High selectivity of MCM-22 for cyclopentanol formation in liquid-phase cyclopentene hydration

Article abstract of DOI:10.1016/S0021-9517(02)00052-0

Highly effective formation of cyclopentanol through the liquid-phase hydration of cyclopentene has been attempted on various zeolites catalysts. MCM-22 zeolite was the most selective catalyst, which actively converted cyclopentene to cyclopentanol with a selectivity up to 99%. The effects on the hydration of catalyst preparation method, reaction atmosphere and temperature have been investigated for the MCM-22 catalysts. On the basis of the effect of reaction atmosphere, the mechanism of liquid-phase cyclopentene hydration was proposed. The thermodynamic equilibrium between cyclopentene and cyclopentanol was suggested to control greatly the cyclopentene conversion. The cyclopentene conversion was increased to 10% by increasing the water/cyclopentene ratio. Poisoning using organic amines with different molecular sizes revealed that the hydration occurred mainly in the 10-membered ring channels of MWW structure, which had an elliptic aperture smaller than that of MFI structure, exhibiting a significant shape selectivity by suppressing the etherification cyclopentanol.

Full text of DOI:10.1016/S0021-9517(02)00052-0

Journal of Catalysis 213 (2003) 272–280
www.elsevier.com/locate/jcat
High selectivity of MCM-22 for cyclopentanol formation
in liquid-phase cyclopentene hydration
Duangamol Nuntasri, Peng Wu, and Takashi Tatsumi ∗
Division of Materials Science & Chemical Engineering, Graduate School of Engineering, Yokohama National University,
79-5 Tokiwadai, Hodogaya-ku, Yokohama 240-8501, Japan
Received 26 July 2002; revised 7 October 2002; accepted 7 October 2002
Abstract
Highly effective formation of cyclopentanol through the liquid-phase hydration of cyclopentene has been attempted on various zeolites
catalysts. MCM-22 zeolite was the most selective catalyst, which actively converted cyclopentene to cyclopentanol with a selectivity up
to 99%. The effects on the hydration of catalyst preparation method, reaction atmosphere and temperature have been investigated for the
MCM-22 catalysts. On the basis of the effect of reaction atmosphere, the mechanism of liquid-phase cyclopentene hydration was proposed.
The thermodynamic equilibrium between cyclopentene and cyclopentanol was suggested to control greatly the cyclopentene conversion. The
cyclopentene conversion was increased to 10% by increasing the water/cyclopentene ratio. Poisoning using organic amines with different
molecular sizes revealed that the hydration occurred mainly in the 10-membered ring channels of MWW structure, which had an elliptic
aperture smaller than that of MFI structure, exhibiting a significant shape selectivity by suppressing the etherification cyclopentanol.
2002 Elsevier Science (USA). All rights reserved.
Keywords: Liquid-phase hydration; Cyclopentene; MCM-22; Shape selectivity
1. Introduction
catalyst activity. Furthermore, such catalysts generally need
to be used at temperatures above about 533 K to obtain
satisfactory yields and selectivities [1]. There have been
some attempts to improve the hydration catalyst using
tungsten oxide, molybdenum promoted silica-alumina and
cation-exchange resins; however, their activities are not
satisfactory enough to replace the heterogenised phosphoric
acid, and especially the organic ion-exchange resins lack
thermal stability at temperatures above 423 K [2,3].
Y zeolites ion-exchanged with chromium and rare-earth
cations and highly siliceous zeolites of ZSM series were
reported to successfully catalyze the direct hydration of
C2–C4 olefins under moderate pressure [4,5]. Zeolites,
having unique cation-exchange properties, strong surface
acidity, and good thermal stability, are presumed to update
the industrial processes of gas-phase hydration.
For the liquid phase hydration of olefins, such as cy-
clohexene, hydration has been shown to proceed over both
homogeneous and heterogeneous acidic catalysts. Sulfuric
acid, phosphoric acid, or p-toluenesulfonic acid, in the pres-
ence of additives such as copper(II) acetate, zinc(II) acetate,
or zinc(II) chloride, produces cyclohexanol with 70% selec-
tivity at ca. 35% conversion of cyclohexene [6]. This process
suffers, however, the disadvantage of forming high molec-
The hydration of olefins to alcohols is a typical acid-
catalyzed reaction which has been conducted in many ways.
The conventional process using sulphuric acid is widely
practiced. Nevertheless, it has several serious disadvantages,
such as the corrosion of the apparatus, the necessity to
dilute the acid with large amount of water to recover
the alcohol product, and the consequent requirement to
reconcentrate the acid before being recycled. To overcome
these disadvantages, it has been proposed to hydrate olefins
directly on various solid acid catalysts instead of liquid
acids. Phosphoric acid deposited on either silica gel, clay
or tungsten oxide, is typical example of such previously
attempted solid catalysts. Although these catalysts obviate
the handling problems imposed by sulfuric acid, they in
turn tend to introduce new complications, notably the
comparative production of a large amount of undesirable
polymer and ketone. Phosphoric acid in particular is easily
eluted during the reaction and then requires that makeup
acid should be added to the feed to maintain a constant
*
Corresponding author.
E-mail address: ttatsumi@ynu.ac.jp (T. Tatsumi).
0021-9517/02/$ – see front matter
PII: S0021-9517(02)00052-0
2002 Elsevier Science (USA). All rights reserved.

D. Nuntasri et al. / Journal of Catalysis 213 (2003) 272–280
273
ular weight by-products which are difficult to be removed
from the reactor. Another problem encountered is that since
cyclohexanol and catalyst exist in the aqueous acid phase,
the dehydration of cyclohexanol back to cyclohexene occurs
to a great extent during the distillation of this phase. The
selectivity of cyclohexene hydration depends significantly
on the competition between the desirable hydration reac-
tion and other side reactions, such as oligomerization, etc. In
order to decrease the extent of undesirable reactions, micro-
porous inorganic, or organic catalysts with shape-selectivity
if possible, are preferred to be applied as heterogeneous acid
catalysts.
to the small molecular size of cyclopentene molecules com-
pared to 10-membered ring (MR) channels of MFI structure.
We have revealed recently that MCM-22 zeolite is a promis-
ing catalyst for the selective formation of cyclopentanol
from cyclopentene hydration [10]. This process may lead
to the development of a new environmentally benign in-
dustrial process using zeolite catalyst. In this study, we
have investigated in details how reaction conditions, syn-
thesis method, and acidity influence the performance of
MCM-22 catalysts. The main purpose is to clarify why high
shape-selectivity was obtained with the structurally diverse
MCM-22.
Although the hydration of cyclohexene catalyzed by
polymer funtionalized with perfluorosulfonic acid or strong
acid polymer resulted in ca. 4% conversion and 80%
selectivity at 458 K [7], the real breakthrough in this reaction
has not been made until the last decade. Asahi Chem.
Ind. has industrialized the production of cyclohexanol from
benzene via partial hydrogenation and subsequent hydration
which are catalyzed by Ru and zeolite catalysts, respectively.
The new process for cyclohexanol production is superior to
the conventional two-step oxidation process in the respects
of selectivity, safety, corrosion, and energy consumption.
The key points of the second step, i.e., the hydration of
cyclohexene, are to take the advantages of ZSM-5 zeolite
catalyst which is hydrophobic due to the siliceous character
and exhibits a distinct shape-selectivity derived from its
medium pores. ZSM-5 catalyst performed in the liquid
bi-phase system at 393 K gives 13 wt.% yield and 99%
selectivity for cyclohexanol [8,9]. This innovative process
has fundamentally changed the methods for producing
the intermediates of adipic acid and ε-caprolactam in the
manufacture of nylon 6.
Liquid-phase cyclohexene hydration is up to now one
of only a few slurry processes using zeolite catalysts.
Making full use of various zeolite catalysts is expected to
develop new process for the production of other alcohols
such as cyclopentanol, which is an important intermediate
chemical in fragrant industries and a starting material for the
production of δ-valerolactam in nylon 5 manufacture.
However, ZSM-5 seems to lack the shape selectivity for
formation of cyclopentanol in cyclopentene hydration due
2. Experimental
2.1. Preparation of zeolite catalysts and characterization
The MCM-22 zeolites were hydrothermally synthesized
by modifying the procedures reported previously [11]. Syn-
thesis gels were prepared from hexamethyleneimine (HMI,
TCI) or piperidine (PI, Wako) as a structure directing
agent (SDA), fumed silica (Cab-o-Sil M7D or Aerosil 200),
sodium aluminate (52.1 wt.% NaOH, 27.7 wt.% Al, Wako),
sodium hydroxide (Koso), and deionized water. The follow-
ing example, corresponding to the catalyst No. 1 in Table 1,
represents the synthesis procedure. To a mixture containing
26.5 g of silica, 3.84 g of NaOH, and 158 g of H2O was
added 39.3 g of HMI dropwise under vigorous stirring. A so-
lution of 2.82 g of NaAlO2 in 158 g of water was added
after 1 h, and the mixture was stirred continuously until
producing a homogenous gel with a molar composition of
SiO2 : 0.033 Al2O3 : 0.3 NaOH : 0.9 HMI : 40 H2O. The re-
sulting gel was transferred into an autoclave which was then
rotated at 60 rpm and heated at 413 K for 7 days. The solid
product was filtered, washed, and dried at 353 K overnight
to obtain a lamellar precusor of MCM-22. Then the lamel-
lar product was calcined at 813 K for 8 h to remove SDA to
obtain MCM-22. Other MCM-22 samples were synthesized
in similar ways from the gels with molar composition shown
in Table 1. Y, Beta, mordenite (MOR), ZSM-5 and ferrierite
(FER) were obtained commercially.
Table 1
a
Synthesis and characterization of various MCM-22 zeolites
b
Sample name
Type of
Gel composition
SiO /Al O
N adsorption
2
2
2
3
c
2
−1
3
−1
)
SDA
SiO /Al O
Na/SiO
2
As-syn.
Ion-exchanged
A
Lang
(m g
)
V
(cm g
p
2
2
3
MCM-22-HMI-30
MCM-22-HMI-50
MCM-22-PI-50
MCM-22-HMI-60
MCM-22-PI-60
HMI
HMI
PI
HMI
PI
30
50
50
60
60
0.3
0.2
0.19
0.15
0.15
0.18
23
44
43
54
55
99
23
43
44
54
55
98
685
635
614
610
565
526
0.16
0.15
0.14
0.14
0.13
0.12
MCM-22-HMI-100
HMI
100
a
Crystalization conditions: temperature, 423 K; time, 7 days; rotating, 60 rpm.
b
c
SDA/SiO = 0.9, H O/SiO = 40.
2
2
2
HMI: Hexamethyleneimine, PI: Piperidine.

274
D. Nuntasri et al. / Journal of Catalysis 213 (2003) 272–280
All the zeolites were transformed into proton-form by
the conventional ion-exchange method. A calcined sample
was treated with 0.2 N NH4NO3 solution for 2 h at 353 K
five times. Ion-exchanged samples were calcined at 773 K
(except for Y, which was calcined at 723 K) for 5 h before
being employed as catalysts in the hydration reaction.
MCM-22 samples were characterized by SEM, XRD,
ICP elemental analysis, IR spectroscopy, temperature pro-
grammed desorption of ammonia (NH3-TPD), and N2 ad-
sorption. XRD patterns were collected using a MAC Science
MX-Labo X-ray diffractometer (Cu-Kα radiation). Scan-
ning electron micrograph (SEM) images were recorded on a
JEOL Electron Probe X-ray microanalyser (JXA-8900RL).
The chemical composition of samples was analyzed by a
Shimadzu ICPS-8000E ICP atomic emission spectrometer.
The IR spectra were collected on an FT-IR Perkin Elmer
1700 series spectrometer. NH3-TPD was carried out on a Bel
Japan REX-P2000 instrument, equipped with a quadrupole
mass spectrometer (M-QA100F, ANELVA). N2 adsorption
isotherms were obtained on a Bel Japan BEL SORP 28SA
instrument.
Fig. 1. NH -TPD profiles of MCM-22 catalysts.
3
showed no significant difference between the as-synthesized
and the ion-exchanged MCM-22 samples (Table 1). The
IR spectra of MCM-22 in the hydroxyl vibration region
showed a strong band at 3745 cm−1 due to the terminal
silanol groups, [SiOH], and a band at 3610 cm−1 due
to the bridging hydroxyl groups, [Si(OH)Al] (not shown).
The 3610 cm−1 band decreased in intensity gradually with
decreasing Al content, indicating a decrease in the amount
of Brønsted acid sites.
NH3-TPD spectra of MCM-22 samples consist of two
peaks at 423–573 K and 573–723 K (Fig. 1), correspond-
ing to weak and strong acid sites, respectively. The lower
temperature peak decreased remarkably with decreasing Al
content. This peak is probably attributed to weak Lewis
acid sites generated by partial dehydroxylation of struc-
tural Si(OH)Al groups at high temperature. Indeed, this type
of Lewis acid sites reversibly return to Brønsted acid sites
upon adsorption of water [14]. The higher temperature peak
mainly due to the Brønsted acid sites of framework Al
decreased gradually with decreasing Al content. The spe-
cific surface area and pore volume of MCM-22 samples,
measured from N2 adsorption isotherms at 77 K, where in
the range of 526–685 m2 g−1 and 0.12–0.16 cm3 g−1, re-
spectively (Table 1). Although the surface area decreased
slightly with increasing SiO2/Al2O3 ratio due to the diffi-
culty in crystallizing high silica MCM-22, the values over
500 m2 g−1 indicated that all the samples had good quality.
2.2. Cycloalkene hydration
The liquid-phase hydration of cycloalkene was carried
out batch-wise in a 50 cm3 teflon-lined autoclave. For a
typical run, 0.3 mol of water, 0.06 mol of cycloalkene
and 1 or 2 g of catalyst were mixed in the reactor. The
reaction was carried out by heating the reactor under
vigorous stirring at 373–413 K for 6–120 h. In some
cases, the reaction is performed under Ar atmosphere in
a glove box. After the reaction the product mixture was
extracted with ether, in which cycloheptanone as an internal
standard was added. The mixture was then analyzed on
a gas chromatograph (Shimadzu 14A) equipped with an
OV-1 capillary column (0.25 mm, 50 m long, df 1.5 µm).
The products were determined on a gas chromatograph-mass
spectrometer (JEOL DATUM-JMS-AX 500).
3. Results and discussion
3.1. Characterization of MCM-22 catalysts
MCM-22 is generally synthesized by using hexamethyl-
eneimine (HMI) as an SDA [11]. ERB-1 with the same
topology as MCM-22, on the other hand, is synthesized with
piperidine (PI) but requires the presence of crystallization
supporting agent of boric acid [12]. We have tried here from
the first time the synthesis of MCM-22 using PI as an SDA
under boric acid-free conditions.
MCM-22 catalysts exhibited similar aggregated platelet
morphology irrespective of SDAs as revealed by SEM
images (not shown). The XRD patterns of all the samples
were identical to those given in Ref. [13], identifying them
as MCM-22. The SiO2/Al2O3 ratios determined by ICP
3.2. Hydration of cyclohexene and cyclopentene over
various zeolite catalysts
Cyclohexene and cyclopentene were hydrated in the
liquid-phase with a proton-type zeolite catalysts to give
the corresponding alcohols as shown in Tables 2 and 3.
These experiments were carried out in the air. However,
the presence of oxygen greatly influenced the product se-
lectivity as described in the later section. The products
were predominately cycloalcohol, dicycloalkyl ether and cy-
cloalkene dimers. Table 2 shows that ZSM-5 was the most
effective catalyst for the cyclohexene hydration among the

D. Nuntasri et al. / Journal of Catalysis 213 (2003) 272–280
275
Table 2
Cyclohexene hydration catalyzed by various zeolite catalysts
a
Catalyst
Conv./
Selectivity/mol %
(SiO /Al O ratio)
mol %
2
2 3
Y (14)
Beta (25)
MOR (17)
ZSM-5 (25)
FER (20)
MCM-22 (30)–HMI
2.4
9.2
3.0
13.4
0.5
5.9
83.8
64.5
92.5
97.8
82.4
96.8
6.9
22.0
2.5
0.1
1.7
4.8
13.0
3.0
0.1
2.2
4.6
0.6
2.0
1.9
13.7
1.7
b
0.4
1.1
a
Reaction conditions: catalyst, 2 g; cyclohexene : H O : catalyst (weight ratio) = 4.8 : 5.4 : 2.0; temperature 393 K; time, 18 h; under air atmosphere.
2
b
HMI: hexamethyleneimine was used as SDA.
zeolite catalysts investigated since it showed the highest
in the channels. FER zeolite with smaller pore size showed
relatively high selectivity for alcohol, but cyclopentene con-
version was low probably because of significantly restricted
diffusion. Different from the cyclohexene hydration, only
MCM-22 was an effective catalyst from the viewpoint of
selective production of alcohol in the cyclopentene hydra-
tion. This was seen more obviously in the time courses of
cyclopentene hydration (Fig. 2). The conversion of MCM-
22 increased up to 6 h, but was nearly the same when the
reaction time was prolonged to 18 h. On the other hand,
the conversion always increased with the reaction time in the
case of ZSM-5 and Beta. However, in contrast to MCM-22
that maintained the alcohol selectivity above 92% indepen-
dent of the time, ZSM-5 and Beta showed the selectivity
decrease from 77 to 40% and from 58 to 30%, respec-
tively, when the hydration was extended from 2 to 18 h. It
should be noted that the conversion on MCM-22 was al-
most unchanged when the catalyst amount was increased
and similar conversion was observed when twice the amount
of cyclopentene/water was used. Thus, the lower cyclopen-
tene conversion on MCM-22 is attributed to neither low
activity nor easy deactivation. A progressive cyclopentene
conversion on ZSM-5 and Beta only occurred through the
consumption of cyclopentanol to ether, suggesting that the
thermodynamic equilibrium between cyclopentene and cy-
clopentanol may control the cyclopentene conversion on
conversion and more importantly, the highest selectivity
to cyclohexanol. The advantages of ZSM-5 must originate
from the medium pores of 10-MR and the hydrophobic na-
ture because of the high silica form. The latter property
allows ZSM-5 to serve as a water-tolerant solid acid cata-
lyst showing high catalytic activity, while the former makes
ZSM-5 exhibit a shape selectivity by suppressing the suc-
cessive reaction of cyclohexanol with cyclohexene to form
ether. These findings are fully consistent with what has
been reported by Ishida et al. [15]. MOR and FER zeo-
lites, imposing more significant diffusion limitation for the
reactant molecules because of practically one dimensional
channels or smaller pores, respectively, were less active than
the other zeolites. Y and beta zeolites having the 12-MR
channels where dicyclohexyl ether formation occurred eas-
ily showed much lower alcohol selectivity than the other
catalysts. MCM-22 exhibited cyclohexanol selectivity com-
parable to ZSM-5, although it showed only a half conversion
of cyclohexene.
As shown in Table 3, Beta and ZSM-5 zeolite catalysts
converted cyclopentene effectively, but showed lower cy-
clopentanol selectivity than MCM-22 catalyst. MOR and
Y zeolites also lack the cyclopentanol selectivity. These
results indicate that their pore sizes are too large to sup-
press the intermolecular dehydration of cyclopentanol or
the etherification between cyclopentene and cyclopentanol
Table 3
a
Cyclopentene hydration catalyzed by various zeolite catalysts
Catalyst
Conv./
Selectivity/mol %
(SiO /Al O ratio)
mol %
2
2 3
Y (14)
Beta (25)
MOR (17)
ZSM-5 (25)
FER (20)
MCM-22 (30)–HMI
a
2.6
13.0
3.8
8.1
0.7
73.3
29.5
57.8
41.0
91.4
92.3
0
0
0
0
1.7
0.2
19.6
64.4
30.5
51.9
0.8
2.9
0.8
3.5
0.4
5.8
3.9
3.6
5.3
6.9
1.9
0.3
0.6
b
2.9
3.0
Reaction conditions: catalyst, 2 g; cyclopentene : H O : catalyst (weight ratio) = 4.0 : 5.4 : 2.0; temperature, 393 K; time, 18 h; under air atmosphere.
2
b
HMI: hexamethyleneimine was used as SDA.

276
D. Nuntasri et al. / Journal of Catalysis 213 (2003) 272–280
lectivity of dimeric by-product, 1,3ꢀ-bicyclopentenyl while
the cyclopentene conversion on MCM-22 was comparable
(Table 4). This result can be interpreted by the reaction
mechanism proposed in Scheme 1. The main product of
cyclopentanol is produced through the interaction of water
with the carbenium intermediate species formed by the pro-
tonation of cyclopentene on the zeolitic Brønsted acid sites
(Scheme 1a). The carbenium ion would also react with the
product cyclopentanol and the reactant cyclopentene to give
cyclopentyl ether and the cyclopentene dimer, respectively.
Both the side reactions produce the bulky by-products, re-
quiring the open reaction space without steric constraint.
The formation of the ether in greater quantities relative to
the dimer product is probably because the nucleophilic sub-
stitution reaction of alcohol to form ether could occur more
easily than the dimerization reaction, which makes the se-
lectivity of dicyclopentyl ether higher than the cyclopentene
dimer. The noticeable formation of 1,3ꢀ-bicyclopentenyl and
2-cyclopentenone in the presence of oxygen is presumed to
occur via the oxidative route (Scheme 1b). Non-catalytic
oxidation at the allylic positions of cyclopentene occurs
easily at elevated temperature to produce a hydroperoxide
intermediate. With the assistance of zeolite acidity, the hy-
droperoxide with high reactivity changes rapidly into the
dimer or unsaturated ketone by-products. Therefore, in or-
der to obtain high alcohol selectivity, it is necessary to carry
out the hydration reaction in inert gas atmosphere to sup-
press the oxidation products. The oxidation pathway shown
for cyclopentene substrate (Scheme 1b) is not applicable to
cyclohexene probably because the greater ring strain of cy-
clopentene results in higher levels of autoxidation relative to
cyclohexene.
Fig. 2. Dependence of cyclopentene conversion (A) and cyclopentanol
selectivity (B) on the reaction time. For the reaction conditions, see Table 3.
3.2.2. Effect of SiO2/Al2O3 ratio and MCM-22 preparation
methods
MCM-22. Detailed consideration of this subject will be
given below.
Table 5 shows the results of cyclopentene hydration
over MCM-22 catalysts prepared using HMI or PI as SDA
under Ar atmosphere. Only the ether was the significant by-
product. All the MCM-22 catalysts gave the cyclopentanol
selectivity over 96% independent of the SiO2/Al2O3 ratio
and the preparation method. Except for MCM-22-HMI-60
which showed a lower conversion of 2.6%, the catalytic ac-
tivity declined slightly for both series of MCM-22 catalysts
with increasing SiO2/Al2O3 ratio, namely decreasing Al
content. This is as expected because MCM-22 with higher
SiO2/Al2O3 ratio contained less acid sites as revealed by
The above results indicated that MCM-22 exhibited an
outstanding cyclopentanol selectivity not shared by the other
zeolite catalysts. Thus, detailed studies have been carried out
to investigate how various parameters affected the catalytic
performance of MCM-22 in the cyclopentene hydration.
3.2.1. Effect of reaction atmosphere
When the cyclopentene hydration was carried out under
Ar instead of air, the cyclopentanol selectivity increased to
as high as 98% as a result of a dramatic decrease in the se-
Table 4
a
Effect of reaction atmosphere on cyclopentene hydration
Reaction
Conv./
Selectivity/mol %
atmosphere
mol %
Air
Ar
3.3
3.5
93.2
98.4
0.6
0
1.2
0.7
4.6
0.6
0.3
0.2
a
Reaction conditions: MCM-22 (SAR = 22), 1 g; cyclopentene : H O : catalyst (weight ratio) = 4.0 : 5.4 : 1.0; temperature, 393 K; time, 18 h.
2

D. Nuntasri et al. / Journal of Catalysis 213 (2003) 272–280
277
Scheme 1. Reaction routes of cyclopentene hydration (a), and the side reactions in air (b).
NH3-TPD measurements (Fig. 1). It is surprising that MCM-
3.2.3. Effects of hydration temperature
22-HMI-100 showed a relatively high conversion of 3.1%
considering it possesses the smallest amount of acid sites.
An increase in hydrophobicity at high SiO2/Al2O3 ratio, the
property important for the effective function of solid acid
in aqueous solution, may partially account for this behavior.
However, the potential ability of MCM-22 seemingly has not
been displayed fully even at high Al content due to the ther-
modynamic limitation on the hydration.
The cyclopentene hydration is an exothermic reaction,
which was confirmed by the lower conversion obtained at
413 K than at 393 K (Table 6). High conversion was thus
expected if the reaction is carried out at lower temperature
which would be favorable for the equilibrium conversion to
cylopentanol. However, at 373 K the conversion increased
with the time very slowly and reached only 4.1% even after
the reaction for 120 h. The hydration occurred more slowly
at reaction temperature lower than 373 K (not shown). The
acid sites of zeolites hardly act as catalytic centers for
the liquid-phase olefin hydration at temperature lower than
373 K probably because of the strong adsorption of water
molecules. Another possible reason may be that the energy
barrier in the protonation of alkenes is hardly overcome at
low temperature.
MCM-22-PI catalysts showed somewhat higher conver-
sion than those prepared by HMI at similar SiO2/Al2O3
ratios (Table 5, no. 2, 3, 5, and 6). The amounts of strong
Brønsted acid sites determined by NH3-TPD for MCM-22-
PI-50 and MCM-22-HMI-50 were 0.55 and 0.44 mmol g−1
,
respectively. This suggests that PI is more effective than
HMI in incorporatingAl species into the frameworks to form
the Brønsted acid sites.
Table 5
a
Cyclopentene hydration catalyzed by MCM-22 catalyst prepared by HMI and PI
b
No.
Catalyst
SiO /Al O
2 3
Conv./
Selectivity/mol %
2
mol %
1
2
3
4
5
6
MCM-22-HMI-30
MCM-22-HMI-50
MCM-22-HMI-60
MCM-22-HMI-100
MCM-22-PI-50
23
43
54
98
44
55
3.7
3.2
2.6
3.1
3.7
3.4
98.2
97.2
96.7
95.7
97.4
97.4
0.8
1.7
2.1
3.4
1.6
1.7
0.7
0.8
0.8
0.7
0.7
0.6
0.3
0.3
0.4
0.3
0.3
0.3
MCM-22-PI-60
a
Reaction conditions: catalyst 1 g; cyclopentene : H O : catalyst (weight ratio) = 4.0 : 5.4 : 1.0; temperature, 393 K; time, 18 h; under Ar atmosphere.
2
b
HMI (hexamethyleneimine) or PI (piperidine) was used as SDA.

278
D. Nuntasri et al. / Journal of Catalysis 213 (2003) 272–280
Table 6
a
Effect of temperature and time on the cyclopentene hydration over MCM-22 catalyst
Temp./
Reaction
Conv./
Selectivity/mol %
K
times/h
mol %
393
413
373
373
373
18
18
18
72
120
4.2
3.4
1.5
3.3
4.1
98.3
95.8
98.9
99.1
98.7
1.1
2.8
0.3
0.4
0.8
0.4
1.0
0.6
0.4
0.4
0.2
0.4
0.1
0.1
0.1
a
Reaction conditions: MCM-22 (SiO /Al O = 24), 1 g; cyclopentene : H O : catalyst (weight ratio) = 4.0 : 5.4 : 1.0; under Ar atmosphere.
2
2
3
2
3.2.4. Effect of thermodynamic equilibrium
reached the end due to the possible deactivation of ZSM-5,
95% cyclopentanol conversion (5% cyclopentanol remain-
ing) matched well with the maximum conversion about (4%)
attained in the cyclopentene hydration. Hence, the thermo-
dynamic equilibrium seems to control the liquid-phase hy-
dration of cyclopentene, which makes the conversion higher
than 4% impossible under the present conditions. Despite
the equilibrium, ZSM-5 and beta zeolite catalysts exhibited
8 and 13% cyclopentene conversion, respectively (Table 3).
This is mainly because the alcohol product further reacts
with starting material of alkene to form ether.
We have made great efforts to increase the cyclopentanol
yield on MCM-22 catalyst by varying the SiO2/Al2O3 ra-
tio, preparation methods and reaction conditions such as
time and temperature as described above, but the conver-
sion varied little and the cyclopentanol yield never exceeded
4%. This maximum yield was also the case in the reaction
catalyzed by the other zeolites. Thus, these results bring
out the problem of thermodynamic equilibrium between cy-
clopentene and cyclopentanol. Since thermodynamic data
depending on temperature and pressure are lacking and
the complicated situations for the reaction do not allow
us to determine correctly the existing phases for the reac-
tants and products, it is hard to calculate the equilibrium.
We have tried to estimate the equilibrium composition by
carrying out the dehydration of cyclopentanol, i.e., the reac-
tion reverse to cyclopentene hydration (Table 7). The same
conditions as in the hydration were employed for the de-
hydration of cyclopentanol except for the replacement of
cyclopentene with cyclopentanol. The dehydration of cy-
clopentanol in the presence of H2O occurred much more
rapidly than the etherification over the three zeolites, while
large pore Beta reasonably produced bulky ether more than
ZSM-5 and MCM-22. Obviously, the dehydration proceeded
more rapidly than the hydration. The lower effectiveness of
Beta and MCM-22 compared to ZSM-5 in the dehydration
is suggested to be due to the higher hydrophilicity of Beta
and the smaller 10 MR channels of MCM-22. The high de-
hydration level observed on ZSM-5 strongly indicates that
the equilibrium is greatly shifted to the cyclopentene side at
393 K. Considering that the dehydration may still have not
3.2.5. Effect of water/olefin ratio
In order to shift the equilibrium to the cyclopentanol
side to increase the conversion in the cyclopentene hy-
dration, the water/olefin molar ratio was increased while
keeping the amount of MCM-22 catalyst constant. The cy-
clopentene conversion increased from 4% at the normal
water/cyclopentene ratio of 5 to 10% at the ratio of 30 while
the alcohol selectivity higher than 98% was retained (Fig. 3).
It is noteworthy that the industrial process for cyclohexene
hydration is operated at similar conversion levels. MCM-22
showed much higher selectivity of cyclopentanol formation
Table 7
a
Cyclopentanol dehydration catalyzed by zeolite catalysts
Catalyst
SiO / Conv./
Selectivity/mol %
2
Al O
2
mol %
3
ratio
Beta
ZSM-5
MCM-22
25
25
24
50
95
54
81.0
93.5
99.9
19.0
6.4
0.1
0.05
0.1
0
Fig. 3. Effect of water/cyclopentene ratio on the cyclopentene hydration
over MCM-22 reaction conditions: MCM-22 (SiO /Al O = 24) catalyst
2
2 3
a
Reaction conditions: catalyst,
1 g; cyclopentanol : H O : catalyst
2
1 g; cyclopentene : catalyst (weight ratio) = 4.0 : 1.0; temperature, 393 K;
(weight ratio) = 5.2 : 5.4 : 1.0 temperature, 393 K; time, 18 h.
time 18 h; under Ar atmosphere.

D. Nuntasri et al. / Journal of Catalysis 213 (2003) 272–280
279
Fig. 5. Channel system of MCM-22.
to the outer surface through the interlayer 10-MR channels.
The elliptic aperture of these 10-MR channels with a free
diameter of 4.0 × 5.5 Å (between layer) and 4.1 × 5.1 Å
(within layer), respectively, smaller than the 10-MR chan-
nels of MFI structure (5.1 × 5.5 Å and 5.3 × 5.6 Å) [17],
would successfully suppress the etherification of cyclopen-
tanol forming bulky ethers. The acid sites at external side
pockets would scarcely promote this hydration reaction be-
cause the external silanol groups are surrounded by water.
MCM-22 is then considered to be characteristic of its 10-
MR channels in the liquid-phase hydration. The fact that the
poisoning with bulky 2,4-DMQ increased the conversion as
well as the selectivity for cyclopentanol would be accounted
for by assuming the suppression of the formation of poly-
meric products which may block the internal active site.
Fig. 4. Cyclopentene hydration over MCM-22, ZSM-5 and Beta; reac-
tion conditions: cyclopentene : H O : catalyst (weight ratio) = 4.0 : 5.4 −
2
31.6 : 1.0 − 2.0; temperature, 393 K.
than ZSM-5 and Beta zeolite catalysts at high cyclopentene
conversions (Fig. 4).
3.2.6. An investigation into the high cyclopentanol
selectivity of MCM-22
The MWW topology of MCM-22 exhibits structural di-
versity of side pockets and two independent 10-MR channel
systems one of which contains the supercages (Fig. 5). The
existence of open reaction spaces which would impose min-
imal shape selectivity makes it difficult to explain the high
alcohol selectivity observed with MCM-22. In order to clar-
ify the MCM-22 active sites contributing to the cyclopen-
tene hydration reaction, poisoning experiments with amines
have been carried out. In the presence of trimethylamine
(Me3N) which is expected to diffuse into all the acid sites of
MCM-22, the cyclopentene conversion decreased to 0.6%
(Table 8), indicating the hydration was severely retarded.
On the other hand, 2,4-dimethylquinoline (2,4-DMQ) which
is believed not to enter into the 10-MR channels but selec-
tively deactivates the acid site within the 12-MR side pockets
[16], both the cyclopentene conversion and the alcohol se-
lectivity increased. These results strongly suggest that the
cyclopentene hydration occurs mainly within the two 10-
MR channels, and the intracrystalline supercages connecting
4. Conclusion
MCM-22 proves to be an active and highly selective
catalyst for the liquid-phase cyclopentene hydration. The
product selectivity depends greatly on the reaction at-
mosphere; the dimer and unsaturated ketone can be dra-
matically decreased when the reaction is conducted under
Ar atmosphere. MCM-22 synthesized with PI is more ef-
fective in catalyzing the hydration than that prepared with
HMI. A thermodynamic equilibrium limits the conversion
level. However, the alkene conversion and alcohol selectiv-
ity can be improved to as high as 10 and 99%, respectively,
Table 8
a
Poisoning of cyclopentene hydration over MCM-22
Poisoning
reagent
Conv./
Selectivity/mol %
mol %
No
Me N
3
2,4-DMQ
3.9
0.6
4.5
97.5
84.4
99.6
1.5
15.0
0.1
0.7
0
0.1
0.3
0.6
0.1
b
a
Reaction conditions: MCM-22 (SiO /Al O = 24) catalyst, 1 g; cyclopentene : H O : catalyst (weight ratio) = 4.0 : 5.4 : 1.0; amine, 3 mmol; temperature,
2
2
3
2
393 K; time, 18 h.
b
2,4-DMQ: 2,4-dimethylquinoline.

280
D. Nuntasri et al. / Journal of Catalysis 213 (2003) 272–280
by increasing the water to olefin molar ratio to 30. The high
cyclopentanol selectivity obtained with MCM-22 essentially
derives from its 10-MR channels exhibiting high shape se-
lectivity. The side pockets are not considered to significantly
contribute to the hydration.
[6] F. Mizukami, R.N.S. Macyama, T. Tsuchiya, K. Shinuzu, J. Imamura,
Sekiyu Gakkaishi 28 (1985) 172;
Chem. Abstr. 103 (1985) 87527.
[7] F.J. Waller, US Patent 4,595,786, 1986.
[8] K. Yamashita, H. Obana, T. Kai, US Patent 5,302,762, 1994.
[9] M. Misono, T. Inui, Catal. Today 51 (1999) 369.
[10] D. Nuntasri, P. Wu, T. Tatsumi, Chem. Lett. 2 (2002) 224.
[11] A. Corma, C. Corell, J. Pérez-Pariente, Zeolites 15 (1995) 2.
[12] R. Millini, G. Perego, W.O. Parker Jr., G. Bellussi, L. Carluccio,
Micropor. Mater. 4 (1995) 221.
[13] M.K. Rubin, B. Cynwyd, P. Chu, US Patent 4,954,325, 1990.
[14] M. Niwa, N. Katada, Catal. Surv. Jpn. (1997) 215.
[15] M. Kohno, Y. Fukuoka, O. Mitsui, H. Ishida, Nippon Kagaka Kaishi
(1989) 521.
References
[1] C.D. Chang, N.J. Morgan, US Patent 4,214,107, 1980.
[2] F. Fajula, R. Ibarra, F. Figueras, C. Gueguen, J. Catal. 89 (1984) 60.
[3] H. Momose, K. Kusumoto, Y. Izumi, Y. Mizutani, J. Catal. 77 (1982)
554.
[16] P. Wu, T. Komatsu, T. Yashima, Micropor. Mesopor. Mater. 22 (1998)
343.
[17] Ch. Baerlocher, W.M. Meier, D.H. Olson, Atlas of Zeolite Framework
Types, Elsevier, Amsterdam, 2001.
[4] M. Iwamoto, M. Tajima, S. Kagawa, J. Catal. 101 (1986) 195.
[5] K. Tao, W. Li, H. Li, X. Qi, Appl. Catal. A 139 (1996) 43.