Characterization of γ-Ga2O3-Al 2O3 prepared by solvothermal method and its performance for methane-SCR of NO

Article abstract of DOI:10.1021/jp901569s

The γ-Ga₂O₃-Al₂O₃ mixed oxides with a spinel structure were prepared by the solvothermal reaction of gallium acetylacetonate and aluminum isopropoxide in diethylenetriamine. In the crystal structures of the cata

Full text of DOI:10.1021/jp901569s

J. Phys. Chem. A 2009, 113, 7021–7029
7021
Characterization of γ-Ga2O3-Al2O3 Prepared by Solvothermal Method and Its Performance
for Methane-SCR of NO
Tetsu Nakatani,† Tsunenori Watanabe,†,‡ Masaru Takahashi,† Yuya Miyahara,†
Hiroshi Deguchi,‡ Shinji Iwamoto,† Hiroyoshi Kanai,† and Masashi Inoue*,†
Department of Energy and Hydrocarbon Chemistry, Graduate School of Engineering, Kyoto UniVersity,
Katsura, Kyoto 615-8510, Japan, and Power Engineering R&D Center, The Kansai Electric Power Company,
Inc., 3-11-20, Nakoji, Amagasaki 661-0974, Japan
ReceiVed: February 20, 2009; ReVised Manuscript ReceiVed: April 26, 2009
The γ-Ga2O3-Al2O3 mixed oxides with a spinel structure were prepared by the solvothermal reaction of
gallium acetylacetonate and aluminum isopropoxide in diethylenetriamine. In the crystal structures of the
catalysts obtained by the calcination of these mixed oxides, Ga3+ and Al3+ ions preferentially occupied
tetrahedral and octahedral sites, respectively. The catalysts with low Ga contents had a unique structure with
high surface areas and a concentration gradient of decreasing Ga content from the surface to the bulk. In
methane-selective catalytic reduction (SCR) of NO, higher NO conversion to N2 was attained on the catalyst
with high occupation of Ga3+ ions at tetrahedral sites and Al3+ ions at octahedral sites. For the γ-Ga2O3-Al2O3
mixed oxide with a charged Ga molar content of 0.3 (ST(0.3)), tetrahedral and octahedral sites were solely
occupied by Ga3+ and Al3+ ions, respectively, and the catalyst exhibited the highest NO conversion to N2.
Therefore, it was concluded that the active site for methane-SCR of NO is tetrahedral Ga3+ ion and octahedral
Al3+ ion, which are linked to each other. Nitrogen monoxide is adsorbed on the isolated hydroxyl group
attached to Al3+ ions and then oxidized by O2 yielding surface nitrate species. Tetrahedral Ga3+ ions work as
Lewis acid sites for the activation of methane because of their coordinative unsaturation. The Ga3+ ions in
the γ-Ga2O3-Al2O3 catalyst have a redox property, which plays important roles in both the oxidation of NO
to surface nitrate species and the activation of methane. The most important factor for this catalyst is that the
sites for the formation of surface nitrate species reside next to the methane activation sites, which facilitates
the reaction between surface nitrate species and the activated species derived from methane, thus mitigating
the consumption of methane by simple combustion with O2. Therefore, ST(0.3), which has the largest number
of ensembles of the tetrahedral Ga3+ ions and octahedral Al3+ ions, shows the highest activity for methane-
SCR of NO.
1. Introduction
reported that Ga2O3-Al2O3 prepared by a sol-gel method has
high performance for SCR of NOx with propene.4
Among the various strategies for removing NOx from exhaust
gases, selective catalytic reduction (SCR) using methane as a
reductant1-5 is an attractive process because methane is the main
component of natural gas, an abundant resource that is used as
a fuel for power stations. Moreover, methane can be easily
removed by catalytic combustion even if methane leaks from
the SCR devices.
Although various zeolite catalysts have high activities for
methane-SCR of NOx, they have poor hydrothermal stabilities;
their frameworks are destroyed by exposure to high-temperature
steam. Therefore, the development of active oxide-type catalysts
with higher hydrothermal stabilities is necessary because the
water content in the exhaust gases from stationary combustion
plants is rather high.
Gallium oxide is an effective catalyst for dehydrogenation
of paraffin to olefin,6 and Ga-containing zeolites are active for
NOx removal using hydrocarbons as reducing agents.7 Shimizu
et al. reported that γ-alumina-supported gallium oxide (Ga2O3/
γ-Al2O3) catalyst prepared by an impregnation method showed
a high activity for SCR of NOx with methane.2,3 Haneda et al.
Although various aqueous medium-based methods for syn-
thesizing metal oxides such as sol-gel, coprecipitation, and
hydrothermal methods have been studied, the synthesis of metal
oxide using organic solvents has not been widely explored. Our
laboratory has been exploring solvothermal synthesis of metal
oxides in organic media,8 where solvothermal reaction means
the reaction in liquid media at high temperatures. Previously,
we reported that the γ-Ga2O3-Al2O3 solid solution catalysts
prepared by solvothermal methods showed higher activities than
the impregnation catalysts for methane-SCR of NO.9-12 The
activities of the catalysts prepared by using various media were
examined, and it was found that the catalyst activity increased
with the increase in the crystallite size. Diethylenetriamine was
found to be the best solvent for the solvothermal synthesis of
the γ-Ga2O3-Al2O3 catalyst having high activities under both
dry and wet conditions.12 The influence of reducing agents
(C1-C3 hydrocarbons) upon the SCR of NO on γ-Ga2O3-
Al2O3 catalyst was investigated, and methane was found to be
the least reactive reductant of the hydrocarbons examined.13
Under wet conditions, NO conversion on the catalyst decreased
for all hydrocarbons. The reactivity of paraffins, especially
methane, was severely affected by the presence of water in the
* To whom correspondence should be addressed. Tel: +81 75 383 2478.
Fax: +81 75 383 2479. E-mail: inoue@scl.kyoto-u.ac.jp.
† Kyoto University.
‡ The Kansai Electric Power Company.
10.1021/jp901569s CCC: $40.75  2009 American Chemical Society
Published on Web 05/29/2009

7022 J. Phys. Chem. A, Vol. 113, No. 25, 2009
Nakatani et al.
feed. This result was explained by preferential adsorption of
water to the adsorption sites for hydrocarbons.13
was placed. After the air in the autoclave was purged with
nitrogen, the autoclave was heated to 300 °C at a rate of 2.5
°C·min-1, kept at 300 °C for 2 h, and then cooled to room
temperature. The products were washed with acetone by
vigorous mixing and centrifuging, air dried, and calcined at 700
°C for 30 min to remove the remaining organic impurities. The
yields of the catalysts were over 95%. In this article, the catalyst
is designated as ST(x), where x stands for the Ga/(Ga + Al)
ratio in the feed composition for the solvothermal treatment.
2.2. Activity Test for the Catalysts. Catalyst tests for the
SCR of NO with methane were carried out in a fixed-bed flow
reactor. The catalysts were tabletted, pulverized into 10-22
mesh size, and placed into the reactor. The catalyst bed was
heated to 650 °C in a helium flow and held at that temperature
for 30 min. Then, the reaction gas composed of 1000 ppm NO,
2000 ppm CH4, 6.7% O2, and 2.5% H2O with helium balance
was introduced to the catalyst bed at W/F ) 0.3 g · s · mL-1
(SV ) ∼11 000 h-1). The reaction temperature was decreased
from 650 to 450 °C at 5 °C·min-1 and kept for 15 min at every
50 °C interval to attain the stationary state. The effluent gases
from the reactor were analyzed with an online gas chromato-
graph (GL Science Micro GC CP-2002) equipped with a 4 m
Molsieve 5A column (80 °C) and a 10 m Poraplot Q column
(40 °C). NO conversion was calculated, as shown in eq 1
The mechanisms of NO reduction by hydrocarbon over
gallia-alumina catalysts have been discussed by several re-
searchers. Shimizu et al. examined the methane-SCR activity
of Ga2O3/Al2O3 prepared by an impregnation method and
concluded that Ga cations in the surface tetrahedral sites are
the origin of the high performance of their catalyst.3 They also
proposed reaction mechanisms for hydrocarbon-SCR on Al2O3-
based catalysts.14 By using IR spectroscopy, He et al. investi-
gated the reaction pathway of propane-SCR of NO over Ga2O3/
Al2O3 prepared by an impregnation method.15 They concluded
that the propane-SCR of NO starts with partial oxidation of
propane mainly by O2 to surface carboxylates such as acetate
and formate, which are highly reactive toward adsorbed NO
(or NO-derived species) yielding nitrogen-containing organic
surface species (e.g., isocyanides and cyanides). The coupling
of nitrogen atoms to form N2 was suggested to occur via the
disproportionation reaction between such nitrogen-containing
species and NO. They also concluded that the rate-determining
step of the whole reaction cycle is the partial oxidation of
propane.15
By using GC-MS and IR spectroscopy, Haneda et al.
investigated the mechanism of propane-SCR of NO over
Ga2O3-Al2O3 prepared by a sol-gel method.16,17 They sug-
gested that the reaction is initiated by the formation of adsorbed
nitrate species via NO oxidation by O2. The reaction between
adsorbed nitrate species and C3H6 causes the formation of the
allyl species, which reacts with NOx yielding organic nitro
compounds such as CH3NO2, C2H3NO2, and C3H7NO2. Hy-
drolysis of these nitro compounds is assumed to form nitriles
(e.g., CH3CN and CH2dCH-CN), isocyanates (-NCO), and
oxygenates. Further hydrolysis of isocyanates (or nitriles) yields
organic amines and CO2 (or carboxylic acid and NH3), which
react with NOx to form N2, as is well known for NH3-SCR.
Haneda et al. proposed that the formation of nitriles and
isocyanates is effectively promoted by Ga2O3.16,17
We found that surface allyl species are formed upon the
exposure of propylene to both tetrahedral and octahedral Ga
sites of Ga2O3-Al2O3 catalyst, and the amount of allyl species
adsorbed on the tetrahedral Ga sites correlates with the activity
of the catalyst.18 However, methane is much less reactive than
C3 hydrocarbons,13 and surface allyl species were not formed
from methane.18 Moreover, surface formate species, a possible
intermediate for methane-SCR, is reported to be less stable than
surface acetate species.19
Therefore, the reaction mechanisms for the methane-SCR of
NO may be different from propane- or propene-SCR. In this
work, methane-SCR of NO was carried out on the γ-Ga2O3-
Al2O3 catalysts prepared by the solvothermal method using
diethylenetriamine as a solvent. The factors controlling the
catalytic activity were explored, and the reaction mechanisms
will be discussed focusing on the individual roles of the Ga3+
and Al3+ ions.
NO conversion (%) ) (2[N2]/[NO])100
(1)
where [N2] is the concentration of N2 in the effluent and [NO]
is the concentration of NO in the fed gas. Note, however, that
no NOx species other than NO were detected during methane-
SCR, as analyzed with a Pfeiffer Vacuum Omnistar GSD 301
O 1 quadrupole mass spectrometer (Q-Mass).
2.3. Characterization of the Catalysts. Powder X-ray
diffraction (XRD) patterns were measured on a Shimadzu XD-
D1 diffractometer using 30 kV, 900 W Cu KR radiation and a
carbon monochromator. Crystallite size was calculated from the
half-height width of the (440) diffraction peak of the spinel
structure using the Scherrer equation.
The BET surface area was calculated by the single-point
method, and the samples were dried in a 30% N2/He flow at
300 °C for 30 min prior to measurement.
The bulk Ga/Al ratios were analyzed with a Shimadzu ICPS-
1000IV inductively coupled plasma atomic emission spectrom-
eter (ICP-AES). The catalysts were dissolved in phosphoric acid
at 80 °C and then diluted to about 0.2 g·L-1 with water. The
surface Ga/Al ratios were determined with an ULVAC-PHI 5500
X-ray photoelectron spectrometer (XPS) with a hemispherical
energy analyzer. Samples were mounted on indium foil and then
transferred to an XPS analyzer chamber. The residual gas
pressure in the chamber during data acquisition was less than 1
× 10-8 Torr (1 Torr ) 133.3 N ·m-2). The spectra were
measured at room temperature using Mg KR radiation (15 kV,
400 W). The electron takeoff angle was set at 45°. Binding
energies were referenced to the calibrated on the basis of C 1s
of residual carbon at 284.6 eV.20 Sputtering was performed by
an Xe+ ion beam generated at 3.0 kV.
X-ray absorption fine structure (XAFS) at the Ga K-edge was
measured by a transmission mode at room temperature using
the SPring-8 BL16B2 beamline. The X-ray was monochroma-
tized by a Si(111) monochromator. The data were analyzed
using the software “Athena and Artemis”.21 As standard samples,
ZnGa2O4 in which Ga ions occupy octahedral sites and ST(0.1)
in which Ga ions preferentially occupy tetrahedral sites were
2. Experimental Section
2.1. Preparation of the Catalysts. The catalysts were
prepared by the solvothermal method developed in our labora-
tory.8 Gallium triacetylacetonate (Ga(acac)3; Mitsuwa Chemical)
and aluminum triisopropoxide (AIP; Nacalai Tesque) with
various Ga/Al ratios were suspended in 80 mL of diethylene-
triamine (dEtA; Nacalai Tesque) in a test tube, and the tube
was placed in a 200 mL autoclave. In the gap between the
autoclave wall and the test tube, an additional 30 mL of dEtA

Methane-SCR of NO on γ-Ga2O3-Al2O3
J. Phys. Chem. A, Vol. 113, No. 25, 2009 7023
TABLE 1: Surface Areas and Compositions of
Ga2O3-Al2O3 Catalysts
Ga/(Ga+Al) ratio
sputter time
surface area
catalyst
(m2 ·g-1
)
bulk 0 min 1 min 3 min 5 min
ST(0.0)
ST(0.1)
ST(0.25)
ST(0.3)
ST(0.375)
ST(0.5)
ST(0.75)
ST(1.0)
212
179
177
154
131
162
133
112
0.077
0.218 0.244
0.254 0.27
0.379
0.511 0.509 0.499 0.501 0.512
0.773 0.775 0.783 0.771 0.777
0.265 0.252 0.239
used. The latter sample was prepared solvothermally in 2-methyl-
aminoethanol.10
Figure 1. Average coordination numbers of Ga in the γ-Ga2O3-Al2O3
catalysts calculated from the EXAFS spectra of the catalysts: Fitting
parameters are as follows: FT range, 3.3-14.0 Å-1; FT k weight, 3;
fitting space, R space; fitting R range, 1.0 to 1.8 Å; fitting k weight, 3.
2.4. Analysis of the Species Adsorbed on the Catalysts.
Temperature-programmed desorption (TPD) of the adsorbed
species except for NH3 was carried out in a fixed-bed flow
reactor. The catalyst was heated at 650 °C for 30 min in He
and cooled to 100 °C, and a gas with a prescribed composition
was allowed to flow over the catalyst at W/F ) 0.3 g ·s·mL-1
for 1 h. After the excess adsorbate gas was purged with a He
flow, the catalyst was heated to 700 °C at a rate of 5 °C·min-1
in a He flow, and the desorbed species were analyzed with the
Q-Mass detector.
compositions as those of the starting materials for solvothermal
synthesis, whereas for the catalysts with low Ga contents, Ga3+
ion contents in the products were smaller than those expected
from the feed composition.
The XPS analysis indicated that the catalysts with high Ga
contents (ST(0.75) and ST(0.5)) had constant Ga concentrations
from surface to bulk, whereas those with small Ga contents
(ST(0.3) and ST(0.25)) had Ga-rich surface compositions,
indicating the presence of a Ga concentration gradient inside
the particles. Actually, ST(0.3) exhibited an obvious Ga
concentration gradient; the Ga concentration decreased gradually
at deeper points from the surface. Therefore, the Ga2O3-Al2O3
catalysts with small Ga contents had larger numbers of surface
Ga sites than those expected from the bulk compositions.
Figure 1 shows the average coordination numbers of Ga3+
ions calculated from the EXAFS spectra. The normalized
absorption data of XANES spectra are shown in Figure S3 in
the Supporting Information. For the value at Ga/(Ga + Al) )
1.0, the datum for pure γ-Ga2O3 is plotted because ST(1.0) was
composed of γ and ꢀ phases. The average coordination number
of Ga3+ ion in pure γ-Ga2O3 was calculated to be 4.9, which is
slightly smaller than the theoretical value. The average coor-
dination number decreased with the decrease in the Ga content
and converged at about 4 for the catalysts with low Ga contents.
This indicates that Ga3+ ions predominantly occupy the tetra-
hedral sites; therefore, the γ-Ga2O3-Al2O3 binary oxides are
not simple solid solutions. Nearly all of the Ga3+ ions in ST(0.3)
are located in the tetrahedral sites, whereas Al3+ ions predomi-
nantly occupy the octahedral sites in the spinel structure. Similar
results have been previously reported.3,11,22 Spinels belong to
the Fd3m space group. There are eight formula units per cubic
unit cell. Anions are located in Wycoff 32e sites whose
coordinates are (u, u, u). The anion sublattice of ideal spinel is
in cubic closest-packed (ccp) arrangement. Because the coor-
dinates of the anions at 32e sites are not special, they can vary
according to a single parameter, u. For a perfect ccp anion
arrangement, u ) 0.375. As u increases from its ideal value,
anions move away from the tetrahedrally coordinated A-site
cations along the <111> directions toward the center of three
B-site cations, which increases the volume of each A-site
interstice. Therefore, the spinel structure can accommodate large
cations in the tetrahedral sites.23
The NH3 TPD measurement was carried out with a Bel-Japan
TPD-1-AT analyzer. The powdered sample (0.4 g) was placed
in the TPD cell and pretreated in a 50 mL ·min-1 He flow at
500 °C for 1 h. Then, the sample was exposed to 2.7 kPa NH3
at 100 °C for 10 min. After evacuation for 1 h, the sample was
heated to 600 °C at a rate of 10 °C·min-1 in a He flow, and the
effluent gases were analyzed with the Q-Mass detector.
A Nicolet Magna-IR system 560 FT-IR diffuse reflectance
spectrometer (resolution: 4 cm-1) was used to detect the
adsorbed species on the catalyst surface. The sample was
pretreated at 500 °C in a 30 mL·min-1 He flow for 1 h; then,
the background spectrum was obtained. The spectra of the
adsorbed species are given as difference spectra obtained by
subtraction of the background.
3. Results and Discussion
3.1. Crystal Structures of the Catalysts. The XRD pattern
of ST(0.0) indicates that the alumina synthesized by the
solvothermal method in diethylenetriamine is amorphous (Figure
S1 in the Supporting Information). The Ga2O3-containing mixed
oxides exhibited the patterns due to the spinel structure, and
their peaks shifted to the lower angle side with an increase in
Ga content, indicating that substitution of Al3+ ions with larger
Ga3+ ions resulted in an enlargement of the unit cell parameter.
Therefore, all of the products except for ST(0.0) and ST(1.0)
are solid solutions of γ-Ga2O3-Al2O3.
Upon calcination at 700 °C, a part of ST(1.0) was transformed
into the ꢀ phase of Ga2O3, whereas ST(0.0) gave poorly
crystallized Al2O3. The other mixed oxides preserved the
γ-phase structure after calcination at 700 °C, although their
crystallite sizes were slightly enlarged (Figure S2 in the
Supporting Information).
The surface areas and chemical compositions of the catalysts
(calcined samples) are summarized in Table 1. The surface area
of the catalyst tends to decrease with the increase in the Ga
content.
The ICP analysis showed that the catalysts with high Ga
contents (e.g., ST(0.75) and ST(0.5)) had almost the same
Because γ-Ga2O3-Al2O3 is made of two metal elements both
with the same valence of 3+ (Ga3+ and Al3+), cation vacancies

7024 J. Phys. Chem. A, Vol. 113, No. 25, 2009
Nakatani et al.
Figure 3. TPD profiles of (a) O2, (b) NO, (c) NO-O2, (d) CH4, and
(e) CH4-O2 adsorbed on ST(0.3) catalyst. The catalyst was pretreated
in He at 650 °C for 30 min, and then the reactant gas diluted to a
desired concentration with He was allowed to flow at 100 °C for 1 h.
After the gas was purged with a He flow, desorption was carried out
in a 100 mL ·min-1 He flow at a heating rate of 5 °C·min-1. Gases:
1000 ppm NO, 2000 ppm CH4, and/or 6.7% O2.
Figure 2. Effect of Ga/(Ga + Al) ratios of the γ-Ga2O3-Al2O3
catalysts on their NO conversion to N2. Reaction conditions: 1000 ppm
NO, 2000 ppm CH4, 6.7% O2, and 2.5% H2O with He balance; SV )
11 000 h-1
.
should be incorporated in the spinel structure because of charge
imbalance. Therefore, one-ninth of the total cation sites available
for the spinel structure are vacant (M3+8/301/3O4, M3+: Ga3+ or
Al3+). Here we will deal with γ-Ga2O3-Al2O3 with an extreme
structure in which Ga3+ and Al3+ ions solely occupy tetrahedral
and octahedral sites, respectively. When the cation vacancies
exist only in the tetrahedral sites ([Ga2/301/3]tet[Al2]octO4), the
Ga/(Ga + Al) ratio is calculated to be 0.25. When the vacancies
exist solely in the octahedral sites ([Ga1]tet[Al5/301/3]octO4), the
Ga/(Ga + Al) ratio would be 0.375. Therefore, the data for
ST(0.3) shown in Figure 1 and Table 1 (coordination number
of Ga is ca. 4 for Ga/(Ga + Al) ratio of around 0.25) indicate
that the cation vacancies mainly exist in the tetrahedral sites of
the spinel structure.
3.2. Catalytic Activity. Figure 2 shows the activity change
of the catalysts for NO conversion to N2 as a function of the
Ga/(Ga + Al) ratio. Both ST(0.0) and ST(1.0) had very low
activity, indicating that the combination of Ga and Al ions is a
crucial factor for the catalytic activity. The conversion of NO
to N2 reached a maximum for ST(0.3) in which Ga3+ and Al3+
ions preferentially occupy the tetrahedral and octahedral sites,
respectively. Each oxygen anion in the spinel (AB2O4) coordi-
nates to one tetrahedral A cation and three octahedral B cations.
Therefore, the catalyst that exhibited the highest activity has
the largest number of tetrahedral Ga3+ ions with octahedral Al3+
ions at the next-nearest neighbor sites. When the Ga/(Ga+Al)
ratio exceeds 0.3, the population of Gatet-O-Gaoct bonds
increases.
and latter peaks to the surface nitrite and nitrate species,
respectively. It is interesting to note that a large low-temperature
peak was observed when NO and O2 were coadsorbed. Because
the formation of nitrite species from NO is an oxidation process,
it is reasonable to assume that oxygen is required for the
formation of surface nitrite species, and the following sequence
was proposed for this process24
1
NO + /2O2 f NO2
(2)
(3)
(4)
NO + NO2 f N2O3
M-O-M + N2O3 f 2MONO
Equation 4 may be rewritten as follows
2MOH + N2O3 f 2MONO + H2O
(5)
If these arguments are correct, then the desorption process
of NO would be accompanied by the formation of O2. This was
not the case, as is shown in Figure 3c. Therefore, we temporarily
attribute the desorption of NO to the formation of NO and
surface nitrate species from the adsorbed NO2 as shown in
eq 6
3.3. Desorption of the Reactants Adsorbed on the Cata-
lysts. Figure 3 shows the TPD profiles of the reactants adsorbed
on ST(0.3) that showed the highest activity. When the catalyst
was exposed to an NO/He flow (Figure 3b), two desorption
peaks of NO were observed at about 350 and 450 °C, and the
latter peak was associated with desorption of a trace amount of
O2. When NO and O2 were coadsorbed on the catalyst, a much
larger amount of NO was adsorbed, and the desorption peaks
of NO were observed at around 200 and 400 °C. The latter
peak was associated with a desorption peak of O2 (Figure 3c)
despite the fact that O2 itself was not adsorbed on the catalyst
(Figure 3a). These results indicate that O2 had an ability to
change the state of NO so as to be easily adsorbed on the catalyst
surface. The amount of desorbed O2 at 450 °C was about half
of the amount of desorbed NO at the same temperature. Haneda
et al.4 reported similar NO-O2 TPD profiles for Ga2O3-Al2O3
catalysts prepared by a sol-gel method and attributed the former
3NO2(ad) + 2MOH f NO + 2MONO2 + H2O (6)
This reaction is essentially identical to the formation of nitric
acid by the reaction of NO2 with water. However, this reaction
would contribute to the SCR of NO by the formation of surface
nitrate species and would not be directly connected with the
SCR of NO because the desorption of NO takes place at a much
lower temperature than the working temperature of the catalyst
for SCR.
The amount of adsorbed CH4 was quite small, as shown in
Figure 3d. This result is due to the fact that a higher temperature
is required for adsorption of methane, as will be discussed later.
However, different phenomena were observed for ethylene or
propylene TPD, where significant uptakes were observed even
when the catalyst was exposed to these gases at 100 °C.13,18 As

Methane-SCR of NO on γ-Ga2O3-Al2O3
J. Phys. Chem. A, Vol. 113, No. 25, 2009 7025
Figure 4. FT-IR spectra of the ST(0.3) treated with NO in the presence
and absence of O2. Adsorption was carried out with (a) 1000 ppm NO
and (b) 1000 ppm NO-6.7% O2, both with He balance at 100 °C for
90 min. Then, the catalyst was heated in situ in a He flow at the desired
temperature specified in the Figure.
Figure 6. Reaction of CH4 with NO adsorbed on ST(0.3) in the
presence of O2. ST (0.3) was exposed to 1000 ppm NO-6.7% O2 with
He balance at 100 °C for 1 h, purged with a He flow, and then heated
at a rate of 5 °C·min-1 in a 100 mL ·min-1 flow of 2000 ppm CH4/He.
The IR spectrum of this surface species resembled that of the
surface nitrate species (Figure 4b); however, peaks are observed
at higher wave numbers. Therefore, predominant species was
assigned as adsorbed NO2 strongly interacting with the surface
hydroxyl groups. Although we believe that a part of the adsorbed
NO2 was converted into surface nitrate species in this stage,
complete conversion into surface nitrate species can be ruled
out because of the formation of NO in the NO-O2 TPD profile
because the condition adopted for obtaining the top spectrum
in Figure 5 (i.e., exposure to a gas composed of 1000 ppm NO
and 6.7% O2 with helium balance at 100 °C for 1 h) is essentially
identical to the adsorption conditions for the TPD experiment.
These results show a sharp contrast with the result reported by
Haneda et al.,4 who described that nitrite species are predominant
when their Ga2O3-Al2O3 catalyst is exposed to NO-O2 at
temperatures below 473 K. As surface nitrate species developed,
a negative band appeared at 3768 cm-1, and simultaneously, a
broadband at 3600-3000 cm-1 grew. The former band is
assigned to the isolated hydroxyl group29,30 and the latter band
is assigned to adsorbed water.31 The formation of adsorbed water
can be explained by eq 6 or by the following reaction
Figure 5. Time course change in the FT-IR spectra of NO species
adsorbed on ST(0.3) in the presence of O2. ST(0.3) was exposed to
1000 ppm NO and 6.7% O2 with He balance at 100 °C.
shown in Figure 3e, CH4 uptake did not change, even in the
presence O2.
The adsorbed state of NO was investigated by FT-IR
spectroscopy (Figure 4). When NO was introduced in the
absence of O2, an intense band developed at 1236 cm-1 together
with less intense bands at 1313 and around 1550 cm-1. Most
researchers assign the peak at 1230 cm-1 to a bridging nitrite
species.25,26 These bands decreased in intensity with an increase
in the temperature and almost disappeared at 500 °C. These
results are in good agreement with the TPD results; however,
small absorption bands that can be attributed to nitrate species
were still observed at 500 °C.
Different spectra were observed when NO was fed in the
presence of O2; five bands appeared at 1244, 1298, ∼1560,
∼1600, and ∼1620 cm-1 at 300 °C. Because the wave numbers
are essentially identical to those of the bands observed when
NO2 was adsorbed on alumina,27 all of the observed bands can
be assigned to nitrate species. The intensities of the bands
decreased as the temperature increased, and species that exhibit
bands at 1244 and ∼1560 cm-1 remained after heat treatment
at 500 °C. Although the assignment of the bands of surface
nitrate species is a controversial issue,15,17 we temporarily
assigned these two bands to asymmetric and symmetric vibration
modes of bridging bidentate nitrate species because this species
is the most stable among the possible structures of surface nitrate
species.26
2NO + 3/2O2 + 2M-OH f 2M-O-NO2 + H2O
(7)
3.4. Reaction between Methane and Surface Species
Derived from NO-O2. The reaction between methane and
surface species derived from NO-O2 was investigated. After
ST(0.3) was treated with NO in the presence of O2 at 100 °C,
CH4/He was introduced on the catalyst, and the composition of
the effluent gas was analyzed by the Q-Mass detector (Figure
6). Similar to the NO-O2 TPD profile (Figure 3c), desorption
of NO was observed at around 200 °C, indicating that a higher
temperature is required to proceed the reaction of CH4 with
adsorbed NO2. However, the concentration of NO in the effluent
gas drastically decreased above 250 °C with a concomitant
decrease in CH4, suggesting that adsorbed NO2 was reduced
with CH4. In the temperature range from 250 to 450 °C, N2,
N2O, CO2, and CO were formed, each giving two peaks at low
and high temperatures. The lower-temperature peaks may be
due to the reaction of methane with adsorbed NO2, whereas
higher-temperature peaks can be attributed to the reaction of
methane with surface nitrate species. These results show a sharp
contrast with the results reported by He et al.,15 who showed
Figure 5 shows the time-course change in the FT-IR spectra
of surface species formed on ST(0.3) by exposure to an
NO-O2-He flow at 100 °C. Nitrite species were formed first,
which were converted to other species after 1 h of exposure.

7026 J. Phys. Chem. A, Vol. 113, No. 25, 2009
Nakatani et al.
Figure 8. TPD profiles of NO on γ-Ga2O3-Al2O3 catalysts. The
catalysts were pretreated in a He flow at 650 °C for 30 min and exposed
to the gas mixture (1000 ppm NO and 6.7% O2 with He balance) at
100 °C for 1 h. After NO was purged with a He flow, the desorption
was carried out in the He flow (100 mL ·min-1) at a heating rate of 5
Figure 7. Reaction of methane with various oxidants on ST(0.3).
Reaction conditions: 2000 ppm CH4, 1000 ppm NO or NO2, and 2.5%
H2O with/without 6.7% O2 balanced with He; SV ) 11 000 h-1
.
that IR spectra of the surface nitrate species adsorbed on their
Ga2O3/Al2O3 catalyst were not changed after exposure to C3H8
at 300 °C. From the results, they concluded that those surface
nitrate species are “spectators” and are not involved in the SCR
reaction. However, the present results clearly showed that
surface nitrate species do react with methane in the temperature
range where the catalyst showed high activity for methane-SCR
of NO.
The same result was obtained when ST(0.3) impregnated with
a small amount of Al(NO3)3 or Ga(NO3)3 was heated in a CH4/
He flow (data not shown). However, low-temperature desorption
of NO without the formation of O2 was not observed. When
NO was adsorbed in the absence of O2, the formation of N2
was scarcely observed because NO has only low oxidizing
ability. Therefore, surface nitrate species play a crucial role in
methane-SCR of NO in the presence of O2.
Figure 7 shows methane oxidation with various oxidants on
ST(0.3). Whereas NO had essentially no reactivity, even at 600
°C, methane was oxidized by NO2 at a temperature as low as
350 °C, even in the absence of O2. Although the concentration
of NO2 fed to the reactor was much lower than that of O2, the
amount of methane oxidized by NO2 at 350 °C was much larger
than that oxidized by O2, indicating that the oxidation ability
of NO2 was superior to that of O2. Therefore, NO2 is highly
reactive for the oxidation of CH4. At higher temperatures,
however, the methane conversion by NO2 rather decreased,
probably because of its decomposition to NO and O2 (eq 8)
°C·min-1
.
tion of surface nitrate species. This argument is supported by
the fact that both of the reactions (NO2 decomposition and
decomposition of surface nitrate species yielding NO and O2)
started at the same temperature (∼375 °C).
In the presence of O2, both NO and NO2 exhibited high
reactivity for the oxidation of methane in the whole temperature
range. This fact again shows the importance of oxygen in
methane-SCR of NO in the temperature range of 350-500 °C.
In this temperature range, NO2 + O2 was slightly more reactive
toward methane than NO + O2, indicating that the participation
of NO2 is important at lower temperatures.
3.5. Sites for the Formation of Surface Nitrate Species.
Because the peak at around 450 °C in the NO-O2 TPD profiles
is due to surface nitrate species, the number of surface sites
giving these species can be estimated from the intensity of this
peak. Figure 8 shows the NO-O2 TPD profiles of the
γ-Ga2O3-Al2O3 catalysts with various Ga/Al ratios. The largest
peak was observed for ST(0.3), indicating that this catalyst had
the largest number of the sites for the formation of surface nitrate
species. The order of the peak intensity, ST(0.3) > ST(0.5) >
ST(0.25) > ST(0.0) > ST(0.75) > ST(1.0), was in good
agreement with the order of catalytic activity (Figure 2),
indicating that the sites for the formation of surface nitrate
species play an important role in the methane-SCR of NO.
Similar results were reported by Haneda et al.4 for the
Ga2O3-Al2O3 catalyst prepared by a sol-gel method. However,
the optimal composition is different between theirs and ours
presumably because of the difference in the catalyst synthesis
processes.
As the Al content in the catalyst increased, the desorption
peak shifted to the higher temperature side, indicating that Al
interacts more strongly with surface nitrate species than Ga.
Moreover, the amounts of NO desorbed from the Ga-rich
catalysts were remarkably smaller than those from the other
catalysts. This result as well as the lower desorption temperature
suggests that surface nitrate species anchored on the Ga sites
have a character that is significantly different from those of the
nitrate species anchored on the Al sites.
2NO2 f 2NO + O2
(8)
This reaction can occur either in the gas phase or on the
catalyst, and the gas-phase reaction is well established because
the reverse reaction of eq 8 is an important reaction for the
synthesis of nitric acid.28 To verify the catalytic activity for
reaction eq 8, we examined NO2 decomposition on the catalyst
(Figure S4 in the Supporting Information); the result clearly
shows that the present catalyst significantly facilitates the
reaction. Therefore, the decrease in CH4 conversion at high
temperatures is caused by the decomposition of NO2 on the
catalyst. From the microscopic reversibility, the catalyst should
also facilitate the reverse reaction of eq 8, which is an important
process for the formation of surface nitrate species from NO
and O2. As discussed earlier, surface nitrate species, which can
be formed by exposure of the catalyst to either NO + O2 or
NO2, decompose into NO and O2 at higher temperatures;
therefore, decomposition of NO2 can take place via decomposi-
Because the formation of surface nitrate species is ac-
companied by consumption of the isolated hydroxyl groups, the
FT-IR spectra in the OH stretching vibration region of the
catalysts with various Ga contents were examined after exposure
to NO-O2 (Figure S5 in the Supporting Information). Although
the spectra are not well resolved, two negative bands are

Methane-SCR of NO on γ-Ga2O3-Al2O3
J. Phys. Chem. A, Vol. 113, No. 25, 2009 7027
CH4 + 2O2 f CO2 + 2H2O (9)
Therefore, the acidity of the catalysts used in this work was
assessed by the NH3-TPD method (data not shown). For the
catalysts with low Ga contents, the acid density monotonically
increased with the increase in the Ga content, and ST(0.5) had
the largest number of acid sites. This result explains the reason
why ST(0.5) exhibited the highest methane conversions at higher
temperatures, as shown in Figure 9.
Lewis acid properties of Ga2O3-Al2O3 catalysts as well as
Al2O3 and Ga2O3 have been well documented.22,32-34 These
Lewis acid properties originate from coordinatively unsaturated
ions and tetrahedral Ga3+ ions emerging at the extended crystal
surface.32-34
Figure 9. CH4 conversion in methane-SCR of NO on the
γ-Ga2O3-Al2O3 catalysts. Reaction conditions are shown in the caption
of Figure 2.
The acid strength was assessed by the desorption temperature
of NH3. ST(0.3) exhibited the highest acid strength (peak
temperature at which the maximum rate for desorption of NH3
was observed was 255 °C). The high performance of this catalyst
for methane-SCR can be explained by the fact that these highly
acidic sites are effective for methane activation relating to
methane-SCR of NO. When pure γ-Ga2O3 and γ-Al2O3 are
compared (peak temperatures were 237 and 207 °C, respec-
tively), γ-Ga2O3 has higher acid strength. However, the acid
strengths of the mixed oxides were higher than those of the
single-component catalysts, indicating that the synergistic effect
of Ga and Al determines the acid nature of the catalysts.
To examine the interaction of methane with the catalysts,
2000 ppm CH4 in He was fed on the catalysts with increasing
reaction temperature of 5 °C·min-1 (Figure 10). The methane
concentration first decreased (at ∼200 °C) and then increased
(at ∼400 °C), indicating that adsorption and desorption of
methane took place at these temperatures. The relatively high
adsorption temperature suggests that activation energy is
required for methane to be adsorbed on the catalyst. The largest
methane uptake was observed for ST(0.5) that has the largest
number of acid sites. It must be noted that methane was not
adsorbed on ST(0.0) at all, indicating that the adsorption of
methane takes place on the Ga sites rather than on the Al sites.
The decrease in methane concentration was also observed at
above 600 °C, and CO and H2 were formed at these temperatures
(Figure S8 in the Supporting Information). This result indicates
that the reduction of the catalyst occurred. However, ST(0.0)
showed neither consumption of methane (Figure 10) nor CO
formation (Figure S8 in the Supporting Information), indicating
that Al2O3 does not have a redox property. The consumption
of methane at 700 °C increased with the increase in Ga content
in the catalyst, and accordingly, the formation of CO increased.
Therefore, Ga2O3 does have a redox property, but Al2O3 does
not.
apparently observed at 3768 and 3686 cm-1. The band at 3768
cm-1 was observed for the Al-rich catalysts but not for the Ga-
rich catalysts. On the contrary, the 3686 cm-1 band was observed
for the Ga-rich catalyst but not for the Al-rich catalyst.
Therefore, the former band was assigned to isolated Al-OH
and the latter band was assigned to isolated Ga-OH.15,30
3.6. Base Property of Isolated Hydroxyl Group. The
basicity of the catalysts was investigated by CO2-TPD (Figure
S6 in the Supporting Information). Pure Ga2O3 adsorbed only
a small amount of CO2, and the CO2 uptake corresponding to
the number of base sites increased monotonically with the
increase in Al content in the catalyst (Figure S6 in the
Supporting Information).
The FT-IR spectra of CO2 species adsorbed on the catalysts
with various Ga contents exhibited bands due to hydrogen
carbonate species together with a negative band due to isolated
Al-OH for the catalysts containing Al (Figure S7 in the
Supporting Information). Hydrogen carbonate is supposed to
be formed when the hydroxyl group with a basic nature interacts
with CO2.32 For pure Ga2O3, no negative band due to isolated
GaOH was observed (Figure S7 in the Supporting Information)
because of low CO2 uptake on this catalyst (Figure S6 in the
Supporting Information). Because both the nitrate and hydrogen
carbonate species seem to be formed on the same Al-OH sites,
the basicity of the group plays an important role in the formation
of surface nitrate species. However, basicity is not the sole factor
because the maximum amount of surface nitrate species was
formed on ST(0.3) despite the fact that the maximum CO2
uptake was observed for ST(0.0); the combination between base
sites and Ga sites seems to be important.
Although Al2O3 is reduced to Al2O or AlO·Al2O3 by
treatment with Al metal at high temperature,35 the reduction of
Al2O3 cannot occur under the present reaction conditions
because the reduction potential of Al3+ is highly negative.
However, Ga3+ has a less-negative reduction potential, and
Ga2O3 can be reduced to gallium oxides with lower oxidation
states under conditions similar to the present one.36 He et al.15
discussed the reaction mechanism of propane-SCR of NO on
the assumption that neither Ga2O3 nor Al2O3 are redox-active
materials. However, Ga2O3 has a redox property. This property
seems to be important not only for the activation of methane
but also for the oxidation of NO to surface nitrate species. The
smaller amount of surface nitrate species formed on the catalysts
with high Al contents may be due to the lack of this redox
property despite the fact that they had much larger numbers of
base sites.
3.7. Activation of Methane. Figure 9 depicts the conversion
of methane in methane-SCR of NO shown in Figure 2. At 550
and 600 °C, the highest methane conversion was observed on
ST(0.5), whereas ST(0.3) exhibited the highest methane conver-
sion at 450 and 500 °C. In the previous work,10 we examined
the correlation between the performance for methane-SCR and
the acidity of the catalysts prepared solvothermally in various
media and found that the surface acid density of the catalyst is
predominantly determined by the reaction medium used for the
catalyst synthesis. Diethylenetriamine adopted in this work gives
the catalyst with the smallest surface acid density. We also found
that the combustion of methane occurs competitively with
methane-SCR of NO, especially at high temperatures, and that
a higher acid density facilitates the combustion of methane, thus
leading to lower efficiency of methane used for SCR.10

7028 J. Phys. Chem. A, Vol. 113, No. 25, 2009
Nakatani et al.
property developed by Ga ions. This redox property is important
for both the oxidation of NO to surface nitrate species and the
activation of methane. The most important factor of this catalyst
is the presence of the site for the formation of surface nitrate
species (octahedral Al) near the methane activation site (tetra-
hedral Ga). Therefore, the ST(0.3) that has the highest number
of Ga-Al ensembles exhibited the highest performance for
methane-SCR of NO.
Acknowledgment. This work was partially supported by a
Grant-in-Aid for Scientific Research from the Ministry of
Education, Culture, Sports, Science and Technology, Japan. We
are grateful to Prof. S. Imamura for his kind help and
discussions. The XAFS experiments were performed at BL16B2
in SPring-8 with the approval of the Japan Synchrotron
Radiation Research Institute (JASRI) (proposal no. C05A16B2-
4050-N).
Figure 10. Temperature-programmed reaction of CH4 with the
γ-Ga2O3-Al2O3 catalysts. The catalysts (0.5 g) were heated in a 100
mL·min-1 flow (SV ) 11 000 h-1) of 2000 ppm CH4/He at a heating
rate of 5 °C·min-1
.
Supporting Information Available: XRD patterns, XAFS
spectra, decomposition of NO2, FT-IR spectra, TPD profiles,
and formation of CO. This material is available free of charge
via the Internet at http://pubs.acs.org.
References and Notes
Figure 11. Proposed active site of the γ-Ga2O3-Al2O3 catalyst for
methane-SCR of NO.
(1) (a) Iwamoto, M.; Yahiro, H. Catal. Today 1994, 22, 5. (b) Armor,
J. N. Catal. Today 1995, 26, 147. (c) Li, Y.; Armor, J. N. J. Catal. 1995,
145, 1.
3.8. Active Site of the Catalyst for Methane-SCR of NO.
The active site of the catalyst for methane-SCR of NO was
proposed as depicted in Figure 11. First, NO is adsorbed on
the isolated hydroxyl group bonded to Al that acts as a base
site. Then, the adsorbed NO is oxidized to surface nitrate species
by O2; here the presence of Ga with a redox property is quite
important. However, methane is activated on the tetrahedral Ga
site (coordinatively unsaturated site); this site acts not only as
a Lewis acid site but also as a redox site. The most important
factor is the presence of the surface nitrate species formed on
the Al sites neighboring the tetrahedral Ga sites that activate
methane. When the site for the formation of surface nitrate
species is not present near the methane activation site, the ac-
tivated methane will react with only O2 in the gas phase
(combustion). Although ST(0.5) has a larger number of acid
sites (that is, methane activation sites), the number of Ga-Al
ensembles suitable for the cooperation is small compared with
ST(0.3). Therefore, ST(0.3) has higher performance than ST(0.5)
despite the fact that the higher methane conversion was observed
on the latter catalyst.
(2) Shimizu, K.; Satsuma, A.; Hattori, T. Appl. Catal., B 1998, 16,
319.
(3) Shimizu, K.; Takamatsu, M.; Nishi, K.; Yoshida, H.; Satsuma, A.;
Tanaka, T.; Yoshida, S.; Hattori, T. J. Phys. Chem. B 1999, 103, 1542.
(4) Haneda, M.; Kintaichi, Y.; Mizushima, T.; Kakuta, N.; Hamada,
H. Appl. Catal., B 2001, 31, 81.
(5) (a) Lezcano, M.; Ribotta, A.; Miro, E.; Lombardo, E.; Petunchi,
J.; Moreaux, C.; Dereppe, J. M. J. Catal. 1997, 168, 511. (b) Sowade, T.;
Schu¨tze, F.-W.; Berndt, H.; Gru¨nert, W. Chem. Eng. Technol. 2004, 27,
1277.
(6) Schulz, P.; Baerns, M. Appl. Catal. 1991, 78, 15.
(7) Tabata, T.; Kokitsu, M.; Okada, O. Appl. Catal., B 1995, 6, 225.
(8) (a) Inoue, M.; Kominami, H.; Inui, T. J. Chem. Soc., Dalton Trans.
1991, 3331. (b) Inoue, M. J. Phys.: Condens. Matter 2004, 16, S1291. (c)
Iwamoto, S.; Inoue, M. J. Jpn. Pet. Inst. 2008, 51, 143.
(9) (a) Inoue, M.; Inoue, N.; Yasuda, T.; Takeguchi, T.; Iwamoto, S.
AdV. Sci. Technol. 2000, 29, 1421. (b) Takahashi, M.; Inoue, N.; Nakatani,
T.; Takeguchi, T.; Iwamoto, S.; watanabe, T.; Inoue, M. Appl. Catal., B
2006, 65, 142.
(10) Takahashi, M.; Nakatani, T.; Iwamoto, S.; Watanabe, T.; Inoue,
M. Appl. Catal., B 2007, 70, 73.
(11) Takahashi, M.; Inoue, N.; Takeguchi, T.; Iwamoto, S.; Watanabe,
T.; Inoue, M. J. Am. Ceram. Soc. 2006, 89, 2158.
(12) (a) Takahashi, M.; Nakatani, T.; Iwamoto, S.; Watanabe, T.; Inoue,
M. Ind. Eng. Chem. Res. 2006, 45, 3678. (b) Takahashi, M.; Nakatani, T.;
Iwamoto, S.; Watanabe, T.; Inoue, M. J. Phys.: Condens. Matter 2006, 18,
5745.
(13) Miyahara, Y.; Takahashi, M.; Masuda, T.; Imamura, S.; Kanai, H.;
Iwamoto, S.; Watanabe, T.; Inoue, M. Appl. Catal., B 2008, 84, 289.
(14) Shimizu, K.; Satsuma, A.; Hattori, T. Catal. SurV. Jpn. 2000, 4,
115.
(15) He, C.; Paulus, M.; Find, J.; Nickl, J. A.; Eberle, H.; Spengler, J.;
Chu, W.; Ko¨hler, K. J. Phys. Chem. B 2005, 109, 15906.
(16) Haneda, M.; Joubert, E.; Me´ne´zo, J.; Duprez, D.; Barbier, J.; Bion,
N.; Daturi, M.; Saussey, J.; Lavalley, J.; Hamada, H. J. Mol. Catal. A:
Chem. 2001, 175, 179.
(17) Haneda, M.; Bion, N.; Daturi, M.; Saussey, J.; Lavalley, J.; Duprez,
D.; Hamada, H. J. Catal. 2002, 206, 114.
(18) Miyahara, Y.; Watanabe, T.; Masuda, T.; Kanai, H.; Deguchi, H.;
Inoue, M. J. Catal. 2008, 259, 36.
(19) Sault, A. G.; Madix, R. J. J. Phys. Chem. 1986, 90, 4723.
(20) Moulder, F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Handbook
of X-ray Photoelectron Specrtroscopy; Perkin-Elmer Co.: Eden Prairie, MN,
1992.
4. Conclusions
The γ-Ga2O3-Al2O3 catalysts prepared by the solvothermal
reaction of gallium acetylacetonate and aluminum isopropoxide
in diethylenetriamine had a spinel structure in which Ga3+ and
Al3+ ions tend to occupy the tetrahedral and octahedral sites,
respectively. When the Ga/(Ga+Al) ratio was 0.3, almost all
tetrahedral sites and octahedral sites were occupied by Ga3+
and Al3+ ions, respectively, and the catalyst exhibited the highest
performance for methane-SCR of NO. The active site was
deduced to be related to the combination of the tetrahedral Ga3+
ion with octahedral Al3+ ions in the next-nearest-neighbor sites.
The following reaction mechanism was proposed: On one hand,
NO is adsorbed on the isolated hydroxyl groups on Al ions that
act as the base sites and then oxidized to the surface nitrate
species by O2. On the other hand, methane is activated on Lewis
acid sites that are supposed to originate from tetrahedral Ga
ions. In addition, the γ-Ga2O3-Al2O3 catalyst has a redox
(21) Ravel, B.; Newville, M. J. Synchrotron Radiat. 2005, 12, 537.
(22) Otero Area´n, C.; Rodr´ıguez Delgado, M.; Montouillout, V.; Massiot,
D. Z. Anorg. Allg. Chem. 2005, 631, 2121.
(23) Sickafus, K. E.; Wills, J. M.; Grimes, N. M. J. Am. Ceram. Soc.
1999, 82, 3279.

Methane-SCR of NO on γ-Ga2O3-Al2O3
J. Phys. Chem. A, Vol. 113, No. 25, 2009 7029
(24) Huang, H. Y.; Yang, R. T. Langmuir 2001, 17, 4997.
(32) Vimont, A.; Lavalley, J. C.; Sahibed-Dine, A.; Otero Area´n, C.;
Rodr´ıguez Delgado, M.; Daturi, M. R. J. Phys. Chem. B 2005, 109, 9656.
(33) (a) Rodr´ıguez Delgado, M.; Morterra, C.; Cerrato, G.; Magnacca,
G.; Otero Acea´n, C. Langmuir 2002, 18, 10255. (b) Lavalley, J. C.; Daturi,
M.; Montouillout, V.; Clet, G.; Otero Area´n, C.; Rodríguez Delgado, M.;
Sahibed-dine, A. Phys. Chem. Chem. Phys. 2003, 5, 1301.
(34) Pushkar, Y. N.; Sinitsky, A.; Parenago, O. O.; Kharlanov, A. N.;
Lunina, E. V. Appl. Surf. Sci. 2000, 167, 69.
(35) Hoch, M.; Johnston, H. L. J. Am. Chem. Soc. 1954, 76, 2560.
(36) (a) Biscardi, J. A.; Iglesia, E. Catal. Today 1996, 31, 207. (b)
Meitzner, G. D.; Iglesia, E.; Baumgartner, J. E.; Huang, E. S. J. Catal.
1993, 140, 209.
(25) (a) Centi, G.; Perathoner, S.; Biglino, D.; Giamello, E. J. Catal.
ˇ
1995, 151, 75. (b) Sazama, P.; Capek, L.; Drobna´, H.; Sobal´ık, Z.; Deˇdecˇek,
J.; Arve, K.; Wichterlova´, B. J. Catal. 2005, 232, 302.
(26) Hadjiivanov, K. I. Catal. ReV.-Sci. Eng. 2000, 42, 71.
(27) Anderson, J. A.; Millar, G. J.; Rochester, C. H. J. Chem. Soc.,
Faraday Trans. 1990, 86, 571.
(28) Tsukahara, H.; Ishida, T.; Mayumi, M. Nitric Oxide 1991, 3, 191.
(29) Busca, G.; Lorenzelli, V.; Ramis, G.; Willey, R. J. Langmuir 1993,
9, 1492.
(30) Otero Area´n, C.; Bellan, A. L.; Mentruit, M. P.; Rodr´ıguez Delgado,
M.; Palomino, G. T. Microporous Mesoporous Mater. 2000, 40, 35.
(31) Nortier, P.; Fourre, P.; Saad, A.; Saur, O.; Lavalley, J. C. Appl.
Catal. 1990, 61, 141.
JP901569S