Selective reduction of NO with propene over Ga2O3-Al2O3: Effect of sol-gel method on the catalytic performance

Article abstract of DOI:10.1006/jcat.2000.2831

SCR of NO_x to N₂ by hydrocarbons has received extensive attention because of its potential for practical applications to diesel and lean-burn engine exhausts. The advantage of this process is the ability to reduce NO_x in t

Full text of DOI:10.1006/jcat.2000.2831

Journal of Catalysis 192, 137–148 (2000)
doi:10.1006/jcat.2000.2831, available online at http://www.idealibrary.com on
Selective Reduction of NO with Propene over Ga2O3–Al2O3:
Effect of Sol–Gel Method on the Catalytic Performance
Masaaki Haneda,1 Yoshiaki Kintaichi, Hiromichi Shimada, and Hideaki Hamada
National Institute of Materials and Chemical Research, 1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan
Received September 29, 1999; revised January 21, 2000; accepted January 25, 2000
Since Cu–ZSM-5 was reported as a highly active catalyst
by Held et al. (4) and lwamoto et al. (5) following earlier
patents from Volkswagen (6) and Toyota (7), a lot of in-
vestigations have focused on zeolite-based catalysts (8–11).
Although metal ion-exchanged zeolites showed excellent
activity at high space velocities, their hydrothermal insta-
bility due to dealumination from the zeolite framework is
a big problem (12, 13).
In addition to zeolite-based catalysts, metal oxide-based
catalysts such as single metal oxides (14–17), binary ox-
ides (18), and metal-supported oxides (19–38) have also
been found to catalyze the selective reduction of NO with
hydrocarbons or oxygenated hydrocarbons. In particular,
supported alumina catalysts such as Co/Al2O3 (20–22, 26,
33), Sn/Al2O3 (23, 25), and Ag/Al2O3 (27, 29, 30–32, 34, 35)
are regarded as promising candidates for practical applica-
tions.
The catalytic activity and selectivity of metal-supported
alumina are well known to depend strongly upon the dis-
persion, loading, and local structure of metal additives.
For example, the activity of Co/Al2O3, which is one of the
catalysts studied most extensively, is affected strongly by
the preparation method and the calcination temperature.
When Co/Al2O3 was prepared by the sol–gel method (21)
or was heated at temperature higher than 873 K (21, 22,
26, 33), fairly high catalytic activity was attained. The high
activity should be due to highly dispersed cobalt species
on alumina. Recently, Yan et al. (26) proposed the impor-
tance of the coordination number of Co2+ ions in CoAl2O4
and small Co3O4 particles as a factor to determine the NO
reduction activity. According to their report, dispersed oc-
tahedrally coordinated Co2+ ions are responsible for the
catalytic activity.
Shimizu et al. (36, 37) reported interesting catalytic be-
havior of Ga/Al2O3 prepared by the impregnation method
for NO reduction by methane. They found that the activity
of Ga/Al2O3 increased with increasing gallium loading up
to about 30 wt% , at which a theoretical monolayer coverage
of gallium oxide on alumina is formed. According to their
considerations, gallium oxide highly dispersed on alumina
is responsible for the high activity and selectivity. The
The effect of the preparation method on the activity of Ga2O3–
Al2O3 for the selective reduction of NO with propene was inves-
tigated. Ga2O3–Al2O3 prepared by the sol–gel method (Ga2O3–
Al2O3(S)) exhibited excellent activity for NO reduction, compared
with Al2O3, Ga2O3, and impregnated Ga2O3/Al2O3 (Ga2O3/
Al2O3(I)). XRD measurements of Ga2O3–Al2O3(S) indicated that a
part of the Al3+ ions in Al2O3 are substituted by Ga3+ ions, resulting
in the formation of a composite oxide, [GaxAl(1 x)]2O3 (x < 1). On
the other hand, Ga2O3 particles and Al2O3 particles were found to
be present separately for Ga2O3/Al2O3(I). The high catalytic activity
of Ga2O3–Al2O3(S) was accounted for by the high surface area and
the nature of the [GaxAl(1 x)]2O3 (x < 1) composite oxide, which was
formed uniformly by the present sol–gel method. The comparison
of the reactivity between NO and NO2 suggested the participation
of NO2 in NO reduction. The kinetic order for NO reduction was
found to be 0.3 with respect to propene and 0.7 with respect to
oxygen on both Ga2O3/Al2O3(I) and Ga2O3–Al2O3(S). However, a
significant difference in the reaction order with respect to NO was
observed between Ga2O3/Al2O3(I) and Ga2O3–Al2O3(S), 0.4 order
on the former catalyst, and zero order on the latter one, indicat-
ing the presence of strongly adsorbed NOx species on the catalyst
surface. On the basis of these results, a reaction mechanism was pro-
posed in which the NOx adspecies formed through NO oxidation is
reduced to nitrogen via organic intermediates containing nitrogen
c
and oxygen.
2000 Academic Press
Key Words: nitrogen monoxide; selective reduction; propene;
Ga2O3–Al2O3; sol–gel method; kinetic studies.
1. INTRODUCTION
Selective catalytic reduction of nitrogen oxides (NOx)
to N2 by hydrocarbons has recently received extensive at-
tention because of its potential for practical applications to
diesel and lean-burn engine exhausts (1). The principal ad-
vantage of this process is its ability to reduce NOx in the
presence of oxygen (2, 3) and the easiness of handling hy-
drocarbons as reductant compared with ammonia which is
currently used for NOx removal for stationary engines and
boilers.
1 To whom correspondence should be addressed. Fax: +81-298-54-4487.
E-mail: hane@nimc.go.jp.
137
0021-9517/00 $35.00
c
Copyright
2000 by Academic Press
All rights of reproduction in any form reserved.

138
HANEDA ET AL.
preparation method is also presumed to affect the catalytic or 9.1% H2O diluted in helium as a balance gas was fed to
1
activity of Ga/Al2O3. In fact, we found that Ga2O3–Al2O3 0.2 g of a catalyst at a rate of 66 cm3 min (W/F = 0.18 g
prepared by the sol–gel method exhibited much higher ac- scm 3). H2O was introduced into the reaction gas mixture
tivity for NO reduction by propene than that prepared by with a micropump. When H2O was used, the feed gas flow
the impregnation method (38). In the present study, we rate and the concentrations of the other gas components
have investigated various factors affecting the catalytic ac- were kept the same by controlling the helium flow rate.
tivity of Ga2O3–Al2O3 for the selective reduction of NO The reaction temperature was changed stepwise from 873
with propene in detail. Several characterization techniques to 573 K with a 50 K step. The effluent gas was analyzed
were employed to reveal important factors related to the by gas chromatogrphy (Shimadzu GC8A TCD). A molec-
catalytic activity. We also performed kinetic studies to dis- ular sieve 5A column was used for the analysis of N2 and
cuss the mechanistic features of NO reduction on Ga2O3– CO and a Porapak Q column for that of N2O, CO2, and
Al2O3.
C3H6. The catalytic activity was evaluated in terms of NOx
(NO or NO2) conversion to N2 and that of propene to COx
(CO + CO2). The formation of N2O was found negligible
in the present study.
2. EXPERIMENTAL
2.1. Catalyst Preparation
2.2.2. Kinetic studies. The reaction rate of N2 forma-
tion was measured by changing the catalyst weight from
0.01 to 0.2 g to obtain the conversion of NO in the range of
10–30% . The kinetic parameters were determined at 723 K
by assuming the following rate equation:
Al2O3 was synthesized from aluminum boehmite sol
prepared by hydrolysis of aluminum(III) triisopropoxide
(AIP) in hot water (363 K) with a small amount of nitric
acid (39–41). The sol solution thus obtained was heated un-
der reduced pressure to remove the solvents. The residue
was dried at 383 K in an oven and then calcined at 873 K
for 5 h in flowing air. Ga2O3 was prepared from gallium
hydroxide which was obtained by adding urea to an aque-
ous solution of gallium(III) nitrate and stirring at 363 K for
10 h. The precipitate thus obtained was washed with dis-
tilled water, followed by drying at 383 K and calcination at
873 K for 5 h in flowing air.
Ga2O3–Al2O3 catalysts were prepared by two differ-
ent methods (abbreviated as Ga2O3/Al2O3(I) and Ga2O3–
Al2O3(S)). Ga2O3/Al2O3(I) was prepared by impregnat-
ing the above-mentioned alumina with an aqueous so-
lution of gallium(III) nitrate. Ga2O3–Al2O3(S) was pre-
pared by the sol–gel method. A solution of gallium(III)
nitrate dissolved in ethylene glycol was first added into
the aluminum boehmite sol solution produced in the
manner as described above. After the solution was
stirred at room temperature for 1 day, the solvents were
eliminated by heating under reduced pressure. The re-
sulting catalyst precursors were dried at 383 K, fol-
lowed by calcination at 873 K for 5 h in flowing air.
Some catalyst samples were further calcined at 1073 and
1273 K for 5 h in air. Basically, the gallium loading
for Ga2O3/Al2O3(I) and Ga2O3–Al2O3(S) was fixed at
30 wt% as Ga2O3, although Ga2O3–Al2O3(S) with dif-
ferent Ga2O3 content from 5 wt% to 70 wt% was also
prepared.
rN = kPNO
P
PO exp( Ea/RT ).
[1]
2
C3H6
2
For this purpose, the concentrations of the reactants were
changed in the range of 200–1500 ppm for NO, 200–
1500 ppm for C3H6, and 3–10% for O2. The total flow rate
was 66 cm3 min 1. The standard reaction conditions were
900 ppm NO, 900 ppm C3H6, and 10% oxygen. The activa-
tion energy, Ea in the rate equation [1], was calculated at
temperatures ranging from 673 to 773 K under the standard
reaction conditions.
2.3. Catalyst Characterization
BET surface area was measured with a conventional flow
apparatus (Micromeritics, Flowsorb II 2300) by nitrogen
adsorption at 77 K. The nitrogen adsorption isotherm at
77 K was measured volumetrically with a BELSORP 28SA.
The pore size distribution curves of the catalyst were calcu-
lated, on the basis of Dollimore–Heal (D–H) theory, from
the desorption branches of the isotherms. The crystal struc-
ture was identified by XRD (Shimadzu XD-D1) measure-
ments using Cu K radiation at 40 kV and 40 mA. The sur-
face morphology of Ga2O3–Al2O3 was observed by SEM
(Hitachi S-5000) operated at an accelerating voltage of
15 kV. TEM analysis was made with a JEOL JEM-2000FX
and at an accelerating voltage of 200 kV.
XPS spectra in the region of Ga 2p and Al 2p were
recorded by using a VG Scientific ESCA-5500 with an
Mg K X-ray source (h = 1253.6 eV) operated at 15 kV
2.2. Catalytic Activity Measurements
2.2.1. Evaluation of NO reduction activity. The cata- and 20 mA. The binding energy was corrected by the
lyticactivitywasmeasured byusinga fixed-bed flowreactor. contaminated carbons (284.6 eV). The peak area ratio
A feed gas mixture containing nitrogen oxides (900 ppm (Ga 2p/Al 2p) was corrected using relative atomic sensitive
NO or 930 ppm NO2), 900 ppm C3H6, 10% oxygen, and 0 factors.

NO REDUCTION BY PROPENE OVER Ga2O3–Al2O3
139
2.4. NO TPD Measurements
Temperature-programmed desorption (TPD) experi-
ments of NOx were carried out by using 50 mg of a sample.
The sample was pretreated in a flow of 10% O2/He at 873 K
for 1 h and then cooled to room temperature. NOx adsorp-
tion was performed by passing a gas mixture containing
1000 ppm NO and 10% O2 diluted in He through the sample
bed at room temperature for 2 h. After the adsorption gas
was purged with He until no NO was detected in the efflu-
ent, the TPD measurement was carried out up to 873 K with
a heating rate of 5 K min 1 in flowing He. The gas flow rate
was fixed at 60 cm3 min 1. The desorbed species with mass
numbers of 30 (NO), 32 (O2), 44 (N2O), and 46 (NO2) were
monitored continuously by a quadrupole mass spectrome-
ter (ANELVA M-QA200TS) as a function of temperature.
3. RESULTS
3.1. Catalytic Activity of Ga2O3–Al2O3
3.1.1. Effect of preparation method. Figure 1 compares
the catalytic activities of Al2O3, Ga2O3/Al2O3(I), Ga2O3–
Al2O3(S), and Ga2O3 for NO reduction by propene in the
absence of H2O. The NO reduction activity of Al2O3 was
higher than that of Ga2O3, although the active tempera-
ture region was almost the same. Apparently, the addition
of Ga2O3 into Al2O3 by the impregnation method caused
an enhancement of NO conversion. No great difference in
propene conversion on these three catalysts was observed.
On the other hand, Ga2O3–Al2O3 prepared by the sol–
FIG. 2. Catalytic activity of Al2O3 (᭺), Ga2O3/Al2O3(I) (᭡), Ga2O3–
Al2O3(S) (ٗ), and Ga2O3 (᭜) for the selective reduction of NO with pro-
pene in the presence of H2O. Conditions: NO = 900 ppm, C3H6 = 900 ppm,
3
O2 = 10% , H2O = 9.1% , catalyst weight = 0.2 g, W/F = 0.18 g scm
.
gel method showed catalytic behavior different from that
of the other catalysts. It is obvious that Ga2O3–Al2O3(S)
shows extremely high activity for NO reduction to N2 in the
wide temperature region between 673 and 823 K. Propene
conversion on Ga2O3–Al2O3(S) was also much higher than
that on the other catalysts. These results indicate that the
preparation method affects strongly the catalytic activity of
Ga2O3–Al2O3.
Figure 2 shows the catalytic activity of Al2O3, Ga2O3/
Al2O3(I), Ga2O3–Al2O3(S), and Ga2O3 for NO reduction
by propene in the presence of H2O. It can be seen by com-
paring Figs. 1 and 2 that the activity of Ga2O3 was depressed
considerablybyaddition ofH2O. CoexistingH2O wasfound
to decrease the N2 formation rate over Ga2O3 at 723 K
by about 20% . Al2O3 and Ga2O3/Al2O3(I) gave almost
the same maximum NO conversion with different effective
temperatures for NO reduction. Their N2 formation rate
at 723 K in the presence of H2O was about 30% of that in
its absence. Apparently, Ga2O3–Al2O3(S) showed excellent
catalyticperformance compared to Al2O3, Ga2O3/Al2O3(I),
and Ga2O3, although its activity was also decreased by H2O.
In this case, the extent of the decrease of the N2 formation
rate at 723 K by coexisting H2O was estimated to be about
30% . This means that the tolerances toward H2O were very
similar for Al2O3, Ga2O3/Al2O3(I), and Ga2O3–Al2O3(S).
It was also found that the order of propene conversion on
the catalysts is the same irrespective of coexisting H2O.
The effect of H2O concentration on the activity of
Ga2O3–Al2O3(S) was examined. The reaction rate for N2
FIG. 1. Catalytic activity of Al2O3 (᭺), Ga2O3/Al2O3(I) (᭡), Ga2O3–
Al2O3(S) (ٗ), and Ga2O3 (᭜) for the selective reduction of NO with pro-
pene in the absence of H2O. Conditions: NO = 900 ppm, C3H6 = 900 ppm,
3
O2 = 10% , H2O = 0% , catalyst weight = 0.2 g, W/F = 0.18 g scm
.

140
HANEDA ET AL.
FIG. 4. Catalytic activity of Al2O3 (᭺), Ga2O3/Al2O3(I) (᭡), Ga2O3–
FIG. 3. Effect of calcination temperature on the activity of (a) Ga2O3/
Al2O3(S) (ٗ), and Ga2O3 (᭜) for the selective reduction of NO2 with
propene in the absence of H2O. Conditions: NO2 = 930 ppm, C3H6 =
Al2O3(I) and (b) Ga2O3–Al2O3(S) for NO reduction by propene in the
presence of H2O. The reaction conditions are the same as for Fig. 2. Cal-
cination temperature: (᭺) 873 K, (᭡) 1073 K, and (ٗ) 1273 K.
900 ppm, O2 = 10% , H2O = 0% , catalyst weight = 0.2 g, W/F = 0.18 g
3
scm
.
formation at 723 K decreased with increasing H2O con-
1
1
(3, 8, 10, 22, 33). Figure 5 shows the activity of Al2O3,
Ga2O3/Al2O3(I), Ga2O3–Al2O3(S), and Ga2O3 for NO oxi-
dation to NO2. Ga2O3–Al2O3(S) catalyzed most effectively
the oxidation of NO by O2 to NO2, while Al2O3, Ga2O3/
Al2O3(I), and Ga2O3 were not effective. This is a quite sur-
prising result because the latter three catalysts catalyzed
effectively NO reduction by propene to N2 but did not cata-
lyze NO oxidation to NO2. This result seems to contradict
the model proposed in which NO oxidation into NO2 is
one of the reaction steps. The low activity for NO oxidation
could be explained by the poisoning of the catalyst by NO2,
as suggested in the case of Ag/Al2O3 (29) and Co/Al2O3
centration as follows: 18.6 mol min
g
(H2O, 0% ),
1
1
1
1
6.81 mol min
g
(H2O, 0.9% ), 5.90 mol min
g
1
1
(H2O, 4.1% ), and 5.48 mol min
g
(H2O, 9.1% ). The
catalytic performance seems to be affected considerably by
the presence of a small amount of H2O.
3.1.2. Effect of calcination temperature. Figure 3 com-
pares the activity of Ga2O3/Al2O3(I) and Ga2O3-Al2O3(S)
calcined at 873, 1073, and 1273 K for NO reduction in the
presence of H2O. Obviously, both catalysts showed similar
catalytic behavior. No considerable activity depression was
observed after the calcination at 1073 K, while the calcina-
tion at 1273 K caused a decrease of the activity over the
entire temperature range.
3.1.3. Reduction of NO2 with propene. Figure 4 shows
the activities of Al2O3, Ga2O3/Al2O3(I), Ga2O3–Al2O3(S),
and Ga2O3 for NO2 reduction by propene in the absence
of H2O. As can be seen by comparing Fig. 1 with Fig. 4,
the replacement of NO with NO2 resulted in a consider-
able increase in N2 formation as well as in COx formation
over all the catalysts. The order of NO2 reduction activity
is as follows: Ga2O3 < Al2O3 < Ga2O3/Al2O3(I) < Ga2O3–
Al2O3(S). This is well consistent with the results for NO
reduction, suggesting the participation of NO2 in NO
reduction.
FIG. 5. Catalytic activity of Al2O3 (᭺), Ga2O3/Al2O3(I) (᭡), Ga2O3–
Al2O3(S) (ٗ), and Ga2O3 (᭜) for NO oxidation into NO2 in the absence of
3.1.4. NO oxidation to NO2. It is well known that NO
oxidation into NO2 is the initial and indispensable step
for NO reduction by hydrocarbons over several catalysts
H2O. Conditions: NO = 900 ppm, O2 = 10% , H2O = 0% , catalyst weight =
3
0.2 g, W/F = 0.18 g scm
.

NO REDUCTION BY PROPENE OVER Ga2O3–Al2O3
141
TABLE 1
Physical Properties of Al2O3, Ga2O3/Al2O3(I),
Ga2O3–Al2O3(S), and Ga2O3
BET surface area/
Mean pore
size/nm
Pore volume/
1
1
Catalyst
Al2O3
Ga2O3/Al2O3(I)
Ga2O3–Al2O3(S)
Ga2O3
m2 g
ml g
205
140
188
18
4.1
5.3
9.6
8.1
0.25
0.21
0.87
0.06
Note: Mean pore size and pore volume were calculated using
Dollimore–Heal (DH) theory from the desorption curve. Mean pore size
represents pore radius, where the peak intensity in Fig. 4 is at a maximum.
Pore volume is estimated in the region of pore radius less than 20 nm.
(42). A similar phenomenon might take place on the present
catalysts.
FIG. 7. XRD patterns of (a) Al2O3, (b) Ga2O3/Al2O3(I), (c) Ga2O3–
Al2O3(S), (d) Ga2O3, and (e) -Al2O3 and -Ga2O3 from JCPDS data
as reference (No. 10-425 and No. 20-426, respectively). (4) for -Al2O3,
(᭹) for -Ga2O3, (ٗ) for -Ga2O3, (᭜) for -Ga2O3, and (♦) for solid
solution (unidentified).
3.2. Catalyst Characterization
In Table 1 are given the surface area, mean pore size, and
pore volume of Al2O3, Ga2O3/Al2O3(I), Ga2O3–Al2O3(S),
and Ga2O3. The pore size and pore volume distribution
curves are shown in Fig. 6. The high surface area of Ga2O3–
Al2O3(S) was comparable to that of Al2O3, while the sur-
face area of Ga2O3/Al2O3(I) was much smaller than that of
Al2O3 probably because of a blockage of the pores of Al2O3
by supported Ga2O3. In fact, an enlargement of the mean
pore size and a decrease of the pore volume of Al2O3 were
recognized after impregnation of gallium (Fig. 6b). As can
be seen in Fig. 6c, Ga2O3–Al2O3(S) showed a broad pore
size distribution curve with no distinct mean pore size, sug-
gesting that no micro- and mesopores but only macropores
are present. It was also found that Ga2O3–Al2O3(S) has a
comparatively large pore volume. The BET surface area
of Ga2O3 was found to be 18 m2 g 1, which is very small
compared with that of the other catalysts.
Figure 7 shows X-ray diffraction patterns of Al2O3,
Ga2O3/Al2O3(I), Ga2O3–Al2O3(S), and Ga2O3. The crystal-
lite structure of Al2O3 was -Al2O3. Two phases, -Ga2O3
and -Ga2O3, were detected for Ga2O3. Two kinds of XRD
peaks assigned to -Al2O3 and -Ga2O3 were observed for
Ga2O3/Al2O3(I), suggesting that Al2O3 and Ga2O3 are pre-
sent independently. On the other hand, Ga2O3–Al2O3(S)
showed an XRD pattern different from that of Ga2O3/
Al2O3(I). Distinct peaks assigned to -Ga2O3 were not
detected, but a shift of diffraction peaks at 2 = 45.5 and
66.5 , which can be assigned to respective -Al2O3(400) and
(440), toward lower angles was recognized, suggesting
the formation of a solid solution of -Al2O3 and -Ga2O3.
SEM observation was performed for Ga2O3/Al2O3(I)
and Ga2O3–Al2O3(S) to examine the texture of the cata-
lyst surface. As depicted in Fig. 8a, the surface of Ga2O3/
Al2O3(I) was very rough and the presence of pores was
recognized. On the other hand, Ga2O3–Al2O3(S) showed
a quite different surface morphology and seems to consist
FIG. 6. Pore size distribution of (a) Al2O3, (b) Ga2O3/Al2O3(I), and
(c) Ga2O3–Al2O3(S).

142
HANEDA ET AL.
FIG. 8. SEM photographs of (a) Ga2O3/Al2O3(I) and (b) Ga2O3–Al2O3(S).
of fibrillar particles. Figure 9 shows TEM photographs of on the surface of Ga2O3/Al2O3(I). The binding energy of
Ga2O3/Al2O3(I) and Ga2O3–Al2O3(S). In the photograph Ga 2p increased in the following order: Ga2O3 < Ga2O3/
of Ga2O3/Al2O3(I) (Fig. 9a), a lot of nonuniform particles Al2O3(I) < Ga2O3–Al2O3(S). In Table 2 are also given the
with the lattice spacing of ca. 0.25 nm, which can be assigned Auger parameters of gallium species. The concept of an
to -Ga2O3(113), were observed. The presence of distinct Auger parameter, which is the sum of the binding en-
particles was not recognized in the photograph of Ga2O3– ergy of Ga 2p and the Auger kinetic energy of Ga (MNN),
Al2O3(S). This might be because a solid solution of Al2O3 was proposed by Wagner et al. (43). The Auger parameter
and Ga2O3 is formed as presumed from the XRD pattern. of gallium species decreased in the sequence of Ga2O3 >
Interestingly, Ga2O3–Al2O3(S) was also found to consist of Ga2O3/Al2O3(I) > Ga2O3–Al2O3(S). The Auger parame-
fibrillar particles from TEM observation.
ters of a target element are known to be different depending
We performed XPS measurements to reveal the elec- upon the correspondingcompound and the crystalstructure
tronic state of gallium species. XPS parameters obtained (44). A difference in the chemical characteristics of gallium
here are summarized in Table 2. The peak area ratio (Ga 2p/ species such as the crystallite structure and the coordination
Al 2p) for Ga2O3/Al2O3(I) was higher than that for Ga2O3– number between Ga2O3/Al2O3(I) and Ga2O3-Al2O3(S) is
Al2O3(S), suggesting that most of the Ga2O3 was deposited also deduced from XPS measurements.
TABLE 2
Summary of Line Energies Obtained for Ga2O3, Ga2O3/Al2O3(I), and Ga2O3–Al2O3(S)
Peak area ratio
Ga 2p/Al 2p
Binding energy
of Ga 2p/eV
Auger kinetic energy
of Ga(MNN)/eV
Auger
Catalyst
Ga2O3
Ga2O3/Al2O3(I)
Ga2O3–Al2O3(S)
parametera/eV
—
0.189
0.147
1118.10
1118.37
1118.53
1063.00
1061.99
1061.62
2181.10
2180.36
2180.15
a Auger parameters were estimated from the sum of binding energy of Ga 2p and Auger kinetic energy of Ga(MNN).

NO REDUCTION BY PROPENE OVER Ga2O3–Al2O3
143
FIG. 9. TEM photographs of (a) Ga2O3/Al2O3(I) and (b) Ga2O3–Al2O3(S).

144
HANEDA ET AL.
TABLE 4
Summary of Kinetic Parameters of NO Reduction to N2
by Propene at 723 K
Reaction ordera with
respect to
Activation energyb/
1
Catalyst
Al2O3
Ga2O3/Al2O3(I)
Ga2O3–Al2O3(S)
Ga2O3
NO
C3H6
O2
kJ mol
0.3
0.3
0.0
0.4
0.3
0.3
0.3
0.0
0.6
0.6
0.7
0.5
85
92
92
63
a Conditions: NO = 200–1500 ppm, C3H6 = 200–1500 ppm, O2 = 3–
10% , temperature = 723 K, gas flow rate = 66 cm3 min 1, catalyst weight =
0.03–0.08 g for Ga2O3/Al2O3(I), 0.01–0.03 g for Ga2O3–Al2O3(S), 0.1 g for
Al2O3, 0.2 g for Ga2O3.
FIG. 10. TPD profiles of NO on Al2O3 (᭺), Ga2O3/Al2O3(I) (᭡),
Ga2O3–Al2O3(S) (ᮀ), and Ga2O3 (᭜). The sample (50 mg) was pretreated
in flowing 10% O2/He at 873 K for 1 h, followed by NO adsorption in
flowing 1000 ppm NO/10% O2/He at room temperature for 2 h. TPD
measurements were carried out up to 873 K with a heating rate of 5 K
min 1 in flowing He (60 cm3 min 1).
b Activation energies were calculated by varying the temperatures from
673 to 773 K under the following reaction conditions: NO = 900 ppm,
1
C3H6 = 900 ppm, O2 = 10% , H2O = 0% , gas flow rate = 66 cm3 min
.
3.3. NO TPD Studies
Ga2O3 gave the highest value, suggesting that the adsorp-
tion site of NO on Ga2O3 is different from that on the other
catalysts.
Figure 10 illustrates TPD profiles of NO on the catalysts
prepared here. In this measurement, no desorption peaks
ascribed to N2O (mass number, 44) and NO2 (mass number,
46) were observed. However, a desorption peak ascribed to
O2 (mass number, 32) from all the catalysts was observed
simultaneously with desorption of NO. This fact suggests
that nitrate species (NO3 ) are formed on the catalyst
surface during adsorption of NO + O2 at room tempera-
ture and desorbed NO results from the decomposition of
nitrates.
3.4. Kinetic Studies of NO Reduction by Propene
over Ga2O3–Al2O3
The kinetic parameters and the activation energies for
N2 formation are summarized in Table 4, where the former
was determined from ln–ln plots of the rate against the
partial pressure of NO, C3H6, or O2 and the latter from
the Arrhenius plots. Good linear correlation was obtained
within experimental error. Al2O3 and Ga2O3/Al2O3(I) gave
similar kinetic parameters including activation energy. The
rate equation is expressed as
All the catalysts, Al2O3, Ga2O3/Al2O3(I), and Ga2O3–
Al2O3(S), except for Ga2O3 showed similar TPD profiles
in which the desorption peak of NO was centered around
690 K, as shown in Fig. 10. The desorption peak on Ga2O3
was observed at a relatively low temperature, 650 K, com-
pared with those of the other catalysts. NO desorption from
Ga2O3 was initiated at the lowest temperature. As summa-
rized in Table 3, the amount of desorbed NO ( mol of NO
rN = kP0.3,0.4
P
PO0.6 exp( 82000, 92000/RT ). [2]
0.3
2
C3H6
NO
2
A significant difference in the reaction order with respect to
NO was recognized between Ga2O3/Al2O3(I) and Ga2O3–
Al2O3(S), although the kinetic orders for C3H6 and O2 as
well as activation energy were almost the same. The re-
action order with respect to NO on Ga2O3–Al2O3(S) was
zero, suggesting that the catalyst surface is covered with
NOx species, probably NO3 adspecies. The rate equation
is as follows:
g
1) from the catalyst increased in the sequence Ga2O3 <
Ga2O3/Al2O3(I) < Al2O3 < Ga2O3–Al2O3(S). The amounts
of desorbed NO normalized by the BET surface area ( mol
of NO m 2) are also summarized in Table 3. Apparently,
TABLE 3
Summary of NO TPD Measurements
0.0 0.3
rN = kP
P
PO0.7 exp( 92000/RT ).
[3]
2
NO C3H6
2
Temperature of
NO desorption
Amount of desorbed NO
On the other hand, Ga2O3 gave a great different kinetic
order with respect to C3H6, which is 0.1, leading to the
presumption that C3H6 is adsorbed strongly on Ga2O3 and
inhibits the N2 formation. The rate equation is described by
1
2
Catalyst
Al2O3
peak/K
mol of NO g
mol of NO m
695
690
690
650
119.7
92.8
170.5
24.7
0.584
0.663
0.907
1.372
Ga2O3/Al2O3(I)
Ga2O3–Al2O3(S)
Ga2O3
0.4
NO
rN = kP
P
0.1 PO0.5 exp( 63000/RT ).
[4]
2
C3H6
2

NO REDUCTION BY PROPENE OVER Ga2O3–Al2O3
145
4. DISCUSSION
kind of aluminium boehmite sols should be formed. Since
the crystal phase of alumina is similar to that of gallium
oxide (47), the formation of Ga–O–Al bondings would be
expected without changes in the shape of fibrillar pseudo-
boehmite sols, resulting in the unique pore structure and
particle shape of Ga2O3–Al2O3(S).
4.1. Relationship between Catalytic Performance
and Physical Properties of Ga2O3–Al2O3
4.1.1. Activity of Ga2O3–Al2O3 for NO reduction by
propene. As shown in Fig. 1, it was found that the catalytic
performance of Ga2O3–Al2O3 for the selective reduction
of NO with propene is quite dependent upon the catalyst
preparation method even though the compositions are the
same. Its tendency did not change in the presence of H2O,
although the activity was depressed by H2O (Fig. 2). Appar-
ently, Ga2O3–Al2O3(S) showed the highest catalytic activity
of all the Ga2O3–Al2O3 catalysts.
The sol–gel technique is a useful method to prepare
catalysts with highly dispersed species, compared to the im-
pregnation method (21, 39, 45). It is also possible to prepare
catalysts with interesting crystal structure by employing
the sol–gel method (39). In the present study, the disper-
sion state and the crystal structure of gallium species of
Ga2O3/Al2O3(I) are presumed to be different from those
of Ga2O3–Al2O3(S). Therefore, we will discuss the differ-
ence of the catalytic performance, especially by focusing
on the dispersion state and the crystal structure of gallium
species, below.
(B) Identification of crystal structure. Since several
phases of gallium oxide, for example, -, -, -, -, and
ε-Ga2O3 (47), closely resemble those of Al2O3, a substi-
tution of Al ion and Ga ion at the interface of Al2O3 and
Ga2O3 particles is expected. In fact, as shown in Fig. 7, a shift
of X-ray diffraction peaks at 2 = 45.5 and 66.5 , which
can be assigned to respective -Al2O3(400) and (440),
toward lower angles was recognized for Ga2O3–Al2O3(S).
This means that Ga ions are incorporated into the -Al2O3
lattice, namely, the formation of a solid solution, where the
radius of the Ga3+ ion (0.062 nm) is larger than that of
the Al3+ ion (0.051 nm). On the other hand, as for Ga2O3/
Al2O3(I), the substitution of Ga ions with Al ions does
not seem to occur, since -Ga2O3 and -Al2O3 are present
independently.
In order to examine in detail the crystal structure of
Ga2O3–Al2O3(S), changes in the XRD peak assigned to
-Al2O3(440) and its lattice constant with Ga2O3 content
(mol% ) were measured. The results are shown in Fig. 11.
The linear correlation was observed in the region of Ga2O3
content below ca. 27 mol% (40 wt% ). It is noted that
the diffraction peaks for Ga2O3–Al2O3(S) consist of sin-
gle peaks without shoulder peaks and split peaks (Fig. 7c).
Taking into account these results, we can consider that com-
posite oxides such as [GaxAl(1 x)]2O3 (x < 1) are formed
uniformly in Ga2O3–Al2O3(S).
4.1.2. Structural characterization of Ga2O3–Al2O3. (A)
Physical properties. A great difference in the shapes
of Ga2O3/Al2O3(I) and Ga2O3–Al2O3(S) was observed by
SEM and TEM measurements. As depicted in Figs. 8 and 9,
Ga2O3–Al2O3(S) wasfound to consist ofa lot offibrillar par-
ticles, while the presence ofnonuniform particleswasrecog-
nized in the photographs of Ga2O3/Al2O3(I). In addition, as
can be seen in Fig. 6, the pore structure of Ga2O3–Al2O3(S)
is different from that of Al2O3 and Ga2O3/Al2O3(I). A
broad distribution curve was observed for the former cata-
lyst and a sharp one centered around 4–5 nm for the latter
two catalysts.
The structural differences between Ga2O3/Al2O3(I) and
Ga2O3–Al2O3(S) might be also explained by the Auger pa-
rameters determined from XPS measurements. As can be
seen in Table 2, the highest Auger parameter was obtained
for Ga2O3 and the Auger parameter of gallium species
Since one of the features of the sol–gel method is to form
M1–O–M2 bondings in the process of gelation (45), the for-
mation of M1–O–M2 bondings is supposed to affect the pore
structure and the particle shape. However, we reported that
the pore structure of Co/Al2O3 (21) and CeO2/Al2O3 (46)
prepared by the same sol–gel method as the present work
is almost the same as that of Al2O3 but obviously differ-
ent from that of Ga2O3–Al2O3(S). Unique properties of
Ga2O3–Al2O3(S) might result from the combination of the
use of the sol–gel method and the similar properties of
Ga2O3 and Al2O3. Ueno et al. (40, 41) reported that the alu-
mina precursors produced by hydrolysis of AIP in hot wa-
ter are composed of fibrillar pseudo-boehmite sols, where
their length and thickness were estimated to be 40–80 and
8–10 nm, respectively, according to TEM observation. Since
we employed almost the same method as that used by Ueno
FIG. 11. Change in the peak position assigned to -Al2O3(440) (᭺)
and the lattice constant of Al2O3 (᭿) in Ga2O3–Al2O3(S) with Ga2O3
et al. for catalyst preparation in the present work, the same loading.

146
HANEDA ET AL.
is higher for Ga2O3/Al2O3(I) than for Ga2O3–Al2O3(S).
I
Ga/IAl evaluated by XPS and the coverage of Ga species
Wagner et al. (44) measured the Auger parameters of on the surface of Al2O3. On the basis of their calculation
aluminum species for several aluminum compounds, and method, the coverage of Ga species was estimated to be
revealed that their Auger parameter decreases in the 0.452 and 0.357 for Ga2O3/Al2O3(I) and Ga2O3-Al2O3(S),
sequence of Al2O3 > SiO2–Al2O3 > zeolites. This suggests respectively. The number of surface Ga ions was also cal-
that the more uniform a composite oxide is, the smaller the culated from the coverage of Ga species and the number of
Auger parameter is. Therefore, gallium species in Ga2O3– surface cation sites of Al2O3.
Al2O3(S) are presumed to be dispersed more uniformly
than those in Ga2O3/Al2O3(I).
It was found that both the TOF-NO and the TOF-M show
the same tendency, although the former TOF was much
Consequently, we can conclude that the difference in higher than the latter one. This is probably because NO is
the crystal structure between Ga2O3/Al2O3(I) and Ga2O3– adsorbed not only on metal ions but also on the neighboring
Al2O3(S) is as follows: (i) in Ga2O3/Al2O3(I), -Ga2O3 par- oxygen basic sites. As can be seen in Table 5, the turnover
ticles are dispersed on the surface of -Al2O3 and their frequency of Al2O3 and Ga2O3 was much less than that
particle size is relatively large, and (ii) in Ga2O3–Al2O3(S), of Ga2O3-Al2O3(S), indicating a synergistic effect between
a composite oxide such as [GaxAl(1 x)]2O3 (x < 1) is formed Ga2O3 and Al2O3. The TOF-NO and TOF-M of Ga2O3-
uniformly in the catalyst.
Al2O3(S) are higher than those of Ga2O3/Al2O3(I) by a fac-
tor of about 1.5 and 3.6, respectively. The difference in the
crystal structures of Ga2O3/Al2O3(I) and Ga2O3-Al2O3(S)
seems to be responsible for the difference in the specific
activity and the turnover frequency. It is noted that the
turnover frequency for Ga2O3 is almost the same as that
for Ga2O3/Al2O3(I). Shimizu et al. (36, 49) reported that
the local structure of Ga atoms, GaO4 tetrahedra highly
dispersed in the surface spinel, is very important to deter-
mine the catalytic activity of Ga2O3/Al2O3 for NO reduc-
tion by methane. In the case of Ga2O3/Al2O3(I), similar
reaction sites for NO reduction by propene would be possi-
ble. Namely, the gallium species dispersed on the surface is
the reaction sites for Ga2O3/Al2O3(I). On the other hand, in
the case of Ga2O3–Al2O3(S), new reaction sites created by
a synergistic effect between Ga and Al ions (represented by
“Ga–Al sites”) in a composite oxide [GaxAl(1 x)]2O3 (x < 1)
would be proposed.
4.1.3. Active sites of Ga2O3–Al2O3 for NO reduction by
propene. In order to compare in detail the activities, we
calculated reaction rate ( mol min 1 g 1) and specific activ-
ity (nmol min 1 m 2), normalized by BET surface area, for
NO reduction to N2 at 723 K. As summarized in Table 5, the
specific activity of Al2O3 was increased by the presence of
Ga2O3 irrespective of preparation method. Relatively high
specific activity was obtained for Ga2O3 itself, suggesting
that the presence of highly dispersed Ga2O3 is necessary to
attain a high catalytic activity. Ga2O3–Al2O3(S) showed the
highest specific activity.
Table 5 also shows two kinds of turnover frequency
(TOF) for N2 formation at 723 K. The TOF-NO and TOF-M
were calculated on the basis of the amount of desorbed
NO (Table 3) and the number of surface metal ions (Al or
Ga), respectively. The number of surface Al and Ga ions
on Al2O3 and Ga2O3 was estimated from the BET surface
area combined with the theoretical surface density of Al
4.2. Proposed Reaction Mechanism
2
(14.5 nm for the (111) face of -Al2O3 (48)) and Ga
(14.0 nm 2 for the (001) face of -Ga2O3 (36)), respectively.
The selective reduction of NO with hydrocarbons is a
Recently, Shimizu et al. (49) reported the equation con- very complex reaction comprising several parallel and/or
cerning the relationship between the peak intensity ratio of consecutive reaction steps. As for the initial step of this
TABLE 5
Reaction Rates, Specific Activities, and Turnover Frequencies (TOF) of Al2O3, Ga2O3/Al2O3(I), Ga2O3–Al2O3(S),
and Ga2O3 for NO Reduction to N2 by Propene in the Absence of H2O at 723 Ka
Number of surface Al
and Ga ions/mmol g
Reaction rate/
Specific activity/
TOF-NOd/
10 2 min
TOF-Me/
10 3 min
1
1
2
1
1
Catalyst
Al2O3
Ga2O3/Al2O3(I)
Ga2O3–Al2O3(S)
Ga2O3
mol min 1 g
nmol min 1 m
4.94b
1.56c
1.23c
0.42c
3.04
6.61
18.6
1.75
14.8
47.2
98.9
97.2
2.54
7.12
10.9
7.09
0.62
4.23
15.1
4.18
1
a Reaction conditions: NO = 900 ppm, C3H6 = 900 ppm, O2 = 10% , H2O = 0% , temperature = 723 K, gas flow rate = 66 cm3 min
b Number of surface Al ions.
.
c Number of surface Ga ions.
d The TOF-NO was calculated on the basis of the amount of desorbed NO (Table 3).
e The TOF-M was calculated on the basis of the number of surface metal ions (Al or Ga).

NO REDUCTION BY PROPENE OVER Ga2O3–Al2O3
147
reaction, many discussions have been published so far. In et al. (51) also reported that NO reduction by alkanes over
many cases, NO reduction by hydrocarbons is believed to Cu–ZSM-5 proceeds via the formation of 2-nitrosopropane
proceed through reaction of hydrocarbons with NO2, which by the reaction of NO with a sec-propyl radical formed by
is produced by oxidation of NO by oxygen (3, 8, 10, 22, 33, H-abstraction of alkane. Taking these reports into account,
50, 51). In thisstudy, it wasfound from the comparison ofthe NO reduction by propene over Ga2O3–Al2O3 should pro-
reactivity between NO and NO2 that the reactivity of NO2 ceed through the formation of organic compounds contain-
is much higher than that of NO and that propene conver- ing nitrogen and oxygen by the reaction of NO2 formed
sion is enhanced considerably by using NO2 instead of NO via NOx adspecies with C3H6-derived species such as allyl
(Fig. 4). These findings suggest the participation of NO2 in species. Since propene conversion on Ga2O3–Al2O3(S) was
NO reduction. On the other hand, Al2O3, Ga2O3/Al2O3(I), much higher than that on Ga2O3/Al2O3(I) irrespective of
and Ga2O3 did not show activity for NO oxidation into NO2 NOx sources (NO or NO2) (Figs. 1 and 4), the reaction of
(Fig. 5), whereas these three catalysts were effective for NO2 with C3H6-derived species would be much faster on
NOx reduction into N2 (Figs. 1 and 4). This result contra- Ga2O3–Al2O3(S) than on Ga2O3/Al2O3(I). Finally, the re-
dicts the reaction pathway proposed so far, in which NO sulting intermediate would be reduced to N2 via several
oxidation to NO2 is the initial and indispensable step for steps by the reaction with NOx and/or O2.
NO reduction by hydrocarbons.
Takeda and Iwamoto (52) reported the following rate
5. CONCLUSIONS
equation for N2 formation in the NO + O2 + C2H4 reaction
over Al2O3:
Ga2O3–Al2O3 prepared by the sol–gel method (Ga2O3–
1.1 0.1
rN = kP
P
PO0.5 exp( 294000/RT ).
[4]
Al2O3(S)) exhibited excellent activity for NO reduction by
propene in the presence of oxygen, compared with Al2O3,
2
NO C2H4
2
They proposed by comparing the rate equations obtained Ga2O3, and Ga2O3/Al2O3 prepared by the impregnation
for NO + O2 + C2H4 and NO2 + O2 + C2H4 reactions that method (Ga2O3/Al2O3(I)). The high catalytic activity of
the formation of NO2 is the key step on Al2O3. On the other Ga2O3–Al2O3(S) was accounted for by the high surface
hand, in the present study, the reaction order with respect area and the presence of [GaxAl(1 x)]2O3 (x < 1) composite
to NO was less than unity for all the catalysts tested here oxides. The formation of [GaxAl(1 x)]2O3 (x < 1) compos-
(Table 4). In particular, the reaction on Ga2O3–Al2O3(S) ite oxides was presumed to result from supporting gallium
was found to proceed with zero-order kinetics for NO species on the surface of aluminium boehmite needles as
(Eq. [3]). These results suggest that the catalyst surface is alumina precursor.
covered with NOx adspecies such as nitrate species (NO3 ).
Thisisconsistent with the resultsof NO TPD measurements NO and propene conversion was enhanced considerably
(Fig. 10). by using NO2 instead of NO, suggesting the participation
The reactivity of NO2 was much higher than that of
NOx adspecies, probably NO3 , formed on the catalyst of NO2 in NO reduction. On the other hand, direct oxida-
surface is known to play an important role in NO reduc- tion of NO to NO2 was found unimportant. A significant
tion (53, 54). Meunier et al. (29) reported that the forma- difference in the reaction order with respect to NO was
tion of organo-nitrite species followed by decomposition/ recognized between Ga2O3/Al2O3(I) and Ga2O3–Al2O3(S),
oxidation but not the direct oxidation of NO with O2 is the while the other kinetic parameters were almost the same.
main route for the formation of NO2 in NO reduction over It should be noted that the reaction order with respect to
Al2O3 and Ag/Al2O3. In the present study, the formation NO was less than unity for all the catalysts, indicating that
of NO2 might proceed via similar reaction steps, although the catalyst surface is covered with NOx adspecies such as
there is no evidence. Ga2O3–Al2O3(S) seems to promote ef- nitrate species (NO3 ). From these findings, the formation
fectively the formation of inorganic NOx adspecies, which of NOx adspecies, probably nitrates, was considered to be
is the precursor or organo-NOx compounds, as evidenced one of the important steps in NO reduction by propene
by NOx TPD measurements (Fig. 10).
over Ga2O3–Al2O3. The following reaction mechanism was
It can be seen in Table 4 that the kinetic order with re- suggested: NO2 formed via NOx adspecies is reduced to N2
spect to C3H6 on Ga2O3/Al2O3(I) and Ga2O3–Al2O3(S) is through the formation of organic compounds containing
0.3, suggesting that C3H6 is adsorbed on the catalyst sur- nitrogen and oxygen by the reaction with propene-derived
face and then reacts with NOx and/or O2 to form inter- species.
mediates. Various organic compounds containing nitrogen
and oxygen have been reported as candidates for the re-
action intermediates based on organic chemistry literature.
REFERENCES
For example, Smits and Iwasawa (50) proposed the for-
mation of the 1-nitro-sec-propyl radical by the reaction
1. Shelef, M., Chem. Rev. 95, 209 (1995).
2. Iwamoto, M., Mizuno, N., and Yahiro, H., Sekiyu Gakkaishi 34, 375
of NO2 with oxygenated hydrocarbon (C3H6O). Adelman
(1991).

148
HANEDA ET AL.
3. Hamada, H., Kintaichi, Y., Sasaki, M., Ito, T., and Tabata, M., Appl.
Catal. 70, L15 (1991).
29. Meunier, F. C., Breen, J. P., Zuzaniuk, V., Olsson, M., and Ross,
J. R. H., J. Catal. 187, 493 (1999).
4. Held, W., Koenig, A., Richter, T., and Puppe, L., SAE Paper 900496
(1990).
5. Iwamoto, M., Yahiro, H., Yu-u, Y., Sundo, S., and Mizuno, N., Shokubai
(Catalyst) 32, 430 (1990).
30. Miyadera, T., Appl. Catal. B: Environ. 2, 199 (1993).
31. Miyadera, T., and Yoshida, K., Chem. Lett. 1483 (1993).
32. Aoyama, N., Yoshida, K., Abe, A., and Miyadera, T., Catal. Lett. 43,
249 (1997).
6. Ger. Offen. DE 3,642,018 (1987) to Volkswagen A.G.
7. Ger. Offen. DE 3,735,151 (1988) to Toyota Co.
8. Yokoyama, C., and Misono, M., J. Catal. 150, 9 (1994).
9. Iwamoto, M., Yahiro, H., Shundo, S., Yu-u, Y., and Mizuno, N., Appl.
Catal. 69, L15 (1991).
33. Okazaki, N., Katoh, Y., Shiina, Y., Tada, A., and Iwamoto, M., Chem.
Lett. 889 (1997).
34. Hoost, T. E., Kudla, R. J., Collins, K. M., and Chattha, M. S., Appl.
Catal. B: Environ. 13, 59 (1997).
35. Jen, H. W., Catal. Today 42, 37 (1998).
10. Kikuchi, E., and Yogo, K., Catal. Today 22, 73 (1996).
11. Tabata, T., Kokitsu, M., Ohtsuka, H., Okada, O., Sabatino, L. M. F.,
and Bellussi, G., Catal. Today 27, 91 (1996).
12. Grinstedt, R. A., Jen, H.-W., Montreuil, C. N., Rokosz, M. J., and
Shelef, M., Zeolites 13, 602 (1993).
13. Keiski, R., Raisanen, H., Harkonen, M., Maunula, T., and Niemisto,
P. , Catal. Today 26, 85 (1996).
14. Kintaichi, Y., Hamada, H., Tabata, M., Sasaki, M., and Ito, T., Catal.
Lett. 6, 239 (1990).
15. Teraoka, Y., Harada, T., Iwasaki, T., Ikeda, T., and Kagawa, S., Chem.
Lett. 773 (1993).
16. Zhang, X., Walters, A. B., and Vannice, M. A., J. Catal. 155, 290
(1995).
17. Okazaki, N., Shiina, Y., Itoh, H., Tada, A., and Iwamoto, M., Catal.
Lett. 49, 169 (1997).
18. Tabata, M., Hamada, H., Kintaichi, Y., Sasaki, M., and Ito, T., Sekiyu
Gakkaishi 36, 191 (1993).
36. Shimizu, K., Takamatsu, M., Nishi, K., Yoshida, H., Satsuma, A., and
Hattori, T., J. Chem. Soc., Chem. Commun. 1827 (1996).
37. Shimizu, K., Satsuma, A., and Hattori, T., Appl. Catal. B: Environ. 16,
319 (1998).
38. Haneda, M., Kintaichi, Y., Shimada, H., and Hamada, H., Chem. Lett.
181 (1998).
39. Haneda, M., Mizushima, T., Kakuta, N., Ueno, A., Sato, Y., Matsuura,
S., Kasahara, K., and Sato, M., Bull. Chem. Soc. Jpn. 66, 1279 (1993).
40. Ishiguro, K., Ishikawa, T., Kakuta, N., Ueno, A., Mitarai, Y., and Kamo,
T., J. Catal. 123, 523 (1990).
41. Ishikawa, T., Ohashi, R., Nakabayashi, H., Kakuta, N., Ueno, A., and
Furuta, A., J. Catal. 134, 87 (1992).
42. Yan, J. Y., Kung, H. H., Sachtler, W. M. H., and Kung, M. C., J. Catal.
175, 294 (1998).
43. Wagner, C. D., Gale, L. H., and Raymond, R. H., Anal. Chem. 51, 466
(1979).
44. Wagner, C. D., Six, H. A., Jansen, W. T., and Tayler, J. A., Appl. Surf.
Sci. 9, 203 (1981).
45. Ueno, A., Suzuki, H., and Kotera, Y., J. Chem. Soc., Faraday Trans. 1
79, 127 (1983).
19. Torikai, Y., Yahiro, H., Mizuno, N., and Iwamoto, M., Catal. Lett. 9, 91
(1991).
20. Hamada, H., Kintaichi, Y., Sasaki, M., Ito, T., and Tabata, M., Appl.
Catal. 75, L1 (1991).
46. Haneda, M., Ph.D. Thesis, Toyohashi University of Technology, Japan,
1995.
21. Nanba, T., Uemura, A., Ueno, A., Haneda, M., Hamada, H., Kakuta,
N., Miura, H., Ohfune, H., and Udagawa, Y., Bull. Chem. Soc. Jpn. 71, 47. Roy, R., Hill, V. G., and Osborn, E. F., J. Am. Chem. Soc. 74, 719 (1952).
2331 (1998).
48. Knozinger, H., and Ratnasamy, P., Catal. Rev. Sci. Eng. 17, 31 (1978).
49. Shimizu, K., Takamatsu, M., Nishi, K., Yoshida, H., Satsuma, A.,
Tanaka, T., Yoshida, S., and Hattori, T., J. Phys. Chem. B 103, 1542
(1999).
50. Smits, R. H. H., and Iwasawa, Y., Appl. Catal. B: Environ. 6, L201
(1995).
51. Adelman, B. J., Beutel, T., Lei, G.-D., and Sachtler, W. M. H., Appl.
Catal. B: Environ. 11, L1 (1996).
52. Takeda, H., and Iwamoto, M., Bull. Chem. Soc. Jpn. 69, 2735 (1996).
22. Hamada, H., Kintaichi, Y., Inaba, M., Tabata, M., Yoshinari, T., and
Tsuchida, H., Catal. Today 29, 53 (1996).
23. Tabata, M., Hamada, H., Suganuma, F., Yoshinari, T., Tsuchida, H.,
Kintaichi, Y., Sasaki, M., and Ito, T., Catal. Lett. 25, 55 (1994).
24. Haneda, M., Kintaichi, Y., Inaba, M., and Hamada, H., Bull. Chem.
Soc. Jpn. 70, 499 (1997).
25. Park, P. W., Kung, H. H., Kim, D.-W., and Kung, M. C., J. Catal. 184,
440 (1999).
26. Yan, J. Y., Kung, M. C., Sachtler, W. M. H., and Kung, H. H., J. Catal. 53. Satsuma, A., Yamada, K., Sato, K., Shimizu, K., Hattori, T., and
172, 178 (1997).
Murakami, Y., Catal. Lett. 45, 267 (1997).
54. Shimizu, K., Kawabata, H., Satsuma, A., and Hattori, T., J. Phys. Chem.
B 103, 5240 (1999).
27. Bethke, K. A., and Kung, H. H., J. Catal. 172, 93 (1997).
28. Bethke, K. A., Alt, D., and Kung, M. C., Catal. Lett. 25, 37 (1994).