Acidic properties of sulfonic acid-functionalized FSM-16 mesoporous silica and its catalytic efficiency for acetalization of carbonyl compounds

Article abstract of DOI:10.1016/j.jcat.2005.01.017

Propyl-sulfonic acid-functionalized FSM-16 mesoporous silica (SO ₃H-FSM) is prepared by a conventional post-modification method. For the acetalization of carbonyl compounds with ethylene glycol, SO ₃H-FSM shows a higher rate and 1,3-dioxolane yield than conventional heterogeneous solid acids such as zeolites, montmorillonite K10 clay, silica-alumina, and the sulfonic resin. SO₃H-FSM is stable during the reaction, with no leaching and deactivation of sulfonic acid groups, and is reusable without loss of its activity. The acidity and hydrophilicity of SO ₃H-FSM are well characterized by the microcalorimetry of NH ₃ adsorption, NH₃-TPD, and H₂O-TPD, and the result is compared with those for various aluminosilicate zeolites (HZSM5, HBEA, HY) and K10 clay. It is found that NH₃-TPD is not suitable for characterizing the acidity of SO₃H-FSM, because the decomposition of SO₃H groups on SO₃H-FSM begins above 200°C. An NH ₃ adsorption microcalorimetric experiment at 150°C shows that, compared with HZSM5, SO₃H-FSM has a smaller number of acid sites but has a similar number of strong acid sites with ammonia adsorption heat above 140 kJ mol^-1. Comparison of the structural properties and catalytic results shows that a large pore diameter and low hydrophilicity are required to obtain high activity. Bronsted acid sites with a relatively strong acid strength are more suitable for this reaction, but the high acid concentration is not indispensable. The high activity of SO₃H-FSM should be caused by the presence of the strong Bronsted acid sites in the mesopore with a relatively low hydrophilicity, where both reactants can smoothly access the acid sites.

Full text of DOI:10.1016/j.jcat.2005.01.017

Journal of Catalysis 231 (2005) 131–138
www.elsevier.com/locate/jcat
Acidic properties of sulfonic acid-functionalized FSM-16 mesoporous
silica and its catalytic efficiency for acetalization of carbonyl compounds
Ken-ichi Shimizu a,∗,1, Eidai Hayashi a, Tsuyoshi Hatamachi b, Tatsuya Kodama b,
Tomoya Higuchi c, Atsushi Satsuma c, Yoshie Kitayama b
a
Graduate School of Science and Technology, Niigata University, Ikarashi-2, Niigata 950-2181, Japan
b
Department of Chemistry & Chemical Engineering, Faculty of Engineering, Niigata University, Ikarashi-2, Niigata 950-2181, Japan
c
Department of Applied Chemistry, Graduate School of Engineering, Nagoya University, Chikusa-ku, Nagoya 464-8603, Japan
Received 18 November 2004; revised 4 January 2005; accepted 16 January 2005
Abstract
Propyl-sulfonic acid-functionalized FSM-16 mesoporous silica (SO H-FSM) is prepared by a conventional post-modification method. For
3
the acetalization of carbonyl compounds with ethylene glycol, SO H-FSM shows a higher rate and 1,3-dioxolane yield than conventional
3
heterogeneous solid acids such as zeolites, montmorillonite K10 clay, silica-alumina, and the sulfonic resin. SO H-FSM is stable during the
3
reaction, with no leaching and deactivation of sulfonic acid groups, and is reusable without loss of its activity. The acidity and hydrophilicity
of SO H-FSM are well characterized by the microcalorimetry of NH adsorption, NH -TPD, and H O-TPD, and the result is compared with
3
3
3
2
those for various aluminosilicate zeolites (HZSM5, HBEA, HY) and K10 clay. It is found that NH -TPD is not suitable for characterizing the
3
◦
acidity of SO H-FSM, because the decomposition of SO H groups on SO H-FSM begins above 200 C. An NH adsorption microcalori-
3
3
3
3
◦
metric experiment at 150 C shows that, compared with HZSM5, SO H-FSM has a smaller number of acid sites but has a similar number of
strong acid sites with ammonia adsorption heat above 140 kJ mol . Comparison of the structural properties and catalytic results shows that
3
−1
a large pore diameter and low hydrophilicity are required to obtain high activity. Brønsted acid sites with a relatively strong acid strength are
more suitable for this reaction, but the high acid concentration is not indispensable. The high activity of SO H-FSM should be caused by the
3
presence of the strong Brønsted acid sites in the mesopore with a relatively low hydrophilicity, where both reactants can smoothly access the
acid sites.
2005 Elsevier Inc. All rights reserved.
Keywords: Acetalization; Mesoporous silica; Sulfonic acid; TPD; Microcalorimetry
1. Introduction
(PTSA) and pyridinium salts [2–5], these homogeneous cat-
alysts present limitations due to the use of toxic and cor-
rosive reagents, the tedious workup procedure, and the ne-
cessity of neutralization of the strong acid media, producing
undesired wastes. To overcome these problems, various het-
erogeneous catalysts such as zeolites [6–8], functionalized
zeolite [9], Ti-exchanged clay [10], and alumina [11] have
been employed for the synthesis of 1,3-dioxolane from car-
bonyl compounds and ethylene glycol. However, some of
these procedures suffer because of a high catalyst/substrate
ratio and a long reaction time. For zeolites, the microporous
structure can limit the accessibility of substances to the inter-
nal acid sites [9], the products or by-products can be strongly
Aldehydes and ketones are frequently protected as ac-
etals in the course of organic synthesis [1]. 1,3-Dioxolane
is one of the most popular protecting groups for this pur-
pose. Although synthesis of 1,3-dioxolane is generally per-
formed with the use of ethylene glycol in the presence of
homogeneous acid catalysts such as p-toluenesulfonic acid
*
Corresponding author. Fax: +81 52 789 3193.
E-mail address: kshimizu@apchem.nagoya-u.ac.jp (K.-i. Shimizu).
Present address: Department of Applied Chemistry, Graduate School
1
of Engineering, Nagoya University, Chikusa-ku, Nagoya 464-8603, Japan.
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2005.01.017
2005 Elsevier Inc. All rights reserved.

132
K.-i. Shimizu et al. / Journal of Catalysis 231 (2005) 131–138
adsorbed to the strong acid sites present in the micropore,
and thus, before catalyst reuse, a regeneration of the zeolite
by high-temperature calcination is required to remove ad-
sorbed species [8].
lowed by calcination at 500 ◦C for 1 h. Amorphous silica
(JRC-SIO-8, a reference catalyst of the Catalysis Society
of Japan; SBET = 303 m2 g−1), silica-alumina (JRC-SAL-
2, SiO2/Al2O3 = 5.6, SBET = 560 m2 g−1), HY zeolite
(JRC-Z-HY4.8, SiO2/Al2O3 = 4.8, SBET = 663 m2 g−1),
HZSM5 zeolite (JRC-Z5-25H, SiO2/Al2O3 = 25), and
HBEA zeolite (JRC-Z-HB25, SiO2/Al2O3 = 25 5) were
supplied by the Catalysis Society of Japan. We prepared
octadecyltriethoxysilane-treated HY zeolite (OD-HY) by re-
fluxing a mixture of the HY zeolite (2 g) dried at 120 ◦C and
octadecyltriethoxysilane (2 mmol) in dry toluene (20 ml) for
6 h, followed by filtering and washing with toluene and ace-
tone, and by drying at 100 ◦C for 1 h [32]. The number of
the octadecyl groups determined by the elemental analysis
The use of organically functionalized mesoporous sil-
ica as a catalyst for organic synthesis has received consid-
erable attention [12–27]. Among these materials, sulfonic
acid-functionalized mesoporous silicas can be alternatives
to commercially available sulfonic resins, which suffer from
low surface area and thermal stability. Although the catalytic
applications of the resins for organic synthesis have been
well established [28,29], relatively few examples of the use
of sulfonic acid mesoporous silica for organic reactions have
been reported [16–24]. For a fundamental understanding of
this new class of the solid acid, it is informative to clar-
ify the structural characteristics, such as the numbers and
strength of acid sites and hydrophilicity in comparison with
conventional solid acid catalysts and to reveal the effect of
these factors on the catalytic activity. However, little atten-
tion has been focused on this aspect. As a part of ongoing
research on the catalysis of functionalized silicas for clean
organic synthesis [25–27,30], we present here sulfonic acid-
functionalized FSM-16 mesoporous silica (SO3H-FSM) as
a highly effective and reusable catalyst for the acetaliza-
tion of various carbonyl compounds with ethylene glycol.
Detailed characterizations show the number and strength of
acid sites and the hydrophilicity of SO3H-FSM in compari-
son with conventional microporous or mesoporous solid acid
catalysts. The results are discussed to reveal the factors af-
fecting the catalytic activity for the acetalization reaction.
was 0.4 mmol g−1
.
Sulfur content in the SO3H-FSM was analyzed with a
CS analyzer. The titration technique was used to determine
the number of H+ on the SO3H-FSM. About 0.05 g of the
sample was added to 15 ml of the 2 M NaCl solution and
allowed to equilibrate. Then the liquid phase was titrated
by the dropwise addition of 0.01 M NaOH(aq) [22]. XRD
patterns were taken with a MX Labo (MAC Science) with
Cu-Kα radiation. We obtained the BET surface area and
pore size distribution by measuring N2 adsorption isotherms
at 77 K, using TriStar 3000 (Micromeritics). The pore size
distribution curve was calculated by the BJH method from
adsorption isotherms.
FTIR measurement of adsorbed ammonia on SO3H-FSM
was performed with a JASCO FT/IR-300. The sample was
pressed into 0.03 g of self-supporting wafers and mounted in
a quartz IR cell with CaF2 windows. After evacuation of the
sample at 150 ◦C for 1 h, followed by ammonia adsorption
(10 Torr) and by evacuation at 150 ◦C for 0.5 h, a difference
spectrum was recorded at room temperature.
2. Experimental
2.1. Catalyst preparation and characterization
A temperature-programmed desorption (TPD) experi-
ment with NH3 or H2O was carried out with TPD equipment
(BEL Japan). To the sample (0.05 g), preheated under vac-
uum at 150 ◦C for 1 h, NH3 (100 Torr) was introduced at
100 ◦C, followed by purging with He at 100 ◦C for 0.5 h.
FSM-16 silica was prepared from kanemite accord-
ing to a method previously described [31]. Sulfonic acid-
functionalized FSM-16 mesoporous silica (SO3H-FSM) was
prepared as follows. The FSM-16 sample (2 g) was dried at
120 ◦C, and then 3-mercaptopropyltriethoxysilane (2 mmol)
in dry toluene (20 ml) was introduced. The mixture was re-
fluxed for 6 h, and the solid was filtered off, washed with
toluene and acetone, and air-dried. The SH groups were
converted into SO3H groups by oxidation in a mixture of
aqueous H2O2 (31%, 5 ml) and CH3CN (30 ml) for 6 h at
a reflux temperature. After filtration and washing with water
and ethanol, the sample was then acidified in 0.1 M H2SO4
solution for 1 h, followed by washing with water until neu-
tral pH and drying at 100 ◦C for 6 h.
NH3-TPD was then carried out under He flow (60 ml min−1
)
at a reduced pressure (200 Torr) from 100 to 650 ◦C at a
heating rate of 10 ◦C min−1, and outlet gases were analyzed
with a mass spectrometer. Although the molecular weight
of NH3 is 17, the fragment with m/e = 16 was used to
quantify NH3, because fragment 17 is affected by the des-
orbed water [33,34]. For H2O-TPD, H2O (20 Torr) was
introduced into the sample (0.05 g) at 25 ◦C, followed by
purging with He at 50 ◦C for 1 h; we performed TPD in
He flow (60 cm3 min−1) at a reduced pressure (200 Torr)
by raising the temperature from 50 to 650 ◦C at a rate of
10 ◦C min−1, and outlet gases (m/e = 18) were analyzed
with a mass spectrometer.
Montmorillonite K10 clay (surface area (SBET) =
220 m2 g−1) and a sulfonic resin (Amberlyst-15; SBET
=
45 m2 g−1) were purchased from Aldrich. To remove resid-
ual metal cations in the cation-exchange site in the clay,
K10 clay (1 g) was treated in an aqueous solution of
NH4NO3 (1 M, 10 ml) at room temperature for 3 h, fol-
Microcalorimetric measurement of NH3 adsorption was
performed at 150 ◦C with a calorimeter (Tokyo Rikou Co.,
HAC-450G) connected to a volumetric glass line (35 ◦C)
with on-line injection system for pulsing NH3 gas. Before

K.-i. Shimizu et al. / Journal of Catalysis 231 (2005) 131–138
133
the adsorption experiments the sample (0.1 g) was pretreated
under vacuum overnight at 150 ◦C (for SO3H-FSM and K10
clay) or 400 ◦C (for HZSM5), respectively.
2.2. Catalytic test
Toluene was dried over 3A sieves. Ethylene glycol
and carbonyl compounds were obtained from commercial
sources and were used without further purification. Typi-
cally, we performed acetalization simply by refluxing the
mixture of carbonyl compound (5 mmol) and ethylene gly-
col (6 mmol) in toluene (5 ml) in air. Under these conditions,
a part of the water produced can remain in the reaction me-
dia, and some water can move to the condenser after the
azeotropic removal of water. For the reaction with SO3H-
FSM catalyst, the catalyst amount was 7 mg, corresponding
to 0.1 mol% with respect to carbonyl compounds (0.001 mol
of SO3H groups/mol of carbonyl compounds). For the ki-
netic test in Fig. 5, the catalyst amount was 7 mg for SO3H-
FSM, FSM-16 silica, 10 mg for Amberlyst-15, and 30 mg
for other catalysts. Yields were determined by GC with
n-dodecane as an internal standard. The products were iden-
Fig. 1. The pore size distribution curves of SO H-FSM (") and
FSM-16 (!).
3
by the sulfonic groups. The type of acid sites in SO3H-
FSM was first confirmed by FTIR measurement of adsorbed
NH3 (result not shown). The spectrum of adsorbed NH3 on
SO3H-FSM showed N–H vibration (in the range of 3000+–
3300 cm−1) and deformation (1420 cm−1) bands of NH4
ion accompanying expanses of the bands due to acidic hy-
droxyl groups in a range of 3600–3700 cm−1 and the asym-
metric stretching band of SO2 moieties at 1370 cm−1. This
1
tified by H NMR. We measured the reproducibility of the
catalytic test by repeating it three times for the acetalization
of acetophenone (5 mmol) with ethylene glycol (6 mmol) in
toluene (5 ml) at reflux temperature by SO3H-FSM (7 mg).
The relative error was estimated to be 6%.
+
indicates that NH4 ions are formed by the interaction of
NH3 with Brønsted acid sites in SO3H groups.
Fig. 2 shows the NH3-TPD curve for SO3H-FSM. A large
desorption peak centered around 170 ◦C was observed. Note
that no desorption peak appeared in the NH3-TPD curve
of the unmodified FSM-16 silica, which confirms that the
peak for SO3H-FSM is assigned to the NH3 desorption from
SO3H groups. The desorption peaks are clearly smaller for
SO3H-FSM than those for zeolites, indicating a smaller acid
amount of SO3H-FSM. The number of NH3 desorbed from
each sample in the range of 100 to 650 ◦C is listed in Ta-
ble 1. The number of NH3 desorbed from SO3H-FSM was
0.43 mmol g−1. The desorption temperature for SO3H-FSM
was lower than those for zeolites (HY, HBEA, and HZSM5),
which may suggest a weaker acid strength for SO3H-FSM
compared with zeolites. However, at a temperature range of
200–600 ◦C, a SO2 (m/e = 64) desorption peak (a dotted
line in Fig. 1) was also observed, which indicates the de-
composition of SO3H groups on this sample above 200 ◦C.
Hence, NH3-TPD is not an appropriate method for char-
acterizing the acidity of SO3H-FSM. The NH3-TPD curve
of octadecyltriethoxysilane-treated HY zeolite (OD-HY) is
also included in Fig. 1. The number of desorbed NH3 from
HY did not significantly decrease after the introduction of
octadecyl groups. Thus, the silanol groups, mainly present
at the external surface of the zeolite, may be preferably con-
sumed by the reaction with octadecyltriethoxysilane.
3. Results and discussion
3.1. Characterization
The XRD pattern of SO3H-FSM showed a typical low-
angle reflection (100) at d100 = 3.65 nm characteristic of
FSM-16 [31], indicating that the long-range ordered struc-
ture of the support was preserved. Fig. 1 shows pore size
distribution curves for SO3H-FSM and FSM-16. Clearly,
introduction of R–SO3H groups resulted in a smaller pore
diameter and a broader distribution curve. For FSM-16, sur-
face area (951 m2 g−1), pore volume (1.2 cm3 g−1), and pore
diameter (around 2.6 nm) were as expected for mesoporous
materials. These values were reduced after the introduc-
tion of R–SO3H groups (surface area = 858 cm3 g−1, pore
volume = 0.90 cm3 g−1, pore diameter = 2.3 nm), which
could be due to the presence of R–SO3H groups in the chan-
nels of FSM-16.
The sulfur content of SO3H-FSM was 0.77 mmol g−1
.
The number of H+ determined by acid-base titration was
0.77 0.02 mmol g−1. This value is very close to the sul-
fur content, indicating that most of the sulfur species on the
sample are in the form of the sulfonic acid groups without
forming disulfides [17]. Note that the amount of H+ for un-
modified FSM-16 silica was below the detection limit of the
titration, confirming that the acidity of SO3H-FSM is caused
Because the decomposition of SO3H groups on SO3H-
FSM begins above 200 ◦C, we performed an NH3 adsorp-
tion microcalorimetric experiment at 150 ◦C to obtain more

134
K.-i. Shimizu et al. / Journal of Catalysis 231 (2005) 131–138
Fig. 2. NH -TPD curves (m/e = 16) for (a) SO H-FSM, (b) K10 clay, (c) HZSM5, (d) HBEA, (e) OD-HY and (f) HY. A dotted line (g) denotes the MS signal
3
3
intensity curves for m/e = 64 (SO ) during the NH -TPD experiment of SO H-FSM.
2
3
3
Table 1
Summary of the acetalization results and structural characteristics of the catalysts
a
b
c
d
e
f
g
h
Catalysts
Rate
Yield
(%)
Q
N
N
N
3 (TPD)
−1
N
H O
2
(TPD)
−1
Pore diameter
(nm)
NH
NH
100
130
3
−1 −1
−1
−1
(mmol g )
−1
(mmol h
g
)
(mmol g
)
(kJ mol
)
(mmol g
)
(mmol g
)
i
SO H-FSM
3
505
60
58
14
5
91
47
45
54
19
13
175
–
–
131
–
187
0.33
–
–
0.10
–
1.00
0.08
–
–
0
–
0.10
0.43
1.04
1.40
0.37
1.43
1.17
0.27
0.32
1.04
0.07
1.29
0.16
2.3 nm
j
j
HBEA
OD-HY
K10 clay
HY
0.76 nm × 0.64 nm
k
(0.74 nm)
i
2.5 nm
j
0.74 nm
0.53 nm × 0.56 nm
HZSM5
a
4
Results in Fig. 5.
b
c
d
e
f
Initial rate of acetal formation.
GC yields after 10 h.
The initial adsorption heat of chemisorbed NH in NH microcalorimetry.
The number of chemisorbed NH having heats of adsorption above 100 kJ mol in NH microcalorimetry.
The number of chemisorbed NH having heats of adsorption above 140 kJ mol in NH microcalorimetry.
3
3
−1
3
3
−1
3
3
g
h
i
◦
The number of desorbed NH in a range 100–650 C in NH -TPD.
3
3
◦
The number of desorbed H O in a range 50–200 C in H O-TPD.
2
2
Pore diameters estimated by N adsorption using the BJH equation.
Pore diameters from crystallographic data.
The pore diameter of the unmodified HY.
2
j
k
tablished that NH3 adsorption on Brønsted acid sites on
several H+-exchanged zeolites produced heats in the range
of 100–150 kJ mol−1 [35–37]. Similar to the published re-
sults [35–37], HZSM5 shows an initial heat of adsorption
of 183 kJ mol−1, a plateau of very similar differential heat
between 130 and 150 kJ mol−1, and a sharp drop in the
heat above the surface coverage of around 1.0 mmol g−1
.
This feature indicates that the majority of Brønsted acid
sites in HZSM5 are homogeneous in terms of the acid
strength, which is in good agreement with a quite narrow
feature of the NH3-TPD curve for HZSM5 (Fig. 2). SO3H-
FSM exhibits an initial heat of approximately 175 kJ mol−1
.
The heat of adsorption gradually decreases with an in-
crease in the amounts of adsorbed NH3, steeply decreases
from 100 to ca. 85 kJ mol−1 above the surface coverage of
0.33 mmol g−1, and then shows a constant value of around
85 kJ mol−1. The amounts of chemisorbed NH3 can be deter-
mined from the curve. From the above-mentioned facts, we
assigned the adsorption with the heat above 100 kJ mol−1
to the chemisorption of NH3 on Brønsted acid sites and that
Fig. 3. Differential heat of adsorption of NH on (!) SO H-FSM, (") K10
clay and (+) HZSM5.
3
3
reliable information on the acidity (number and strength
of acid sites) of this catalyst. Fig. 3 compares the curves
of the differential heat of ammonia adsorption on SO3H-
FSM, K10 clay, and a typical zeolite, HZSM5. It is es-

K.-i. Shimizu et al. / Journal of Catalysis 231 (2005) 131–138
135
Fig. 4. H O-TPD curves (m/e = 18) for (a) SO H-FSM, (b) K10 clay, (c) HZSM5, (d) HBEA, (e) OD-HY and (f) HY.
2
3
with the heat below 100 kJ mol−1 to weak adsorption, pos
sibly via hydrogen bonding, on the site that should not be
catalytically relevant. As listed in Table 1, the number of
chemisorbed NH3 on Brønsted acid sites of SO3H-FSM
and HZSM5 are 0.33 mmol g−1 and 1.0 mmol g−1, respec-
tively, indicating the smaller number of Brønsted acid sites
in SO3H-FSM than that in HZSM5. However, the number of
acid sites with the heat above 140 kJ mol−1 for SO3H-FSM
(0.08 mmol g−1) and HZSM5 (0.10 mmol g−1) are close to
each other, indicating that the number of strong acid sites in
SO3H-FSM is close to that in a typical aluminosilicate zeo-
lite, HZSM5. The strength of the acid sites of SO3H-FSM is
heterogeneous: with increasing volume of adsorbed ammo-
nia, the adsorption heat decreases continuously. The number
of Brønsted acid sites in SO3H-FSM determined by the mi-
crocalorimetric experiment (0.33 mmol g−1) is lower than
the number of H+ determined by acid-base titration (0.77
0.02 meq g−1) and the sulfur content (0.77 mmol g−1). The
reason for these results is not clear, but we believe that inter-
action between a SO3H group and neighboring SO3H groups
or SiOH groups on the support is responsible for the het-
erogeneous distribution of the acid strength of SO3H-FSM.
Fig. 3 includes the curves of the differential heat for K10
clay. K10 clay exhibits an initial heat of 131 kJ mol−1, and
the heat of adsorption decreases with an increase in the
amounts of adsorbed NH3. The number of acid sites with
the heat above 100 kJ mol−1 is 0.10 mmol g−1. This indi-
cates that the number and the strength of acid sites in K10
clay are smaller than those for HZSM5 and SO3H-FSM.
To compare the hydrophilicity of the samples, H2O-TPD
was carried out. As shown in Fig. 4, a desorption peak
centered around 80–120 ◦C is observed for all of the sam-
ples. For SO3H-FSM, a SO2 desorption peak (m/e = 64,
result not shown) was observed in the temperature range of
200–600 ◦C, indicating a thermal decomposition of SO3H
groups. A H2O desorption peak in the range of 200–600 ◦C
should be caused by the thermal decomposition of SO3H
groups and by the dehydration of surface Si–OH groups
on SO3H-FSM. The decomposition temperature of SO3H
groups (200 ◦C) is lower than the reaction temperature for
typical liquid-phase organic reactions catalyzed by acid cat-
alysts, and hence SO3H-FSM can be stable during most of
the reaction. In this study we assumed that H2O desorption
below 200 ◦C mainly originates from the physisorbed H2O,
and the number of H2O desorbed from various samples in a
temperature range of 50–200 ◦C (Table 1) is used to compare
the hydrophilicity of different samples. It is reasonable to as-
sume that the sample with a higher number of desorbed H2O
shows higher hydrophilicity. The hydrophilicity of the sam-
ples changed in the order of K10 clay < HZSM5 < SO3H-
FSM < HBEA < OD-HY < HY.
3.2. Structure-activity relationship for acetalization
The solid acids characterized in the above section were
tested in the acetalization of acetophenone with ethyl-
ene glycol by refluxing in air (Fig. 5). From the slope
of this kinetic plot, the initial rate per gram of the cat-
alyst was estimated. The rate changed in the order of
SO3H-FSM > Amberlyst-15 > HBEA = OD-HY > FSM-
16 > HY > HZSM5 > silica-alumina = amorphous silica.
Tanaka et al. [38] reported that siliceous mesoporous ma-
terial (MCM-41) catalyzed the acetalization of carbonyl
compounds by methanol. We have found that the meso-
porous silica, FSM-16, also catalyzes the acetalization of
acetophenone by ethylene glycol without loading of SO3H.
However, the initial rate per gram of the catalyst was 15
times higher for SO3H-FSM than for FSM-16, and, hence,
the anchored SO3H group is the main catalytic species of
SO3H-FSM. Under the present conditions, SO3H-FSM was
more effective than conventional solid acids such as silica-
alumina, H+-exchanged zeolites (HBEA, HY, and HZSM5),
montmorillonite K10 clay, and a sulfonic resin (Amberlyst-
15). Taking into account the acid amount of Amberlyst-15
(4.8 mmol g−1) [22], the rate per acid site for SO3H-FSM
(656 h−1) is estimated to be 9 times higher than that for the
resin (70 h−1). The lower rate of Amberlyst-15 could be re-
lated to the low accessibility of the reactant to the acid sites
inside the bulk of Amberlyst-15, which has a relatively low
surface area (45 m2 g−1).
A comparison of the catalytic results for microporous
or mesoporous solid acid and structural characteristics are
shown in Table 1. The result shows that low hydrophilic-

136
K.-i. Shimizu et al. / Journal of Catalysis 231 (2005) 131–138
Fig. 5. Plot of GC yield versus time for the acetalization of acetophenone (5 mmol) with ethylene glycol (6 mmol) in toluene (5 ml) at reflux temperature by
various catalysts: (!) SO H-FSM, (") Amberlyst-15, (1) K10 clay, (+) FSM-16 silica, (Q) amorphous silica, (F) silica-alumina, (a) HY, (E) HZSM5,
3
(2) HBEA, and (e) OD-HY. Catalyst amount was 7 mg for SO H-FSM, FSM-16 silica, 10 mg for Amberlyst-15, and 30 mg for other catalysts.
3
ity and large pore diameter are primarily important factors
as described below. Ogawa et al. [32] reported that highly
hydrophobic zeolite was prepared by octadecyltrichlorosi-
lane treatment of HZSM5. To investigate the effect of hy-
drophobicity of the catalyst on the activity, we prepared
octadecyltriethoxysilane-treated H-Y zeolite (OD-HY) and
used it for the acetalization of acetophenone. It is noteworthy
that the initial rate of product formation and the final yield of
HY zeolite were greatly enhanced by the modification with
the octadecyl groups, and the result could be explained as
follows. In the absence of octadecyl groups, the HY zeo-
lite with a low SiO2/Al2O3 ratio (4.8) is highly hydrophilic,
as indicated by H2O-TPD. Hence polar molecules, such as
ethylene glycol and water produced by the reaction, should
be strongly adsorbed on the acid site, resulting in blocking
of the site. As a result of a decrease in the hydrophilicity
with the introduction of octadecyl groups, a less polar sub-
stance, acetophenone, can be more accessible to the acid
site in HY, and thus the activity is improved. The impor-
tance of the hydrophobicity of zeolite for the acetalization
by ethylene glycol was demonstrated by Climent et al., who
observed a relatively low activity of the Al-rich zeolites [8].
HZSM5, with relatively low hydrophilicity, shows the lowest
activity. This can be explained by the small pore diameter of
this sample. The retarding effect of geometrical constraints
on the catalytic activity of zeolites was clearly illustrated by
Climent et al. for the dimethylacetal formation in the study
of the influence of the zeolite crystal size on the reaction
rate [39]. The geometrical constraints could inhibit diffusion
of the reactant and the bulkier reaction products in the pore
of HZSM5, which has the smallest pore size among the cat-
alysts tested in our study. Among the unmodified zeolites,
HBEA shows the highest rate, possibly because the catalyst
has both a large pore diameter (0.76 nm×0.64 nm) and a rel-
atively low hydrophilicity. Taking into account the fact that
SO3H-FSM also has hydrophobic groups (CH2 groups) and
hence the catalyst shows a relatively low hydrophilicity (Ta-
ble 1), it is suggested that the higher activity of SO3H-FSM
compared with zeolites is partly due to its hydrophobic na-
ture and its larger-diameter channels. We checked the effect
of water removal during the reaction by 0.5 g of molecu-
lar sieves (MS4A) placed in the condenser, with HBEA as a
catalyst (not shown). The yield after 10 h of the reaction was
71%, and the yield did not increase after 24 h. This value is
higher than the yield with HBEA under standard conditions
(47%), where a part of the water produced during the reac-
tion can remain in the reaction media. This indicates that a
shift of the equilibrium with the removal of water from the
reaction media is effective in obtaining a high yield. How-
ever, a moderate yield under dry conditions (71%) suggests
that the activity is also retarded by other effects, such as
strong poisoning of water on acid sites in the micropore.
Comparison of the number of acid sites with the reaction
rates (Table 1) clearly indicates that a large number of acid
sites per gram of the catalyst is not necessarily required for
1,3-dioxolane formation; HY and HZSM5, with a relatively
large number of acid sites, show a lower rate than K10 clay
and SO3H-FSM, with a smaller number of acid sites. Al-
though K10 clay exhibits low hydrophilicity and a large pore
diameter, it shows much lower activity than SO3H-FSM. The
higher acid strength of SO3H-FSM than K10 clay, as shown
by microcalorimetric experiment, should be responsible for
the higher activity of SO3H-FSM.
From the above discussion, the factors affecting the ac-
etalization of the carbonyl compound with ethylene glycol
are summarized as follows. Large pore diameter and low hy-
drophilicity are required to obtain a high reaction rate and
a high yield of the 1,3-dioxolane. Brønsted acid sites with a
relatively strong acid strength are suitable for this reaction,
but the high acid concentration is not necessarily required.
The high activity of SO3H-FSM should be caused by the
low hydrophilicity of the strong Brønsted acid sites present
in the mesopore, where both reactants can smoothly access
to the acid sites.
3.3. Catalytic performance of SO3H-FSM for acetalization
As shown in Fig. 5, SO3H-FSM (0.1 mol%) effectively
catalyzed the acetalization of acetophenone with ethylene
glycol. No measurable by-products were observed by TLC,

K.-i. Shimizu et al. / Journal of Catalysis 231 (2005) 131–138
137
Table 2
SO H-FSM catalyzed acetalization of carbonyl compounds with ethylene glycol
a
3
b
c
Entry
Substrate
Time
10
Product
Conversion (%)
Yield (%)
1
2
R = H
96
93
3
4
3
1
R = H
99
98
90
89
t
R = Bu
5
3
99
88
6
3
91
91
7
8
9
1
1
12
R = H
R = Cl
99
98
97
90
94
80
R = OCH
3
10
12
77
74
11
12
8
2
84
99
81
99
d
e
f
d
e
f
13
14
3
1
99, 98 , 99 , 98
99, 97 , 98 , 96
93
88
a
The reaction was performed by simply refluxing the mixture of carbonyl compound (5 mmol) and ethylene glycol (6 mmol) in toluene (5 ml) with a
catalyst (0.1 mol%, 7 mg) in air.
b
Conversions of carbonyl compounds determined by GC using internal standard method.
Yields of acetals determined by GC.
Second cycle.
Third cycle.
c
d
e
f
Fourth cycle.
1
GC, or H NMR analyses. After the reaction, SO3H-FSM
was removed by centrifugation and was washed with ace-
tone, followed by drying at 100 ◦C for 1 h. The titration
experiment of the recovered catalyst showed that the num-
ber of acid sites (0.77 0.01 meq g−1) did not change after
the reaction, which indicates that no leaching or deactiva-
tion of sulfonic acid groups occurred during the reaction.
A possible contribution of homogeneous catalysis was ex-
cluded in the following experiment. When the SO3H-FSM
was removed at an early stage of the reaction (t = 10 min,
20% yield), the reaction did not proceed further, confirming
that the observed catalysis of SO3H-FSM is truly heteroge-
neous in nature. After the first run, the reusability of the
SO3H-FSM was tested. For the acetalization of n-octanal,
the catalyst can be recycled at least three times while keep-
ing almost quantitative conversion without any reactivation
treatment (entry 13 in Table 2).
As summarized in Table 2, the present method was
widely applicable for 1,3-dioxolane synthesis from various
ketones (entries 1–6) and aldehydes (entries 7–14) with good
yields (90–98%), by simply refluxing for 1–12 h. Aromatic
aldehydes, aliphatic aldehydes, and conjugated aldehyde un-
derwent smooth transformation to the corresponding acetal
in good yield. Cyclic, aliphatic, and aromatic ketones reacted
smoothly to give the corresponding acetals. Note that sizes
of the reagents and products listed in Table 2 are smaller
than the mean pore diameters of the SO3H-FSM catalyst
(2.3 nm).
References
[1] T.W. Greene, P.G.M. Wuts, Protecting Groups in Organic Synthesis,
second ed., Wiley, New York, 1991.

138
K.-i. Shimizu et al. / Journal of Catalysis 231 (2005) 131–138
[2] R.A. Daignault, E.L. Eliel, Organic Synthesis, Wiley, New York, 1973.
[3] L.F. Fieser, R. Stevenson, J. Am. Chem. Soc. 76 (1954) 1728.
[21] I.K. Mbaraka, D.R. Radu, V.S. Lin, B.H. Shanks, J. Catal. 219 (2003)
329.
[4] W.G. Dauben, J.M. Gerdes, G.C. Look, J. Org. Chem. 51 (1986) 4964.
[5] N.M. Leonard, M.C. Oswald, D.A. Freiberg, B.A. Nattier, R.C. Smith,
R.S. Mohan, J. Org. Chem. 67 (2002) 5202.
[6] A. Corma, M.J. Climent, H. Garcia, J. Primo, Appl. Catal. 58 (1990)
333.
[7] R. Ballini, G. Boscica, B. Frullanti, R. Maggi, G. Sartori, F. Schroer,
Tetrahedron Lett. 39 (1998) 1615.
[8] M.J. Climent, A. Corma, A. Velty, M. Susarte, J. Catal. 196 (2000)
345.
[9] C.W. Jones, K. Tsuji, M.E. Davis, Nature 393 (1998) 52.
[10] T. Kawabata, T. Mizugaki, K. Ebitani, K. Kaneda, Tetrahedron Lett. 42
(2001) 8329.
[11] Y. Kamitori, M. Hojo, R. Masuda, T. Yoshida, Tetrahedron Lett. 26
(1985) 4767.
[12] J.H. Clark, D.J. Macquarrie, Chem. Commun. (1998) 853.
[13] A.P. Wight, M.E. Davis, Chem. Rev. 102 (2002) 3589.
[14] R.D. Badley, W.T. Ford, J. Org. Chem. 54 (1989) 5437.
[15] W.M. Van Rhijn, D.E. De Vos, B.F. Sels, W.D. Bossaert, P.A. Jacobs,
Chem. Commun. (1998) 317.
[22] M. Chidambaram, D. Curulla-Ferre, A.P. Singh, B.G. Anderson,
J. Catal. 220 (2003) 442.
[23] S. Shylesh, S. Sharma, S.P. Mirajkar, A.P. Singh, J. Mol. Catal. A 212
(2004) 219.
[24] J.A. Melero, R. van Grieken, G. Morales, V. Nuno, Catal. Commun. 5
(2004) 131.
[25] K. Shimizu, E. Hayashi, T. Inokuchi, T. Kodama, H. Hagiwara, Y. Ki-
tayama, Tetrahedron Lett. 43 (2002) 9073.
[26] H. Hagiwara, S. Tsuji, T. Okabe, T. Hoshi, T. Suzuki, H. Suzuki, K.
Shimizu, Y. Kitayama, Green Chemistry 4 (2002) 461.
[27] K. Shimizu, H. Suzuki, E. Hayashi, T. Kodama, Y. Tsuchiya, H. Hagi-
wara, Y. Kitayama, Chem. Commun. (2002) 1068.
[28] G.A. Olah, P.S. Iyer, G.K.S. Prakash, Synthesis (1986) 513.
[29] M.A. Harmer, Q. Sun, Appl. Catal. A 221 (2001) 45.
[30] K. Shimizu, E. Hayashi, T. Hatamachi, T. Kodama, Y. Kitayama, Tetra-
hedron Lett. 45 (2004) 5135.
[31] S. Inagaki, Y. Fukushima, K. Kuroda, J. Chem. Soc., Chem. Commun.
(1993) 680.
[32] H. Ogawa, T. Koh, K. Taya, T. Chihara, J. Catal. 148 (1994) 493.
[33] M. Niwa, N. Katada, Catal. Surv. Jpn. 1 (1997) 215.
[34] N. Katada, M. Niwa, Catal. Surv. Asia 8 (2004) 161.
[35] N. Cardona-Martinez, J.A. Dumesic, Adv. Catal. 38 (1992) 149.
[36] A. Auroux, Top. Catal. 4 (1997) 71.
[16] D. M Margolese, J.A. Melero, S.C. Christiansen, B.F. Chmelka, G.D.
Stucky, Chem. Mater. 12 (2000) 2448.
[17] I. Diaz, C. Marquez-Alvarez, F. Mohino, J. Perez-Pariente, E. Sastre,
J. Catal. 193 (2000) 283.
[37] V. Solinas, I. Ferino, Catal. Today 41 (1998) 179.
[38] Y. Tanaka, N. Sawamura, M. Iwamoto, Tetrahedron Lett. 39 (1998)
9457.
[39] M.J. Climent, A. Corma, S. Iborra, M.C. Navarro, J. Primolty,
J. Catal. 161 (1996) 783.
[18] I. Diaz, C. Marquez-Alvarez, F. Mohino, J. Perez-Pariente, E. Sastre,
J. Catal. 193 (2000) 295.
[19] D. Das, J.-F. Lee, S. Cheng, Chem. Commun. (2001) 2178.
[20] K. Wilson, A.F. Lee, D.J. Macquarrie, J.H. Clark, Appl. Catal. A 228
(2002) 127.