Generation and reactivity of singlet oxygen within zeolites: Remarkable control of hydroperoxidation of alkenes

Article abstract of DOI:10.1039/a700977a

A highly selective oxidation of alkenes included within a Y zeolite is achieved by generating singlet oxygen via excitation of a monomeric dye exchanged within a zeolite; conformational control brought forth by the medium is suggested to be responsible fo

Full text of DOI:10.1039/a700977a

View Article Online / Journal Homepage / Table of Contents for this issue
Generation and reactivity of singlet oxygen within zeolites: remarkable control
of hydroperoxidation of alkenes
Rebecca J. Robbins and V. Ramamurthy*
Department of Chemistry, Tulane University, New Orleans, LA 70118, USA
A highly selective oxidation of alkenes included within a Y
zeolite is achieved by generating singlet oxygen via excitation
of a monomeric dye exchanged within a zeolite; conforma-
tional control brought forth by the medium is suggested to be
responsible for the observed remarkable selectivity.
‘dry’ Na Y/thionin/oxygen generated an emission at 1268 nm
corresponding to 1O2. Irradiation of the dye-exchanged zeolite
in the presence of 2,3-dimethylbut-2-ene, a standard alkene
known to react with singlet oxygen, gave the same hydro-
peroxide as in the isotropic solution medium [Scheme 1, eqn.
(1)].7 This demonstrated that not only was 1O2 generated in the
zeolites, but that it was reactive as well.
Having established that generation of singlet oxygen within a
zeolite is feasible, we tested our conformational control
proposition with the alkenes 1–3a–c. Irradiation took place in a
Kimax tube containing a zeolite and alkene as a slurry in
hexane. Product hydroperoxide(s) were extracted from the
zeolite with tetrahydrofuran and were analysed by GC and
NMR after reduction with PPh3. The only products obtained are
those reported. The isolated yield ranged between 65 and 75%.
Products were compared and found consistent with literature
reports.8 Table 1 compares the ratio of secondary (4) to tertiary
One well known reaction of singlet oxygen is the hydroperoxi-
dation of alkenes containing allylic hydrogens, often referred to
as the ‘ene’ reaction or Schenk reaction.1 Reaction with
substrates containing several distinct allylic hydrogens pro-
duces several distinct products. Although the peroxidation via
singlet oxygen has found much synthetic application, regio- and
stereo-control are still difficult issues.2 We show here that the
interior of a zeolite offers one possible solution to this important
problem. When the oxidation of alkenes 1–3 is conducted
within the supercages of faujasite Y type zeolite a remarkable
selectivity of the hydroperoxidation is observed.
1O2
a R = H
b R = Me
c R = OMe
R
(1)
NaY / thionin
hexane
OOH
1
2
3
O2
R
R
R
Although the mechanistic details of the ene reaction are still
being debated, certain trends have been identified.3 Experi-
ments by several groups have brought out the importance of the
geometrical arrangement of the allylic hydrogens with respect
to the attacking singlet oxygen during the hydroperoxide
formation.4 This suggested to us that if one succeeds in
controlling the conformation of the alkene and thus the
accessibility of the allylic hydrogen(s) to the reagent, singlet
oxygen, one might be able to achieve selectivity in the oxidation
process. We have already established that one can use zeolite as
a medium to achieve conformational control on ketones that
undergo Norrish type reactions.5 With this in mind, we set out
to use a similar strategy to demonstrate that zeolite can play a
vital role in oxidation processes initiated by singlet oxygen.
Indeed, this strategy has yielded remarkable results in terms of
obtaining predominantly a single hydroperoxide from the
alkenes 1–3 within the supramolecular assembly of alkene,
thiazine dye, oxygen and zeolite.
hν
*
(2)
+
NaY / thionin
hexane
O2
OOH
OOH
4
5
Scheme 1
Table 1 A comparison of the ratio of secondary to tertiary hydroperoxides
in various media
Secondary
hydroperoxide
4
Tertiary
hydroperoxide
Alkene
Conditions
5
1
Thionin/MeCN
Rose Bengal/MeCN
NaY/thionin
51
50
72
49
50
28
2
Thionin/MeCN
NaY/thionin
40
60
—
100
In our system, generation of singlet oxygen entails irradiation
of a thiazine dye located within the supercages of the zeolite
lattice. These dyes (thionin, methylene blue and methylene
green) are placed within the zeolite via a cation-exchange
process.6 The exchange involves dropping the zeolite (e.g. 600
mg unactivated NaY) in a prestirred aqueous solution of thionin
(2 mg in 250 ml of water). Continuous stirring for 3 h followed
by filtration and aqueous washes (or Soxhlet extraction) yields
a pink zeolite. The presence of water in the zeolite has a
dramatic effect on the state of aggregation of the organic dye
within the zeolite cage. With coadsorbed water or in ‘wet’
zeolite, thionin exists as a dimer, whereas in ‘dry’ zeolite the
organic dye exists as a monomer and as a result the zeolite turns
blue.6 It is only the monomeric form of the dye which functions
as a sensitizer and generates singlet oxygen. Initial experi-
mentation showed that in the absence of alkene, irradiation of
3a
Thionin/MeCN
Rose Bengal/MeCN
LiY/thionin
NaY/thionin
RbY/thionin
50
54
100
85
80
66
50
46
—
15
20
34
CsY/thionin
3b
3c
Rose Bengal/MeCN
LiY/thionin
NaY/thionin
RbY/thionin
CsY/thionin
56
100
100
94
44
—
—
6
92
8
Rose Bengal/MeCN
LiY/thionin
NaY/thionin
RbY/thionin
CsY/thionin
52
100
100
100
100
48
—
—
—
—
Chem. Commun., 1997
1071

View Article Online
R
Top
Left
Ha
H
Hc
R
Hb
Hc
Hb
H
Ha
H
H
Right
H
H
H
H
H
H
Fig. 2
Top right - - - - - Ha or Hb
Top left - - - - - Hc
consistent with this speculation. A comparison of the product
distribution from 1 and 2 (and 3) makes it clear that in order to
achieve selectivity it is necessary to replace at least one
hydrogen of the methyl group (in 1) by a larger group such as
methyl or phenyl. We recognize that the above model is only a
working hypothesis and further experiments are needed to
understand this unusual selectivity. The use of zeolite as a
medium to achieve selective oxidation has been established and
we are in the process of exploring this strategy with other
alkenes. The ability to generate singlet oxygen and to achieve
regioselective hydroperoxidation of alkenes within a zeolite has
prompted us to consider stereoselective hydroperoxidation of
alkenes with zeolite as the medium.10
Fig. 1
(5) hydroperoxides obtained in an acetonitrile solution to that
obtained in the zeolite oxidation reaction. For additional
confirmation, the tertiary alcohol product of 3a was independ-
ently synthesized as an authentic sample. Retention times and
NMR of the authentic samples are consistent with the observed
products. Also, the hydroperoxides generated from Rose Bengal
sensitized oxidation were stirred overnight in a slurry of hexane
and NaY/thionin. Both products were recovered in the same
ratio as prior to zeolite treatment. These control experiments
confirmed that the observed unprecedented selectivity is not an
experimental artifact.
Formation of both hydroperoxides 4 and 5 in solution has
been rationalized on the basis that singlet oxygen attacks the
alkene from the top-right side (Fig. 1).4 In such an approach, the
transition state is stabilized by secondary interactions between
the oxygen and the allylic hydrogens which are situated parallel
to the p–p orbitals. As per this accepted model the methyl group
on the top-left side (Fig. 1) does not participate in the oxidation
process. Results within the zeolites clearly suggest that the
methylene hydrogens Ha of 2 and 3 (Fig. 1) are not abstracted by
the singlet oxygen. While the lack of formation of 5 within
zeolites is an indication that the methylene hydrogens are
excluded from the reaction, selective formation of 4 does not
indicate which of the two (or both) methyl groups participates in
the oxidation process. Without further experiments that would
distinguish between the gem dimethyl groups, we can only
provide a tentative model for the observed selectivity.
We suggest that the R group in the alkene (Fig. 1) plays a
crucial role in the type of product(s) formed. While in solution,
the most favoured conformation places both the methyl and
methylene hydrogens in an appropriate geometry for abstraction
(Fig. 1), it is quite likely that such a conformation may not be
favoured within a zeolite. In a supramolecular assembly one
will have to consider the interactions that arise between the
adsorbent/guest and the environment. We speculate that within
a zeolite, the alkene will be adsorbed to the surface via cation–p
interactions.9 A rotation of the C3–C4 bond might result under
such conditions to relieve the steric strain that develops between
the bulky R group and the surface. Such a rotation will place the
methylene hydrogens away from the incoming singlet oxygen
(Fig. 2) and therefore no tertiary hydroperoxide would be
formed. The extent of steric repulsion between the surface and
the R group may depend on the distance between the group and
the surface. Larger cations such as Cs ion may place the alkene
slightly further from the surface and thus reduce the steric strain
between the surface and the R group. Formation of small
amounts of tertiary hydroperoxide 5 in the case of 3a and 3b is
The authors thank ACS-PRF and the Division of Chemical
Sciences, Office of Basic Energy Sciences, Office of Energy
Research, US Department of Energy for support of this
program.
Footnote
* E-mail: Murthy@mailhost.tcs.tulane.edu
References
1 Singlet Oxygen, ed. H. H. Wasserman and R. W. Murray, Academic
Press, New York, 1979; Singlet Oxygen, ed. A. A. Frimer, CRC Press,
Boca Raton, FL, 1985, vol. 1–4.
2 M. Prein and W. Adam, Angew. Chem., 1996, 35, 477.
3 A. A. Frimer and L. M. Stephenson, in Singlet Oxygen, ed. A. A. Frimer,
CRC Press, Boca Raton, FL, 1985, pp. 67–92; K. Gollnick and
H. J. Kuhn, in Singlet Oxygen, ed. H. H. Wasserman and R. W. Murray,
Academic Press, New York, 1979, pp. 287–429.
4 L. M. Stephenson, M. J. Grdina and M. Orfanopoulos, Acc. Chem. Res.,
1980, 13, 419; J. R. Hurst and G. B. Schuster, J. Am. Chem. Soc., 1982,
104, 6854; A. A. Gorman, H. Hartman and H. Paur, J. Am. Chem. Soc.,
1982, 104, 7098.
5 D. R. Corbin, D. F. Eaton and V. Ramamurthy, J. Am. Chem. Soc., 1988,
110, 4848; V. Ramamurthy, D. R. Corbin and L. J. Johnston, J. Am.
Chem. Soc., 1992, 114, 3870.
6 V. Ramamurthy, D. F. Eaton and D. R. Sanderson, J. Am. Chem. Soc.,
1993, 115, 10438.
7 C. S. Foote and G. Uhde, in Organic Photochemical Synthesis, ed. R.
Srinivasan, Wiley-Interscience, New York, 1971, vol. 1, pp. 60–61.
8 T. T. Fujimoto, Dissertation, University of California, Los Angeles,
1972, Univ. Microfilm, 72-18125; Diss. Abstr. B., 1972, 32, 6910.
9 M. A. Hepp, V. Ramamurthy, C. R. Corbin and C. Dybowski, J. Phys.
Chem., 1992, 96, 2629; R. H. Staley and J. L. Beauchamp, J. Am. Chem.
Soc., 1975, 97, 5920.
10 M. Leibovitch, G. Olovsson, G. Sundarababu, V. Ramamurthy,
J. R. Scheffer and J. Trotter, J. Am. Chem. Soc., 1996, 118, 1219;
G. Sundarababu, M. Leibovitch, D. R. Corbin, J. R. Scheffer and
V. Ramamurthy, Chem. Commun., 1996, 2049.
Received in Columbia, MO, USA, 11th February 1997; Com.
7/00977A
1072
Chem. Commun., 1997
Typeset and printed by Black Bear Press Limited, Cambridge, England