Fig. 2 Hydrogen peroxide concentration profile upon irradiation (visible
light, l > 350 nm) of TP-NaY (100 mg) in water.
Scheme 3
to blocking of the pores by oligomers or bleaching of Ti atoms
out of the Ti-b framework.
On the other hand, according to Scheme 3, cyclohex-2-enol
and cyclohexane-1,3-diol would be concurrently formed in the
oxidation step while cyclohex-2-enone would arise from the
subsequent overoxidation of cyclohex-2-enol.
Even though the TP-NaY/Ti-zeolite mixtures act as photo-
catalysts, their activity and selectivity are far from optimal. In
order to increase the efficiency of the overall photocatalytic
reaction, a fully-integrated approach was explored. It consists of
the preparation of a single bifunctional photocatalyst that
contains the two centers necessary for the process in the closest
possible proximity. For this purpose, we have carried out the
ship-in-a-bottle synthesis of TP+ in a Ti-Beta zeolite following
the procedure reported for zeolite Y.9 As can be seen in Table 1,
the efficiency of the overall dihydroxylation process is almost
doubled using this novel TP, Ti-b photocatalyst compared to the
mechanical mixture of the two separate components. This
higher activity of TP,Ti-b for the production of cyclohexane-
1,2-diol can be understood as a consequence of the spatial
arrangement of the sites that reduces the chances for undesirable
decomposition of hydroperoxy cyclohexene.
To establish a valid comparison of the efficiency of the TP,
Ti-b photocatalysis, the autooxidation of cyclohexene (2 ml)
initiated by azoisobutyronitrile (30 mg) was also carried in
MeCN (2 ml) in the presence of Ti-b (200 mg) bubling air
through the solution. After 5 h reaction, only 4.5 mg of
cyclohexane-1,2-diol was formed. This value is smaller than
those achieved with the TP,Ti-b photocatalysis (see Table 1).
Even though further improvements are still needed, herein we
have shown that TP,Ti-b acts as a new positive photocatalytic
system that by using photons promotes the dihydroxylation of
cyclohexene using water and oxygen as reagents. The overall
process occurs within the pores of zeolites, whose rigid
structure embeds and protects the active centers, cooperating in
the process in a way reminiscent of the quaternary structure of
enzymes around a prostetic group.
Scheme 2
Large pore Ti molecular sieves are efficient catalysts for the
selective epoxidation of alkenes by organic hydroperoxides.7
We have exploited the ability of TP-NaY to generate a
significant concentration of hydroperoxides from light and
H2O–O2 and coupled it with the subsequent epoxidation
converting the whole process in a one-pot reaction.
Two different strategies have been explored. The simplest
one consists in using a mechanical mixture of two molecular
sieves, one containing TP+ as photocatalyst and the second one
having Ti atoms as epoxidation sites. Table 1 summarizes the
results that have been achieved using a system containing a
mixture of TP-NaY and Ti-b or Ti-MCM-41.7–10 The major
difference between the latter two epoxidation catalysts is the
geometry and dimensions of the internal voids, both having in
common their relative hydrophocity that favors the preferential
adsorption of organic compounds.7 The structure of b zeolite
defines a tridirectional network of oval cages (11 Å major axis)
while MCM-41 is formed by an array of hexagonal channels
(30 Å diameter).
In agreement with the chemical literature,7 Al-free Ti-MCM-
41 was found more active than Al-free Ti-b. The higher activity
of Ti-MCM-41 is easily rationalized based on the larger
dimensions of its pores that suit better to the size of the reagents
involved in the process. When the TP-NaY/Tib mixture was
reused in a second and third experiment with fresh feed a
notable loss of activity was observed (see Table 1, footnotes c
and d). However, this decay in the efficiency of the photo-
catalytic mixture is not caused by the bleaching of TP+ since no
changes were observed in the characteristic UV/vis absorption
of the solid. Rather, it is more likely that the deactivation is due
Notes and references
Table 1 Results after 4 h irradiation (125 W Hg lamp through Pyrex) at
room temperature of cyclohexene (2 ml) in a mixture of MeCN (2 ml)–H2O
(1 ml) in the presence of the corresponding photocatalyst (500 mg, 4%
loading of TP+)
1 O. Legrin, E. Oliveros and A. M. Braun, Chem. Rev., 1993, 93, 671.
2 L. Cermenati, C. Richter and A. Albini, Chem. Commun., 1998, 805.
3 Introduction of Zeolite Science and Practice, ed. H. van Bekkum, E. M.
Flanigen and J. C. Jansen, Elsevier, Amsterdam, 1991.
4 M. A. Miranda and H. Garc´ıa, Chem. Rev., 1994, 94, 1063.
5 A. Sanjua´n, M. Alvaro, G. Aguirre, H. Garc´ıa and J. C. Scaiano, J. Am.
Chem. Soc., 1998, 120, 7351.
Products/mg
cyclohexane-
1,2-diol
6 M. Fattahi, C. Houeelevin, C. Ferradini and P. Jacquier, Radiat. Phys.
Chem., 1992, 40, 167.
Photocatalysta
cyclohex-2-enol cyclohex-2-enone
7 T. Blasco, M. A. Camblor, A. Corma, P. Esteve, J. M. Guil, A. Mart´ınez,
J. A. Perdigo´n-Melo´n and S. Valencia, J. Phys. Chem. B, 1998, 102,
75.
8 TP-NaY was obtained by submitting TP-HY (ref. 9) to exhaustive Na+
exchange with aqueous Na2CO3. Ti-b was prepared by the novel OH2
free procedure (ref. 7). Ti-MCM-41 was prepared as reported (ref.
10).
9 A. Corma, V. Forne´s, H. Garc´ıa, M. A. Miranda, J. Primo and M. J.
Sabater, J. Am. Chem. Soc., 1994, 116, 2276.
10 A. Corma, M. T. Navarro, J. Pe´rez-Pariente, and F. Sa´nchez, Stud. Surf.
Sci. Catal., 1994, 84, 69.
TP-NaYb
1.5
10.9
4.5
6.8
23.7
13.9
13.6
13.9
16.8
13.8
7.4
TP-NaY/Ti-b
c
TP-NaY/Ti-b
d
TP-NaY/Ti-b
2.2
TP-NaY/Ti-
MCM-41
TP, Ti-b
a
13.4
20.2
17.6
39.7
18.7
11.5
Blank controls under Ar atmosphere, in the dark or in the absence of
b
TP-Y did not lead to any product. In the absence of Ti-containing co-
catalyst. c Second run reusing TP-NaY/Ti-b with fresh reagents. d Third run
reusing TP-NaY/Ti-b with fresh reagents.
Communication 9/04519H
1642
Chem. Commun., 1999, 1641–1642