Escolar Documentos
Profissional Documentos
Cultura Documentos
2005
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
by
Gonghu Li
Copyright by
GONGHU LI
2005
All Rights Reserved
Graduate College
The University of Iowa
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
vi
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
adsorption as a reference. In addition, gas-phase absorptions have been
subtracted from each of the FT-IR spectra. ..............................................................27
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). .............................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
adsorption as a reference. In addition, gas-phase absorptions have been
subtracted from each of the FT-IR spectra. ..............................................................29
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
adsorption as a reference. In addition, gas-phase absorptions have been
subtracted from each of the FT-IR spectra. ..............................................................32
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
xii
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
NaY zeolite prior to adsorption as a reference. In addition, gas-phase
absorptions have been subtracted from each of the FT-IR spectra. ..........................77
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
subtracted from each of the FT-IR spectra. ..............................................................96
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 from each of the FT-IR spectra. ..........................99
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
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).
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).
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
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
IR
Reflecting
mirror
Gases
Pressure
transducer
Premix chamber
Liquids
Turbo/Mechanical pump
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.
A
b
s
o
r
b
a
n
c
e
4000
3500
3000
2500
2000
1500
1000
4000
3500
3000
2500
2000
Wavenumber (cm-1 )
Wavenumber (cm-1 )
1000
4000
3500
3000
2500
1500
1000
IR
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).
15
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
25 mg
50 mg
Dry at 353 K
Dehydrate at 623 K
Zeolite hydrosol
Load reactants at r.t.
GC analysis
Product extracted
> 400 nm
using CH3CN
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.
19
Thermal
couple
Heating element
FID
Zeolite sample
He
Figure 2.8 Schematic of the sample cell and heating portion of the TPD reactor.
20
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)
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.
23
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
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
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.
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
81
11
48
86
14
Conversion (%) was calculated by integrating the absorption at 1451 cm-1 in the FTIR spectrum (estimated error = 3%).
28
Coverage (molecules/supercage)
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
3000
2500
2000
1500
1000
-1
W avenumber (cm )
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
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.
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
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
13
11
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
1377
0.4
A
b
s
o
r
b
a
n
c
e
1693
(b) After irradiation with O2
1475
1368
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
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.
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.
35
3.4 Discussion
3.4.1 Influence of Wavelength on Product Distribution
36
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
100
1.5
60
1
40
Thermal oxidation
0.5
20
Thickness (mm)
Percent Conversion
80
50
100
150
200
250
0
300
42
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
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).
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).
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
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
Ammonia
Pyridine
Substituted pyridine
Alkanes, alkenes
Nitrogen dioxide
Carbon monoxide
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
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
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
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
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.
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
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
57
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
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 )
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
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 )
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
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
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.
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.
4.6 Acknowledgements
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 Experimental Section
5.2.1 Preparation of Nanocrystalline NaY Zeolite
(Particle Size 80 nm)
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
(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).
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.
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
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
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.
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
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
absorptions have been subtracted from each of the FT-IR spectra.
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
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.
77
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
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
2187
(CN)
RCN
1969
(NO)
Adsorbed NO
[121]
1943
(NO)
Adsorbed N2O3
[121]
1723, 1694
(CO)
Carbonyl species
[106, 125]
(ONO)
1598,1483
Carbonaceous species
[131]
1557,1387
(NO2)
RNO2
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
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)
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
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
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
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
(5.1)
(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-
(5.3)
(5.4a)
(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)
(5.8)
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)
(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
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.
5.6 Acknowledgements
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
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.
6.2 Experimental Section
93
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
18
37
11
13
33
48
Commercial NaY
16
31
31
22
52
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
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.
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)
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
(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
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
from each of the FT-IR spectra.
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
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
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.
104
105
106
6.4 Conclusions
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
(7.1)
(7.2)
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.
7.2 Experimental Section
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
from each of the FT-IR spectra.
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
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
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.
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)
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.
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
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.
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
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
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.
CO2
NO
N2O
Nanocrystalline NaY
56
138
12
Commercial NaY
17
74
Commercial NaY b
53
69
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
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.
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].
2NO2 + H2O HNO3 + HONO
(7.3)
(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)
(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
H
H2O
CO2, NH3
AlOx
SiOH
SiOH
AlOx
AlOx
C
SiOH
OCNH
Figure 7.9 Thermal decomposition of urea on silanol (SiOH) and EFAL (AlOx) sites in
nanocrystalline NaY zeolite.
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)
(7.10)
(7.11)
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
nanocrystalline NaY zeolite.
(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
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.
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
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.
7.6 Acknowledgements
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
*****************************************************
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)
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 )
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 )
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 )
1000
148
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
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
151
Table C.1 Quantification of cyclohexane adsorbed in BaY
Blank
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
0.6
3 K/min
15 K/min
30 K/min
60 K/min
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
-12.5
ln(heat rate/Tm2)
-13
-13.5
-14
-14.5
-15
0.0018
0.0019
0.002
0.0021
0.0022
0.0023
0.0024
-1
1/Tm (K )
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.,
Amsterdam. 2001, p 11.
[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.,
Amsterdam. 2001, p 261.
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.