Promotion effects in the oxidation of CO over zeolite-supported Pt nanoparticles

Article abstract of DOI:10.1021/jp044767f

Well-defined Pt-nanoparticles with an average diameter of 1 nm supported on a series of zeolite Y samples containing different monovalent (H⁺, Na⁺, K⁺, Rb⁺, and Cs⁺) and divalent (Mg²⁺, Ca²⁺, Sr²⁺, and Ba²⁺) cations have been used as model systems to investigate the effect of promotor elements in the oxidation of CO in excess oxygen. Time-resolved infrared spectroscopy measurements allowed us to study the temperature-programmed desorption of CO from supported Pt nanoparticles to monitor the electronic changes in the local environment of adsorbed CO. It was found that the red shift of the linear Pt-coordinated Ca??O vibration compared to that of gas-phase CO increases with an increasing cation radius-to-charge ratio. In addition, a systematic shift from linear (L) to bridge (B) bonded Ca??O was observed for decreasing Lewis acidity, as expressed by the Kamlet-Taft parameter ?±. A decreasing ?± results in an increasing electron charge on the framework oxygen atoms and therefore an increasing electron charge on the supported Pt nanoparticles. This observation was confirmed with X-ray absorption spectroscopy, and the intensity of the experimental Pt atomic XAFS correlates with the Lewis acidity of the cation introduced. Furthermore, it was found that the CO coverage increases with increasing electron density on the Pt nanoparticles. This increasing electron density was found to result in an increased CO oxidation activity; i.e., the T_50% for CO oxidation decreases with decreasing a. In other words, basic promotors facilitate the chemisorption of CO on the Pt particles. The most promoted CO oxidation catalyst is a Pt/K-Y sample, which has a T_50%of 390 K and a L:B intensity ratio of 2.7. The obtained results provide guidelines to design improved CO oxidation catalysts. ? 2005 American Chemical Society.

Full text of DOI:10.1021/jp044767f

3822
J. Phys. Chem. B 2005, 109, 3822-3831
Promotion Effects in the Oxidation of CO over Zeolite-Supported Pt Nanoparticles
Tom Visser,† T. Alexander Nijhuis,† Ad M. J. van der Eerden,† Karin Jenken,† Yaying Ji,†
Wim Bras,‡ Sergei Nikitenko,‡ Yasuo Ikeda,§ Muriel Lepage,§ and Bert M. Weckhuysen*,†
Department of Inorganic Chemistry and Catalysis, Debye Institute, Utrecht UniVersity, Sorbonnelaan 16,
3584 CA Utrecht, The Netherlands, DUBBLE CRG, ESRF, BP 220, 38043 Grenoble Cedex 9, France, and
Material Engineering DiVision, Toyota Motor Engineering & Manufacturing Europe Technical Centre,
Hoge Wei 33B, B-1930 ZaVentem, Belgium
ReceiVed: NoVember 16, 2004; In Final Form: December 23, 2004
Well-defined Pt-nanoparticles with an average diameter of 1 nm supported on a series of zeolite Y samples
containing different monovalent (H+, Na+, K+, Rb+, and Cs+) and divalent (Mg2+, Ca2+, Sr2+, and Ba2+)
cations have been used as model systems to investigate the effect of promotor elements in the oxidation of
CO in excess oxygen. Time-resolved infrared spectroscopy measurements allowed us to study the temperature-
programmed desorption of CO from supported Pt nanoparticles to monitor the electronic changes in the local
environment of adsorbed CO. It was found that the red shift of the linear Pt-coordinated CtO vibration
compared to that of gas-phase CO increases with an increasing cation radius-to-charge ratio. In addition, a
systematic shift from linear (L) to bridge (B) bonded CtO was observed for decreasing Lewis acidity, as
expressed by the Kamlet-Taft parameter R. A decreasing R results in an increasing electron charge on the
framework oxygen atoms and therefore an increasing electron charge on the supported Pt nanoparticles. This
observation was confirmed with X-ray absorption spectroscopy, and the intensity of the experimental Pt atomic
XAFS correlates with the Lewis acidity of the cation introduced. Furthermore, it was found that the CO
coverage increases with increasing electron density on the Pt nanoparticles. This increasing electron density
was found to result in an increased CO oxidation activity; i.e., the T50% for CO oxidation decreases with
decreasing R. In other words, basic promotors facilitate the chemisorption of CO on the Pt particles. The
most promoted CO oxidation catalyst is a Pt/K-Y sample, which has a T50% of 390 K and a L:B intensity
ratio of 2.7. The obtained results provide guidelines to design improved CO oxidation catalysts.
Introduction
linear and bridged bonded bands (L:B ratio) can be used as a
measure for the electronic state of the Pt nanoparticles.6 It has
been found that Pt nanoclusters on basic supports are relatively
more electron-rich than on acidic supports, which is reflected
by an increased red shift of the CO stretching frequency and a
decreased L:B ratio.6-8 Another attractive feature of the IR
technique is that it can be used to perform real-time temperature
programmed desorption (TPD) studies to obtain direct informa-
tion on changes in the Pt-C(O) bond strength. Surprisingly,
only a few IR thermal desorption studies of Pt-CO systems
have been reported in the literature, mainly dealing with stepped
single Pt crystals.4,9-14 To our best knowledge, IR-CO-TPD
studies on supported Pt catalysts have only been reported by
Barth et al.4,13 and Gandao et al.14 The former group compared
the CO desorption process from Pt/SiO2 and Pt/Al2O3 upon
heating and cooling, whereas the latter team used the technique
to determine the presence of three types of Pt bonded CO on
Pt/Mg(Al)O catalysts.
Understanding the effect of a support oxide on the properties
of noble metal particles is a challenging subject because it opens
a way toward modeling and tuning of the catalytic properties
by a deliberate choice of the support. The catalytic activity of
supported Pt nanoparticles is well-known to be sensitive to the
support composition, acid-base properties, and pore curvature,1-3
but despite many studies that have been undertaken to elucidate
the exact nature of the relationship, many questions still remain
open.
A very useful, but often underestimated, technique is infrared
spectroscopy (IR) because the IR spectrum of chemisorbed
carbon monoxide (CO) can be regarded as a sensitive and local
probe of the electronic properties of supported Pt nanopar-
ticles.4,5 Its application is based on the fact that differences in
electronic state are reflected in changes in the vibrational
characteristics of the Pt-CO adduct. The result is a red shift of
the CO stretching frequency compared to that of the gas phase
and the appearance of two principal IR bands: one in the region
2100-1900 cm-1, which is assigned to CO that is coordinated
linear to Pt, and a bridged bonded one between 1900 and 1700
cm-1, which is attributed to 2-, 3-, or even 4-fold-coordinated
CO. Next to the stretching frequency, the intensity ratio of the
It is clear that systematic studies on well-defined Pt nano-
particles supported on a wide range of porous oxides are lacking
in the literature, preventing the generalization of some of the
above-mentioned conclusions. As a consequence, the aim of
the present research is to investigate the effect of monovalent
(H+, Na+, K+, Rb+, and Cs+) and divalent cations (Mg2+, Ca2+
,
Sr2+, and Ba2+) on the electronic properties of Pt particles
encaged in the supercages of zeolite Y by means of the IR-
CO-TPD technique. The results will be discussed in relation
to the corresponding CO oxidation activity and spectroscopic
* To whom correspondence should be addressed. E-mail:
b.m.weckhuysen@chem.uu.nl.
† Utrecht University.
‡ DUBBLE CRG.
§ Toyota Motor Engineering & Manufacturing Europe Technical Centre.
10.1021/jp044767f CCC: $30.25 © 2005 American Chemical Society
Published on Web 02/09/2005

CO Oxidation over Zeolite-Supported Pt Nanoparticles
J. Phys. Chem. B, Vol. 109, No. 9, 2005 3823
TABLE 1: Overview of the Supported Pt Catalysts under Study, Together with Some Physicochemical Properties As
Determined with N2/H2 Chemisorption, X-ray Fluorescence, High-Resolution Transmission Electron Microscopy, and NH3
Temperature Programmed Desorption
dispersion (%)
as determined with
H2 chemisorption
H
+ (mmol/g)
as determined with
NH3 TPD
Pt particle size (nm) and their relative
abundance as determined with HRTEM
surface area
(m2/g)
pore vol
(mL/g)
sample name
wt % Pt
Pt/H-Y
Pt/Na-Y
Pt/K-Y
Pt/Rb-Y
Pt/Cs-Y
Pt/Mg-Y
Pt/Ca-Y
Pt/Sr-Y
Pt/Ba-Y
1.2
1.0
1.1
1.0
0.8
1.3
1.2
0.9
1.0
e1 (exclusive)
e1 (exclusive)
700
686
690
678
542
679
687
690
696
0.28
0.27
0.27
0.27
0.23
0.27
0.27
0.27
0.27
93
95
82
88
69
82
83
82
92
1.97
0.16
0.15
0.16
0.18
0.22
0.17
0.15
0.18
e1 (main fraction); 3 nm (traces)
e1 (main fraction); 3 nm (traces)
e1 (fraction); 5 nm (significant fraction)
e1 (main fraction); 10 nm (traces)
e1 (main fraction); 5 nm (traces)
e1 (main fraction); 5 nm (traces)
e1 (exclusive)
Pt atomic XAFS (AXAFS) data of these materials. It will be
shown that an increasing electron charge on the supported Pt
nanoparticles indirectly induced by alkaline and earth alkaline
metal ions via the framework oxygen atoms promotes the CO
oxidation activity due to an increased CO surface coverage.
Furthermore, the systems under investigation can be envisaged
as model systems for the NOx storage-reduction catalysts (e.g.,
Pt-BaO/Al2O3) currently used in cars. The results of the present
study show that BaO plays most likely a dual role because it
acts as a NOx trap but also promotes the CO oxidation activity
of the supported Pt nanoparticles. The work therefore provides
guidelines for the development of improved three-way exhaust
catalytic converters.
drying at 353 K in N2 (quality 4.0; Hoekloos) for 12 h,
calcination was carried out by drying in a high airflow (30 mL/
min) for 12 h followed by increasing the temperature to 573 K
at a heating rate of 0.2 K/min to achieve complete removal of
NH3 prior to reduction. Reduction was performed in pure H2
(quality 5.0; Hoekloos) with a flow of 10 mL/min at 573 K for
2 h. After reduction and flushing with N2 at room temperature,
passivation was carried out by admitting a small amount of air
(10 mL/min) into the system to prevent aggregation of Pt
particles. Table 1 summarizes the different supported Pt catalysts
under study in this work.
2. Catalyst Characterization. X-ray fluorescence (XRF) was
carried out on all samples to determine the Pt loadings using a
Spectro X-lab 2000 instrument. N2 physisorption was performed
at 77 K with a Micromeretics ASAP 2400 apparatus (quality
4.0; Hoekloos). Prior to these measurements, the samples were
degassed for 24 h at 573 K in vacuum. Surface area, pore
volume and pore size distribution were calculated with standard
BET theory. Hydrogen chemisorption measurements were
performed in a conventional static volume apparatus (Micromer-
etics ASAP 2010C). The samples were first dried under
evacuation at 373 K overnight, then reduced in pure H2 at 573
K for 1 h (ramp: 5 K/min). The samples were cooled in H2
down to 523 K and evacuated at this temperature for 1 h.
Subsequently, the samples were cooled in a vacuum to room
temperature and H2 adsorption isotherms were taken at 309 K.
At room temperature, the samples were evacuated and a second
isotherm was recorded. The hydrogen chemisorption capacity
was calculated at the difference between the two isotherms
extrapolated to zero pressure. The Pt dispersions were calculated
on the basis of the assumption that one hydrogen atom is
adsorbed per platinum surface atom. The number of support
protons was determined by temperature programmed desorption
(NH3 TPD) for catalysts containing only chemisorbed NH3.
Approximately 1 g of catalyst was saturated in a flow of 5%
NH3/N2 at room temperature. To remove physisorbed NH3, the
catalyst was washed three times in 50 mL of H2O at 353 K,
filtered, and dried at 373 K. The quantity of desorbed NH3 was
determined by titration with a standard solution of 0.1 M HCl.
The number of protons was calculated on the basis of the
assumption that one NH3 molecule was chemisorbed per acid
site. High-resolution transmission electron microscopy (HR-
TEM) was done with a Philips CM 30 UT electron microscope
equipped with a field emission gun as the source of Elecron
operated at 300 kV. Samples were mounted on a microgrid
carbon polymer supported on a copper grid by placing a few
droplets of a suspension of ground sample in ethanol on the
grid, followed by drying at ambient conditions. Extended X-ray
absorption fine structure (EXAFS) measurements were carried
out at Hasylab (Germany) at station ×1.1 and at the ESRF
Experimental Section
Prior to preparation and characterization, it should be
emphasized that the CO stretching vibration of Pt-chemisorbed
CO is not only sensitive to the support characteristics but also
to a large number of experimental parameters. According to
literature, the loading, dispersion, particle size, and coordination
number of Pt,9,15-19 the reduction temperature,20,21 and CO
exposure pressure22,23 can affect the observed IR data, whereas
sample pretreatment,24 CO exposure temperature,12 CO adsorp-
tion,25 and CO interaction with support cations7,26 may play a
role as well. To prevent erroneous assignments and conclusions,
special attention has been paid to keep all experimental
parameters and instrumental settings the same.
1. Catalyst Preparation. H-Y (Si:Al ratio of 2.71) has been
obtained from Linde, whereas the starting material for ion
exchange was a Na-Y (Si:Al ratio of 2.49) material from
Ventron. The zeolite materials were put in their Na+, K+, Rb+,
Cs+, Mg2+, Ca2+, Sr2+, and Ba2+ form by four successive ion
exchanges for 12 h with an aqueous 1 M solution of NaCl
(Aldrich, p.a.), KCl (Aldrich, p.a.), RbCl (Merck, p.a.), CsCl
(Merck, p.a.), MgCl2 (UCB, p.a.), CaCl2 (UCB, p.a.), SrCl2
(Aldrich, p.a.), and BaCl2 (Aldrich, p.a.), respectively. The
samples were washed chlorine-free and dried in air at room-
temperature overnight. The crystallinity before and after ion
exchange was confirmed by XRD and SEM techniques. Chemi-
cal analysis of the zeolite materials indicated that after four
successive ion exchange steps the degree of exchange was above
98% for all materials under study. Details of the zeolite materials
can also be found elsewhere.27 As will be shown, the zeolite
materials, with the exception of H-Y, are almost free of
Bro¨nsted acid sites, ensuring that the ion exchange of the
promoter cations does not introduce H+-sites in the catalyst
materials.
Supported Pt particles (1 wt %) were prepared via a dry
impregnation step of the support materials with the appropriate
aqueous solutions of Pt(NH3)4(NO3)2. After impregnation and

3824 J. Phys. Chem. B, Vol. 109, No. 9, 2005
Visser et al.
TABLE 2: EXAFS Fit Parameters of the Supported Pt
(France) at CRG DUBBLE (BM26A). Both beamlines were
equipped with a double crystal monochromator with Si111
crystals. Measurements were performed in transmission, and
ion chambers were used for detection with a gas fill to absorb
20% in the first and 80% in the second ion chamber. Reduction
of the higher harmonic radiation at Hasylab was established by
detuning the second crystal to 50% of the maximum intensity,
and at the DUBBLE station by using a vertically focusing mirror
installed behind the monochromator. The samples were prepared
by pressing 0.120 to 0.140 g powder to a self-supporting wafer.
Next, the wafer was placed in a treatment cell. Prior to
measurement, the sample was dried at 423 K for 30 min in a
flow of dry He and then slowly heated to 573 K under a flow
of 10% H2 in He to reduce the passivated Pt. Next, the samples
were measured at liquid N2 temperature. EXAFS data analysis
was carried out with the XDAP program.28
Catalysts under Study, Together with the Corresponding
AXAFS Intensities
∆σ2
(Å2) (Å)
R
E0
(eV)
AXAFS intensity
sample
scatterer
N
(×10-2
)
Pt/H-Y
Pt
O
Pt
O
Pt
O
Pt
O
Pt
O
Pt
O
Pt
O
Pt
O
Pt
O
5.4 3.4 2.74
0.2 3.0 2.10 -12.3
5.7 4.6 2.73 2.5
0.6 3.8 2.07 -15.0
6.2 4.4 2.74 2.2
0.2 3.4 2.07 -13.0
7.1 6.1 2.72 1.9
0.4 5.6 2.00 -12.0
6.6 3.8 2.74 2.2
0.5 1.4 2.05 -12.0
5.8 6.1 2.73 2.0
0.4 1.0 2.19 -10.3
6.7 5.1 2.74 2.5
0.3 5.3 2.03 -13.0
5.1 3.1 2.74 1.9
0.5 1.3 2.10 -12.0
5.1 4.7 2.73 2.7
0.4 2.1 2.09 -13.0
1.9
2.20
Pt/Na-Y
Pt/K-Y
1.06
0.94
1.37
2.26
1.95
1.25
1.30
1.16
Pt/Rb-Y
Pt/Cs-Y
Pt/Mg-Y
Pt/Ca-Y
Pt/Sr-Y
Pt/Ba-Y
IR measurements were performed on self-supported catalyst
wafers that were pressed from 0.01 to 0.020 g of calcined and
reduced sample material. A pressure of 3 bar was applied during
10 s to prevent destruction of the pore structure of the support.24
The wafer was placed in an IR transmission cell equipped with
CaF2 windows as previously described.29 The cell was evacuated
to 10-8 bar followed by drying of the wafer at 393 K overnight
prior to reduction by a 10 mL/min flow of H2 (quality 4.6;
Hoekloos). During reduction the temperature was raised from
393 K with 3 K/min to 573 K where it was maintained for 1 h.
Next the system was switched to vacuum and after 1 h at 573
K, cooled to 323 K at a rate of 3 K/min where it was maintained
for 1h. We believe on the basis of the absence of a Pt-H
stretching vibration at around 2050 cm-1 in the Pt-loaded
samples that this vacuum treatment removed all the hydrogen
from the supported Pt nanoparticles. Next the system was
switched to 10% CO (quality 4.7; Hoekloos) in He (quality 4.6;
Hoekloos) at a starting pressure of 0.2 bar. After 30 min of
static pressure, the system was reevacuated and after another
30 min, temperature programmed desorption (TPD) was started
by increasing the temperature from 323 to 573 K (3 K/min)
where it was maintained for 30 min. IR spectra were recorded
on a Perkin-Elmer 2000 FTIR instrument with a data point
resolution of 4 cm-1. For each spectrum 25 scans were co-
added. As a background, the spectrum of the catalyst wafer was
taken after drying, reduction and cooling to 323 K, 2 min prior
to CO exposure. During adsorption and TPD, time-resolved
scanning was carried out by automatically acquiring spectral
data every 2 min, using Perkin-Elmer Time-Base software. To
obtain spectra of the adsorbed CO during CO exposure, the gas-
phase spectrum of CO was subtracted from the spectrum that
was taken prior to reevacuation of the cell. Intensity data were
acquired by calculating the integrated area of baseline corrected
spectra using the Perkin-Elmer Spectrum software. The repro-
ducibility of all IR-TPD measurements was verified by carrying
out the experiments in duplo in two ways: (1) by analyzing
two different wafers from the same sample and (2) by repeating
the exposure and desorption process to the same wafer without
opening the IR cell.
Results and Discussion
1. Catalyst Characterization. Table 1 summarizes some
physicochemical properties of the materials under study as
obtained with N2 and H2 physisorption, NH3 TPD, and XRF.
The surface area and pore volume of the Pt catalysts are typical
for zeolite Y. Neither the surface area, nor the pore volume of
the zeolite support is significantly affected by the introduction
of Pt nanoparticles in the micropores of zeolite Y. An exception
is the Pt/Cs-Y sample, which shows a decreased surface area
and pore volume. A possible explanation could be pore blocking
by Pt particles at the outer surface of the zeolite crystals or
zeolite destruction due to extensive Pt particle growth.31 The
Pt content of the different catalysts as determined by XRF varied
between 0.7 (Pt/Cs-Y) and 1.3 (Pt/Mg-Y) wt % Pt, but most
of the Pt loadings are close to the target value of 1 wt %. The
H2 chemisorption results indicate that the Pt nanoparticles are
well-dispersed, although Pt/Cs-Y shows the lowest Pt disper-
sion, an observation in line with the N2 chemisorption data.
Finally, NH3 TPD was used to estimate the number of acid sites
in the Pt-loaded zeolite materials. It was found that with the
exception of Pt/H-Y the catalyst materials do not contain
substantial amounts of protons. More importantly, we do not
see any systematic trend as a function of the introduced promoter
cations.
The size and size distribution of the supported Pt nanoparticles
have been determined with EXAFS and HRTEM. Whereas
EXAFS determines an average number for the Pt size (via the
number of Pt scatterers), HRTEM gives insight in the size
distribution of the supported Pt nanoparticles.
Table 2 summarizes the results of the EXAFS analysis on
all the samples under study. The data show that for most samples
a Pt-Pt coordination number of around 5.5-6.0 was found,
which corresponds to a spherical Pt particle size of around 1
nm, most probably located in the supercage of zeolite Y. The
three exceptions to this observation are the Pt-Pt coordination
number of 6.7 for the Pt/Cs-Y and Pt/Ca-Y samples and the
Pt-Pt coordination number of 7.1 for the Pt/Rb-Y sample. We
did not observe scattering against the introduced co-cations,
indicating that there is no direct chemical contact between the
Pt nanoparticles and the promoter ions. However, because some
of the cations introduced (e.g., Ba and Sr) can be regarded as
high Z elements, such a Pt absorber-Z backscatter pair still
may exist, although we did not try to fit the higher shell
3. Catalytic Testing. The Pt-catalysts were tested for the
oxidation of CO in a 6-flow reactor setup, descibed in detail
elsewhere.30 For this purpose, 20 mg of each catalyst diluted
with 200 mg of SiC was loaded in the reactors. The reacting
gas consists of 1000 ppm of CO in an excess of 10% of O2.
The flow rate was 250 mL/min and the heating and cooling
rates were 0.3 K/min. The catalysts were tested reversibly and
CO and CO2 gas analysis was done with a nondispersive IR
analyzer.

CO Oxidation over Zeolite-Supported Pt Nanoparticles
J. Phys. Chem. B, Vol. 109, No. 9, 2005 3825
Figure 1. HRTEM micrographs of Pt/Na-Y (a) and Pt/Cs-Y (b) catalyst samples.
Figure 2. Time-resolved IR spectra during temperature programmed CO desorption from Pt/H-Y: (a) raw spectra and (b) baseline corrected
spectra.
contributions of the EXAFS data. In this respect, it is worthwhile
to mention the work of Vaarkamp et al. in which 1.4 Ba atoms
were found at a distance of 3.76 Å from a Pt particle with a
coordination number of 3.7 in a Pt/BaKL zeolite material.32 The
small amount of oxygen in the EXAFS fits (Table 2) is a support
oxygen originating from the zeolite material. A formal Pt-O
bond does not exist because the small Pt clusters are reduced
in hydrogen. Therefore, an oxygen atom from the support
material is the observed scatterer. This results in an increased
Pt-O bond distance in comparison with platinum oxide. The
spread in the Pt-O distances is due to the changes induced by
the presence of cations as well reflects the affinity of the Pt
particles for the support. Finally, the relative high E0 or inner
potential values have to be addressed for this scatterer. There
is an ongoing discussion on the accepted variance in E0 values.
The standard, used in the current calculations, is the Pt-O bond
in Na2Pt(OH)6. For the type of bond this is the best possible
standard, but not fully covering the identity of the zeolite oxygen
scatterer, which results in a more then normally accepted
variance in E0 value.
Figure 1 shows as examples the HRTEM pictures of Pt/Na-Y
and Pt/Cs-Y. Whereas in the case of Pt/Na-Y no large Pt
nanoparticles can be observed, HRTEM of Pt/Cs-Y shows a
significant portion of Pt nanoparticles of 1-5 nm on top of the
zeolite Y crystals. This observation explains, most probably,
the decreased surface area and pore volume of this material as
well as the higher Pt-Pt coordination number. Another explana-
tion could be that the Pt particles further grow in the Pt/Cs-Y
sample exceeding the supercage size, leading to mesopore
formation. Evidence for such effects has been reported for the
formation of CDS and ZnS in Na-X, Ir clusters in Na-X and
Pt clusters in ZSM-5 zeolites.31 Table 1 summarizes the
observations made by HRTEM. It can be concluded that all
samples, with the exception of the Pt/Cs-Y matial, contain
exclusively or as a main fraction Pt nanoparticles of 1 nm or
smaller. As expected, the Pt/Ca-Y and Pt/Rb-Y samples
contain Pt particles of 3-5 nm as well. Thus, although some
differences are noticed between the different samples, EXAFS,
H2 chemisorption and HRTEM indicate that well-defined Pt-
loaded zeolite Y samples have been prepared. These samples
contain predominantly 1 nm Pt nanoparticles, which are most
probably occluded in the supercages of zeolite Y.
2. Infrared Spectroscopy. Figures 2-4 show the time- and
temperature-resolved IR spectra of the different supported Pt
catalysts obtained during the CO TPD procedure. It is important
to recall that the results of two subsequent measurements of
the same self-supported wafer were practically identical, which
implies that the samples were not affected by the treatment and
that reproducible results were obtained. We also verified the
stability of the Pt-zeolite samples during the CO TPD measure-
ments and have noticed with HR-TEM no disintegration or
aggregation of the supported Pt nanoparticles. In other words,
the applied methodology did not affect the Pt size and size
distribution of the Pt nanoparticles in the different zeolite
samples.
Figure 2 shows the CO IR TPD spectra of Pt/H-Y. The 3D-
set of spectra in Figure 2a are the raw IR data. It is clear that
the spectra show a strong IR absorption band at around 2000
cm-1 and a weaker one at around 1800 cm-1. The former band
is assigned to the stretching vibration of a linearly Pt-coordinated
CO, whereas the latter band is due to bridged Pt-CO stretching
vibrations.20,33-35 Both IR absorption bands decrease in intensity
with increasing TPD temperature due to CO desorption. In
addition, a background absorption is arising, which complicates
the analysis at higher TPD temperatures. One can apply a
baseline correction leading to the spectra shown in Figure 2b.
On the basis of Figures 2-4 one can make the following
qualitative observations:

3826 J. Phys. Chem. B, Vol. 109, No. 9, 2005
Visser et al.
Figure 3. Time-resolved IR spectra during temperature programmed CO desorption from (a) Pt/Na-Y, (b) Pt/K-Y, (c) Pt/Rb-Y, and (d) Pt/
Cs-Y. No baseline correction has been applied.
Figure 4. Time-resolved IR spectra during temperature programmed CO desorption from (a) Pt/Mg-Y, (b) Pt/Ca-Y, (c) Pt/Sr-Y, and (d) Pt/
Ba-Y. No baseline correction has been applied.
(1) The linear CO absorption bands of Pt/H-Y (Figure 2)
are sharp and symmetrical, whereas alkaline metal (Figure 3)
and earth alkaline metal (Figure 4) cation-containing samples
possess broad asymmetrical bands with shoulders. Indeed, the
bandwidth is small (half bandwidth (HBW) of 49 cm-1) for
Pt/H-Y compared to those of the alkaline metal cation-
exchanged (HBW of 90-110 cm-1) and the earth alkaline metal
cation-exchanged (HBW of 55-70 cm-1) zeolite Y samples.
In the case of Pt/H-Y, the HBW points to the presence of a
single type of linear-coordinated CO and hence, to a uniform
coordination number for the supported Pt particles.33 On the
other hand, the large HBWs for the cation-containing zeolite Y
samples implies that either different Pt adsorption sites are
present or that different CO orientations are possible as a result
of local interactions within the zeolite supercage. The former
is rather unlikely, because the Pt particle sizes are practically
the same for all the samples, with the exception of Pt/Cs-Y
(Table 1 and Figure 1). The most plausible explanation is to
assume the presence of nonbonding electrostatic interactions.
Two types of nonbonding interaction of the CO-oxygen can
affect the CO vibration and result in different CO orientations:
(a) a repulsive electrostatic interaction with oxygen atoms of
the framework36,37 and (2) an attractive ion-dipole interaction
with framework cations.7 Compared to the Pt/H-Y sample, the
presence of cations seems crucial in the band broadening effect
and so it is most likely that the attractive ion-dipole interactions
play the largest role. Next, the presence of cation-affected and
“free” CO positions explains to some extent the complex linear
CO band in the IR. In addition, reorientation of CO as a result
of changing interactions might explain the increasing band
complexity at higher temperature. This phenomenon is particu-
larly visible in the IR-CO-TPD pattern of Pt/Na-Y by
the increasing intensity of the shoulder at 2000 cm-1 upon
heating.

CO Oxidation over Zeolite-Supported Pt Nanoparticles
J. Phys. Chem. B, Vol. 109, No. 9, 2005 3827
TABLE 3: Positions (cm-1) of the Absorption Bands of
Linearly and Bridged Pt-Coordinated CO for the Supported
Pt Catalysts Measured at Different Temperatures
linear Pt-coordinated
CtO stretching band
bridged Pt-coordinated
dCdO stretching band
sample 323 K 373 K 473 K 573 K 323 K 373 K 473 K 573 K
Pt/H-Y 2083 2078 2072 2066 1856 1854 1751 1750
Pt/Na-Y 2073 2036 1995 1957 1823 1806 1775 1720
Pt/K-Y 2045 2022 1996 1964 1821 1797 1736 1723
Pt/Rb-Y 1992 1984 1971 1963 1803 1792 1725 1718
Pt/Cs-Y 2001 1959 1937 1918 1813 1753 1720 1719
Pt/Mg-Y 2087 2075 2066 2054 1892
1837 1855 1855
Figure 5. Red shift of linearly bonded CtO on supported Pt
nanoparticles compared to gas-phase CtO as a function of the promotor
cation radius-to-charge ratio. No baseline correction has been applied.
1808 1786 1784
Pt/Ca-Y 2093 2083 2069 2047 1887
1855 1855
1798 1786
TABLE 4: Overview of the Infrared L:B Ratios and
Catalytic Activity Data of the Supported Pt Catalysts under
Studya
Pt/Sr-Y
2076 2070 2059 2041 1872 1868 1868 1897
1844 1845 1845
1802 1773
Pt/Ba-Y 2076 2049 2026 2021 1845 1843 1834 1813
infrared L:B ratio
at 323 K
T
50% (K) for
Eact (kJ/mol) for
CO conversion
1788 1785
sample
CO conversion
1760 1754 1728 1723
Pt/H-Y
Pt/Na-Y
Pt/K-Y
Pt/Rb-Y
Pt/Cs-Y
Pt/Mg-Y
Pt/Ca-Y
Pt/Sr-Y
Pt/Ba-Y
10.0
2.9
2.7
2.8
2.8
8.0
4.5
4.0
3.5
460
410
390
405
355
450
420
395
420
92
92
93
97
98
94
94
85
79
(2) The differences and shifts of the band maxima for both
linear and bridged Pt-coordinated CO for the different samples
upon TPD are summarized in Table 3. Comparison of the linear
Pt-coordinated CO stretching band position under experimental
conditions demonstrates that the maximum decreases upon
substitution of the framework protons by alkali or earth alkali
metal ions more or less in the same order as their position in
the periodic system, i.e., H > Na > K > Rb ≈ Cs and Mg ≈
Ca > Sr > Ba. Similar trends can be observed for the bridged
Pt-coordinated CO stretching band. Besides, the spread of about
100 cm-1 for alkali metal cation-containing samples and of
about 20 cm-1 for alkaline earth metal cation-containing samples
indicates a large difference in the CO bond strength. These
results point to a correlation of the vCtO with the electronic
properties of M+/M2+. Assuming an inversely proportional
relationship of vCtO and the PtsC(O) bond strength, one would
expect an order in the CO-desorption rate of Pt/CssY >
Pt/RbsY > Pt/KsY > Pt/NasY > Pt/HsY. A closer
examination of the IR-desorption patterns in 3D-projection
(Figures 3 and 4) does not show such trend, indicating that there
is no simple correlation between the vCtO with the PtsC(O)
bond strength. Apparently, a high CO stretching frequency is
not by definition related to a weak Pt-C(O) bonding an vice
versa. This is in line with a recent conclusion of Wasileski et
al. that there is no simple correlation between the field dependent
Pt bonded CO vibrational frequency and the PtsC(O) bonding
energy.38
(3) Both alkaline and earth alkaline metal ion-containing
samples possess a complex bridged Pt-coordinated CO stretching
band pattern and at least three broad and ill-resolved absorption
bands are observed (Table 3). This is the most evident for
materials loaded with the divalent cations. As can be seen from
Figure 4, the intensity decrease of the bridged bonded CO band
at around 1800 cm-1 is partially offset by an increase of the
baseline and in principle the latter might also be due to the
formation of 3- or 4-fold-coordinated CO. However, upon
cooling, the baseline decreases to its initial values whereas the
linear and 2-fold-coordinated bands do not increase in intensity.
For that reason, we reject the formation of the latter type of Pt
bonded CO. It implies that the intensity of the bridged band
decreases faster than that of the linear PtsCtO band at around
2000 cm-1. Comparison among Figures 2-4 also indicate that
the IR band intensities of linear and bridged Pt-coordinated CO
decrease with increasing measurement temperature and the
a The L:B ratios have been obtained from the IR spectra after baseline
correction.
relative extent of the band intensity decrease is influenced by
the presence of alkaline and earth alkaline metal cations.
Furthermore, the ratio of the IR band intensities of linear to
bridged Pt-coordinated CO decreases in the order: Pt/HsY >
Pt/M2+sY > Pt/M+sY.
In what follows, we will elaborate on these differences and
present a more quantitative comparison between the materials
under investigation. For this purpose, the summed integrated
intensities of linear and bridged Pt-coordinated CtO stretching
bands have been used after normalizing for the Pt contents of
the self-supported wafers. Figure 5 relates the red shift of linearly
Pt-coordinated CtO compared to gas-phase CtO (positioned
at 2143 cm-1) with the cation radius-to-charge ratio of the
different cations under study. It is evident that there exists a
correlation and a high red shift corresponds to a high cation
radius-to-charge ratio. Similar trends are observed for the
bridged Pt-coordinated CtO stretching band, although due to
the complexity and broadness of these bands less straightforward
conclusions can be drawn.
Next, the effect of the different co-cations on the intensity
ratio of the linear (L) to bridge (B) coordinated bands was
examined, and the results are summarized in Table 4. These
values were obtained from the IR TPD spectra obtained at 323
K after baseline correction. Comparison of the IR L:B intensity
ratios of the spectra recorded at 323 K reveal much lower L:B
values of 2.7-2.9 for the alkali ion-exchanged Pt-zeolites
compared to 10 for the Pt/HsY sample. Earth alkaline ion-
exchanged Pt-zeolites have intermediate L:B values of 3.5-
8.0, but importantly, the values follow the trend expected from
the ordering in the periodic table. It should be noted that the
inaccuracy of the L:B values of the Pt/M+sY and Pt/M2+/Y
catalysts is larger than for Pt/HsY due to the asymmetry of
the bands, but the trends remain clear. In addition, we could
not find a simple relationship between the red shift for linear

3828 J. Phys. Chem. B, Vol. 109, No. 9, 2005
Visser et al.
Figure 6. Relationship between IR L:B intensity ratios and Lewis acid
properties of the cations introduced in zeolite Y as expressed by the
Kamlet-Taft parameter R. The Pt/H-Y sample could not be introduced
in this plot because no R-values for H+ are available.
Figure 7. CO coverage on supported Pt nanoparticles as a function of
the desorption temperature: (a) Pt/K-Y, (b) Pt/Ca-Y, (c) Pt/Mg-Y,
and (d) Pt/H-Y.
bonded CtO to Pt and the IR L:B intensity ratio. Instead, we
found a relationship between the L:B intensity ratios and the
Lewis acid properties of the cations introduced in zeolite Y.
The results are summarized in Figure 6. A parameter, describing
the Lewis acid behavior of the cations under investigation, is
the Kamlet-Taft parameter R.39 This parameter expresses the
ability of a cation to accept an electron pair and a high R value
corresponds with a strong Lewis acid character. It is clear that
a weak Lewis acid, such as K+ (R ) 0.85), results in an
increasing electron charge on the framework oxygen atoms and
therefore an increasing electron charge on the supported Pt
nanoparticles. This increase results in an increasing occupancy
of the d-levels, resulting in an increase in back-donation in the
molecular orbital with 2π* character and therefore an increasing
red shift of the linear CtO band and a decreasing infrared L:B
ratio. The reverse situation can be rationalized for a strong Lewis
acid site, such as Mg2+ (R ) 4.66).
Finally, one can use the integrated band intensities of the
linearly and bridged Pt-coordinated CO molecules to determine
the surface coverage of these species as a function of the
desorption temperature in the different supported Pt nano-
particles. The obvious assumptions are that (1) the extinction
coefficients of linearly and bridged Pt-coordinated CtO
molecules do not change with increasing desorption temperature,
(2) the CO surface coverage does not effect too much the CO
band intensities, and (3) the extinction coefficients are not
affected to a great extent by the electron properties of the
supported Pt nanoparticle. The second assumption is only valid
if Pt nanoparticles of similar size and shape are under study,
which is the case on the basis of our EXAFS and HRTEM data,
with the exception of Pt/CssY. With respect to the third
assumption, it should be mentioned that extinction coefficients
of a carbonyl function (a CdO unit comparable to CtO) in
aldehydes and ketones are not dramatically altered when the
surrounding R groups are changed. Examples are the molecules
CH3C(dO)CH3 (ꢀ ) 8.5 l/mol‚cm2) and ClCH2C(dO)C6H5 (ꢀ
) 7.9 l/mol‚cm2).40 In other words, changes in the electron
properties of the supported Pt particle due to the presence of a
different promoter element are only expected to slightly
influence the extinction coefficient of an adsorbed CO molecule.
In a future study we will attempt to determine the extinction
coefficients of adsorbed CO on supported Pt nanoparticles.
Figure 7 illustrates the CO surface coverage for Pt/KsY, Pt/
CasY, Pt/MgsY, and Pt/HsY zeolites as a function of the
desorption temperature. The values have been obtained by taking
into account the weight and Pt content of the IR wafers for
each sample as well as by assuming that the extinction
coefficient of linearly Pt-coordinated CtO molecules is twice
as large as that of the bridged Pt-coordinated CtO molecules.
Figure 8. CO coverage on supported Pt nanoparticles measured at
450 K as a function of the IR L:B intensity ratio. The Pt/Ba-Y sample
behaves atypically and has a very high CO coverage.
The latter assumption has been obtained from the work of
Vannice and Twu.41 Figure 7 shows that in each zeolite sample
the CO coverage on the supported Pt particles decreases with
increasing desorption temperature. In addition, the relative
amount of adsorbed CO is the highest on Pt/KsY and the lowest
on Pt/HsY, whereas the two other Pt-loaded zeolites have
intermediate CO coverages. To elaborate further on this point,
we have plotted the CO coverage values at 450 K as a function
of the corresponding IR L:B ratios of the different samples.
The result is illustrated in Figure 8. It seems that the CO
coverages on supported Pt nanoparticles increase with decreasing
IR L:B ratio. In addition, a decreasing CO coverage leads to an
increasing effect of the electronic charge of the Pt particles on
the red shift of the CtO stretching band for each adsorbed CO
molecule. In other words, the νCO will shift due to an increased
2π* back-bonding to lower energy for decreasing CO coverage,
which is experimentally observed, as shown in Table 3.
3. Pt AXAFS Measurements. Recently, Koningsberger et
al. developed a new tool for obtaining structural information
from XAFS data of metal-supported catalysts, such as Pt-loaded
zeolites.42,43 A previously nonutilized feature called AXAFS has
been shown to be highly sensitive to the composition of the
support oxide. The change in intensity of the Fourier transformed
AXAFS peak can be directly related to a change in the
interatomic potential of the Pt atoms averaged over the whole
Pt metal particle. The change in interatomic potential is caused
by an increase or decrease in electron charge (more or less
electron-rich) of the support oxygen atoms. The electron charge
of the support oxygen atoms in turn is determined by the
Madelung potential of the support oxide, which depends on the
composition of the support material. Thus, a high intensity of
the AXAFS contribution corresponds to an electron-poor Pt
particle and vice versa.44,45

CO Oxidation over Zeolite-Supported Pt Nanoparticles
J. Phys. Chem. B, Vol. 109, No. 9, 2005 3829
Figure 9. Fourier transform (k1, ∆k ) 2.5-8 Å-1) of atomic XAFS of (a) Pt/K-Y, (b) Pt/Ca-Y, (c) Pt/Mg-Y, and (d) Pt/H-Y.
Figure 10. AXAFS peak intensity of supported Pt nanoparticles as a
Figure 11. CO oxidation activity over supported Pt nanoparticles
function of the IR L:B intensity ratio.
expressed by T50% as a function of the IR L:B intensity ratio.
The AXAFS contribution can be isolated by subtracting the
calculated Pt-Pt and Pt-O contributions from the raw EXAFS
data.42 Details on the procedure for obtaining AXAFS spectra
have been published elsewhere.43 Figure 9 shows the AXAFS
contributions in the R-range from 0 to 1.5 Å for Pt/K-Y, Pt/
Ca-Y, Pt/Mg-Y, and Pt/H-Y zeolites. One can notice from
Figure 9 that the AXAFS data of the different Pt-zeolites differ
in their peak intensity, as well as in the peak centroid. A decrease
in the AXAFS intensity results in a shift of the peak centroid
to higher R values and vice versa. The AXAFS intensities
obtained from the XAFS data of the different Pt-zeolites are
included in Table 2.
whereas the Pt/K-Y sample has the lowest AXAFS peak
intensity and IR L:B value. It is fair to mention that because
Pt/Cs-Y contains larger Pt particles, we have not included this
data point in Figure 10.
4. CO Oxidation Activity. The supported Pt catalysts under
investigation show clearly different catalytic behavior in the
oxidation of CO in the presence of excess oxygen. The
temperatures at which 50% of the CO is converted (T50%) for
the different materials are summarized in Table 4. It is evident
that depending on the cation present in the parent zeolite Y the
T50% can differ with more than 100 K. The catalytic results can
Figure 10 plots the AXAFS peak intensities of the Pt-loaded
zeolites versus their corresponding IR L:B ratios and it is evident
that the AXAFS intensity decreases with decreasing IR L:B
ratio. In other words, both the AXAFS peak intensity and IR
L:B ratio are measures for the electron charge of the Pt
nanoparticles and as a consequence of the electron charge of
the support oxygen atoms. Thus, the AXAFS peak intensity and
the IR L:B ratio decreases with increasing electron richness of
the Pt nanoparticles. The Pt/H-Y sample has the highest
AXAFS peak intensity as well as the highest IR L:B value,
also be plotted as a function of the corresponding IR L:B
intensity ratios of the materials, which expresses the electron
charge on the supported Pt nanoparticles. The obtained cor-
relation is shown in Figure 11. It is evident that electron-rich
supported Pt nanoparticles promote CO oxidation activity,
whereas electron-poor supported Pt nanoparticles oxidize CO
at much higher activity. It is important to stress that because
the Pt/Cs-Y sample contains larger Pt particles, we have not
included this data point in Figure 11; however, the trend remains
clear. A more detailed analysis of the catalytic data also revealed

3830 J. Phys. Chem. B, Vol. 109, No. 9, 2005
Visser et al.
TABLE 5: Relation between IR, XAFS, and CO Oxidation
Results on Supported Pt Catalysts
that the activation energy for CO conversions between 1 and
10% was between 79 and 102 kJ/mol (Table 4). This is in line
with literature data,46 indicating that there was no mass transfer
limitations in the catalytic systems under study.
properties related to a
specific measuring technique
observation
Au: ref 49 was not cited; citation added below to “47,48”;
relocate as needed
IR
CO L:B ratio
CO coverage
AXAFS peak intensity
AXAFS peak centroid
CO oxidation activity
low
high
low
long
high
high
low
high
short
low
It is well-known that the CO oxidation on Pt proceeds via a
XAFS
three-step Langmuir-Hinshelwood reaction scheme:47-49
Catalysis
COgas T COads
(O2)gas f 2 Oads
(1)
(2)
(3)
the effect of the addition of alkaline and alkaline earth metal
cations to the electronic properties of Pt. An interesting question
remains if the same trends can be obtained for other support
types and compositions as well and further work should be
directed to reveal such trends and hopefully design more active
catalyst systems for car exhaust applications.
COads + Oads f (CO2)gas
where the indices “gas” and “ads” refer to the gas-phase and
adsorbed species, respectively. Under ultrahigh vacuum condi-
tions CO adsorbs associatively and starts to desorb above about
350 K, whereas O2 adsorbs dissociatively above about 100 K
and desorbs associatively above 720 K.59,50 CO diffuses rapidly
over the surface and reacts with O to yield CO2, which
immediately desorbs into the gas phase. On the basis of the
reaction scheme (1)-(3), one can assume that in an excess of
oxygen (as is the case for our catalytic experiments) and at
temperatures around the desorption temperature of CO (350 K),
the rate of the reaction r is determined by the relative surface
concentration of adsorbed CO, i.e., the CO surface coverage
(θCtO). One could then write the reaction equation as follows:
Conclusions
Table 5 relates the different observations obtained by the CO
adsorption infrared, XAFS, and CO oxidation experiments. On
the basis of this work the following conclusions can be made:
(1) Time-resolved infrared spectroscopy proves to be a
feasible tool to monitor desorption of CO from supported highly
dispersed Pt catalysts as a function of temperature and time.
Next to the surface coverage, the position and shape of the linear
and bridge Pt bonded CO stretching band appear to be very
sensitive to the chemical composition of the support and the
desorption temperature. The work also demonstrates that a high
CO stretching frequency is not by definition related to the
desorption rate of CO, which is not correlated to a strong Pts
CtO bonding and vice versa.
(2) The red shift of the linear Pt-coordinated CtO vibration
compared to that of gas-phase CO increases with an increasing
cation radius-to-charge ratio. In addition, a systematic shift from
linear to bridge bonded CO was observed for decreasing Lewis
acidity, as expressed by the Kamlet-Taft parameter R. A
decreasing R results in an increasing electron charge on the
framework oxygen atoms and therefore an increasing electron
charge on the supported Pt nanoparticles.
r ) -{d[(CO2)gas]/dt} ) k′θCtO
(4)
As shown before, we were able on the basis of the IR
measurements to estimate values for the different catalysts under
study. The combined information from Figures 8 and 11 allows
us to state that there exists a relationship between T50% and
θ
CtO; i.e., CO oxidation over supported Pt particles is facilitated
by a higher CO coverage and vice versa. This is in line with
reaction eq 4. This effect can be achieved by adding the adequate
promotor element to the catalyst system. K+ seems to be the
best promoting element for tuning the electronic properties of
the supported Pt nanoparticles.
(3) The CO coverage on the supported Pt nanoparticles
increases with increasing electron density on Pt. This increasing
electron density was found to result in an increased CO oxidation
activity, and basic promotors facilitate the chemisorption of CO
on the Pt nanoparticles. This adsorbed CO reacts with adsorbed
O according to a Langmuir-Hinshelwood mechanism to give
rise to the formation of CO2. The most promoted CO oxidation
catalyst is a Pt/K-Y sample, which has a T50% of 390 K and a
L:B ratio of 2.7.
One should, however, be aware that the spectroscopic and
catalytic measurements are not done under the same environ-
mental conditions, i.e., reducing vs oxidizing environments.
Indeed, the characterization by AXAFS and IR does not
necessarily represent the state of the active catalyst material and
the reduced Pt particles can be (partially) oxidized under the
catalytic conditions employed. An alternative explanation could
be that the differences in electron density of the Pt nanoparticles
on the different support materials lead to a higher fraction of
surface oxidation. An alternative explanation could therefore
be that differences in oxygen coverage on the Pt nanoparticles
lead to the observed differences in the CO oxidation perfor-
mances. In any case, the electronic properties of the supported
Pt nanoparticles influence the catalytic behavior of the materials
under study.
Finally, it should be emphasized that the addition of promotor
elements has, besides electronic effects, also structural or
stabilization effects, which have not been the subject of this
work. Much literature exists in which BaO has been studied in
relation to its NOx storage properties.51-59 Here, we show that
besides this NOx adsorption capability, BaO also induces an
electronic effect on the supported Pt nanoparticles. In this
respect, zeolite Y-supported Pt particles of uniform sizes can
be regarded as model systems to study in a controlled manner
Acknowledgment. This work has been funded by Toyota
Motor Research. We kindly thank fruitful discussions with I.
Masaru, K. Ishibashi, and S. Matsumoto of Toyota Motor
Corporation Japan, Y. Nagai, T. Tanabe, T. Nonaka, H.
Sobukawa, and M. Sugiura of Toyota Central R&D Labs Japan,
and O. Kito of the Material Engineering Division of Toyota
Motor Europe. P. J. Kooyman (Delft University of Technology)
is acknowledged for performing HRTEM measurements, D. C.
Koningsberger (Utrecht University) for interesting discussions
and help with the EXAFS as well as AXAFS analysis and A.
Mens and V. Koot (Utrecht University) for carrying out the N2
and H2 sorption measurements. We also acknowledge beamtime
grants from the DUBBLE Grenoble and X1.1 Hasylab beamline
stations.

CO Oxidation over Zeolite-Supported Pt Nanoparticles
J. Phys. Chem. B, Vol. 109, No. 9, 2005 3831
(30) Nijhuis, T. A.; Makkee, M.; van Langeveld, A. D.; Moulijn, J. A.
Appl. Catal. A: Gen. 1997, 164, 237.
References and Notes
(1) (a) Arena, G. E.; Centi, G. Top. Catal. 2004, 30-31, 147. (b) Centi,
G.; Arena, G. E. J. Mol. Catal. A Chem. 2003, 204, 663. (c) Fornasari, G.;
Trifiro, F.; Vaccari, A.; Prinetto, F.; Ghiotti, G.; Centi, G. Catal. Today
2002, 75, 421. (d) Liotta, L. F.; Macaluso, A.; Arena, G. E.; Livi, M.; Centi,
G. Catal. Today 2002, 75, 439.
(31) Schulz-Ekloff, G. Stud. Surf. Sci. Catal. 1991, 69, 65. Wark, M.;
Schulz-Ekloff, J. N. I.; Zukal, A. Stud. Surf. Sci. Catal. 1991, 69, 189.
(32) Vaarkamp, M.; Grondelle, J. V.; Miller, J. T.; Sajkowski, D. J.;
Modica, F. S.; Lane, G. S.; Gates, B. C.; Koningsberger, D. C. Catal. Lett.
1990, 6, 369.
(2) (a) Schmauke, T.; Eichel, R. A.; Schweiger, A.; Roduner, E. Phys.
Chem. Chem. Phys. 2003, 5, 3076. (b) Schmauke, T.; Menzel, M.; Roduner,
E. J. Mol. Catal. A Chem. 2003, 194, 211. (c) Zheng, J.; Schmauke, T.;
Roduner, E.; Dong, J. L.; Xu, Q. H. J. Mol. Catal. A Chem. 2001, 171,
181. (d) Schmauke, T.; Moller, E.; Roduner, E. Chem. Commun. 1998, 2589.
(3) (a) Miller, J. T.; Koningsberger, D. C. J. Catal. 1996, 162, 209.
(b) Koningsberger, D. C.; Ramaker, D. E.; Miller, J. T.; De Graaf, J.; Mojet,
B. L. Top. Catal. 2001, 15, 35. (c) De Graaf, J.; van Dillen, A. J.; de Jong,
K. P.; Koningsberger, D. C. J. Catal. 2001, 203, 307.
(4) Barth, R.; Pitchai, R.; Anderson, R. L.; Verykios, X. J. Catal. 1989,
116, 61.
(33) Sheppard, N.; Nguyen, T. T. AdV. Infrared Raman Spectrosc. 1978,
5, 67.
(34) Bourane, A.; Dulaurent, O.; Bianchi, D. J. Catal. 2000, 196, 115.
(35) Greenler, R. G.; Brandt, R. K. Colloids Surf. A: Physicochem. Eng.
Asp. 1995, 105, 19.
(36) Larin, A. V.; Leherte, L.; Vercauteren, D. P. Phys. Chem. Chem.
Phys. 2002, 4, 2416.
(37) Larin, A. V.; Vercauteren, D. P.; Lamberti, C.; Bordiga, S.;
Zecchina, A. Phys. Chem. Chem. Phys. 2002, 4, 2424.
(38) Wasileski, S. A.; Weaver, M. A.; Koper, M. T. M. J. Electroanal.
Chem. 2001, 500, 344.
(39) Marcus, Y. Ion properties; Marcel Dekker: New York, Basel, Hong
Kong, 1997; p 193.
(5) Blyholder, G. J. J. Phys. Chem. 1968, 60, 2772.
(6) Mojet, B. L.; Miller, J. T.; Koningsberger, D. C. J. Phys. Chem. B
1999, 103, 2724.
(7) Kappers, M. J.; Vaarkamp, M.; Miller, J. T.; Modica, F. S.; Barr,
M. K.; van der Maas, J. H.; Koningsberger, D. C. Catal. Lett. 1993, 21,
235.
(8) Ramaker, D. E.; de Graaf, J.; van Veen, J. A. R.; Koningsberger,
D. C. J. Catal. 2001, 203, 7.
(40) Renema, J. Structural information from carbonyl stretching frequen-
cies. Ph.D. Thesis, Utrecht University, 1972.
(41) Vannice, M. A.; Twu, C. C. J. Chem. Phys. 1981, 75, 5944.
(42) Koningsberger, D. C.; Mojet, B. L.; van Dorssen, G. E.; Ramaker,
D. E. Top. Catal. 2000, 10, 143.
(43) Ramaker, D. E.; van Dorssen, G. E.; Mojet, B. L.; Koningsberger,
D. C. Top. Catal. 2000, 10, 157.
(44) Koningsberger, D. C.; de Graaf, J.; Mojet, B. L.; Ramaker, D. E.;
Miller, J. T. Appl. Catal. A: General 2000, 191, 205.
(45) Mojet, B. L.; Miller, J. T.; Ramaker, D. E.; Koningsberger, D. C.
J. Catal. 1999, 186, 373.
(46) Venderbosch, R. H.; Prins, W.; van Swaaij, W. P. M. Chem. Eng.
Sci. 1998, 53, 3355.
(47) Ertl, G.; Engel, T. AdV. Catal. 1979, 28, 1.
(48) Ertl, G. AdV. Catal. 1990, 37, 1.
(49) Nolan, P. D.; Lutz, B. R.; Tanaka, P. L.; Davis, J. E.; Mullins, C.
B. J. Chem. Phys. 1999, 111, 3696.
(9) Xu, J.; Yates, J. T. Surf. Sci. 1995, 327, 193.
(10) Xu, J.; Hendriksen, P. N.; Yates, J. T. Langmuir 1994, 10, 3663.
(11) Hayden, B. E.; Kretzschmar, K.; Bradshaw, A. M. Surf. Sci. 1985,
149, 394.
(12) Mukerji, R. J.; Bolina, A. S.; Brown, W. A. Surf. Sci. 2003, 527,
198.
(13) Barth, A.; Ramachandran, A. J. Catal. 1990, 125, 467.
(14) Gandao, Z.; Coq, B.; de Me´norval, L. C.; Tichit, D. Appl. Catal.
A: Gen. 1996, 147, 395.
(15) Bourane, A.; Bianchi, D. J. Catal. 2003, 218, 447.
(16) Welch, P. C.; Mills, P. S. W.; Mason, C.; Hollins, P. J. Electron.
Spectrosc. Relat. Phenom. 1993, 64/65, 151.
(17) Hollins, P. Surf. Sci. Rep. 1992, 16, 51.
(18) Koper, M. T. M.; van Santen, R. A.; Wasileski, S. A.; Weaver, M.
J. J. Chem. Phys. 2000, 113, 4392.
(19) Greenler, R. G.; Brandt, R. K. Physicochem. Eng. Aspects 1995,
105, 19.
(20) Hippe, C.; Lamber, R.; Schulz-Ekloff, G.; Schubert, U. Catal. Lett.
1997, 43, 195.
(21) Silvestre-Albero, J.; Sepulveda-Escribano, A.; Rodriguez-Reinoso,
F.; Anderson, J. A. Phys. Chem. Chem. Phys. 2003, 5, 208.
(22) Sharma, R. K.; Brown, W. A.; King, D. A. Surf. Sci. 1998, 414,
68.
(23) Stoop, F.; Toolenaar, F. J. C. M.; Ponec, V. J. Catal. 1982, 73, 50.
(24) Geidel, E.; Lechert, H.; Dobler, J.; Jobic, H.; Calazferri, G.; Bauer,
F. Micropor. Mesopor. Mater. 2003, 65, 31.
(25) Otero Arean, C.; Turnes Palomino, G.; Zecchina, A.; Spoto, G.;
Bordiga, S.; Roy, P. Phys. Chem. Chem. Phys. 1999, 1, 4139.
(26) De Mallman, A.; Barthomeuf, D. Stud. Surf. Sci. Catal. 1989, 46,
429.
(27) Gijzeman, O. L. J.; Mens, A. J. M.; van Lenthe, J. H.; Mortier, W.
J.; Weckhuysen, B. M. J. Phys. Chem. B 2003, 107, 678.
(28) Vaarkamp, M.; Linders, J. C.; Koningsberger, D. C. Physica B 1995,
208/209, 9825.
(29) Vaarkamp, M.; Mojet, B. L.; Kappers, M. J.; Miller, J. T.;
Koningsberger, D. C. J. Phys. Chem. 1995, 99, 16067.
(50) Artsyukhovich, A. N.; Ukraintsev, V. A.; Harrison, I. Surf. Sci.
1996, 347, 303.
(51) Broqvist, P.; Gronbeck, H.; Fridell, E.; Panas, I. J. Phys. Chem.
2004, 108, 3523.
(52) Nova, I.; Castoldi, L.; Prinetto, F.; Dal Santo, V.; Lietti, L.;
Tronconi, E.; Forzatti, P.; Ghiotti, G.; Psaro, R.; Recchia, S. Top. Catal.
2004, 30-31, 181.
(53) Fridell, E.; Skoglundh, M.; Westerberg, B.; Johansson, S.; Smedler,
G. J. Catal. 1999, 183, 196.
(54) Olsson, L.; Persson, H.; Fridell, E.; Skoglundh, M.; Andersson, B.
J. Phys. Chem. B 2001, 105, 6895.
(55) Lesage, T.; Verrier, C.; Bazin, P.; Saussey, J.; Daturi, M. Phys.
Chem. Chem. Phys. 2003, 5, 4435.
(56) Lesage, T.; Verrier, C.; Bazin, P.; Saussey, J.; Malo, S.; Hedouin,
C.; Blanchard, G.; Daturi, M. Top. Catal. 2004, 30-31, 31.
(57) Shinjoh, H.; Tanabe, T.; Sobukawa, H.; Sugiura, M. Top. Catal.
2001, 16-17, 95.
(58) Takahashi, N.; Shinjoh, H.; Iijima, T.; Suzuki, T.; Yamazaki, K.;
Yokota, K.; Suzuki, H.; Miyoshi, N.; Matsumoto, S.; Tanizawa, T.; Tanaka,
T.; Tateishi, S.; Kasahara, K. Catal. Today 1996, 27, 63.
(59) Shinjoh, H.; Takahashi, N.; Yokota, K.; Sugiura, M. Appl. Catal.
B EnViron. 1998, 15, 189.