Synthesis and catalytic properties of porous Nb-Mo oxide solid acid

Article abstract of DOI:10.1016/j.cattod.2010.10.049

Several porous Nb_xMo_10-x mixed oxides prepared from NbCl₅ and MoCl₅ in the presence of block copolymer surfactant Pluronic P-123 were examined as potential solid acid catalysts. Amorphous mesoporous structures w

Full text of DOI:10.1016/j.cattod.2010.10.049

Catalysis Today 164 (2011) 358–363
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Synthesis and catalytic properties of porous Nb–Mo oxide solid acid
a
a
b
a
b
Caio Tagusagawa , Atsushi Takagaki , Ai Iguchi , Kazuhiro Takanabe , Junko N. Kondo ,
c
b
a,∗
Kohki Ebitani , Takashi Tatsumi , Kazunari Domen
Department of Chemical System Engineering, School of Engineering, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-7656, Japan
Chemical Resources Laboratory, Tokyo Institute of Technology, 4259 Nagatsuta Midori-ku, Yokohama 226-8503, Japan
a
b
c
School of Materials Science, Japan Advanced Institute of Science and Technology (JAIST), 1-1 Asahidai, Nomi, Ishikawa 923-1292, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 11 November 2010
Several porous Nb Mo
10−x
mixed oxides prepared from NbCl5 and MoCl5 in the presence of block copoly-
x
mer surfactant Pluronic P-123 were examined as potential solid acid catalysts. Amorphous mesoporous
structures were observed in samples with x from 10 to 9, whereas samples with higher Mo concentra-
tions (x = 3–8) formed large pores by inter-particle voids and a non-porous structures with crystallized
molybdenum oxide (MoO3) were observed in samples with x from 0 to 2. The acid strength increased
with increasing the Mo content in NbxMo10−x oxides, which was determined by NH3 temperature-
programmed desorption (NH3-TPD) and Fourier transform infrared (FT-IR) spectroscopy using pyridine as
a probe molecule. Porous Nb3Mo7 oxide exhibited the highest acid-catalytic activity for the Friedel–Crafts
alkylation of anisole due not only to mole ratio of Nb/Mo but also to pore structure.
Keywords:
Solid acid
Porous metal oxide
Friedel–Crafts alkylation
FT-IR
©
2010 Elsevier B.V. All rights reserved.
1
. Introduction
[13], which is known in the zeolites, or formation of heteropoly
compounds like the H PO –WO –Nb O5 [14,15].
3
4
3
2
The acidities of bulk metal oxides and mixed oxides have been
To improve the rates of metal oxide acid catalysis, mesoporous
structured solid acids have been widely studied [16–21]. High sur-
face area mesopores in oxide enable the reactants to access acid
sites, exhibiting high activity and selectivity, principally for sul-
fated mesoporous niobium [20] and tantalum oxides [21,22]. Our
previous studies on mesoporous Nb–W and Ta–W oxide solid acids
[23,24] have revealed that wormhole-type mesoporous Nb–W and
Ta–W oxides can be applied as recyclable and highly active solid
acid catalysts for Friedel–Crafts alkylation, hydrolysis of disaccha-
rides, and esterification. The reaction rate and the acid strength
increased gradually with the addition of W, reaching the highest
reaction rate with mesoporous Nb W7 and Ta W7 oxides, which
widely studied as a promising substitute for liquid acid. Hydrated
niobium oxide, also called niobic acid is a well known bulk metal
oxide with strong Brønsted acidity and water tolerant solid acid.
Therefore deep study on niobium oxide and tantalum oxide has
been investigated on their acid sites characterizations and cat-
alytic activities [1–8]. However, its application to acid-catalyzed
reactions has not yet been performed sufficiently.
Mixed metal oxides are known to exhibit higher reaction activ-
ity and stronger acidity than single metal oxides. To explain the
acid generation of binary oxides [9,10], Tanabe et al. proposed the
hypothesis, where acid site generation is caused by an excess of a
negative or positive charge in the model structure of a binary oxide
for chemically mixed oxides. Actually, the hypothesis agrees with
3
3
exceeded the reaction rate of ion-exchange resins, zeolites, and
non-mesoporous metal oxides.
9
0% of the 31 kinds of binary oxides composed principally by TiO2,
In the present study, porous Nb–Mo mixed oxides with differ-
ent Nb and Mo concentrations were examined as novel solid acid
catalysts. Previous study demonstrated that isomorphous replace-
ZnO, Al O , SiO and ZrO2 [11]. Nb O5–MoO3 and Nb O5–WO3
2
3
2
2
2
reported by Niwa and co-workers exhibit very strong acid strength
and catalytic activity for Friedel–Crafts alkylation of anisole with
benzyl alcohol [12]. Their activities are much higher than that of
other niobium containing metal oxides (Nb O5/SiO , Nb O5/TiO ,
Nb O5/MgO and Nb O5). They suggested two hypotheses for the
ment of Nb5+ and Ta by higher-valence W cations formed strong
Brønsted acid sites in W-enriched samples, resulting in a highly
active mesoporous solid acid [23,24]. Therefore, porous Nb–Mo
mixed oxide was tested in an attempt to obtain a more effective
solid acid than non-porous Nb O5–MoO oxide. The acid properties
5+
6+
2
2
2
2
2
2
formation of active species; an isomorphous replacement model
2
3
of porous Nb–Mo oxides were evaluated by NH3 temperature-
programmed desorption (NH -TPD) and Fourier transform infrared
3
(
FT-IR) spectroscopy. The acid catalytic activity of porous Nb–Mo
∗ Correspondence author: Tel.: +81 3 5841 1148; fax: +81 3 5841 8838.
E-mail address: domen@chemsys.t.u-tokyo.ac.jp (K. Domen).
oxides with different Nb–Mo concentrations were examined dur-
0
920-5861/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2010.10.049

C. Tagusagawa et al. / Catalysis Today 164 (2011) 358–363
359
for adsorption, and finally heated at 2 K min 1. For IR spectroscopic
measurements, the samples were evacuated in a conventional
gas-circulation system at 423 K for 1 h, and then pyridine was intro-
duced into the system to be adsorbed on the acid sites for the study
of the acid properties.
−
ing liquid-phase Friedel–Crafts alkylation of anisole and the results
are compared with those of conventional solid acids.
2
. Experimental
2.1. Preparation of porous metal oxides
2.3. Acid-catalyzed reactions
Porous NbxMo(
oxides were prepared from niobium pen-
10−x)
tachloride NbCl5 (99.99%, Kojundo Chemical Laboratory Co., Ltd.)
and molybdenum pentachloride MoCl5 (99.99%, Kojundo Chemical
Laboratory Co., Ltd.). Poly block copolymer surfactant Pluronic
Friedel–Crafts alkylation was performed using 0.2 g of cat-
alyst, 100 mmol of anisole, and 10 mmol of benzyl alcohol in
an oil bath at 373 K for 1 h. The products (ortho-benzylanisole,
para-benzylanisole, and dibenzylether) were analyzed by flame
ionization gas chromatography (GC-2014, Shimadzu), using a cap-
illary column (J&W Scientific-DB-FFAP, length 30 m, i.d. 0.25 mm,
and film 0.25 ␮m).
P-123
Aldrich) was used as a structure-directing agent (SDA). Porous
NbxMo( oxide was synthesized by dissolving 1 g of P-123 in
((HO(CH CH O) –(CH CH(CH )O) –(CH CH O)20H,
2 2 20 2 3 70 2 2
10−x)
1
0 g of dehydrated 1-propanol, and adding NbCl5 and MoCl5 (total
of 6 mmol metal chloride) for porous NbxMo( and 7 mmol
The activities of niobic acid (Nb2O5•nH2O), Nb2O5–MoO3, ion-
exchange resins (Amberlyst-15 and Nafion NR50), and H-Beta
zeolite (SiO2/Al2O3 = 25, JRC-Z-HB25 supplied from the Cataly-
sis Society of Japan (Japan Reference Catalyst)) were used for
comparison. Non-porous Nb2O5–MoO3 was prepared as follows.
Nb2O5•nH2O (n = 3.8, measured by TG) kindly supplied by CBMM
Co. and (NH ) Mo7O •4H O were dissolved in oxalic acid solution
10−x)
NbCl5 for mesoporous Nb oxide with vigorous stirring. Water
30 mmol) was added to the solution with further stirring. The
(
resulting sol was gelled in a Petri dish at 313 K for 10–14 days. The
aged gel samples were then treated at 673 K for 10 h in static air to
remove the SDA.
4
6
24
2
and distilled water, respectively, at 353 K. The solutions were then
mixed and stirred vigorously at 353 K. After drying, the obtained
materials were calcined at 673 K for 3 h to yield Nb–Mo mixed
oxides.
2.2. Characterization
Samples were characterized by N2 desorption (BEL Japan,
BELSORP-miniII), X-ray diffraction (XRD; RINT-UltimaIII, Rigaku),
and scanning electron microscopy (SEM; S-4700, Hitachi). The
Brunauer–Emmett–Teller (BET) surface area was estimated over a
3. Results and discussion
relative pressure (P/P ) range of 0.05–0.30. The pore size distribu-
0
tion was obtained from analysis of the adsorption branches of the
isotherms using the Barrett–Joyner–Halenda (BJH) method and the
small-angle powder X-ray diffraction pattern.
3.1. Structure of porous NbxMo(
metal oxides
10−x)
The presence of a porous structure was evaluated from N2
sorption isotherms (Fig. 1), XRD patterns (Fig. 2) and SEM images
(Fig. 3). N2 sorption isotherms (Fig. 1) indicated the presence of
porous structures in samples with x = 3–10. These exhibited a typi-
cal type-IV pattern with an H1-type hysteresis loop at high relative
pressures for mesopores with high Nb (x = 9–10) concentrations.
Hysteresis loops similar to the H2-type were observed for sam-
The acid properties of the samples were determined by NH -TPD
3
(
acid strength and concentration of sites) using a TPD-1-AT instru-
ment (BEL Japan) equipped with a quadrupole mass spectrometer
and FT-IR (acid type: Brønsted and Lewis acid) spectroscopy (FT-
IR; Jasco, FT/IR-6200). For TPD measurement, a 20 mg sample was
heated at 423 K for 1 h under helium flow, exposed to NH3 at 373 K
8
7
6
5
4
3
2
1
00
00
00
00
00
00
00
00
0
800
(k) x=0 Mo
(
e) x=6 Nb Mo4
6
700
600
500
400
300
200
100
0
(j) x=1 Nb Mo9
1
(
d) x=7 Nb Mo3
7
(
i) x=2 Nb Mo8
(
c) x=8 Nb Mo2
2
8
(
(
h) x=3 Nb Mo7
3
(
b) x=9 Nb Mo1
9
g) x=4 Nb Mo6
4
(
a) x=10 Nb
0.2
(
f) x=5 Nb Mo5
5
0
0.4
0.6
0.8
1.0
0
0.2
0.4
P P0-1
0.6
0.8
1.0
P P -1
0
Fig. 1. N2 sorption isotherms of mesoporous (a) Nb, (b) Nb9Mo1, porous (c) Nb8Mo2, (d) Nb7Mo3, (e) Nb6Mo4, (f) Nb5Mo5, (g) Nb4Mo6, (h) Nb3Mo7 and non-porous (i)
Nb2Mo8, (j) Nb1Mo9 and (k) Mo oxides.

3
60
C. Tagusagawa et al. / Catalysis Today 164 (2011) 358–363
(A)
(B)
#
#
#
# #
(
(
k) x=0 Mo
j) x=1 Nb Mo
(
k) x=0 Mo
+
1
9
8
+
+
+
#
(
(
(
(
(
i) x=2 Nb Mo
2
(j) x=1 Nb Mo
1
9
+
h) x=3 Nb Mo
3
7
# +
+
+
+
+
(
i) x=2 Nb Mo
2
g) x=4 Nb Mo
8
4
6
+
f) x=5 Nb Mo
+
5
5
#
(
(
h) x=3 Nb Mo
3
7
e) x=6 Nb Mo
6
4
g) x=4 Nb Mo
4
6
(
(
d) x=7 Nb Mo
7
3
(
(
(
f) x=5 Nb Mo
5
5
e) x=6 Nb Mo
c) x=8 Nb Mo
6
4
8
2
d) x=7 Nb Mo
7 3
(
(
b) x=9 Nb Mo
9
1
(c) x=8 Nb Mo
8 2
(
(
b) x=9 Nb Mo
9
1
a) x=10 Nb
a) x=10 Nb
0
1
2
3
4
5
0
10
20
30
2θ / degree
40
50
60
2
θ / degree
Fig. 2. (A) Low and (B) wide-angle powder XRD patterns for mesoporous (a) Nb, (b) Nb9Mo1 and non-mesoporous (c) Nb8Mo2, (d) Nb7Mo3, (e) Nb6Mo4, (f) Nb5Mo5, (g)
Nb4Mo6, (h) Nb3Mo7, (i) Nb2Mo8 oxides, (j) Nb1Mo9 and (k) Mo oxides (+, Mo3Nb2O14; #, MoO3).
ples with x values from 3 to 8. The small-angle XRD patterns
Fig. 2) contained peaks attributable to mesopores for NbxMo(
XRD pattern. Nb7Mo , Nb5Mo5, and Nb Mo7 oxides had large pores
3 3
(
formed by inter-particle voids, different from mesopores observed
at mesoporous Nb oxide, supporting both N2 sorption isotherms
(formation of pores) and XRD patterns results (no mesopores). A
crystallized MoO3 was observed in the Mo oxides, supporting the
XRD results.
10−x)
oxides with x values from 9 to 10. Peaks attributed to (1 1 0) and
2 0 0) of the two-dimensional hexagonal structure were observed
(
from an x = 10 sample (mesoporous Nb O5), which was consistent
2
with previous studies [23]. The small-angle XRD peaks indicat-
ing mesoporous structure could not be observed for samples with
x values form 0 to 8. Wide-angle powder XRD patterns revealed
the presence of an amorphous structure in samples with x = 4–10,
Mo Nb O in samples with x = 4–1 and crystallized molybdenum
Characterization of the pores was conducted by both
Brunauer–Emmett–Teller (BET) (surface area estimation) and
Barrett–Joyner–Halenda (BJH) (pore size distributions) methods.
Fig. 4 shows the variation in surface area and pore volume versus
Mo concentration. The BET surface area decreased gradually from
3
2
14
oxide (MoO ) in Mo-rich samples with x = 3–0. SEM images of the
3
2
−1
prepared porous oxides are shown in Fig. 3. The mesoporous Nb
oxide had hexagonally structured mesopores, as observed in the
200 (mesoporous Nb oxide) to 21 m g (non-porous Mo oxide)
with increasing Mo content, up to x = 0, due to the formation of
Fig. 3. SEM images of mesoporous (a) Nb, porous (b) Nb7Mo3, (c) Nb5Mo5, (d) Nb3Mo7 oxides and non-porous e) Mo oxides.

C. Tagusagawa et al. / Catalysis Today 164 (2011) 358–363
361
2
1
1
40
80
20
0.4
4
4
45 K
0.3
0.2
0.1
(
c)
6
0
0
40 K
0
0
20
40
60
80
100
Mo mol%
2
Fig. 4. BET surface areas (m /g) and pore volumes (␮L/g) of porous NbxMo1−x oxides.
crystallized Mo Nb O and MoO3 oxides in the Mo-rich samples.
Aggregation and crystallization resulted in the destruction of the
original mesoporous structure and the development of larger
3
2
14
(b)
4
35 K
pores (between 10.6 and 18.6 nm) for NbxMo(
oxides (x = 1–8)
10−x)
due to inter-particle voids. The thermal stability improvement of
the material in the amorphous status caused by addition of the
transition metal Nb to the Mo oxide should not have been sufficient
to elevate the crystallization temperature beyond that required to
completely remove the mesoporous template for these samples
with x = 1–8, forming inter-particle voids in spite of mesopores.
And the decrease on Nb concentration should have decreased
(
a)
the crystallization temperature, forming Mo Nb O and MoO3
for Mo enriched samples with x = 0–3. The pore volume obtained
by BJH decreased up to x = 0 due to formation of crystallized
3
2
14
350
450
550
650
Temperature / K
Mo Nb O and MoO , as can be observed in the XRD patterns
3
2
14
3
Fig. 5. NH3-TPD (m/e = 16, 10 K min−1) curves for porous (a) Nb7Mo3, (b) Nb5Mo5,
and (c) Nb3Mo7.
(
Fig. 2) and SEM image of non-porous Mo oxide (Fig. 3). The
BJH pore size distributions and BET surface area of samples with
different concentrations obtained from N2 sorption isotherms are
summarized in Table 1. The BJH pore size distributions indicate
increase on pores sizes for Mo-enriched samples.
The acid properties of NbxMo(
were also evaluated by probing the vibrational spectra of adsorbed
pyridine using IR spectroscopic measurements, as shown in Fig. 6.
oxides (x = 3, 5, 7, and 10)
10−x)
NbxMo(
oxides (x = 3, 5, 7, and 10) samples were confirmed to
10−x)
3.2. Characterization of acid sites
−1
−1
posses Lewis acid site 1440 cm and Brønsted acid site 1532 cm .
The IR spectroscopic measurements indicated that the peak inten-
sity of the bond at 1532 cm , attributed to pyridinium ions
formed on Brønsted acid sites [12,13,25–27], was enhanced by
increasing the Mo content. The increase in Brønsted acid sites
The acid properties of porous NbxMo(
0) were evaluated by NH3 temperature-programmed desorption
oxides (x = 3, 5, 7, and
−1
10−x)
1
(
NH -TPD) and Fourier transform infrared (FT-IR) spectroscopic
3
measurements of adsorbed pyridine. The NH -TPD profiles for
3
porous NbxMo(
(x = 3, 5, 7) exhibited a broad main peak
10−x)
at 435–445 K (Fig. 5). The order of acid strength of the porous
NbxMo( oxides (x = 3, 5, 7) increased with increasing Mo
B
L
10−x)
oxide concentration. The main peak was observed at 435 K for
−1
porous Nb7Mo oxide (0.29 mmol g ), at 440 K for porous Nb5Mo5
3
−
1
3
(d) x=3 Nb Mo7
oxide (0.27 mmol g ), and at 445 K for porous Nb Mo7 oxide
3
−
1
(
0.24 mmol g ).
(
c) x=5 Nb Mo5
5
Table 1
Structural properties of mesoporous NbxMo(10−x) oxides.
SBET (m2
g
−1
)
Pore size
(nm)
Pore volume
Mesoporous
NbxMo(10−x) oxides
−
1
(b) x=7 Nb Mo3
(mL g
)
7
Nb0Mo10
Nb1Mo9
Nb2Mo8
Nb3Mo7
Nb4Mo6
Nb5Mo5
Nb6Mo4
Nb7Mo3
Nb8Mo2
Nb9Mo1
Nb10Mo0
21
43
52
57
68
70
77
81
99
–
–
18.6
13.9
13.3
13.9
10.6
12.1
13.9
10.6
8.0
0.15
0.16
0.20
0.20
0.19
0.20
0.22
0.24
0.29
0.35
(a) x=10 Nb10
1650
1600
1550
1500
1450
1400
-1
Wavenumber / cm
Fig. 6. FT-IR spectra for pyridine adsorbed porous (a) Nb, (b) Nb7Mo3, (c) Nb5Mo5,
and (d) Nb3Mo7 oxides (B, Brønsted acid site; L, Lewis acid site). Assignments:
166
200
7.0
−
1
−
1
1
532 cm (Brønsted acid site), 1435 cm (Lewis acid site).

3
62
C. Tagusagawa et al. / Catalysis Today 164 (2011) 358–363
1
00
−1
gradually for Mo enriched samples: Nb7Mo (33 h ) < Nb5Mo5
(
3
− −1
83 h ) < Nb Mo7 (158 h ). This increase in the TOF indicates that
3
1
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
the acid sites on porous Nb Mo7 have higher activity than that of
3
porous Nb7Mo3 and Nb5Mo5 oxides. Both the reaction yield and
−
1
TOF (h ) increased for Mo-enriched porous NbxMo(
oxide
10−x)
samples, consistent with the acid strength determined by NH -
3
TPD and with the increase in Brønsted acid sites, which promotes
the Friedel–Crafts alkylation reaction [28], observed by IR spectro-
scopic measurements.
The acid catalytic activities of porous Nb–Mo oxides for
Friedel–Crafts alkylation were also compared to those of con-
ventional solid acids. The acid concentration and surface area
of the tested solid acids and the results of the Friedel–Crafts
alkylation of anisole are summarized in Table 2. Bulk (Non-
porous) Nb O5–MoO3 oxides prepared from Nb O5•nH O and
2
2
2
(
NH ) Mo7O •4H O with different Nb–Mo concentrations exhib-
4
6
24
2
ited lower reaction yield compared with porous NbxMo
oxides, especially much different between bulk Nb3Mo7 oxide and
(10−x)
0
10
20
30
40
50
60
70
80
90 100
Mo mol%
porous Nb Mo7 oxide.
3
After the reaction, ortho-benzylanisole, para-benzylanisole, and
dibenzylether were formed. The selectivity of ortho-benzylanisole
Fig. 7. (a) Friedel–Crafts alkylation of anisole over porous NbxMo(10−x) oxides (reac-
tion conditions: anisole (100 mmol), benzyl alcohol (10 mmol), catalyst (0.2 g),
via para-benzylanisole observed for porous NbxMo(
oxides
10−x)
3
73 K, 1 h).
for Friedel–Crafts alkylation gradually increased with increasing
Mo content (38.9, 42.3, and 44.6% for porous Nb7Mo , Nb5Mo5,
3
corresponded with the model structure of a binary oxide for chem-
ically mixed oxides. The excess of a negative charge caused by
increase of Mo concentration should have generated the Brønsted
acid sites.
and Nb Mo7 oxides, respectively). The same selectivity of o-
3
benzylanisole was observed for bulk Nb O –MoO (Nb Mo ,
5
7
2
3
3
Nb Mo , Nb Mo ) oxide (39.0, 40.1 and 43.8%). The variation
5
5
7
3
of selectivity was probably not caused by the porous struc-
ture, but by variations of the acid properties (Brønsted acid
and Lewis acid sites) due to changing Nb and Mo concentra-
tions [23,24]. The selectivity for dibenzylether, a by-product
3.3. Acid-catalyzed reactions
of benzyl alcohol, was 13%, 10%, and 7% for porous Nb7Mo3,
The acid catalyzed reactions were tested on porous NbxMo(
10−x)
Nb5Mo5, and Nb Mo7 oxides respectively. These results are con-
oxides using liquid-phase Friedel–Crafts alkylation of anisole with
benzyl alcohol. A plot of the product yield for different Nb and
Mo contents in the Friedel–Crafts alkylation of anisole, reacted
for 1 h at 373 K is shown in Fig. 7. Variation of the Nb and Mo
content resulted in remarkably different reaction rates of ben-
zylanisole formation in alkylation. The reaction rates increased
gradually with increasing Mo content, starting from 0% yield for
porous Nb oxide and reaching the highest alkylation yield (75.7%
3
sistent with the IR spectroscopic measurements results, which
−1
increase on Brønsted acid sites 1532 cm
promotes decrease
on selectivity of dibenzyl ether, a by-product formed by the
Lewis acids. However, dibenzyl ether is also a good alkylat-
ing agent, and its concentration decreased as it was consumed,
along with the benzyl alcohol, at the end of the alkylation reac-
tion.
Considering the similarity between the acid properties of bulk
of benzylanisole) for porous Nb Mo7 oxide. The yield decreased
3
Nb O5–MoO and porous NbxMo oxides, the decrease on reac-
drastically for non-porous oxide samples with x from 2 to 0, reach-
ing 0% for Mo oxide. The Mo enriched samples show formation
2
3
(10−x)
2
tion yield of Mo-enriched bulk Nb O5–MoO oxides should have
3
been caused by the surface area reduction. The surface areas of bulk
of Mo Nb O and MoO3 with no acidity and low surface areas,
3
2
14
2
−1
−
1
oxides decrease about 20 times from bulk Nb Mo (60 m g ) to
resulting in the decrease of catalytic activities. The TOF (h ) val-
ues (Table 2) obtained from the rate of produced benzylanisole
7
3
2
−1
bulk Nb Mo7 (3 m g ). Therefore the formation of porous struc-
3
ture for Mo-enriched samples (Nb Mo7 and Nb5Mo5) should have
(
1 h) and acid amount of porous NbxMo(
oxides also increase
3
10−x)
Table 2
Friedel–Crafts alkylation of anisole over several solid acid catalysts.
SBET (m2
g
−1
)
Acid Amount (mmol g
−1
)
Alkylation of anisolea
Catalyst
Selectivity of o-benzylanisole (%)b
TOF (h )c
−1
Yield (%)
Mesoporous Nb3Mo7
Mesoporous Nb5Mo5
Mesoporous Nb7Mo3
Mesoporous Nb
Bulk Nb3Mo7
57
70
81
200
3
16
0.24
0.27
0.29
0.54
0.04
0.10
0.16
0.40
1.0
75.7
45.0
19.1
N.D.
6.9
44.6
42.3
38.9
N.D.
43.8
40.1
39.0
N.D.
42.1
45.0
49.2
158
83
33
N.D.
86
45
53
N.D.
15
4
24
Bulk Nb5Mo5
Bulk Nb7Mo3
8.9
60
17.0
N.D.
30.6
42.1
42.3
Nb2O5•nH2O
128
420
50
d
H-beta
Amberlyst-15
Nafion NR50
4.8
0.9
0.02
a
Reaction conditions: anisole (100 mmol), benzyl alcohol (10 mmol), catalyst (0.2 g), 373 K, 1 h.
Selectivity (%) = (o-benzylanisole)/(o-benzylanisole + p-benzylanisole) × 100.
Calculated from the rate of produced benzylanisole (1 h) and acid amount.
SiO2/Al2O3 = 25, JRC-Z-HB25; N.D., not detected.
b
c
d

C. Tagusagawa et al. / Catalysis Today 164 (2011) 358–363
363
helped to maintain the high porous surface area, resulting in high
Acknowledgments
activity for porous oxides.
The high catalytic performance of porous Nb Mo7 oxide was
This work was supported by the Development in a New Inter-
disciplinary Field Based on Nanotechnology and Materials Science
program of the Ministry of Education, Culture, Sports, Science and
Technology (MEXT) of Japan and the Global Center of Excellence
Program for Chemistry.
3
due to the high surface area and formation of strong Brønsted
5+
acid sites by the isomorphous replacement of Nb by higher-
6
+
valence Mo
cations in these Mo-enriched samples like that
observed at MoO /ZrO2 [29]. It is reported that replacement of
3
ZrO2 by MoO3 forms acids sites similar to that of SO /ZrO2 [29]
4
with strong acid sites. The highest activity of MoO /ZrO2 has
References
3
been obtained at the range of surface molybdenum densities,
which maximizes the quantity of amorphous surface molyb-
[
[
[
1] I. Nowak, M. Ziolek, Chem. Rev. 99 (1999) 3603.
2] K. Tanabe, S. Okazaki, Appl. Catal. 133 (1995) 191.
3] K. Hanaoka, T. Takeuchi, T. Matsuzaki, Y. Sugi, Catal. Today 8 (1990) 123.
denum species [30]. Similar to MoO /ZrO2 and Nb O5–MoO ,
3
2
3
formation of strong Brønsted acid sites could be observed at
IR spectroscopic measurements results of porous NbxMo(
oxides. The Brønsted acid sites could have formed led to the
high surface molybdenum densities in niobium matrix. Higher
concentrations of Mo oxide formed the crystallized MoO3 and
[4] K. Hanaoka, T. Takeuchi, T. Matsuzaki, Y. Sugi, Catal. Lett. 5 (1990) 13.
[
[
[
5] K. Ogasawara, T. Iizuka, K. Tanabe, Chem. Lett. (1984) 645.
6] O. Okazaki, H. Harada, Chem. Lett. (1988) 1313.
7] J.J. Spivey, M.R. Gogate, J.R. Zoeller, R.D. Colberg, Ind. Eng. Chem. Res. 36 (1997)
4600.
10−x)
[8] T. Ushikubo, K. Wada, Appl. Catal. 67 (1990) 25.
9] L. Li, Y. Yoshinaga, T. Okuhara, Phys. Chem. Chem. Phys. 1 (1999) 4913.
10] K. Shibata, T. Kiyoura, J. Kitagawa, T. Sumiyoshi, K. Tanabe, Bull. Chem. Soc. Jpn.
6 (1973) 2985.
11] K. Tanabe, T. Sumiyoshi, K. Shibata, T. Kiyoura, J. Kitagawa, Bull. Chem. Soc. Jpn.
7 (1974) 1064.
[
Mo Nb O decreasing the reaction rate for Mo rich samples
3
2
14
[
[
(
x = 0–2).
4
4
4
. Conclusion
[
[
12] K. Yamashita, M. Hirano, K. Okumura, M. Niwa, Catal. Today 118 (2006) 385.
13] M. Hino, M. Kurashige, K. Arata, Catal. Commun. 5 (2004) 107.
Porous NbxMo(
oxides were found to function as highly
[14] K. Okumura, K. Yamashita, M. Hirano, M. Niwa, Chem. Lett. 34 (2005) 716.
10−x)
[
[
15] K. Okumura, K. Yamashita, M. Hirano, M. Niwa, J. Catal. 234 (2005) 300.
16] Y.-Y. Huang, T.J. McCarthy, W.M.H. Sachtler, Appl. Catal. A 148 (1996) 135.
active mixed metal oxide solid acid catalysts for Friedel–Crafts alky-
lation, exceeding the reaction rate of non-porous Nb O5–MoO3
oxide. Different from mesoporous NbxW(
2
[17] M. Chidambaram, D. Curulla-Ferre, A.P. Singh, B.G. Anderson, J. Catal. 220 (2003)
442.
and TaxW
(10−x)
10−x)
[
18] S.M. Landge, M. Chidambaram, A.P. Singh, J. Mol. Catal. A: Chem. 213 (2004)
57.
19] C.-C. Hwang, C.-Y. Mou, J. Phys. Chem. C 113 (2009) 5212.
oxides, NbxMo(
oxide formed mesopores only for Nb rich
10−x)
2
samples (x = 9–10), forming a large pores formed by inter-particle
voids for samples with x from 3 to 8. The formation of crystal-
lized MoO3 and Mo Nb O were observed in Mo-rich samples
[
[20] Y. Rao, M. Trudeau, D. Antonelli, J. Am. Chem. Soc. 128 (2006) 13996.
[
[
[
21] Y. Rao, J. Kang, D. Antonelli, J. Am. Chem. Soc. 130 (2008) 394.
22] J. Kang, Y. Rao, M. Trudeau, D. Antonelli, Angew. Chem. Int. Ed. 47 (2008) 4896.
23] C. Tagusagawa, A. Takagaki, A. Iguchi, K. Takanabe, J.N. Kondo, K. Ebitani, S.
Hayashi, T. Tatsumi, K. Domen, Angew. Chem. Int. Ed. 49 (2010) 1128.
24] C. Tagusagawa, A. Takagaki, A. Iguchi, K. Takanabe, J.N. Kondo, K. Ebitani, T.
Tatsumi, K. Domen, Chem. Mater. 22 (2010) 3072.
3
2
14
(
x = 0–4). The Nb–Mo interaction seems to be responsible for
the formation of inter-particle voids and crystallized structures.
The reaction rate and acid strength increased gradually with
the addition of Mo, reaching the highest reaction rate with
[
[
25] E.P. Parry, J. Catal. 2 (1963) 371.
porous Nb Mo7 oxide, which exceeded the reaction rate of
[26] T.R. Hughes, H.M. White, J. Phys. Chem. 71 (1967) 2192.
[27] M. Hino, M. Kurashige, H. Matsuhashi, K. Arata, Appl. Catal. A 310 (2009) 190.
3
ion-exchange resins, zeolites, and non-porous metal oxide. The
high catalytic performance of porous Nb Mo7 oxide was due
to the formation of strong Brønsted acid sites and high surface
area.
[28] T. Shishido, T. Kitano, K. Teramura, T. Tanaka, Catal. Lett. 129 (2009) 383.
[29] K. Arata, H. Matsuhashi, M. Hino, H. Nakamura, Catal. Today 81 (2003) 17.
[30] S.K. Maity, M.S. Rana, B.N. Srinivas, S.K. Bej, G. Murali Dhar, T.S.R. Prasada Rao,
J. Mol. Catal. Chem. 153 (2000) 121.
3