.H. Grassian, J. Vac. Sci. Tech. A 1998, 16,
2585.
[123] A.L. Goodman, G.M. Underwood and V.H. Grassian, J. Phys. Chem. A 1999, 103,
7217.
[124] K. Hadjiivanov, D. Klissurski, G. Ramis and G. Busca, Appl. Catal. B 1996, 7,
251.
[125] A. Lee Smith, Applied Infrared Spectroscopy: Fundamentals, Techniques, and
Analytical Problem Solving, 1979.
[126] M. Haneda, N. Bion, M. Daturi, J. Saussey, J. Lavalley, D. Duprez and H.
Hamada, J. Catal. 2002, 206, 114.
[127] S.V. Gerei, E.V. Rozhkova and Y.B. Gorokhovatskii, J. Catal. 1973, 28, 341.
[128] B.A. Morrow and I.A. Cody, J. Chem. Soc., Faraday Trans. 1975, 71, 1021.
[129] F. Solymosi and T. Bansagi, J. Catal. 1995, 156, 75.
[130] V.H. Grassian, J. Phys. Chem. A 2002, 106, 860.
[131] J.M. Garcia-Cortes, J. Perez-Ramirez, J.N. Rouzaud, A.R. Vaccaro, M.J. IllanGomez and C. Salinas-Martinez De Lecea, J. Catal. 2003, 218, 111.
161
[132] K. Hadjiivanov and L. Dimitrov, Micro. Meso. Mater. 1999, 27, 49.
[133] I. Kiricsi, H. Foerster and G. Tasi, J. Mol. Catal. 1991, 65, L29.
[134] T.E. Hoost, K.A. Laframboise and K. Otto, Appl. Catal. B 1995, 7, 79.
[135] Y. Ukisu, S. Sato, A. Abe and K. Yoshida, Appl. Catal. B 1993, 2, 147.
[136] G.R. Bamwenda, A. Ogata, A. Obuchi, J. Oi, K. Mizuno and J. Skrzypek, Appl.
Catal. B 1995, 6, 311.
[137] W. Schiesser, H. Vinek and A. Jentys, Appl. Catal. B 2001, 33, 263.
[138] S. Sumiya, H. He, A. Abe, N. Takezawa and K. Yoshida, J. Chem. Soc., Faraday
Trans. 1998, 94, 2217.
[139] T. Weingand, S. Kuba, K. Hadjiivanov and H. Knoezinger, J. Catal. 2002, 209,
539.
[140] N.W. Hayes, R.W. Joyner and E.S. Shpiro, Appl. Catal. B 1996, 8, 343.
[141] T. Beutel, B. Adelman and W.M. H. Sachtler, Catal. Lett. 1996, 37, 125.
[142] J.L. d'Itri and W.M.H. Sachtler, Appl. Catal. B 1993, 2, L7.
[143] H. Chen, T. Voskoboinikov and W.M.H. Sachtler, J. Catal. 1999, 186, 91.
[144] E. Finocchio, G. Busca, V. Lorenzelli and R.J. Willey, J. Catal. 1995, 151, 204.
[145] J. Vassallo, E. Miro and J. Petunchi, Appl. Catal. B 1995, 7, 65.
[146] A.D. Cowan, R. Duempelmann and N.W. Cant, J. Catal. 1995, 151, 356.
[147] P.A.S. Smith, The Organic Chemistry of Open-Chain Nitrogen Compounds, 1965.
[148] H. Chen, T. Voskoboinikov and W.M.H. Sachtler, J. Catal. 1998, 180, 171.
[149] N.W. Cant, D.C. Chambers, A.D. Cowan, I.O. Y. Liu and A. Satsuma, Top. Catal.
2000, 10, 13.
[150] M. Ogura, T. Ohsaki and E. Kikuchi, Micro. Meso. Mater. 1998, 21, 533.
[151] A. Shichi, K. Katagi, A. Satsuma and T. Hattori, Appl. Catal. B 2000, 24, 97.
[152] A. Shichi, A. Satsuma and T. Hattori, Appl. Catal. B 2001, 30, 25.
[153] T. Tabata and H. Ohtsuka, Catal. Lett. 1997, 48, 203.
[154] P. Praserthdam, N. Mongkolsiri and P. Kanchanawanichkun, Catal. Comm. 2002,
3, 191.
[155] M. Haneda, Y. Kintaichi, N. Bion and H. Hamada, Appl. Catal. B 2003, 42, 57.
[156] P. Forzatti, Appl. Catal. A 2001, 222, 221.
162
[157] M. Koebel, M. Elsener and M. Kleemann, Catal. Today 2000, 59, 335.
[158] E. Seker, N. Yasyerli, E. Gulari, C. Lambert and R.H. Hammerle, Appl. Catal. B
2002, 37, 27.
[159] M. Koebel and E. Strutz, Ind. Eng. Chem. Res. 2003, 42, 2093.
[160] R. van Helden, R. Verbeek, F. Willems and R. van der Welle, Society of
Automotive Engineers, 2004, SP-1861 25.
[161] B.K. Gullett, P.W. Groff, M. Linda Lin and J.M. Chen, Air and Waste 1994, 44,
1188.
[162] P.L.T. Gabrielsson, Top. Catal. 2004, 28, 177.
[163] T. Morimune, H. Yamaguchi and Y. Yasukawa, Exp. Thermal Fluid Sci. 1998, 18,
220.
[164] D.H.E. Seher, M. Reichelt and S. Wickert, Society of Automotive Engineers, 2003,
SP-1811, 21.
[165] P.Tennison, C. Lambert and M. Levin, Society of Automotive Engineers, 2004, SP1860, 349.
[166] H.L. Fang and H.F.M. DaCosta, Appl. Catal. B 2003, 46, 17.
[167] S. Yim, S. Kim, J. Baik, I. Nam, Y. Mok, J. Lee, B. Cho and S. Oh, Ind. Eng.
Chem. Res. 2004, 43, 4856.
[168] M. Larrubia, G. Ramis and G. Busca, Appl. Catal. B 2000, 27, L145.
[169] C.A. Jones, D. Stec and S.C. Larsen, J. Mol. Catal. A 2004, 212, 329.
[170] J. Baik, S. Yim, I. Nam, Y. Mok, J. Lee, B. Cho and S. Oh, Top. Catal. 2004, 30,
37.
[171] U. Eickhoff and F. Temps, Phys. Chem. Chem. Phys. 1999, 1, 243.
[172] A. Lapinski, J. Spanget-Larsen, J. Waluk and J. George Radziszewski, J. Chem.
Phys. 2001, 115, 1757.
[173] R.E. Baren, M.A. Erickson and J.F. Hershberger, Int. J. Chem. Kinet. 2002, 34, 12.
[174] M.S. Lowenthal, R.K. Khanna and M.H. Moore, Spectrochimica Acta Part A
2002, 58, 73.
[175] N.B. Colthup, L.H. Daly and S.E. Wiberley, Introduction to Infrared and Raman
Spectroscopy, 1990.
[176] D.T. Burns, A. Townshend and A.H. Carter, Inorganic Reaction Chemistry (II):
Reactions of the Elements and Their Compounds (Part A), 1981.

FT-IR Studies of Zeolite Materials - Characterization and Environm PDF

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

FT-IR Studies of Zeolite Materials - Characterization and Environm PDF

Enviado por

Direitos autorais:

Formatos disponíveis

University of Iowa

Iowa Research Online

FT-IR studies of zeolite materials: characterization

Copyright 2005 Gonghu Li

Follow this and additional works at: http://ir.uiowa.edu/etd

FT-IR STUDIES OF ZEOLITE MATERIALS: CHARACTERIZATION

FT-IR STUDIES OF ZEOLITE MATERIALS: CHARACTERIZATION

A thesis submitted in partial fulfillment

CHAPTER 2 EXPERIMENTAL METHODS .................................................................10

CHAPTER 4 CHARACTERIZATION OF NANOCRYSTALLINE NaY

CHAPTER 7 SELECTIVE CATALYTIC REDUCTION OF NO2 WITH UREA

Figure 4.5 Representative SEM images of synthesized nanocrystalline NaY with a

Figure 4.15 FT-IR spectra of SO2 adsorbed in nanocrystalline NaY at 298 K as a

Figure B.2 FT-IR spectra of gas-phase C3H6, CO2 and CO............................................147

Nanocrystalline zeolites are zeolites with discrete, uniform crystals with

Hydroxyl nests can be produced in the process of steaming (heating to high

of cyclohexane and cyclohexene in BaY zeolite. The selectivity to the formation of

instrumental parameters of FT-IR measurements and macro programs for automatic

Dry air purge area

Figure 2.3 Experimental setup for transmission FT-IR studies.

2.1.3 Protocol of In Situ FT-IR Measurements

(b) Clean zeolite

(c) Zeolite with

(d) Surface species

(a) Gas phase

2.1.4 Potassium Bromide Pellets Technique

2.2 Experimental Setup for Photooxidation Studies

Dry air purge area

2.2.2 Batch Reactor Studies

100 mg 150 mg 200 mg 300 mg

Cover with Aluminum foil

A zeolite thin film is prepared by depositing zeolite sample on a 5.5 cm 0.7 cm

With a temperature controller (Omega i/16 Temperature & Process Controller),

A thermal couple is placed closely to the zeolite sample to monitor the

elemental composition of individual particles. Atomic force microscopy (AFM, Digital

Figure 3.1 Examples of selective photooxidation of hydrocarbons with molecular oxygen

In the proposed mechanism, the oxidation is initiated by excitation of a

63]. Cyclohexyl hydroperoxide is produced as an intermediate following proton

Figure 3.3 FT-IR spectra of cyclohexane adsorbed in BaY as a function of cyclohexane

1000 1500 2000 2500 3000 3500 4000

The photooxidation of cyclohexane in BaY with O2 was investigated using FT-IR

O2, 325nm, 100 mTorr, 18 h

O2, 455nm, 100 mTorr, 18 h

The results indicate that ~80% of the products formed in photooxidation of

The FT-IR spectra of cyclohexene in BaY as a function of cyclohexene pressure

are shown in Figure 3.6.

500 1000 1500 2000 2500 3000 3500 4000

Expansion Pressure (mTorr)

Figure 3.6 FT-IR spectra of cyclohexene adsorbed in BaY as a function of cyclohexene

O2, 325nm, 23.5 h

O2, 325nm, 23.5 h

O2, 455nm, 23.5 h

TBHP is a major industrial oxidant for the transformation of small olefins to

(a) Isobutane in BaY

The product distributions determined by GC and GC-MS analysis are listed in

For cyclohexane, GC product analysis indicates that photooxidation with O2 at

The photooxidation of cyclohexane in BaY with molecular oxygen yields

relative activity of H-abstraction by peroxide radicals from cyclohexene relative to

The main product formed in oxidation of cyclohexene with TBHP is 1,2-epoxy

In the photooxidation of cyclohexane and cyclohexene in BaY using molecular

The above comparison demonstrates that nanocrystalline zeolites are superior

M ass of BaY (mg)

Figure 4.4 Percent conversion in photooxidation of cyclohexane in BaY (solid cycle) as