An additional experiment was carried out with dry TBHP and
dry catalyst Ti-MCM-41. After 30 min the conversion was 91%
and the selectivity was 100%. These results are the best
conversion and selectivity to epoxide ever reported using Ti-
based catalysts for epoxidation processes.
We conclude that the presence of water is responsible for the
ring opening of the epoxide. However, its influence goes
beyond the selectivity effect, since the diols resulting from the
ring opening of the epoxide strongly decrease the catalytic
conversion. We have seen this effect by performing the
epoxidation reaction using dried TBHP, but adding 3.07 mmol
of cyclohexanediol which corresponds to a typical amount of
the diol formed during the catalytic experiments when non-
dried TBHP is used. Under these conditions, the conversion is
significantly reduced to 13% after 30 min of reaction. This
result suggests that the increase in activity observed when using
either silylated Ti-MCM-41 or non silylated Ti-MCM-41 with
dry reactants is due to the significant decrease in the formation
of diols which act as catalyst poisons for the Ti sites.
In conclusion we have presented two different strategies that
can be used to obtain remarkably active and selective epoxida-
tion catalysts based on Ti-MCM-41. The first strategy relies on
the use of highly silylated samples and greater than 40%
silylation coverage is required. In this case water can be present
in the reaction media up to levels of 3 wt.%. On the other hand,
one can use non-silylated catalysts, but in this case water must
be removed from the reaction media.
It is also concluded in this work that the increase in activity
observed with the silylated hydrophobic catalyst or with the
absence of water in the reaction medium, is probably not due to
a change of the intrinsic activity of the Ti catalytic sites but
rather to a decrease in catalyst deactivation. An increase in
catalyst stability is obtained by reducing the formation of diols
that are produced by ring opening of the epoxide in the above
conditions.
Financial support by the Spanish MAT97-1016-CO2-01 and
MAT97-1207-CO3-01 is gratefully acknowledged. J. L. J. and
M. D. thank the M.E.C. and M.E.A., respectively, for
supporting their doctoral fellowships. M. T. N. thanks the CSIC
for the postdoctoral grant. B. M. C. and L. T. N. acknowledge
UOP for financial support.
Fig. 1 Catalytic activity in the epoxidation of cyclohexene with TBHP of
silylated Ti-MCM-41 at different degrees of surface coverage; (a) catalytic
conversion after 30 min of reaction, (b) selectivity to epoxide and efficiency
of TBHP at 85% cyclohexene conversion
Nevertheless, it was surprising to find that every hydrophobic
Ti-MCM-41 catalysts were required in order to have highly
active and selective epoxidation catalysts even when only
organic reactants were used. This motivated us to analyze the
amount of water contained in the reactants, and it was found by
1
means of H NMR that the TBHP contained 8 wt.% water.
Then, we carried out the epoxidation of cyclohexene with
1
TBHP which was dried using 4 Å molecular sieves. With H
NMR we determined that no decomposition of TBHP occurred
and the remaining amount of water was below the detection
limit of H NMR. Under these conditions and using the non-
silylated catalyst Ti-MCM-41, which contains 13.5 wt.% water
adsorbed on the catalyst, the total conversion and selectivity to
the epoxide obtained after 30 min of reaction were 85 and 97%,
respectively. These results are very similar to those obtained
with the completely silylated sample Ti-MCM-41S4 when
using the as-received commercial TBHP.
Notes and References
1
1 M. Taramasso, G. Perego and B. Notari, US Pat., 4 410 501, 1983.
2 M. A. Camblor, A. Corma, A. Mart ´ı nez and J. P e´ rez-Pariente, J. Chem.
Soc., Chem. Commun., 1992, 589.
3
A. Corma, M. T. Navarro and J. P e´ rez-Pariente, J. Chem. Soc., Chem.
Commun., 1994, 197.
4
5
6
T. J. Pinnavaia and W. Zhang, Stud. Surf. Sci. Catal., 1998, 117, 23.
S. Gontier and A. Tuel, J. Catal., 1995, 157, 124.
A. Corma, M. T. Navarro, J. P e´ rez-Pariente and F. S a´ nchez, Stud. Surf.
Sci. Catal., 1995, 84, 69.
7
8
9
T. Blasco, M. A. Camblor, A. Corma, P. Esteve, J. M. Guil, A. Mart ´ı nez,
J. A. Perdig o´ n and S. Valencia, J. Phys. Chem. B, 1998, 102, 75.
M. A. Camblor, A. Corma, P. Esteve, A. Mart ´ı nez and S. Valencia,
Chem. Commun., 1997, 795.
H. P. Wulff, US Pat., 3 923 843, 1975.
1
1
1
1
1
1
0 K. A. Koyano, T. Tastumi, Y. Tanaka and S. Nakata, J. Phys. Chem.,
997, 101, 9436.
1 S. L. Burkett, S. D. Simis and S. Mann, Chem. Commun., 1996,
367.
2 C. E. Fowler, S. L. Burkett and S. Mann, Chem. Commun., 1997,
769.
3 A. Corma, J. L. Jord a´ , M. T. Navarro and F. Rey, Chem. Commun.,
998, 1899.
4 T. Tatsumi, K. A. Koyano and N. Igarashi, Chem. Commun., 1998,
25.
5 T. Blasco, A. Corma, M. T. Navarro and J. P e´ rez-Pariente, J. Catal.,
995, 156, 65.
1
1
1
1
3
1
1
1
6 A. Corma, Chem. Rev., 1997, 97, 2373.
7 Insight II Molecular User Guide, San Diego, Biosym/MSI, 1995.
Fig. 2 Variation of the hydrophobicity, calculated as the weight loss at
150 °C by thermogravimetry, with the degree of silylation of Ti-MCM-41
catalysts
Received in Bath, UK, 20th August 1998; 8/06702C
2212
Chem. Commun., 1998