A dramatic improvement of epoxide selectivity of [Ti,Al]-beta by ion-exchange with quaternary ammonium salts

Article abstract of DOI:10.1039/b105088p

Ion-exchange of [Ti,Al]-beta with quaternary ammonium acetates greatly enhances the epoxide selectivity in the oxidation of alkenes with hydrogen peroxide; this is due to the selective poisoning of the acid sites without suppressing the oxidation activity of Ti sites.

Full text of DOI:10.1039/b105088p

View Article Online / Journal Homepage / Table of Contents for this issue
A dramatic improvement of epoxide selectivity of [Ti,Al]-beta by
ion-exchange with quaternary ammonium salts
Yasuhide Goa,a Peng Wub and Takashi Tatsumi*b
a Department of Applied Chemistry, Graduate School of Engineering, The University of Tokyo, 7-3-1,
Hongo, Bunkyo-ku, Tokyo, 113-8656, Japan
b Division of Materials Science & Engineering, Graduate School of Engineering, Yokohama National
University, 79-5, Tokiwadai, Hodogaya-ku, Yokohama, 240-8501, Japan. E-mail: ttatsumi@ynu.ac.jp
Received (in Cambridge, UK) 11th June 2001, Accepted 30th July 2001
First published as an Advance Article on the web 29th August 2001
Ion-exchange of [Ti,Al]-beta with quaternary ammonium
acetates greatly enhances the epoxide selectivity in the
oxidation of alkenes with hydrogen peroxide; this is due to
the selective poisoning of the acid sites without suppressing
the oxidation activity of Ti sites.
[Ti,Al]-beta zeolite was hydrothermally synthesized by
modifying the method reported by Camblor et al.2,3 Tetrabutyl
orthotitanate [Ti(OBu)4] was employed as a Ti source instead of
tetraethyl orthotitanate [Ti(OEt)4] and was hydrolyzed in
aqueous H2O2 before adding to the SDA solution to suppress the
formation of extraframework anatase TiO2. The mother gel
composition was SiO2+0.33 TiO2+0.005 Al2O3+0.56
TEAOH+12.5 H2O. The crystallization was carried out by
agitating the gel in a Teflon-lined stainless steel autoclave at
413 K for 4 days. The resultant [Ti,Al]-beta was calcined at 793
K for 10 h in a stream of O2 (30 ml min21 g-solid21) to remove
the SDA. Calcined [Ti,Al]-beta was treated with 1 M aqueous
solutions of various ammonium acetates (NR4OAc, R = Me,
Et, Pr and H) or 1 M aqueous potassium acetate (KOAc)
solution at 353 K for 4 h and then filtered off, washed, dried and
finally calcined of 473 K for 10 h. The liquid phase epoxidation
of alkenes with H2O2 was performed in a flask.
Since the first report of TS-1 by Taramasso et al. in 1983,
titanosilicates have attracted the attention of many researchers.1
Ti-beta zeolite has a three-dimensionally connected pore system
with 12-membered-ring channels and is verified to be more
active towards bulky substances such as branched and cyclic
alkanes and alkenes using aqueous H2O2 or tert-butyl hydro-
peroxide (TBHP) as an oxidant than medium pore TS-1 because
of the reduced diffusion restriction imposed by the large pores.
A small amount of Al (Si/Al < 200), as crystallization-promot-
ing agent, is generally needed in the conventional hydrothermal
synthesis method for Ti-beta using tetraethylammonium hy-
droxide (TEAOH) as a structure-directing agent (SDA).2,3 The
presence of Al leads to the formation of acid sites, which
promotes the successive ring-opening solvolysis of the epoxide
generated in the catalytic reactions to lower its selectivity
extensively.4 Afterwards, several novel synthesis methods for
Al-free or low-Al Ti-beta were also developed, such as the
seeding method,5 cogel method,6 fluoride method7,8 and dry-gel
conversion method;9,10 however none of these were very
efficient for enhancing epoxide selectivity in alcoholic solvents.
This is probably because the very complex, severely intergrown
nature of the beta structure results in high concentration of weak
acid sites such as silanol groups at defect sites. Although ion-
exchange with alkaline or alkaline earth metal cations inhibits
the acidity of titanosilicates, it also suppresses the oxidation
activity of Ti species.11,12 We report here a new method for
dramatically enhancing the yield of epoxide in the alkene
oxidation reactions catalyzed by Ti-beta even in protic alcoholic
solvents without losing its oxidation activity.
Table 1 shows the results of epoxidation of alkenes. Calcined
[Ti,Al]-beta exhibited a low selectivity to epoxide for oxidation
reactions of cyclohexene and hex-2-ene (entries 1 and 2). The
products were mainly diol and monomethyl glycol ether
produced by the acid-catalyzed solvolysis of epoxide, together
with a small amount of allylic oxidation products. The
negligible formation of epoxide strongly indicates that there
were a considerable amount of acid sites in [Ti,Al]-beta, which
contributed to the solvolysis of epoxide.
When the catalyst ion-exchanged with tetramethylammon-
ium acetate was employed (entries 4 and 5), the epoxide
selectivity was greatly enhanced owing to the suppression of
ring-opening reactions, and the selectivity to the allylic
oxidation did not change significantly. The enhancement was
more obvious for hex-2-ene which gave an epoxide/glycol ratio
of 28. Solvent effects were also tested. It is generally known that
aprotic solvents such as acetonitrile enhance the selectivity to
the epoxide. Employment of the weakly basic solvent, acetoni-
Table 1 Catalytic oxidation of cyclohexene and hex-2-enea
Maximum
conv.
(mol%)
Selectivityb (mol%)
H2O2 (mol%)
Entry
Ion-exchange
Substrate
N+/Al
Epox.
Glyc.
Allyl
Conv.
Efficiency
1
2
No
No
No
NMe4OAc
NMe4OAc
NMe4OAc
NEt4OAc
NPr4OAc
NH4OAc
KOAc
Cyclohexene
Hex-2-enec
Cyclohexene
Cyclohexene
Hex-2-enec
Cyclohexene
Cyclohexene
Cyclohexene
Cyclohexene
Cyclohexene
–
–
–
1.8
1.8
1.8
1.8
1.5
1.6
2.2e
76.1
49.5
30.4
75.2
60.7
57.1
47.2
24.5
57.4
18.7
0.0
0.2
97.6
98.0
14.9
43.2
3.3
2.4
0.0
15.8
1.8
4.1
6.1
1.9
2.0
3.0
2.3
83
76
81
95
78
86
69
42
60
48
90
64
41
79
72
69
67
59
97
40
3d
4
69.3
55.1
92.4
88.5
71.6
51.4
0.1
5
6d
7
8
9
5.3
26.5
46.6
96.9
34.2
10
63.5
a Reaction conditions: 333 K, 3 h, 8.25 mmol of alkene, 2.5 mmol of H2O2 (31 wt% in water), 50 mg of catalyst (Si/Ti = 35, Si/Al = 38) and 5 ml of
methanol. b Epox. = cyclohexene oxide or 2,3-epoxyhexane; Glyc. = cyclohexanediol or hexane-2,3-diol and their monomethyl ethers; Allyl = alcohol and
ketone formed by oxidation. c cis/trans Isomer ratio = 41/59. d Acetonitrile was employed as solvent. e K+/Al.
1714
Chem. Commun., 2001, 1714–1715
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b105088p

View Article Online
band disappeared and the silanol bands significantly decreased
in intensity [Fig. 1(b)]. When the ion-exchanged sample was
calcined at 773 K, the intensities of these bands were almost
recovered [Fig. 1(c)]. The pH of aqueous NMe4OAc used for
the ion-exchange was high enough (pH ~ 9) to allow the silanol
groups to be exchanged with NMe4+ cations since the isoelectric
point of SiO2 is 1.0–2.0. These findings indicate that a part of
silanol groups with weaker acidity as well as all of the bridged
hydroxy groups with strong acidity were ion-exchanged with
+
NMe4 .
The pH of the NMe4OAc solution was varied from 8 to 12 to
control the ion-exchange ratio. The oxidation activity gradually
increased with increasing pH. Treatment at high pH resulted in
the formation of mesopores, as revealed by N2 adsorption
measurements, which could promote the diffusion of the
substances. However, the catalyst ion-exchanged at high pH
exhibited low selectivity for epoxide; probably the zeolite
framework dissolves to form silanol groups, which would
catalyze undesirable reactions.
When the ion-exchanged [Ti,Al]-beta catalyst was reused for
the oxidation of cyclohexene after the catalyst was calcined at
473 K for 10 h, the epoxide selectively was retained, although
the oxidation activity gradually decreased to 37% for the third
run probably because of the blocking of the Ti species by
unremovable reaction products such as polymers with high
boiling points. Nevertheless, after re-calcination at 793 K and
repeated ion-exchange of the deactivated catalyst with
NMe4OAc, both the oxidation activity and the epoxide
selectivity were recovered to the initial value, which strongly
indicates that the present method is effective in developing an
active, selective and reusable catalyst.
Fig. 1 FTIR spectra of Ti-beta in the OH region; (a) without ion-exchange
after evacuation at 773 K, (b) ion-exchanged with NMe4OAc and evacuated
at 473 K, (c) ion-exchanged and evacuated at 773 K.
trile, further enhanced the epoxide selectivity (entries 3 and
6).
Use of larger quaternary ammonium cations led to decreased
activity, probably because the bulky cations hinder the reactant
molecules from approaching the Ti sites (entries 4, 7 and 8).
Ammonium acetate treatment led to no enhancement of the
epoxide selectivity (entry 9). FT-IR spectra of the ion-
exchanged [Ti,Al]-beta taken before and after the catalytic runs
+
+
indicated that NH4 was oxidatively degradated while NMe4
Notes and references
remained intact. Ion-exchange with KOAc also increased the
selectivity to the epoxide; however the oxidation activity was
adversely affected (entry 10). This is consistent with the
observations that ion-exchange with sodium carbonate
(Na2CO3) or sodium azide (NaN3) on Al-containing TS-1 gave
rise to a higher epoxide selectivity but a very low activity in
allyl alcohol oxidation,11 that ion-exchange with potassium
carbonate severely lowered the oxidation turnover in un-
saturated alcohol oxidation12 and that TS-1 synthesized in the
presence of large amounts of alkali and alkaline earth ions
showed low oxidation activity.13 Elemental analyses indicated
that ion-exchanged catalyst contained NMe4+ cations in excess
1 M. Taramasso, G. Perego and B. Notari, US Pat., 4,410,501, 1983.
2 M. A. Camblor, A. Corma, A. Martínez and J. Pérez-Pariente, Chem.
Commun., 1992, 589.
3 M. A. Camblor, A. Corma and J. Pérez-Pariente, Zeolites, 1993, 13,
83.
4 A. Corma, P. Esteve, A. Martínez and S. Valencia, J. Catal., 1995, 152,
18.
5 M. A. Camblor, M. Constantini, A. Corma, L. Gilbert, P. Esteve, A.
Martínez and S. Valencia, Chem. Commun., 1996, 1339.
6 M. A. Camblor, M. Constantini, A. Corma, P. Esteve, L. Guilbert, A.
Martínez and S. Valencia, Appl. Catal., 1995, 133, L185.
7 T. Blasco, M. A. Camblor, A. Corma, P. Esteve, A. Martínez, C. Prieto
and S. Valencia, Chem. Commun., 1996, 2367.
8 T. Blasco, M. A. Camblor, A. Corma, P. Esteve, J. M. Guil, A. Martínez,
J. A. Perdigon-Melon and S. Valencia, J. Phys. Chem. B, 1998, 102,
7126
9 N. Jappar, Q.-H. Xia and T. Tatsumi, J. Catal, 1998, 180, 132.
10 T. Tatsumi and N. Jappar, J. Phys. Chem. B, 1998, 102, 7126.
11 G. J. Hutchings, D. F. Lee and A. R. Minihan, Catal. Lett., 1996, 39,
83.
+
of Al, which suggests that the NMe4 blocks not only all the
bridging hydroxy groups but also other less acidic sites.
Calcined [Ti,Al]-beta exhibited two bands in its FTIR spectra
in the OH vibration region [Fig. 1(a)]. A peak at 3609 cm21 is
assigned to bridged hydroxy groups [Si–(OH)–Al] and the other
peak around 3740 cm21 to terminal silanols (SiOH) composed
of several types of silanol groups:14 silanols of amorphous
silica–alumina species (3747 cm21), silanols of amorphous
silica (3745 cm21) and terminal silanols attached to the zeolite
lattice (3736 cm21) After ion-exchange with NMe4OAc
followed by calcination at 473 K, the bridged hydroxy groups
12 T. Tatsumi, K. A. Koyano and Y. Shimizu, Appl. Catal. A, 2000, 200,
125.
13 C. B. Khouw and M. E. Davis, J. Catal., 1995, 151, 71.
14 A. Janin, M. Maache, J. C. Lavalley, J. F. Joly, F. Raatz and N.
Szydlowsky, Zeolites, 1991, 11, 391.
Chem. Commun., 2001, 1714–1715
1715