selectivity of the reaction whereas the PEO contributes to the
conversion and thus a careful balance of hydrophobic–
hydrophilic characteristics§ of a silica particle can be uniquely
used to maximize both the activity and selectivity of a supported
catalyst. Although the effect of the polyether composition on
activity and selectivity clearly indicated that the reaction was
occuring in the polyether phase, additional verification showing
that MTO was not dissolved in either of the reactants was
possible by mixing 10% PEO, 10% PPO-SiO2, 30% H2O2 and
MTO for 10 min. After filtration, the filtrate (30% H2O2)
showed no catalytic activity upon addition of cyclohexene.
Similarly, mixing 10% PEO, 10% PPO-SiO2, cyclohexene and
MTO for 10 min, followed by filtration and addition of 30%
H2O2 to the filtrate, showed no cyclohexene conversion.
In order to demonstrate the generality of this reaction, the
oxidation of various other alkenes using the optimal hete-
rogeneous system (MTO on 10% PEO–10% PPO-SiO2) was
compared to a typical homogeneous reaction using MeOH as
solvent, (Table 2). Compared to the homogeneous standard,
where quantitative conversions were the norm, the polyether
tethered silica support showed somewhat lower activity even
with more oxidant (three-fold excess) but very considerable
selectivity to the epoxide, even in the case of indene whose
epoxide is believed unstable under acidic conditions.6 Oxida-
tion of styrene also yielded benzaldehyde, the product of
carbon–carbon bond cleavage. The homogeneous reaction
system gave almost exclusively the ring opened products which
were identified by GC–MS, except in the case of norbornene
where epoxide was also observed. Finally, initial experiments
were carried out to test the possibility of catalyst recycling. At
the end of each reaction period the catalytic assembly was
filtered under vacuum and washed with 1 ml of cold Et2O,
followed by addition of more substrate and oxidant. For four
cycles the catalytic activity (85 ± 7 mol% conversion) and
selectivity (89 ± 6 mol% epoxide) remained high, although
there were rather considerable fluctuations. From the fifth cycle
onward, both conversion and selectivity dropped considerably.
Analysis by GC–MS of the reaction mixture after each cycle
showed some polyether remnants, indicating that deactivation
was by polyether detachment from the silica particle.
MTO supported on silica particles with polyether tethers is an
effective catalytic assembly for the epoxidation of alkenes with
H2O2 under environmentally benign conditions, i.e. no organic
solvent. The ability to design and control the hydrophobic–
hydrophilic balance of the attached polyether phase holds much
promise for the development of new heterogeneous catalytic
systems, especially where reactants are immiscible.
Table 2 Comparative oxidation of alkenes catalysed by methyltrioxo-
rhenium supported on 10% PEO–10% PPO-SiO2 vs. a homogeneous
reactiona
This research was supported by grant No. 95-00076 from the
United States-Israel Binational Science Foundation (BSF),
Jerusalem, Israel.
Conversion Selectivity Other products
Substrate
(mol%)b
(mol%)c
(mol%)
OH
OH
Footnotes and References
99.9 (100)
86.4 (0)
OH
OMe
* E-mail: ronny@vms.huji.ac.il
13.6 (7.9)
0 (92.1)
† (MeO)3SiC6H4CH2Cl (50 mmol) was reacted with poly(ethylene glycol)
monomethyl ether (PEGME, MW = 350) (50 mmol) or poly(propylene
glycol) monomethyl ether (PPGME, MW = 200) (50 mmol) in acetone
(200 ml) in the presence of solid K2CO3 (250 mmol) at reflux for 18 h to
form (MeO)3SiC6H4CH2O(CH2CHRO)Me 1a (R
(R = Me).
‡ A mixture of 1 and Si(OEt)4 (total 10 mmol), water (100 mmol) and
dibutyltin dilaurate (0.1 mmol) (polymerization catalyst) in acetone (40 ml)
were heated at 60 °C for 3 h. The solution was kept in a beaker and the
volatile organics were left to slowly evaporate over a period of two days at
ambient temperature. The functionalized silica was then carefully washed
twice with water and EtOH and dried in vacuo. Polymerization was
quantitative ( > 99%) as determined by microanalysis. The polymerization
technique used is generally assumed to lead to a random (mixed) polymer
(ref. 7). A detailed kinetic investigation of the rates of hydrolysis and self-
and cross-condensation of 1 and Si(OEt)4 would be required to verify this
assumption and to exclude the formation of PEO and/or PPO enriched
domains.
OH
71.0 (28.0)
94.3 (0)
94.2 (100)
88.5 (100)
=
H) and 1b
OH
24.2 (64.4)
OH(Me)
OMe(H)
5.7 (100)
OMe
CO2Me
CHO
33.8 (0)
83.8 (100)
66.2 (0)
0 (100)
62.7 (100) 100 (62.9)
OH(Me)
OMe(H)
§ A wetting angle of 55° was measured for a 10% PEO–10% PPO-SiO2
film. This wetting angle is indeed intermediate between hydrophilic
(15–25°) and hydrophobic (80–100°).
0 (37.1)
OH(Me)
OMe(H)
1 F. R. Hartley, Supported Metal Complexes, Reidal, Dordrecht, 1985;
R. A. Sheldon, Curr. Opin. Solid State Mater. Sci., 1996, 1, 101.
2 P. E. Rony, J. Mol. Catal., 1975, 1, 13; M. E. Davis, Chemtech, 1992,
498; K. T. Wan and M. E. Davis, Nature, 1994, 370, 449.
3 R. Neumann and M. Cohen, Angew. Chem., Int. Ed. Engl., 1997, in the
press.
79.4 (0)
93.2 (0)
80.2 (100)
72.8 (100)
20.6 (100)
OH(Me)
4 W. A. Herrmann, R. W. Fisher, M. U. Rauch and W. Schere, J. Mol.
Catal., 1994, 86, 243; A. M. Al-Ajlouni and J. H. Espenson, J. Am. Chem.
Soc., 1995, 117, 9243.
OMe(H)
6.8 (100)
5 W. A. Herrmann, R. W. Fisher, W. Schere and M. U. Rauch, Angew.
Chem. Int. Ed. Engl., 1993, 32, 1157.
6 F. Fringuelli, F. Pizzo, R. Germani and G. Savelli, Org. Prep. Proced.
Int., 1989, 21, 757.
a
Reaction conditions (heterogeneous): alkene (1 mmol), 30% H2O2 (3
mmol), MTO (0.02 mmol) on 10% PEO–10% PPO-SiO2 (100 mg);
(homogeneous): alkene (1 mmol), 30% H2O2 (2 mmol), MTO (0.02 mmol),
MeOH (1 ml); 25 °C, 3 h. Values given in parentheses are for the
homogeneous reaction. Alkene conversion. Selectivity to the epoxide
product.
7 C. J. Brinker, Sol-Gel Science, Academic Press, 1990.
b
c
Received in Cambridge, UK, 26th June 1997; 7/04496H
1916
Chem. Commun., 1997