Properties of Ti-beta zeolites synthesized by dry-gel conversion and hydrothermal methods

Article abstract of DOI:10.1021/jp9816216

The large-pore beta zeolite containing Ti has been synthesized in the presence of alkali cations (Na⁺) by a dry-gel conversion (DGC) technique using tetraethylammonium hydroxide (TEAOH) as a structure-directing agent. The obtained Ti-beta zeolites were compared with hydrothermally synthesized (HTS) Ti-beta. It was found that Ti-beta-DGC samples adsorbed less water (12-13 wt %) than Ti-beta-HTS (19 wt %), indicating that Ti-beta-DGC is more hydrophobic than Ti-beta-HTS. We propose that this leads to a higher activity for the former in the oxidation of cyclohexane. Upon being washed with H₂SO₄, the turnover number (TON) of DGC samples slightly increased for the oxidation of cyclohexane. DGC and HTS samples demonstrated no significant difference in the activity for oxidation reactions of unsaturated alcohols and C₆-C₈ cyclic alcohols.

Full text of DOI:10.1021/jp9816216

7126
J. Phys. Chem. B 1998, 102, 7126-7131
Properties of Ti-Beta Zeolites Synthesized by Dry-Gel Conversion and Hydrothermal
Methods
Takashi Tatsumi* and Nizamidin Jappar
Engineering Research Institute, School of Engineering, The UniVersity of Tokyo, Yayoi, Tokyo 113-8656, Japan
ReceiVed: March 25, 1998; In Final Form: June 10, 1998
+
The large-pore beta zeolite containing Ti has been synthesized in the presence of alkali cations (Na ) by a
dry-gel conversion (DGC) technique using tetraethylammonium hydroxide (TEAOH) as a structure-directing
agent. The obtained Ti-beta zeolites were compared with hydrothermally synthesized (HTS) Ti-beta. It was
found that Ti-beta-DGC samples adsorbed less water (12-13 wt %) than Ti-beta-HTS (19 wt %), indicating
that Ti-beta-DGC is more hydrophobic than Ti-beta-HTS. We propose that this leads to a higher activity for
the former in the oxidation of cyclohexane. Upon being washed with H
DGC samples slightly increased for the oxidation of cyclohexane. DGC and HTS samples demonstrated no
significant difference in the activity for oxidation reactions of unsaturated alcohols and C -C cyclic alcohols.
2 4
SO , the turnover number (TON) of
6
8
1
. Introduction
connectivity defects. Hence the presence of Al and large
concentration of internal and external silanol groups confers a
rather hydrophilic character on Ti-beta in contrast to the
organophilic characteristics of TS-1.
Titanium silicalites (TS-1 and TS-2) are highly efficient for
the selective oxidation of a large number of organic substrates,
such as alkenes, alcohols, aromatics, and alkanes, using H2O2
as the oxidant under mild reaction conditions.1 Unlike the group
The beta structure is thought to be an unstable kinetic phase
which crystallizes merely under very specific conditions. Unlike
the molecular sieve having an MFI structure which may
crystallize using a wide variety of templates, Ti-beta with SiO2/
Al2O3 higher than 200 could not be obtained by the conventional
IV-VI metal oxide catalysts, titanium silicalites are active in
the presence of diluted aqueous solutions of hydrogen peroxide,
whereas organic hydroperoxides, such as tert-butyl hydroper-
oxide (TBHP), are the oxidants of choice for the former
4,7
catalysts.2 The differences can be partially ascribed to the
hydrothermal synthesis method. Even in the case of synthesis
with a lower SiO2/Al2O3 ratio, the product yield was often very
hydrophobic character of titanium silicalites, which favors the
adsorption of the organic substrates over the more polar water
molecules present in the H2O2 solution, thus maintaining the
Ti sites inside the zeolite pores in the organic-rich environment.
The activity for the oxidation of 1-hexene and n-octane on TS-1
is not modified by the use of either aqueous or anhydrous
solutions of H2O2, but activity of amorphous TiO2-SiO2
7
low. There is therefore a strong incentive for the preparation
of Ti-beta with a better zeolite yield by using new templates
and methods.
9
Improved synthesis methods (e.g., Cogel method, seeding
1
0
11
12
techniques, fluoride method, and gas-solid and liquid-
13
solid isomorphous substitution have been developed to obtain
the Ti-beta zeolites with high yields. The first successful
preparation of Ti-beta without the need for Al and alkali cations
in the gel was patented by ARCO using a benzyl-substituted
diquaternary ammonium cation as a novel structure-directing
catalysts is strongly inhibited when the aqueous H2O2 solution
is used as oxidant.3
Recently, a large-pore Ti-containing zeolite, Ti-beta, has been
hydrothermally synthesized from gels containing tetraethylam-
1
4
agent.
Aluminum-free Ti-beta was synthesized using di-
monium hydroxide (TEAOH) as a structure directing agent and
some aluminum as a framework constituent.4 Ti-beta proved
(cyclohexylmethyl)dimethylammonium as the structure-directing
1
5
agent and was used for Meerwein-Ponndorf-Verley-Op-
to be an active catalyst for the oxidation of alkanes and alkenes,
1
6,17
18
either by using H2O2 or TBHP as oxidants.5
-7
penauer reactions
and epoxidation reactions.
In the
Due to its large
crystallization of a variety of zeolites, the vapor-phase transport
pore size, Ti-beta was shown to be more active than TS-1 for
the oxidation of bulky substrates such as cyclic and branched
molecules. However, an additional factor, viz., the presence
of aluminum in the framework of Ti-beta, can contribute to the
1
9-21
(VPT) method is also an efficient one,
which involves
5
,6
crystallization of a dry gel in the presence of volatile structure-
directing agents and steam. However, in many cases, the VPT
method is not applicable since most of the zeolite preparations
need nonvolatile structure-directing agents. Recently, Rao and
different catalytic behavior observed between Ti-beta and TS-
7
1; the framework Al introduces additional acidity into Ti-beta,
22,23
Matsukata
developed a new effective technique named dry-
which is reflected in the lower selectivity to the epoxide during
olefin epoxidation than on TS-1, since residual protons associ-
ated to Al catalyze the opening of the oxirane ring with the
corresponding formation of glycols and glycol derivatives.
Furthermore, aluminum atoms incorporated into the framework
gel conversion (DGC) for the synthesis of beta zeolite with a
high SiO2/Al2O3 molar ratio up to 900 using TEAOH as the
template. More recently, we have succeeded in synthesizing
Ti-beta zeolites in the presence of certain amount of sodium
6
,7
2
4
8
by the DGC method using TEAOH. The product yield was
more than 95%. In the oxidation reaction of cyclohexene, the
Ti-beta exhibited complete conversion and 100% of selectivity
of H2O2 to the oxidized products.
lead to a detrimental effect on the activity for epoxidation. On
the other hand, thus prepared Ti-beta has a high density of
*
To whom correspondence should be addressed.
S1089-5647(98)01621-6 CCC: $15.00 © 1998 American Chemical Society
Published on Web 08/25/1998

Properties of Ti-Beta Zeolites
J. Phys. Chem. B, Vol. 102, No. 37, 1998 7127
TABLE 1: Synthesis of Ti-Beta Materials
in gel (mol %)
in zeolite (mol %)
Na/
H
2
SO
4
washed (mol %)
Na/ Al/
Ti/
Na/
Al/
Ti/
Al/
crystallinity
(%)
Ti/
sample
(TIV + Al) (TIV + Al) (TIV + Al) (TIV + Al) (TIV + Al) (TIV + Al)
(TIV + Al) (TIV + Al) (TIV + Al)
DGC-1
DGC-2
DGC-3
DGC-4
DGC-5
DGC-6
DGC-7
DGC-8
DGC-9
DGC-10
HTS-1
HTS-2
TS-1
3.19
3.20
3.18
3.18
3.17
3.20
3.18
3.20
3.18
3.18
2.43
2.43
3.0
0.98
0.68
0.99
0.98
0.98
0.99
0.98
1.93
0.98
0.39
0.16
0.29
0.29
0.29
0.80
0.16
0.16
0.29
0.13
0.29
0.24
0.24
3.30
3.12
3.12
3.12
3.25
3.30
3.30
3.39
0.52
0.64
0.92
1.02
1.04
0.57
0.66
1.95
0.12
0.30
0.28
0.27
0.62
0.12
0.12
0.30
70
81
74
82
85
80
80
38
0
2.32
2.17
2.29
2.66
3.11
0.23
0.23
0.22
0.18
0.10
0.12
0.12
0.15
0.15
0.49
a
nd
nd
0
3.43
2.17
1.25
0.42
0.65
85
85
100
1.92
1.42
0.23
0.45
a
nd ) no data.
The aim of this work is to further investigate the properties
of the Ti-beta zeolite synthesized by the dry-gel conversion
method (Ti-beta-DGC) in comparison with the Ti-beta synthe-
sized by the hydrothermal method (Ti-beta-HTS).
experiment, 18.7 mmol of alkane, 10 mL of solvent, 0.1 g
catalyst, and 9.3 mmol of hydrogen peroxide (31% aq solution)
were mixed in a Teflon inner placed in the autoclave and heated
to the desired temperature under vigorous agitation. The
unsaturated alcohol oxidation reactions were carried out at 323
K for 3 h in a glass round-bottom flask (50 mL) fitted with a
condenser and a magnetic stirrer. A mixture of 16.5 mmol of
unsaturated alcohol, 10 mL of solvent, 0.1 g catalyst, and 5
mmol of hydrogen peroxide (31% aq solution) was heated in
the flask to the desired temperature under vigorous agitation.
The oxidation of cyclic alcohols was carried out at 338 K. After
the reaction was completed, the catalysts were separated by
filtration and the products were analyzed on a Shimadzu GC
2. Experimental Section
Synthesis and Characterization of Ti-beta Catalyst. Typi-
cally, Ti-beta was synthesized according to the following
procedure: 0.58 g of tetrabutyl orthotitanate (97 wt %, Kanto)
was first added to 4.0 g of distilled water, and to the resulting
suspension was added 2.0 g of H2O2 (31 wt %, MGC) after 1
h. The mixture was stirred at room temperature for 1 h, leading
to solution A containing peroxide titanate. Solution B was
prepared by dissolving 0.0124 g of anhydrous NaAlO2 (Koso)
and 0.015 g of NaOH (96 wt %, Koso) in 8.0 g of TEAOH (40
wt % in water, Alfa) at room temperature with stirring for 1 h.
Solution B was added to solution A, and stirring was continued
for 1.5 h. Then, 3.0 g of fumed silica (Aerosil-200, Nippon
Aerosil) was added under vigorous stirring. A clear homoge-
neous solution obtained after 2 h was heated to 353 K and dried
while stirring. When the gel became dry, it was ground into
fine powder (chemical composition: SiO2:TiO2:Al2O3:Na2O:
TEAOH ) 304:10:0.46:1.55:132.5) and transferred into a special
autoclave, where water as source of steam was poured into the
bottom. The crystallization was carried out in steam first at
1
4A gas chromatograph equipped with a 50 m OV-1 capillary
column. Cycloheptanone or 2-pentanone was used as an internal
standard.
3
. Results and Discussion
Characterization. The yield of Ti-beta zeolite is higher than
95 wt % based on silica for the DGC method in contrast to the
low (∼15 wt %) yield for the HTS method. The Ti-beta-DGC
synthesized from colloidal silica with a high content of sodium
(Na2O/SiO2 ) 0.04) is well crystallized and the compound was
used as a reference of the crystallinity.
The composition of the gel in the synthesis of Ti-beta-DGC
significantly affects the crystallinity and oxidation activity of
the final products. The composition and crystallinity of Ti-
beta-DGC and Ti-beta-HTS are shown in Table 1. The presence
of Na is unfavorable for the synthesis of Ti-beta by the
403 K for 96 h, subsequently at 448 K for 18 h under autogenous
pressure. The obtained Ti-beta zeolite product (Ti-beta-DGC)
was washed with distilled water, dried at 308 K for 10 h, and
calcined at 793 K for 10 h in the flow of air. The hydrother-
mally synthesized Ti-beta (Ti-beta-HTS) was obtained according
to the recipe described in the literature.7 The Ti-beta zeolites
were treated with 1 M H2SO4 at room temperature for 12 h and
then washed with distilled water, dried at 308 K for 10 h, and
calcined at 793 K for 5 h in the flow of air.
X-ray powder diffraction patterns were collected on a Rigaku
Denki RU-200A diffractometer using Cu KR radiation. SEM
micrographs were obtained with a JEOL JSM 5400 microscope
equipped with an ED probe. The Fourier transformed infrared
spectra (FTIR) were recorded in the transmittance mode on a
Perkin-Elmer 1600 spectrometer using conventional heatable
IR cells connected to vacuum-gas manipulation apparata. UV-
vis spectra were obtained on a Hitachi 340 spectrophotometer.
Thermogravimetry (TG) was performed on an ULVAC TGD-
7
hydrothermal method. However, in the case of the DGC
method, the presence of a certain amount of Na species in the
gel is necessary. At Na contents lower than 0.4 mol %, the
products are amorphous. With a high content of Na, the
products exhibit low crystallinity (see, e.g., DGC-8 in Table
1). The amount of Al in the gel also influences the crystallinity
of the products. At Al contents Al/(Si+Ti+Al) lower than 0.16
mol %, only amorphous products were obtained. Upon treat-
ment of sample with 1 M H2SO4 at room temperature, the Al
and Na contents of Ti-beta-DGC are remarkably decreased and
Ti is also slightly lost.
The UV-vis spectra of the calcined Ti-beta synthesized by
the DGC and HTS methods are presented in Figure 1. All the
samples exhibit a band at 205-230 nm, resulting from the
4
+
7
000 differential thermogravimetric analyzer.
charge transfer from oxygen atoms to Ti , characteristic of
tetrahedrally coordinated Ti in the framework and is generally
Catalytic Reactions. The alkane oxidation reactions were
conducted at 373 K for 5 h in an autoclave (30 mL). In a typical
25
observed for Ti-substituted molecular sieves. The absence of

7128 J. Phys. Chem. B, Vol. 102, No. 37, 1998
Tatsumi and Jappar
the framework during hydrothermal syntheses.5 By the DGC
method, however, the existence of Na below 1 mol % does not
affect the incorporation of titanium into the framework.
Figure 2 shows the SEM images of Ti-beta zeolite synthesized
by the DGC and HTS methods. The particle size of the products
for both the DGC and HTS samples is rather uniform and no
amorphous phase can be found in the samples. The average
crystal size of the Ti-beta-DGC (20-40 nm) is much smaller
than that of the Ti-beta-HTS (150-400 nm). The formation
of such small uniform crystals from the dry gel powder is of
particular interest to understand the mechanism of zeolite
formation, possibly indicating a large number of nuclei are
formed during the crystallization of Ti-beta-DGC.
The IR spectra of the hydroxyl groups of the Ti-beta zeolites
recorded as a function of the outgassing temperature are
illustrated in Figure 3. For the DGC sample (Ti-beta-DGC-5
-
1
H2SO4-washed), an intense band at 3740 cm is observed
-
1
together with broad shoulder bands ranging from 3610 cm to
-
1
3
300 cm . For the HTS sample, three types of OH bands can
-
1
be distinguished: a strong band at 3740 cm , a weak band at
-
1
3
606 cm , and shoulder bands ranging from 3600 to 3300
-1
cm . With increasing outgassing temperature, the band at 3740
cm is sharpened and the broad shoulder bands are substantially
decreased in intensity.
The band at 3740 cm- is ascribed to the isolated silanol
-1
1
-
1
groups, the band at 3606 cm to the Si-O-Al groups, and
Figure 1. UV-vis spectra of Ti-beta zeolites. DGC: dry-gel conver-
sion methods sample, HTS: hydrothermally synthesized sample. H:
-1
the broad band (3610-3300 cm ) to the H-bonded OH groups
2
6-28
of silanol groups.
The broad feature of the latter band
2 4
H SO -washed sample.
should be due to the interaction between the hydroxyl groups
at internal defects which results in the formation of hydrogen
bonded chains. The length and configuration of such chains
vary depending upon the structure of the defects of the zeolites.28
Heating at the elevated temperature in vacuo leads to a decrease
in the length of the chain and to an increase in the number of
the isolated silanol species due to the removal of water in the
zeolites and, consequently to the decrease in intensity of the
the band between 310 and 350 nm indicates that anatase-like
phase is not formed in these samples during crystallization.
DGC-8, synthesized in the presence of large amount of alkali
cations, shows a shoulder at 270 nm, which can be attributed
to hexacoordinated Ti species belonging to an amorphous
titanosilicate phase.25 The HTS sample synthesized hydrother-
mally in the absence of sodium shows a band at about 205 nm
and Ti-beta-DGC synthesized in the presence of sodium exhibits
band at 205-230 nm. It has been reported that the presence
of alkali cation could prevent titanium from incorporating into
-1
broad band around 3500 cm . Ti-beta-DGC-5 (H2SO4 washed)
sample contains a negligible amount of Na (Na2O/SiO2 )
0.0005, Na2O/Al2O3 ) 0.2), which cannot influence IR spectra
Figure 2. SEM of Ti-beta synthesized by (A) DGC method and (B) HTS method.

Properties of Ti-Beta Zeolites
J. Phys. Chem. B, Vol. 102, No. 37, 1998 7129
Figure 4. Desorption of water on Ti-beta. DGC: dry-gel conversion
methods sample. HTS: hydrothermally synthesized sample. H: H
2
-
SO -washed sample.
4
activities of Ti-beta samples synthesized by the DGC and HTS
methods in the oxidation of cyclohexane with H2O2 at 373 K.
Cyclohexanol and cyclohexanone are formed from cyclohexane
over the Ti-beta samples. The Ti-beta-DGC samples exhibit
higher turnover number (TON) and H2O2 selectivity than Ti-
beta-HTS samples. The difference is probably related to the
hydrophobic character of Ti-beta zeolites. For hydrophilic Ti-
beta zeolites, the Ti active sites are covered by water molecules,
and therefore, cyclohexane molecules are made to be not easily
accessible to the active sites, as will be discussed below.
Upon treatment of the Ti-beta-DGC samples (Table 1) with
1 M H2SO4 at room temperature, large amounts of Al and Na
are eliminated and Ti is also slightly lost. As a result, the
turnover number (TON) of the samples is significantly increased.
This is in agreement with the previous findings; the catalytic
activities for the hydrothermally synthesized Ti-containing
2 4
Figure 3. FT-IR spectra of Ti-beta. H: H SO -washed sample.
in the O-H region. Compared with the Ti-beta-HTS sample,
the intensity of the broad band for the Ti-beta-DGC sample is
much weaker, which suggests that the Ti-beta-DGC contains
less defects than the Ti-beta-HTS.
The amount of water absorbed on the calcined Ti-beta
samples which were contacted with moisture over saturated
aqueous NH4Cl solution for 24 h is determined thermogravi-
metrically (Figure 4). The weight losses of Ti-beta-DGC
samples are about 12-13%. The weight losses are independent
of the contents of Na or Al. The amount of water desorbed
from Ti-beta-HTS-1 (19 wt %) is larger than that from DGC
samples, implying that Ti-beta-DGC is more hydrophobic than
Ti-beta-HTS, in agreement with the lower density of the defects
on the former.
5
zeolites are decreased with increasing Al for Ti-beta and Na
2
,8,30
Influence of Ti-beta Properties on its Catalytic Activity
in the Oxidation of Cyclohexane. Table 2 shows the catalytic
content for TS-1.
The cyclohexane oxidation activity of
H2SO4-washed Ti-beta-HTS sample is, however, low compared
a
2 2
TABLE 2: Activity of Ti-Beta Catalysts in the Oxidation of Cyclohexane with H O
calcined
H
2
SO
4
washed
c
c
cyclohexane
selectivity (%)
H
O
2 2
(%)
cyclohexane
selectivity (%)
2 2
H O (%)
sample conversion (%) TONb OL ONE EHK conversion selectivity conversion (%) TONb OL ONE EHK conversion selectivity
DGC-1
DGC-2
DGC-3
DGC-4
DGC-5
DGC-6
DGC-7
HTS-1
HTS-2
TS-1
11.1
11.5
11.3
10.1
9.8
12.1
11.2
6.4
46
47
48
42
41
51
47
26
30
79 21
81 19
78 22
84 16
83 17
74 26
77 23
93
89
92
90
89
90
91
90
90
88
29
31
30
26
26
34
30
16
12
23
20
23
17
67
62
69
45
83
82
80
80
12
12
13
12
4.6
5.9
7.1
8.2
91
90
91
92
25
27
31
23
81
84
3.5 16
7.1 0.6
2.2
2.0
8.8 55
9.1 76
43
13 11
3
79
78
4.0
3.0
5.3
0.4
5.4 48 52
1.3
a
2 2
Reaction conditions: reaction temperature ) 373 K, reaction time ) 5 h, 18.7 mmol of cyclohexane, 9.3 mmol of H O (31 wt % in water),
b
1
0 mL of methanol (solvent), 0.1 g of catalyst. Moles of products/moles of Ti in the catalyst. c ONE, cyclohexanone; OL, cyclohexanol; EHK,
cyclohexyl methyl ether, hemiketal, and ketal.

7130 J. Phys. Chem. B, Vol. 102, No. 37, 1998
Tatsumi and Jappar
TABLE 3: Oxidation of Cyclic Alcohols with H
2
O
2
over Ti-Betaa
H
2
O
2
(%)
conversion
(%)
turnover number
(mol/mol-Ti)
cyclic alcohol
cyclohexanol
catalyst
conversion
selectivity
DGC
HTS
DGC
HTS
13.3
13.0
14.6
14.9
39
37
43
42
80
78
80
75
55
55
61
65
cyclooctanol
a
Reaction conditions: Reaction temperature ) 338 K, reaction time ) 3 h, 16.5 mmol of cyclic alcohol, 5 mmol of H
0 mL of acetonitrile (solvent), 0.1 g of catalyst (DGC, Ti-Beta-DGC-7; HTS, Ti-beta-HTS-1).
2 2
O (31 wt % in water),
1
TABLE 4: Oxidation of Unsaturated Alcohols with H
2
O
2
on Ti-Betaa
selectivity (%)
H
2
O
2
(%)
selectivity
conversion
(%)
TON
(mol/mol-Ti)
unsaturated alcohol
-penten-1-ol
catalyst
aldehyde
epoxy-1-aldehyde
epoxy-1-ol
conversion
4
3
DGC
HTS
10.9
10.0
34
29
2.9
5.4
1.3
2.1
96
93
60
58
60
57
-methyl-2-buten-1-ol
DGC
HTS
11.2
9.9
35
30
23
26
17
25
60
49
77
77
56
52
a
2 2
Reaction conditions: reaction temperature 323 K, reaction time ) 3 h, 16.5 mmol of unsaturated alcohol, 5 mmol of H O (31 wt % in water),
1
0 mL of solvent (acetonitrile), 0.1 g of catalyst (DGC, Ti-beta-DGC-5; HTS, Ti-beta-HTS-1).
to that of the unwashed sample, mainly due to the large decrease
in the amount of Ti. The activity of TS-1 for this reaction is
negligible, probably because of cyclohexane is too large to easily
diffuse into the channel of TS-1. When Ti-beta-HTS and H2-
SO4-washed Ti-beta-DGC samples are used as catalysts, the
following acid-catalyzed secondary reactions take place between
the cyclohexanol and cyclohexanone with the solvent MeOH.
In the first case, cyclohexyl methyl ether is formed (a), while
reaction b leads first to the formation of hemiketal followed by
formation of ketal through its reaction with MeOH. In the
reaction conditions, the reactions b are predominant. The extent
of these secondary reactions is small when the unwashed Ti-
beta-DGC is used as the catalyst, because of the resulting
reduced acidity.
the relatively hydrophilic Ti-beta channels decreases with
hydrophobicity of the reactants because of the preferential
adsorption of water, resulting in the decreases in the rate of
oxidation. For relatively hydrophobic zeolites, the hydrophobic
reactants can be competitively adsorbed into the zeolite channels.
However, hydrophilic reactants would have no difficulty in
adsorbing into the zeolite channels in the presence of water
molecules, regardless of hydrophobicity/hydrophilicity of the
zeolite surface.
Another explanation for the higher activity of the Ti-beta-
DGC samples for the cyclohexane oxidation compared to the
Ti-beta-HTS ones may be the small crystal sizes of the former.
However, there is no virtual difference in the activity for
oxidation of alcohols, hydrophilic substrates, between the two
samples prepared by the different methods, indicating the higher
activity of the Ti-beta-DGC samples for the cyclohexane
oxidation is not due to the smaller crystal size but to the higher
hydrophobicity.
Oxidation of Unsaturated Alcohols on Ti-Beta. The effects
of the structure of C5 unsaturated alcohols on the reactivity
toward oxidation are investigated; pertinent data are shown in
Table 4. The experiments are conducted at 323 K in acetonitrile
used as solvent. Similar reactivity is observed between 3-meth-
yl-2-buten-1-ol and 4-penten-1-ol. The presence of the electron-
donating methyl group increased not only electric density of
the CdC double bond of 3-methyl-2-buten-1-ol but also steric
restriction to access to the active site in the Ti-beta, resulting
in the decrease in the selectivity for the epoxidation. The
epoxide/aldehyde ratios in the oxidation of 3-methyl-2-buten-
Oxidation of Cyclic Alcohols on Ti-Beta. We have studied
the oxidation of cyclic alcohols on the Ti-beta catalysts using
hydrogen peroxide as oxidant. The experiments are performed
at 338 K in acetonitrile solvent. Under these reaction conditions,
only the corresponding ketones are formed from cyclic alcohols.
As shown in Table 3, reactivity of cyclooctanol is slightly higher
than that of cyclohexanol. The differences in the conversion
between cyclohexanol and cyclooctanol are similar for the Ti-
beta-DGC and Ti-beta-HTS samples, although the latter has
much larger average crystal size than the former. This indicates
that the conversion is not influenced by diffusional factors due
to the large pore sizes of beta zeolites.
3
1
1
-ol over TS-1 and Ti-beta are 0.23 and 1.4, respectively.
These results are accounted for by assuming that the steric
restriction in the large pore beta zeolite are less than that in
TS-1.
Furthermore, the fact that larger cyclooctanol exhibits higher
reactivity than cyclohexanol suggests that the reactivity of
cyclooctanol is intrinsically higher than that of cyclohexanol.
The Ti-beta-DGC and Ti-beta-HTS samples show similar
activity for oxidation of both alcohols. In contrast, the relatively
hydrophobic Ti-beta-DGC samples exhibit higher activity for
the oxidation of alkanes than Ti-beta-HTS, as described above.
These phenomena may be due to the hydrophobicity/hydrophi-
licity of the substrate. The concentration of the reactants in
The Ti-beta-DGC catalyst shows slightly higher activity and
higher selectivity for epoxidation than the Ti-beta-HTS catalyst.
This might be also related to the higher hydrophobic character
of the Ti-beta-DGC sample than the Ti-beta-HTS sample; more
hydrophobic Ti-beta favors epoxidation against alcohol oxidation
and vice versa. Thus it is possible that the hydrophobicity/
hydrophilicity affect the chemoselectivity as well as the activity
for the liquid-phase oxidation.

Properties of Ti-Beta Zeolites
J. Phys. Chem. B, Vol. 102, No. 37, 1998 7131
4
. Conclusions
(10) Camblor, M. A.; Costantini, M.; Corma, A.; Gilbert, L.; Esteve,
P.; Martinez, A.; Valencia, S. J. Chem. Soc., Chem. Commun. 1996, 1339.
(11) Camblor, M. A.; Corma, A.; Valencia, S. Chem. Commun. 1996,
From the present study, the following conclusions can be
2
365.
(12) Rigutto, M. S.; De Ruiter, R.; Niederer, J. P. M.; Van Bekkum, H.
Stud. Surf. Sci. Catal. 1994, 84, 2245.
drawn: (1) The large-pore beta zeolite containing Ti can be
synthesized in the presence of alkali cations (Na ) by the dry-
gel conversion (DGC) technique using TEAOH as the structure-
directing agent. The presence of Na less than 1.0 mol % does
not affect the incorporation of titanium into the framework of
beta zeolites.
+
(13) Reddy, J. S.; Sayari, A. J. Chem. Soc., Chem. Commun. 1995, 23.
(14) Saxton, R. J.; Crocco, G. L.; Zajacek, J. G.; Wijesekea, K. S. Eur.
Patent 1994, 659 685 A1.
15) van del Waal, J. C.; Lin, P.; Rigutto, M. S.; van Bekkum, H. Stud.
Surf. Sci. Catal. 1997, 105, 1093.
16) van del Waal, J. C.; Tan, K.; van Bekkum, H. Catal. Lett. 1996,
1, 63.
(17) van del Waal, J. C.; Kunkeler, P. J.; Tan, K.; van Bekkum, H. J.
Catal. 1998, 173, 74.
18) van del Waal, J. C.; van Bekkum, H. J. Mol. Catal. 1997, 124,
37.
19) Xu, W.; Dong, J.; Li, J.; Wu, F. J. Chem. Soc., Chem. Commun.
1990, 755.
(20) Kim, M. H.; Li, H. X.; Davis, M. E. Microporous Mater. 1993, 1,
91.
21) Matsukata, M.; Nishiyama, N.; Ueyama, K. Microporous Mater.
993, 1, 219.
22) Hari Prasad Rao, P. R.; Matsukata, M. J. Chem. Soc., Chem.
(
(2) Ti-beta-DGC adsorbs less amount of water than Ti-beta-
(
HTS, indicating the higher hydrophobicity of the former zeolite.
The intensity of the broad infrared band ranging from 3610 to
3
resulting from the defects in the zeolite structure, is lower for
Ti-beta-DGC than for Ti-beta-HTS.
4
-
1
300 cm , associated with hydrogen-bonded hydroxyl groups
(
1
(
(3) In the oxidation of hydrophobic substrate, such as
cyclohexane, the Ti-beta-DGC samples show higher activity than
Ti-beta-HTS, whereas they show similar activity in the oxidation
of alcohols. This is interpreted in tems of high hydrophobicity
of Ti-beta-DGC compared to that of Ti-beta-HTS.
1
(
1
(
Commun. 1996, 1441.
References and Notes
(23) Hari Prasad Rao, P. R.; Ueyama, M.; Matsukata, M. Appl. Catal.
998, 166, 97.
1
(
(
(
(
1) Notari, B. AdV. Catal. 1996, 41, 253.
(24) Tatsumi, T.; Xia, Q.; Nizamidin, J. Chem. Lett. 1997, 677.
2) Sheldon, R. A. J. Mol. Catal. 1980, 7, 107.
(25) Geobaldo, F.; Bordiga, S.; Zecchina, A.; Gianello, E.; Leofanti,
3) Khouw, C. B.; Dartt, C. B.; Davis, M. E. J. Catal. 1994, 149, 195.
4) Camblor, M. A.; Corma, A.; Martinez, A.; Perez-Pariente, J. J.
G.; Petrini, G. Catal. Lett. 1992, 16, 129.
26) Maache, M.; Janin, A.; Lavalley, J. C.; Joly, J. F.; Benazzi, E.
Zeolites 1993, 13, 419.
27) Bourgeat Lami, E.; Fajula, F.; Anglerot D.; Des Courieres, T.
Microporous Mater. 1993, 1, 237.
28) Zecchina, A.; Bordiga, G.; Spoto, G.; Marchese, L.; Petrini, G.;
Leofanti, G.; Padovan, M. J. Phys. Chem. 1992, 96, 4991.
29) Bellussi, G.; Carati, A.; Clerici, M. G.; Esposito, A. Stud. Surf.
(
Chem. Soc., Chem. Commun. 1992, 589.
5) Corma, A.; Camblor, M. A.; Esteve, P.; Martinez, A.; Perez-
Pariente, J. J. Catal. 1994, 145, 151.
6) Corma, A.; Esteve, P.; Martinez, A.; Valencia, S. J. Catal. 1995,
52, 18.
(
(
(
(
1
(7) Camblor, M. A.; Corma, A.; Perez-Pariente, J. Zeolites 1993, 13,
(
8
2.
Sci. Catal. 1991, 63, 421.
(8) Camblor, M. A.; Corma, A.; Martinez, A.; Perez-Pariente, J.; Primo,
J. Stud. Surf. Sci. Catal. 1993, 78, 393.
9) Camblor, M. A.; Costantini, M.; Corma, A.; Gilbert, L.; Esteve,
P.; Martinez, A.; Valencia, S. Appl. Catal. 1995, 133, L185.
(30) Khouw, C. B.; Davis, M. E. J. Catal. 1995, 151, 77.
(31) Tatsumi, T.; Yako, M.; Nakamura, M.; Yuhara, Y.; Tominaga, H.
J. Mol. Catal. 1993, 78, L41.
(