Bimodal cesium hydrogen salts of 12-tungstosilicic acid, Cs xH4-xSiW12O40, as highly active solid acid catalysts for transesterification of glycerol tributyrate with methanol

Article abstract of DOI:10.1016/j.jcat.2014.07.008

Bimodal Cs_xH_4-x[SiW₁₂O ₄₀] (Csx-bimodal) with mesopores interconnected with micropores were synthesized from microporous Cs_xH_4-x[SiW ₁₂O₄₀] (Csx-micro) with x = 1.0-2.5, which were prepared in advance by titrating an aqueous solution of H₄[SiW ₁₂O₄₀] with an aqueous solution of Cs₂CO ₃, followed by treatment in refluxing ethanol to mainly dissolve the H₄[SiW₁₂O₄₀] in the particles. Mesopore size distributions and their pore volumes changed depending on x in Csx-micro. Microporous Cs2.5-micro transformed into bimodal Cs2.5-bimodal with mesopores having average diameters of 3.7 nm and large mesopore volumes. Although Cs2.5-bimodal exhibited only low catalytic activity for the decomposition of isopropyl acetate, post-treatment in H₂SO₄ enhanced the catalytic activity due to substitution of the Cs⁺ ions on the surface with H⁺. H₂SO₄-treated Csx-bimodal showed high activity toward transesterification of glyceryl tributyrate with methanol due to its strong acid strength and mesoporosity.

Full text of DOI:10.1016/j.jcat.2014.07.008

Journal of Catalysis 318 (2014) 34–42
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Bimodal cesium hydrogen salts of 12-tungstosilicic acid,
CsxH4ꢀxSiW12O40, as highly active solid acid catalysts for
transesterification of glycerol tributyrate with methanol
Yukari Iwase a, Shogo Sano b, Lina Mahardiani b, Ryu Abe a, Yuichi Kamiya c,
⇑
a Department of Energy and Hydrocarbon Chemistry, Graduate School of Engineering, Kyoto University, Katsura, Nishikyo-ku, Kyoto 615-8510, Japan
b Graduate School of Environmental Science, Hokkaido University, Nishi 5, Kita 10, Kita-ku, Sapporo 060-0810, Japan
c Research Faculty of Environmental Earth Science, Hokkaido University, Nishi 5, Kita 10, Kita-ku, Sapporo 060-0810, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Bimodal CsxH4ꢀx[SiW12O40
] (Csx-bimodal) with mesopores interconnected with micropores were
Received 7 May 2014
Revised 3 July 2014
Accepted 13 July 2014
synthesized from microporous CsxH4ꢀx[SiW12O40] (Csx-micro) with x = 1.0–2.5, which were prepared
in advance by titrating an aqueous solution of H4[SiW12O40] with an aqueous solution of Cs2CO3, followed
by treatment in refluxing ethanol to mainly dissolve the H4[SiW12O40] in the particles. Mesopore size
distributions and their pore volumes changed depending on x in Csx-micro. Microporous Cs2.5-micro
transformed into bimodal Cs2.5-bimodal with mesopores having average diameters of 3.7 nm and large
mesopore volumes. Although Cs2.5-bimodal exhibited only low catalytic activity for the decomposition
of isopropyl acetate, post-treatment in H2SO4 enhanced the catalytic activity due to substitution of the
Cs+ ions on the surface with H+. H2SO4-treated Csx-bimodal showed high activity toward transesterifica-
tion of glyceryl tributyrate with methanol due to its strong acid strength and mesoporosity.
Ó 2014 Elsevier Inc. All rights reserved.
Keywords:
Heteropolyacid
Mesopore
Bimodal
Biodiesel
Solid acid
Dissolution
1. Introduction
forming mesopores that pass through the zeolite crystallite by
dissolving a part of the zeolite crystallite prepared in advance
Bimodal porous materials having ordered micropores and
ordered mesopores that are interconnected can be used as cata-
lysts for reactions involving bulky reactants and products [1–17].
Bimodal solid catalysts are expected to show high catalytic activi-
ties because they have large numbers of active sites due to high
surface areas arising from the micropores and because of fast diffu-
sion of the reactants in the interconnected mesopores. In addition,
the bimodal pores can suppress catalyst deactivation due to coking
in reactions involving hydrocarbon conversion because of the
smooth release of the products from the area around the active site
through the interconnected mesopores [8,18–20]. Thus, rational
design of hierarchical pore structures and the active sites in the
catalysts is important for developing highly active, selective, and
highly durable solid catalysts.
[26–29]. In the former method, hard (e.g., mesoporous carbons
[21], carbon nanoparticles [24], and carbon nanotubes [23]) and
soft (e.g., micelles formed with surfactants [22]) materials with
ordered mesostructures are used as templates to create mesopores.
Zeolites are crystallized inside the nanospaces of the template or
around the template, followed by the removal of the template to
form the mesopores.
On the other hand, a typical example of the latter method is to
heat zeolites in a strong alkaline solution to dissolve a part of the
zeolite crystallites [26–29]. Ogura et al. have reported that thermal
treatment of MFI zeolites in aqueous NaOH solutions results in the
selective dissolution of siliceous species in the zeolite to form mes-
opores with uniform sizes [27,30]. The treatment doubles the dif-
fusion rate of benzene inside the catalyst particle, and the
obtained bimodal MFI zeolite shows higher catalytic activity for
cumene cracking than the original microporous one does [27].
Although bimodal zeolites with ordered mesopores are useful solid
catalysts, their uses as solid acid catalysts are limited because their
acid strengths are relatively weak. Thus, there has been strong
interest toward the development of bimodal solids with strong
acid sites, other than zeolites, in order to expand the range of their
possible applications to challenging acid-catalyzed reactions.
There has been extensive research to prepare bimodal solid cat-
alysts in crystalline microporous aluminosilicate zeolites with
ordered mesopores. Methods for fabricating such bimodal zeolites
involve (i) forming mesopores in the zeolite particles concurrently
with crystallization of the zeolite framework [21–25] and (ii)
⇑
Corresponding author. Fax: +81 11 706 2217.
E-mail address: kamiya@ees.hokudai.ac.jp (Y. Kamiya).
http://dx.doi.org/10.1016/j.jcat.2014.07.008
0021-9517/Ó 2014 Elsevier Inc. All rights reserved.

Y. Iwase et al. / Journal of Catalysis 318 (2014) 34–42
35
Keggin-type heteropolyacids, Hn[XY12O40], are molecular metal
oxide clusters with discrete molecular structures [31–33]. Keggin-
type heteropolyacids with tungsten as addenda atoms, including
H3[PW12O40] and H4[SiW12O40], have strong Brønsted acid sites,
which are stronger than those of common aluminosilicate zeolites
and are comparable to those of sulfated zirconia [34–36]. In fact,
initial heats of ammonia adsorption for H3[PW12O40] and H4[SiW12-
O40] are 195 and 175 kJ molꢀ1, respectively [37,38], whereas those
for MFI zeolites and amorphous silica-alumina are 150 and
145 kJ molꢀ1, respectively [35,39]. Thus, heteropolyacids are classi-
fied as superacids [31]. Moreover, the superacidity of Keggin-type
heteropolyacids has been demonstrated on the basis of solid-state
31P MAS NMR of trymethyl phosphine oxide (TMPO) adsorbed on
them [40,41].
treatment and structural and compositional analyses of the Cs salts
before and after treatment. Furthermore, we investigated the
post-treatment of the resulting bimodal CsxH4ꢀx[SiW12O40] in
sulfuric acid to increase the number of acid sites and consequently
to improve the catalytic activity toward acid-catalyzed reactions.
Finally, we conducted transesterification of glycerol tributyrate
with methanol over the sulfuric acid-treated bimodal
CsxH4ꢀx[SiW12O40] to demonstrate the advantage of mesopores in
catalysts for reactions involving bulky reactants.
2. Experimental
2.1. Materials
Although solid-state Keggin-type heteropolyacids are non-
porous and thus have small surface areas (<15 m2 gꢀ1), substitution
of H+ with Cs+ leads to an increase in the surface area and to the
formation of pores. The surface area and pore structure change
depending on the degree of the substitution of H+ [42–45]. Thus
far, the syntheses, structural analyses, and catalytic activity of neu-
tral and acidic Cs+ salts of H3[PW12O40] have been extensively
investigated [31,44,46–50]. It has been reported that Cs2.1H0.9
[PW12O40] is microporous and has no mesopores and that it
Dodecatungstosilicic acid (H4[SiW12O40]ꢁnH2O) was supplied by
Nippon Inorganic Color and Chemical Co. After H4[SiW12O40]ꢁnH2O
was recrystallized from water several times, it was dried in a
vacuum at 338 K to obtain H4[SiW12O40]ꢁ8H2O.
Cesium carbonate, ethanol, propanol, isopropyl acetate, n-hexyl
ether, 1,3,5-triisopropyl benzene, n-decane, and sulfuric acid were
obtained from Wako Pure Chem. Ind., Ltd. and used as received.
Glycerol tributyrate (C15H26O6, Tokyo Chem. Ind. Co., Ltd.) and
methanol (Wako Pure Chem. Ind., Ltd.) were treated with anion-
exchange resin (Amberlite IRA400J CL, Organo Co.) and cation-
exchange resin (Amberlite 200C, Organo Co.) to remove both acidic
and basic impurities before use.
acts as
a size-selective catalyst for acid-catalyzed reactions
[44,47–49]. Since Cs2.1H0.9[PW12O40] crystallites themselves have
no micropores [47], their micropores form due to crystallographic
mismatching between the facets of the crystallites [47].
Zeolites including H-Y (JRC-Z-HY5.6), H-b (JRC-Z-HB25H), and
H-ZSM-5 (JRC-Z5-25H) and silica-alumina (JRC-SAH-1) were sup-
plied from the Catalysis Society of Japan as reference catalysts.
Cs2.5H0.5[PW12O40] was prepared by using a previously reported
procedure [50].
On the other hand, the pores of CsxH3ꢀx[PW12O40] with x P 2.3
show bimodal distributions, ranging from microporous to
mesoporous, and the interparticle spaces form the mesopores
[51]. CsxH3ꢀx[PW12O40] with x = 2.5 has a large surface area and
consequently the largest number of surface acid sites among
CsxH3ꢀx[PW12O40] [52]. Due to the large number of surface acid
sites, bimodal pores, and strong acid strengths, the Cs salts exhibit
extremely high catalytic activities for various acid-catalyzed
reactions involving bulky reactants [52–54].
2.2. Synthesis
2.2.1. Microporous CsxH4ꢀx[SiW12O40] (Csx-micro)
Although the molecular structure of H4[SiW12O40] is similar
CsxH4ꢀx[SiW12O40] with different Cs contents (x = 1.0, 1.5, 2.0,
2.5, 3.0, 3.5, and 4.0) were prepared by titrating an aqueous solu-
tion of H4[SiW12O40] (0.08 mol dmꢀ3, 20 cm3) with an aqueous
solution of Cs2CO3 (0.10 mol dmꢀ3) at room temperature while vig-
orously stirring [45]. x was adjusted by changing the amount of the
Cs2CO3 solution added, which was carefully controlled by using an
automatic syringe pump: first, 2 cm3 was added at a rate of
0.033 cm3 minꢀ1, and then, the remaining solution was added at
a rate of 0.2 cm3 minꢀ1. The resulting colloidal solution was
allowed to stand for 48 h at room temperature, and then the sol-
vent was evaporated at 313 K to obtain CsxH4ꢀx[SiW12O40] as a
white solid. After grinding the solid, it was dried in a vacuum at
room temperature. The resulting solids are denoted as Csx-micro,
in which x represents the Cs content.
to that of H3[PW12O40
] and the major difference between
H4[SiW12O40] and H3[PW12O40] is the electronic charges of the
heteropoly anions, which are ꢀ4 and ꢀ3, respectively, the
mechanisms for the formation of the pores and for the changes
in the pore structure on the degree of substitution of H+ with Cs+
in H4[SiW12O40] are quite different from those of H3[PW12O40]. In
the crystal of the Cs salts of H4[SiW12O40] (CsxH4ꢀx[SiW12O40]), het-
eropoly anions ([SiW12O40]
4ꢀ) are packed in a body-centered cubic
(bcc) cell and cations exist between anions. CsxH4ꢀx[SiW12O40] form
microporous solids with low external surface areas (less than 4% of
the total surface area) and do not have mesopores, regardless of the
substitution degree of H+ [45]. Microporous Cs3H[SiW12O40], which
has the highest surface area among CsxH4ꢀx[SiW12O40], acts as a
size-selective catalyst toward acid-catalyzed reactions in the liquid
phase due to its uniform micropores and small external surface area
[45]. The micropores in Cs3H[SiW12O40] originate from defects in
2.2.2. Bimodal CsxH4ꢀx[SiW12O40] (Csx-bimodal)
Csx-micro (2 g) was added to ethanol (30 cm3), and the suspen-
sion was refluxed at 347 K for 4 h with stirring. The remaining solid
the heteropoly anions, which form to avoid mismatches in the
4ꢀ
Cs+/[SiW12O40
]
ratio required for charge balance (=4/1) and for
a bcc structure (=3/1) [45,55,56]. If the ordered mesopores that
are interconnected with the micropores are present in microporous
CsxH4ꢀx[SiW12O40], the resulting bimodal Cs salts should show high
catalytic activity for acid-catalyzed reactions involving bulky reac-
tants. In other words, bimodal porosity will expand the possible
applications of CsxH4ꢀx[SiW12O40] as a solid acid catalyst.
was collected on a membrane filter (£ 0.2 lm) and washed three
times with ethanol. After the solid was dried at room temperature,
it was weighed to estimate the dissolution rate (Eq. (1)).
Dissolution rate of Csx-micro ð%Þ
2-Weight of the solid remaining after the treatment ½gꢂ
¼
ꢃ100 ð1Þ
In the present study, we synthesized bimodal CsxH4ꢀx[SiW12O40
]
Initial weight of the solid ½2gꢂ
with micropores and interconnected mesopores by treating micro-
porous CsxH4ꢀx[SiW12O40] with x = 1.0–2.5 in refluxing ethanol. A
formation mechanism of the mesopores is discussed on the basis
of the dissolution rate of microporous CsxH4ꢀx[SiW12O40] during
The solid was calcined in air at 523 K for 4 h. The resulting
solids are denoted as Csx-bimodal, in which x represents the Cs
content in the original Csx-micro.

36
Y. Iwase et al. / Journal of Catalysis 318 (2014) 34–42
To investigate the influence of the solvent, Cs2.5-micro was
treated in water and 1-propanol in a similar manner to that in
ethanol.
Transesterification of glyceryl tributyrate with methanol was
also performed in the batch type reactor at 333 K. The catalyst
(0.2 g) was pretreated in a flow of N2 at 473 K for 0.5 h. Methanol
(300 mmol), n-hexyl ether (internal standard, 2.5 mmol), and glyc-
eryl tributyrate (10 mmol) were introduced into the reactor, and
then the reactor was heated at 333 K while the solution was stir-
red. After 24 h, the catalyst was separated by centrifugation
(9000 rpm and 10 min), and the supernatant was analyzed by
using a gas chromatograph (Shimadzu, GC-14B) equipped with a
capillary column (Neutrabond-1, 60 m ꢃ 0.25 mm) and a flame
ionization detector (FID-GC). Amberlyst-15 (Organo Co.) was pre-
treated at 373 K to prevent catalyst decomposition. The yield of
methyl butyrate was defined as shown in Eq. (3).
2.2.3. Post-treatment of Cs2.5-bimodal in sulfuric acid
Cs2.5-bimodal (1.0 g) was added to 25% sulfuric acid (15 cm3),
and the suspension was heated at 353 K for 2 h with stirring. The
solid was collected on a membrane filter (£ 0.2 lm) and washed
with water. The solid was dried under ambient conditions for
24 h and then calcined in air at 523 K for 4 h. The resulting solid
is denoted as Cs2.5-bimodal(H2SO4).
2.3. Characterization
½Amount of formed methyl butyrateꢂ
Yield ð%Þ ¼
ꢃ 100
Nitrogen adsorption–desorption isotherms were measured at
77 K on a Belsorp-mini instrument (BEL Japan Inc.) after pretreat-
ment of the samples at 473 K in a flow of N2 for 0.5 h. Mesopore
size distributions and mesopore volumes were determined by
using the Dollimore-Heal (DH) method on an N2 desorption iso-
therm. The micropore-size distribution was determined by using
the Saito-Foley method [57] on an Ar adsorption isotherm at
87 K measured with an automatic apparatus (Belsorp 28SA, BEL
Japan Inc.). Before measurement, the sample was pretreated at
473 K in a vacuum for 10 h.
½Initial amount of glycerol tributyrateꢂ ꢃ 3
ð3Þ
3. Results and discussion
3.1. Formation of mesopores by treatment of Csx-micro in refluxing
ethanol
Fig. 1 shows N2 adsorption–desorption isotherms for Csx-micro
and Csx-bimodal. In the case of Csx-micro, a type I isotherm was
observed regardless of x, indicating that these were microporous
materials [45]. The amount of N2 uptake for Csx-micro increased
with an increase in x to 3.0 and 3.5 (Fig. 1e and f, circles). As a
Powder X-ray diffraction (XRD) was performed using an X-ray
diffractometer (Rigaku Mini Flex) with Cu
Ka radiation
(k = 0.154 nm). Crystallite size was determined by using Scherrer’s
equation with a diffraction line at 2h = 25–26°. Elemental analyses
of the samples were performed on an atomic absorption spectrom-
eter (Hitachi A-2000) and an inductively coupled plasma-atomic
emission spectrometer (Shimadzu ICPS-7000) for Cs and W,
respectively. Scanning electron microscopy (SEM) images were
obtained on an FE-SEM (Hitach, S-4800).
Temperature-programmed desorption of ammonia (NH3-TPD)
was carried out by using a multi-task TPD system (BEL Japan
Inc.) equipped with a quadrupole mass spectrometer. After pre-
treatment at 473 K for 1 h in a flow of He (50 cm3 minꢀ1), the cat-
alyst was exposed to NH3 at 13 kPa and 373 K for 0.5 h, and then,
the excess NH3 was removed in a He flow at 373 K for 0.5 h. The
temperature of the catalyst was increased at a rate of 10 K minꢀ1
to 1073 K, and the desorbed gas was monitored at m/z = 16.
result, Cs3.0-micro (192 m2 gꢀ1) and Cs3.5-micro (187 m2 gꢀ1
)
had the largest surface areas among Csx-micro (Fig. S1). For Csx-
micro with x = 1.0–2.5, the isotherms were dramatically changed
by treatment in refluxing ethanol, and those of the corresponding
Csx-bimodal were type IV isotherms. The results indicate that
the treatment causes the formation of mesopores in Csx-bimodal
with x = 1.0–2.5 (Fig. S2). In the isotherm for Cs2.5-bimodal, a
clear hysteresis loop was observed. In addition, a steep increase
in the N2 adsorption amount at low pressures remained even after
treatment, indicating that the micropores were preserved. The sur-
face areas were slightly increased for the samples with x = 1.0–2.5
by the treatment. In contrast, no change was observed in the
isotherms for the materials with x P 3.
Fig. 2 shows pore-size distributions of the micropores and mes-
opores in Cs2.5-micro and Cs2.5-bimodal. In the micropore-size
distribution for Cs2.5-micro (solid line), an intense peak at
2.4. Catalytic reaction
Decomposition of isopropyl acetate was performed in a batch
type reactor at 373 K. After the catalyst (0.1 g) was pretreated in
a flow of N2 at 473 K for 0.5 h, n-decane (solvent, 28 cm3), 1,3,5-tri-
isopropyl benzene (internal standard, 4 mmol) and isopropyl ace-
tate (8 mmol) were introduced into the reactor. Then, the reactor
was heated at 373 K while the solution was stirred. Samples (ca.
0.1 cm3) were taken from the reaction solution at 0.5, 1, 1.5, and
2 h. The catalyst in the sample solution was immediately separated
by centrifugation (9000 rpm and 10 min), and the supernatant was
analyzed by using a gas chromatograph (Shimadzu, GC-2010)
equipped with a capillary column (Neutrabond-1, 60 m ꢃ 0.25
mm, GL Science Inc.) and a flame ionization detector (FID-GC).
The conversion was defined as Eq. (2).
0.52 nm (Micropore
pore b) were observed. It has been reported that cesium hydrogen
salts of 12-tungstosilicic acid, like Cs3.0H0.3[SiW12O40 0.83
ꢁ3H2O
[56] and Cs3H[SiW12O40] [45], have anion ([SiW12O40
4ꢀ) vacancies
in the crystal lattice to compensate for the excess negative charges
and that these vacancies form the micropores [45]. After the treat-
ment in refluxing ethanol, Micropore b was preserved, whereas
a) and a shoulder at around 0.62 nm (Micro-
]
]
Micropore
a partially collapsed. The changes in the micropore-size
distributions due to the treatment will be discussed later.
In contrast to the micropore-size distributions, the mesopore
size ones were completely different after the treatment. Although
for Cs2.5-micro, there was only a small peak in the mesopore
region, a clear intense peak at 3.7 nm appeared for Cs2.5-bimodal.
½Initial amount of isopropyl acetateꢂ-½Amount of isopropyl actetate at x hꢂ
Conversion ð%Þ ¼
ꢃ 100
ð2Þ
½Initial amount of isopropyl acetateꢂ
No product other than acetic acid formed in the liquid phase, and
propylene separated from the solution.
The mesopore volume of Cs2.5-bimodal was 71 mm3 gꢀ1. Cs2.5-
bimodal was thermally stable in air up to 723 K, which was

Y. Iwase et al. / Journal of Catalysis 318 (2014) 34–42
37
(a)
(b)
10
2
4
6
8
10
0.4 0.6 0.8 1.0
Pore diameter/nm
(c)
(d)
Fig. 2. Micropore and mesopore size distributions in Cs2.5-micro (solid line) and
Cs2.5-bimodal (dashed line).
mesopore volumes, but there were no or only little changes for
the materials with x = 3.5 and 4.
Acid-form Keggin-type heteropolyacids, like H4[SiW12O40], are
usually highly soluble in water and polar organic solvents, includ-
ing lower alcohols [31]. On the other hand, salts of Keggin-type
heteropolyacids with large monovalent cations, such as Cs+, have
low solubility in those solvents [53]. Thus, we believe that the mes-
opores form by dissolution of a part of Csx-micro in ethanol.
Therefore, we determined the dissolution rate of Csx-micro during
treatment in ethanol. There was a distinct correlation between the
dissolution rate and x in Csx-micro, as shown in Fig. 4. Although
H4[SiW12O40] completely dissolved in ethanol under the present
conditions (data not shown), the Cs salts were less soluble. How-
ever, 61% of Cs1.0-micro was soluble in ethanol. The dissolution
rate monotonically decreased with an increase in x. It is noted that
Cs3.5-micro and Cs4.0-micro, both of which did not become
bimodal Cs salts, were basically insoluble in ethanol under the
present conditions.
(e)
(f)
The Cs/polyanion ratio (=Cs/12 W) in the materials before (Csx-
micro) and after (Csx-bimodal) the treatment gave valuable infor-
mation about the component that dissolved during the treatment.
In Fig. 5, Cs/polyanion ratios for Csx-micro and Csx-bimodal are
plotted as a function of x. The dashed line in the figure represents
Cs/polyanion = x. The Cs/polyanion ratios for Csx-bimodal were
higher than those for the corresponding Csx-micro for x = 1.0–
2.5. For example, the Cs/polyanion ratio increased from 2.46 for
Cs2.5-micro to 2.95 for Cs2.5-bimodal. The treatment caused a
larger difference in the Cs/polyanion ratio for the materials with
low x than it did for those with large x. The results clearly indicate
that the part of Csx-micro with low x dissolved into ethanol during
the treatment. In the case of Cs2.5-micro, we estimated from the
Cs/polyanion ratio and dissolution rate that the dissolving part of
Csx-micro had x = 0.65. In contrast, there was no change in the
Cs/polyanion ratio for Cs3.5-micro and Cs4.0-micro upon the
treatment due to no dissolution.
Fig. 6 shows XRD patterns for Csx-micro and Csx-bimodal, and
the XRD pattern for H4[SiW12O40] (Cs0.0) is included as a reference.
All diffraction lines for Cs3.5-micro and Cs4.0-micro were
assigned to a cubic structure, in which heteropoly anions are
packed in a bcc cell. The lattice constant (a) for Cs4.0-micro was
1.177 nm, and this cubic phase is referred to as phase C. The XRD
pattern for H4[SiW12O40] (Cs0.0) was also assigned to a cubic struc-
ture similar to Cs4.0-micro, but a was 1.212 nm (phase A). The
large value of a for phase A is due to occupation of the cation sites
with hydrated protons ((H2O)nH+), which are larger than Cs+. For
Csx-micro with x = 1.5–2.5, diffraction lines were broad, and the
diffraction angles near 2h = 26° were between those of phase C
(2h = 26.20°) and those of phase A (2h = 25.40°). As will be dis-
cussed later, Csx-micro with x = 1.5–2.5 were composed of three
(g)
Fig. 1. N2 adsorption–desorption isotherms at 77 K for Csx-micro and Csx-
bimodal. x = (a) 1.0, (b) 1.5, (c) 2.0, (d) 2.5, (e) 3.0, (f) 3.5, and (g) 4.0. Circle: Csx-
micro, and square: Csx-bimodal. Filled symbol: adsorption branch, and open
symbol: desorption branch.
confirmed from XRD patterns and N2 adsorption–desorption iso-
therms. It seems that the mesopore volume per unit weight for
Cs2.5-bimodal is not large in comparison with that of mesoporous
silica, such as MCM-41 (>700 mm3 gꢀ1) [58]. However, if the heavy
formula weight of the heteropolyacid is taken into account
(FWCs3H[SiW2O40] = 3273 and FWSiO2 = 60), it can be concluded that
that Cs2.5-bimodal is highly porous.
3.2. Formation mechanism of mesopores by the treatment of Csx-
micro in refluxing ethanol
The volumes of the mesopores in Csx-micro and Csx-bimodal
estimated from N2 desorption isotherms are compared in Fig. 3.
Csx-micro with x 6 2.5 transformed into Csx-bimodal with large

38
Y. Iwase et al. / Journal of Catalysis 318 (2014) 34–42
100
50
0
1
2
3
4
x in Csx-micro
Fig. 3. Dependence of mesopore volumes on the Cs content (x) before and after the
treatment in ethanol. (s) Csx-micro and (d) Csx-bimodal.
100
50
0
Fig. 6. XRD patterns of (a) Csx-micro and (b) Csx-bimodal.
(a)
1
2
3
4
x in Csx-micro
Fig. 4. Relationship between the dissolution rate of Csx-micro treated in ethanol
and Cs content (x). Treatment conditions: Csx-micro, 2 g; volume of ethanol,
30 cm3; temperature, 347 K; time, 4 h. After treatment, the remaining solid was
collected on a membrane filter, washed with ethanol, dried in air, and weighted.
100 nm
4
3
2
1
0
(b)
1
2
3
4
x in Csx-micro
Fig. 5. Cs/polyanion (=Cs/12 W) ratio in (s) Csx-micro and (d) Csx-bimodal. The
dashed line represents Cs/polyanion = x.
100 nm
Fig. 7. SEM images of (a) Cs2.5-micro and (b) Cs2.5-bimodal.
phases (phase A, B, and C) and the value of a for phase B was
between those for phases A and C. In phase B, the cation sites were
occupied with Cs+ and hydrated H+. In the XRD pattern for Cs1.0-
micro, two sets of the cubic XRD patterns consistent with phases
A and B were observed.
After treating Cs1.0-micro in ethanol, phase A (H4[SiW12O40])
completely disappeared, and only the XRD pattern for phase B with
a = 1.187 nm was observed. For Csx-bimodal with x = 1.5–2.5, the
diffraction lines became sharp after the treatment. In addition,
the skirts of the diffraction lines broadened toward lower 2h in
Csx-micro disappeared after the treatment. These results support
the fact that H4[SiW12O40] and/or CsxH4ꢀx[SiW12O40] with low x
dissolved in ethanol and that CsxH4ꢀx[SiW12O40] with high x did
not. On the other hand, the XRD patterns for Cs3.5-bimodal and
Cs4.0-bimodal did not change at all.
Dissolution of a part of Csx-micro upon treatment was directly
observed by using SEM (Fig. 7). In an SEM micrograph for Cs2.5-
micro (Fig. 7a), fine particles (primary particle) with a size of ca.
10 nm were observed. The fine particles tightly aggregated to form
secondary particles, eliminating any interstitial voids between the
fine particles. The shape of the secondary particles and the size of
the fine particles for Cs2.5-bimodal were quite similar to those for
Cs2.5-micro. However, it appeared that some fine particles had
fallen off the secondary particles forming interstitial voids in the

Y. Iwase et al. / Journal of Catalysis 318 (2014) 34–42
39
secondary particles. The aperture sizes of the interstitial voids
were less than 10 nm, which roughly agreed with the mesopore
size estimated from the N2 adsorption–desorption isotherm, indi-
cating that the interstitial voids were mesopores.
To elucidate the formation mechanism of the mesopores, the
formation process and microstructure of Csx-micro was clarified
first. Here those of Cs2.5-micro and Cs2.5-bimodal will be
described as examples. During the preparation of Cs2.5-micro,
the addition of Cs2CO3 solution to H4SiW12O40 solution caused a
white solid to precipitate from the milky suspension (Scheme 1).
To obtain Cs2.5-micro, water was removed from the milky suspen-
sion by evaporation. To investigate the structure and composition
of the precipitate and milky solution separately, the precipitate
was removed by filtration, and the resulting milky solution was
evaporated to form a solid (Scheme 1). The precipitate and solid
obtained by evaporation of the milky solution are denoted as
Cs2.5(precipitate) and Cs2.5(solution), respectively. The mass
ratio of Cs2.5(precipitate) to Cs2.5(solution) was 59/41. The Cs/
polyanion ratio of Cs2.5(precipitate) was 3.7, which is much larger
than 2.5. In the XRD pattern for Cs2.5(precipitate), peaks consis-
tent with phase C were observed, giving a Type I isotherm of N2
adsorption–desorption (Fig. 8a), which were the same as those
for Cs3.5-micro and Cs4.0-micro. In addition, Cs2.5(precipitate)
did not dissolve in ethanol at all, and the XRD pattern and N2
adsorption–desorption isotherm were the same after the treat-
ment in refluxing ethanol (Fig. 8b)). The size of the crystallites of
Cs2.5(precipitate) calculated from Scherrer’s equation was
39 nm. From these results, we conclude that Cs2.5(precipitate) is
microporous crystallites of Cs3.7H0.3[SiW12O40] with 39 nm in size.
On the other hand, Cs/polyanion ratio of Cs2.5(solution) was 1.0
Fig. 8. XRD patterns and N2 adsorption–desorption isotherms for Cs2.5(precipi-
tate) and Cs2.5(solution) and for the materials obtained after treatment in ethanol.
(a) Cs2.5(precipitate), (b) material obtained after treatment of Cs2.5(precipitate) in
ethanol, (c) Cs2.5(solution), and (d) material obtained after treatment of
Cs2.5(solution) in ethanol.
type II (Fig. 8d), which was dissimilar to that for Cs2.5-bimoidal.
These results indicate that phase B is Cs2.2H1.8[SiW12O40] with
23 nm in size and that it does not have any micropores.
From these results, we proposed the model shown in Scheme 2.
Cs2.5-micro contained three crystallites with different x and crys-
tallite sizes: H4[SiW12O40], Cs2.2H1.8[SiW12O40](23 nm), and Cs3.7-
H0.3[SiW12O40](39 nm), which correspond to phases A, B, and C,
respectively. During the titration process for preparing Cs2.5-
micro, Cs3.7H0.3[SiW12O40](39 nm) precipitates first. In the milky
supernatant, fine particles of Cs2.2H1.8[SiW12O40](23 nm) are sus-
pended, and H4[SiW12O40] dissolves. By evaporating the suspen-
sion, the fine particles of Cs2.2H1.8[SiW12O40](23 nm) become
physically mixed with those of Cs3.7H0.3[SiW12O40](39 nm), and
the interstitial voids between the crystallites are finally filled with
H4[SiW12O40] as a glue to form the secondary particles (Fig. 7(a)).
When Cs2.5-micro is treated in ethanol, H4[SiW12O40] dissolves,
and thus Cs2.5(solution) contained
a
small amount of Cs+
and from XRD, it was confirmed to be a mixture of phase A
(H4[SiW12O40]) and phase B (Fig. 8c). The amount of N2 uptake
for Cs2.5(solution) was small (Fig. 8c), and no micropores or mes-
opores were present. Upon treating Cs2.5(solution) in refluxing
ethanol, 53% of the material dissolved, and the Cs/polyanion ratio
became 2.2. Only the XRD pattern for phase B with a crystallite size
of 23 nm (Fig. 8d) was observed for the treated Cs2.5(solution),
meaning that phase A (H4[SiW12O40]) was completely eliminated.
The resulting solid gave a N2 adsorption–desorption isotherm of
Scheme 1.

40
Y. Iwase et al. / Journal of Catalysis 318 (2014) 34–42
Scheme 2.
forming interstitial voids at those sites. The interstitial voids are
the mesopores in Cs2.5-bimodal.
Cs2.5-micro
The presence of three crystallites in the appropriate ratio is
crucial for the formation of the mesopores with regular shapes in
Csx-bimodal. When H4[SiW12O40] and Cs2.2H1.8[SiW12O40](23 nm)
are absent, as in the cases of Cs3.5-micro and Cs4.0-micro, no
mesopores form because Csx-micro contains only insoluble
Cs3.7H0.3[SiW12O40](39 nm). On the other hand, when a large
amount of H4[SiW12O40] is present in Csx-micro, as in the case of
Cs1.0-micro, mesopores with wide distributions of sizes form
because the shapes and sizes of the voids formed between the
insoluble crystallites are irregular. In addition, the choice of solvent
used for the treatment is another factor controlling the shape and
volume of the mesopores. When water was used as a solvent for
the treatment, mesopores with wide distributions of sizes formed
(Fig. S3). This is because not only H4[SiW12O40] but also a part of
Cs2.2H1.8[SiW12O40](23 nm) are soluble in water due to its high
polarity. In contrast, the treatment of Cs2.5-bimodal in 1-propanol
led to the formation of mesopores with small volumes due to the
low solubility of H4[SiW12O40] (Fig. S3).
Cs2.5-bimodal
Cs2.5-bimodal(H2SO4)
0
10
20
30
40
Conversion/%
Fig. 9. Results from catalytic decomposition of isopropyl acetate over Cs2.5-micro,
Cs2.5-bimodal, and Cs2.5-bimodal(H2SO4). Reaction conditions: isopropyl acetate,
8 mmol; n-decane (solvent), 28 cm3; 1,3,5-triisopropyl benzene (internal standard),
4 mmol; catalyst weight, 0.1 g; reaction temperature, 373 K; reaction time 2 h.
c
b
During treatment of Cs2.5-micro in ethanol, H4[SiW12O40
]
dissolved, and thus, Cs2.5-bimodal consists of insoluble Cs3.7H0.3
[SiW12O40](39 nm) and Cs2.2H1.8[SiW12O40](23 nm) crystallites.
Since Cs2.2H1.8[SiW12O40](23 nm) and H4[SiW12O40
] have no
micropores, Micropore b (0.62 nm) of Cs2.5-micro and Cs2.5-
bimodal (Fig. 2) must be present in the crystal lattices of Cs3.7H0.3
a
[SiW12O40](39 nm). On the other hand, Micropore
appears to be interparticle voids formed mainly at the interface
between the outer surface of Cs3.7H0.3[SiW12O40](39 nm) and
a (0.52 nm)
673
873
1073
473
Desorption temperature/K
H4[SiW12O40] because Micropore
in refluxing ethanol. However, further studies are needed.
a collapsed during the treatment
Fig. 10. Ammonia-TPD profiles of (a) Cs2.5-micro, (b) Cs2.5-bimodal, and (c)
Cs2.5-bimodal(H2SO4).
3.3. Catalytic properties and post-treatment of Cs2.5-bimodal in
sulfuric acid
Cs2.5-bimodal was only one-third of that of Cs2.5-micro. The
decrease in the catalytic activity is related to the dissolution of H4-
[SiW12O40] into ethanol, by which the mesopores were formed. In
fact, the acid amount for Cs2.5-bimodal estimated from NH3-TPD
(Fig. 10) was 0.010 mmol gꢀ1, which is lower than that for Cs2.5-
micro (0.099 mmol gꢀ1). In addition, the peak around 773 K, which
corresponded to acidic protons, disappeared for Cs2.5-bimodal.
To increase the number of acid sites in Cs2.5-bimodal, Cs2.5-
bimodal was treated in sulfuric acid to replace the Cs+ ions on
the surface with H+. As shown in Fig. 10, the peak due to acidic pro-
tons (773 K) partially recovered after post-treatment in sulfuric
acid, and the acid amount of Cs2.5-bimodal(H2SO4) increased to
0.028 mmol gꢀ1. The crystalline structure of Cs2.5-bimodal was
completely preserved during the post-treatment (Fig. S4) and there
Fig. 9 shows catalytic activities of Cs2.5-micro, Cs2.5-bimodal,
and Cs2.5-bimodal(H2SO4) toward the decomposition of isopropyl
acetate (Eq. (4)) in n-decane at 373 K for 2 h.
ð4Þ
Cs2.5-bimodal(H2SO4) was obtained by post-treatment of
Cs2.5-bimodal in sulfuric acid (see Section 2.2.3). Treatment in
ethanol decreased the catalytic activity; the catalytic activity of

Y. Iwase et al. / Journal of Catalysis 318 (2014) 34–42
41
30
20
10
0
1st Run
Cs2.5-micro
2nd Run
Catalyst was
filtered off.
Cs2.5-bimodal
1st Run
2nd Run
Cs2.5-bimodal(H2SO4)
Cs2.5H0.5[PW12O40
]
H-Y
H-β
H-ZSM-5
0
50
100
150
Reaction time/min
SiO2-Al2O3
Amberlyst-15
Fig. 11. Time courses for decomposition of isopropyl acetate over Cs2.5-
bimodal(H2SO4) with the catalyst (d) present during the entire reaction and (s)
after it was removed from the reaction solution at 60 min. Reaction conditions:
isopropyl acetate, 8 mmol; n-decane (solvent), 28 cm3; 1,3,5-triisopropyl benzene
(internal standard), 4 mmol; catalyst weight 0.1 g, and reaction temperature 373 K.
0
5
10
15
20
Yield of methyl butyrate/%
Fig. 12. Results from the catalytic transesterification of glyceryl tributyrate with
methanol over solid acid catalysts. Reaction conditions: glyceryl tributyrate,
10 mmol; methanol, 9.6 g (300 mmol); n-hexyl ether (internal standard), 2.5 mmol;
catalyst weight, 0.2 g; reaction temperature, 333 K; reaction time, 24 h. For Cs2.5-
micro and Cs2.5-bimodal(H2SO4), their reusabilities were checked. After the first
run, the catalyst was separated by centrifugation and dried at 373 K.
was only little change in the surface areas by the post-treatment.
The micropores were basically preserved during the post-treat-
ment, which was confirmed from N2 adsorption–desorption iso-
therms. Although the mesopores in Cs2.5-bimodal were
preserved during the post-treatment, the mesopore volume
slightly decreased (Fig. S5). The exchange of Cs+ with H+ for
Cs2.5-bimodal(H2SO4) was also confirmed by using elemental
analysis. The Cs/polyanion ratio decreased from 2.9 (Cs2.5-bimo-
dal) to 2.6 (Cs2.5-bimodal(H2SO4)). The catalytic activity of
Cs2.5-bimodal(H2SO4) was about twice that of Cs2.5-bimodal
(Fig. 9). Since only negligible amounts of sulfur were detected in
Cs2.5-bimodal(H2SO4) by using elemental analysis, it was not
modified with sulfate ion, forming sulfated metal oxides, like sul-
fated zirconia.
Fig. 11 shows time courses for the decomposition of isopropyl
acetate over Cs2.5-bimodal(H2SO4) along with those after the cat-
alyst was filtered off at 60 min. The conversion increased with
reaction time, but the reaction completely stopped after the cata-
lyst was filtered off from the reaction solution, indicating that
Cs2.5-bimodal(H2SO4) acted as a heterogeneous catalyst.
One advantage of catalysts with mesopores is that they can be
used in reactions involving large reactants and products, which
are too big for microporous catalysts. To demonstrate the advan-
tage of the mesoporosity and high acid strength of Cs2.5-
bimodal(H2SO4), we carried out the transesterification of glyceryl
tributyrate with methanol (Eq. (5)) over various solid acids and
compared their catalytic performances with that of Cs2.5-
bimodal(H2SO4) (Fig. 12).
highest activity among the catalysts, its catalytic activity signifi-
cantly decreased when it was reused. The high activity for the first
run and low activity for the second run are due to dissolution of the
acidic component of Cs2.5-micro into the polar reaction solution,
which was mainly composed of methanol during the first run. In
separate experiment, it was confirmed that 14.9% of Cs2.5-micro
was dissolved in methanol. In contrast to Cs2.5-micro, the recov-
ered Cs2.5-bimodal(H2SO4) showed similar activity for the reac-
tion to that of the fresh one. In other words, there was little
catalyst deactivation. This was because dissolution of Cs2.5-
bimodal(H2SO4) in methanol was only little (3.9%), which was
basically the same as that of Cs2.5-bimodal.
Cs2.5-bimodal(H2SO4) had a greater activity than Amberlyst-
15 did. Since Amberlyst-15 has a larger number of acid sites
(4.7 mmol gꢀ1
)
than Cs2.5-bimodal(H2SO4) (0.028 mmol gꢀ1
)
does, the strong acid strength as well as mesoporosity of Cs2.5-
bimodal(H2SO4) greatly contributes to the high activity toward
the transesterification of glyceryl tributyrate with methanol.
4. Conclusions
Treatment of microporous CsxH4ꢀxSiW12O40 (Csx-micro) with
x = 1.0–2.5 in refluxing ethanol resulted in the formation of
mesopores in the particle with the preservation of the micropores.
ð5Þ
Silica-alumina and zeolites, including H-Y, H-b, and H-ZSM-5,
showed little or negligible activity due to their weak acid strengths
and/or microporosity. Although fresh Cs2.5-micro showed the
In particular, Cs2.5-micro transformed into a bimodal Cs salt
(Cs2.5-bimodal) with a narrow distribution of mesopore diame-
ters (3.7 nm) and large mesopore volumes. Cs2.5-micro was

42
Y. Iwase et al. / Journal of Catalysis 318 (2014) 34–42
[17] M. Rutkowska, L. Chmielarz, D. Macina, Z. Piwowarska, B. Dudek, A. Adamski,
S. Witkowski, Z. Sojka, L. Obalová, C.J. Van Oers, P. Cool, Appl. Catal. B 146
(2014) 112–122.
[18] F.L. Bleken, K. Barbera, F. Bonino, U. Olsbye, K.P. Lillerud, S. Bordiga, P. Beatp,
T.V.W. Janssens, S. Svelle, J. Catal. 307 (2013) 62–73.
[19] L. Wu, V. Degirmenci, P.C.M.M. Magusin, N.J.H.G.M. Lousberg, E.J.M. Hensen, J.
Catal. 298 (2013) 27–40.
[20] H. Mochizuki, T. Yokoi, H. Imai, S. Namba, J.N. Kondo, T. Tatsumi, Appl. Catal. A
449 (2012) 188–197.
[21] Y. Fang, H. Hu, J. Am. Chem. Soc. 128 (2006) 10636–10637.
[22] M. Choi, H.S. Cho, R. Srivastava, C. Venkatesan, D.-H. Choi, R. Ryoo, Nat. Mater.
5 (2006) 718–723.
[23] L. Schmidt, A. Boisen, E. Gustavsson, K. Ståhl, S. Pehrson, S. Dahl, A. Carlsson,
C.J.H. Jacobsen, Chem. Mater. 13 (2001) 4416–4418.
[24] Y. Tao, H. Kanoh, K. Kaneko, J. Am. Chem. Soc. 125 (2003) 6044–6045.
[25] A.H. Janssen, I. Schmidt, C.J.H. Jacobsen, A.J. Koster, K.P. de Jong, Micropore
Mesopore Mater. 65 (2003) 59–75.
[26] M. Ogura, S. Shinomiya, J. Tateno, Y. Nara, E. Kikuchi, M. Matsukata, Chem. Lett.
(2000) 882–883.
[27] M. Ogura, S. Shinomiya, J. Tateno, Y. Nara, M. Nomura, E. Kikuchi, M.
Matsukata, Appl. Catal. A 219 (2001) 33–43.
composed of three microcrystallites (Cs3.7H0.3[SiW12O40] with an
average crystallite size of 39 nm, Cs2.2H1.7[SiW12O40] with an aver-
age crystallite size of 23 nm, and H4[SiW12O40]), and these aggre-
gated to form secondary particles. Treating Cs2.5-micro in
ethanol caused part of the microcrystallites, mainly H4[SiW12O40],
to dissolve, creating voids between the insoluble parts, where the
H4[SiW12O40] crystallites were. These interstitial voids were the
mesopores in Cs2.5-bimodal.
The catalytic activity of bimodal Cs2.5-bimodal toward the
decomposition of isopropyl acetate in the liquid phase was only
one-third of that of untreated microporous Cs2.5-micro because
the acidic component, mainly H4[SiW12O40], was eliminated from
the solid at the time that the mesopores formed during the treat-
ment. Further thermal treatment of Cs2.5-bimodal in sulfuric acid
led to a threefold increase in the acid amount due to the substitu-
tion of Cs+ ions on the surface with H+ ions, and thus, the catalytic
activity for the decomposition of isopropyl acetate increased.
Cs2.5-bimodal treated with sulfuric acid showed high activity
toward the transesterification of bulky glyceryl tributyrate with
methanol due to its strong acid strength and mesoporosity.
[28] J.C. Groen, J.C. Jansen, J.A. Moulijn, J. Pérez-Ramírez, J. Phys. Chem. B 108
(2004) 13062–13065.
[29] T. Suzuki, T. Okuhara, Micropore Mesopore Mater. 43 (2001) 83–89.
[30] M. Ogura, Top. Catal. 12 (2008) 16–27.
[31] T. Okuhara, N. Mizuno, M. Misono, Adv. Catal. 41 (1996) 113–252.
[32] M.T. Pope, Heteropoly and Isopoly Oxometalates, Springer-Verlag, Berlin,
1983.
[33] I.V. Kozhevnikov, Chem. Rev. 98 (1998) 171–198.
[34] T. Okuhara, T. Nishimura, H. Watanabe, M. Misono, J. Mol. Catal. 74 (1992)
247–256.
Acknowledgment
This work was supported by KAKENHI (24760634).
[35] T. Okuhara, T. Nishimura, M. Misono, Stud. Surf. Sci. Catal. 101 (1996) 581–
590.
[36] M. Misono, N. Mizuno, K. Katamura, A. Kasai, Y. Konishi, K. Sakata, T. Okuhara,
Y. Yoneda, Bull. Chem. Soc. Jpn. 55 (1982) 400–406.
[37] L.C. Jozefowicz, H.G. Karge, E. Vasilyeva, J.M. Moffat, Microporous Mater. 1
(1993) 313–322.
Appendix A. Supplementary material
[38] F. Lefebvre, F.X. Liu-Cai, A. Auroux, J. Mater. Chem. 4 (1994) 125–131.
[39] D.J. Parrollo, C. Lee, R.J. Gorte, D. White, W.E. Farneth, J. Phys. Chem. 99 (1995)
8745–8749.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2014.07.008.
[40] A. Zheng, S.-J. Huang, S.-B. Liu, F. Deng, Phys. Chem. Chem. Phys. 13 (2011)
14889–14901.
[41] N. Feng, A. Zheng, S.-J. Huang, H. Zhang, N. Yu, C.-Y. Yang, S.-B. Liu, F. Deng, J.
Phys. Chem. C 114 (2010) 15464–15472.
References
[1] D.P. Serrano, J. Aguado, J.M. Escola, ACS Catal. 2 (2012) 1924–1941.
[2] D.O. de Zárate, F. Bouyer, H. Zschiedrich, P.J. Kooyman, P. Trens, J. Iapichella, R.
Durand, C. Guillem, E. Prouzet, Chem. Mater. 20 (2008) 1410–1420.
[3] M. Enterría, F. Suírez-García, A. Martínez-Alonso, J.M.D. Tascón, J. Alloys
Compd. 589 (2014) 60–69.
[42] M. Misono, Catal. Rev. Sci. Eng. 29 (1987) 269–321.
[43] J.B. McMonagle, J.B. Moffat, J. Colloid Interface Sci. 101 (1984) 479–488.
[44] T. Okuhara, T. Nishimura, M. Misono, Chem. Lett. (1995) 155–156.
[45] Y. Kamiya, S. Sano, Y. Miura, Y. Uchida, Y. Ogawa, Y. Iwase, T. Okuhara, Chem.
Lett. 39 (2010) 881–883.
[4] C.R. Patil, P.S. Niphadkar, V.V. Bolade, P.N. Joshi, Catal. Commun. 43 (2014)
188–191.
[5] J. Ding, H. Liu, P. Yuan, G. Shi, X. Bao, ChemCatChem 5 (2013) 2258–2269.
[6] Y. Yue, Z.-A. Qiao, P.F. Fulvio, A.J. Binder, C. Tian, J. Chen, K.M. Nelson, X. Zhu, S.
Dai, J. Am. Chem. Soc. 135 (2013) 9572–9575.
[46] I.V. Kozhevnikov, Catal. Rev. Sci. Eng. 37 (1995) 311–352.
[47] T. Yamada, Y. Yoshinaga, T. Okuhara, Bull. Chem. Soc. Jpn. 71 (1998) 2727–
2734.
[48] T. Yamada, K. Johkan, T. Okuhara, Micropore Mesopore Mater. 26 (1998) 109–
115.
[7] L.G. Possato, R.N. Diniz, T. Garetto, S.H. Pulcinelli, C.V. Santilli, L. Martins, J.
Catal. 300 (2013) 102–112.
[49] Y. Yoshinaga, K. Seki, T. Nakato, T. Okuhara, Angew. Chem. Int. Ed. 36 (1997)
2833–2835.
[8] Y.H. Kim, K.H. Lee, C.-M. Nam, J.S. Lee, ChemCatChem 4 (2012) 1143–1153.
[9] S. Schlienger, J. Alauzun, F. Michaux, L. Vidal, J. Parmentier, C. Gervais, F.
Babonneau, S. Bernard, P. Miele, J.B. Parra, Chem. Mater. 24 (2012) 88–96.
[10] X.-Y, Yang, G. Tian, L.-H. Chen, Y. Li, J.C. Rooke, Y.-X. Wei, Z.-M. Liu, Z. Deng,
G.V. Tendeloo, B.-L. Su, Chem. Eur. J. 17 (2011) 14987–14995.
[11] X. Zhang, K. Tao, T. Kubota, T. Shimamura, T. Kawabata, K. Matsuda, S. Ikeno, N.
Tsubaki, Appl. Catal. A 405 (2011) 160–165.
[50] T. Okuhara, H. Watanabe, T. Nishimura, K. Inumaru, M. Misono, Chem. Mater.
12 (2000) 2230–2238.
[51] Y. Yoshinaga, T. Suzuki, M. Yoshimune, T. Okuhara, Top. Catal. 19 (2002) 179–
185.
[52] T. Okuhara, T. Nakato, Catal. Surv. Jpn. 2 (1998) 31–44.
[53] T. Okuhara, Chem. Rev. 102 (2002) 3641–3666.
[54] T. Okuhara, N. Mizuno, M. Misono, Appl. Catal. A 222 (2001) 63–77.
[55] Y. Miura, H. Imai, T. Yokoi, T. Tatsumi, Y. Kamiya, Micropore Mesopore Mater.
174 (2013) 34–43.
[12] N. Danilina, S.A. Castelanelli, E. Troussard, J.A. van Bokhoven, Catal. Today 168
(2011) 80–85.
[13] D.P. Serrano, R. Sanz, P. Pizarro, I. Moreno, Top. Catal. 53 (2010) 1319–1329.
[14] N. Danilina, F. Krumeich, J.A. van Bokhoven, J. Catal. 272 (2010) 37–43.
[15] K.M. Jinka, H.C. Bajaj, R.V. Jasra, E.A. Prasetyanto, S.-E. Park, Top. Catal. 53
(2010) 238–246.
[56] Y. Ogasawara, S. Uchida, T. Maruichi, R. Ishikawa, N. Shibata, Y. Ikugara, N.
Mizuno, Chem. Mater. 25 (2013) 905–911.
[57] A. Saito, H.C. Foley, Microporous Mater. 3 (1995) 531–542.
[58] V.B. Fenelonov, V.N. Romannikov, A.Y. Derevyankin, Micropore Mesopore
Mater. 28 (1999) 57–72.
[16] C. Perego, R. Millini, Chem. Soc. Rev. 42 (2013) 3956–3976.