Site specificity in the photooxidation of some trisubstituted alkenes in thionin-supported zeolite Na-Y. On the role of the alkali metal cation

Article abstract of DOI:10.1021/ol000021y

(Matrix Presented) The less substituted side of some geminal dimethyl trisubstituted alkenes becomes significantly more reactive if photooxygenation takes place in thionin-supported zeolite Na-Y. These results are mainly attributed to a synergistic intera

Full text of DOI:10.1021/ol000021y

ORGANIC
LETTERS
2000
Vol. 2, No. 10
1369-1372
Site Specificity in the Photooxidation of
Some Trisubstituted Alkenes in
Thionin-Supported Zeolite Na−Y. On the
Role of the Alkali Metal Cation
Manolis Stratakis* and George Froudakis
Department of Chemistry, UniVersity of Crete, 71409 Iraklion, Greece
stratakis@chemistry.uch.gr
Received February 2, 2000
ABSTRACT
The less substituted side of some geminal dimethyl trisubstituted alkenes becomes significantly more reactive if photooxygenation takes
place in thionin-supported zeolite Na−Y. These results are mainly attributed to a synergistic interaction between the alkali metal cation, the
alkene, and the oxygen in the transition state of perepoxide formation.
The photooxygenation of alkenes adsorbed on dye-supported
zeolites has recently attracted considerable attention due to
the significant enhancement of product regioselectivity.1-7
Ramamurthy and co-workers have reported a remarkable
change in regioselectivity in the photooxygenation of several
geminal dimethyl trisubstituted alkenes adsorbed on thionin-
supported alkali metal exchanged Y-zeolites compared to
that in solution.2-5 For example, intrazeolite reaction of
2-methyl-2-pentene (R ) methyl in Scheme 1) with singlet
The dramatic change in the regioselectivity had been
attributed originally to two independent models:5 (i) steric
factors inside the cavity place the bulkier alkyl group (R) of
1 in such a conformation that methylene hydrogens are
unreactive and (ii) polarization of the alkene by interaction
of the cation to the olefin double bond in such a way that
the more substituted bears a partial positive charge. Thus,
the electrophilic oxygen attacks the more nucleophilic
disubstituted carbon, and hydrogen abstraction occurs at the
geminal methyls forming the secondary hydroperoxide. More
recently, Clennan and Sram6 proposed a new mechanistic
model based on the regioselectivity results in the photooxy-
genation of some tetrasubstituted alkenes. They argued that
the alkali metal cation forms a complex with the pendant
oxygen in the intermediate perepoxide which leads to greater
Scheme 1
(2) Li, X.; Ramamurthy, V. J. Am. Chem. Soc. 1996, 118, 10666-10667.
(3) Robins, R. J.; Ramamurthy, V. J. Chem. Soc., Chem. Commun. 1997,
1071-1072.
oxygen affords regiospecifically only the secondary allylic
hydroperoxide, whereas when the reaction takes place in
solution, a 60/40 ratio of tertiary/secondary hydroperoxides
is formed.
(4) Tung, C.-H.; Wang, H.; Ying Cheng, Y.-M. J. Am. Chem. Soc. 1998,
120, 5179-5184.
(5) Ramamurthy, V.; Lakshminarasimhan, P. H.; Grey, C. P.; Johnston,
L. J. Chem. Soc., Chem. Commun. 1998, 2411-2418.
(6) Clennan, E. L.; Sram, J. P. Tetrahedron Lett. 1999, 40, 5275-5278.
(7) Sen, S. E.; Smith, S. M.; Sullivan, K. A. Tetrahedron 1999, 55,
12657-12698.
(1) Li, X.; Ramamurthy, V. Tetrahedron Lett. 1996, 37, 5235-5238.
10.1021/ol000021y CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/28/2000

positive charge on the carbon framework, while steric
interactions between the cation and the alkyl substituents
affect the stability of the transition states leading to the
intermediate perepoxides.
To determine the relative reactivity of syn and anti methyl
groups in the geminal dimethyl trisubstituted alkenes 1 and
to shed some light on the reaction mechanism, we performed
the photooxygenation of the specifically labeled alkenes 2
and 3 adsorbed on thionin-supported zeolite Na-Y. Prepara-
tion of 2 and 3 was accomplished in 97% and 96%
geometrical purity, respectively, following the reactions
sequence presented in Scheme 2. The results shown in
analyzed. The mass balances in those experiments were
found to be >85%, and the isolated yields were around 70%.
1
Product distribution was measured by H NMR integration
of the appropriate peaks and by GC analysis of the allylic
alcohols produced after reduction of initially formed allylic
hydroperoxides with excess triphenylphosphine. The percent-
age reactivities of the allylic hydrogens were quite reproduc-
ible (the percentage deviation of three independent measure-
ments was (3%). The dye-supported zeolite was carefully
treated with pyridine9 prior to use to suppress the small extent
of alkenes isomerization (∼2-3%) under the reaction
conditions. It is notable that alkene 3 did not form any
detectable amounts of the tertiary allylic hydroperoxide 3a
as had been reported previously.3 However, a dye-zeolite
sample left in the atmosphere for several days and then used
led to the formation of increased amounts of 3a. This points
out the significance of the moisture on the stereoselectivity
of the ene reaction. In the current experiments we used dye-
zeolite samples that have been vacuum-dried for 3 h at 100
°C prior to use.
Scheme 2
To model the cation-π interaction, we performed theo-
retical calculations on a simplified system. Our initial
theoretical treatment indicates that interaction of the smallest
alkali metal cation (lithium) with the simplest trisubstituted
alkene, trimethylethylene, can form two distinct species in
which the cation lies either on the less or on the more
substituted side of the alkene but not on top of the carbon-
carbon double bond (Figure 1). Calculations were performed
with the GAUSSIAN 94 program package.10 The theoretical
treatment of the structures involved density functional theory
(DFT) with the three-parameter hybrid functional of Becke11
and the use of the Lee-Yang-Parr correlation functional
(B3LYP).12 The level of the calculations used herein was
B3LYP/6-31G*. Many different starting geometries were
used; however, after the optimizations, we found the two
almost isoenergetic local minima presented in Figure 1.
Similar results were found with the MP2 method and the
same basis set. The MP2 method was used recently by Haw
and co-workers to model the adsorption of benzene into
Scheme 3 reveal that the methyl group in the less substituted
side of both alkenes becomes significantly more reactive
within zeolite compared to the reaction in solution (carbon
tetrachloride/tetraphenylporphine).
In a typical experiment 1 g of dye-exchanged zeolite was
added to 15 mL of dry hexane, followed by 10 mg of the
alkene. Thus, the loading levels of alkenes8 were ap-
proximately n ) 0.1-0.15, to ensure that no “free” uncom-
plexed olefin would be oxidized. The slurry was irradiated
with visible light (>450 nm). The loading level of thionin
was maintained at lower than 1 dye molecule per 100
supercages as described in the procedure of Ramamurthy.2
The reaction was almost complete within 5 min. Subse-
quently, 10 mL of moisted tetrahydrofuran was added and
stirring was continued for 2 h. After filtration the solvent
was evaporated under vacuum and the hydroperoxides were
zeolite LiZSM-5.13
Surprisingly, even for the symmetrical
tetramethylethylene, the structure in which Li+ lies toward
Scheme 3. Photooxidation of Alkenes 2 and 3 in Thionin-Supported Zeolite Na-Y or in Solution
1370
Org. Lett., Vol. 2, No. 10, 2000

Figure 1. Calculated structures of Li+ binding to trimethylethylene (lithium cation lies above the olefinic plane at the approximate distance
of 2 Å).
the cis methyl groups was calculated to be more stable by 1
kcal/mol than that in which the cation sits on top in the
middle of the C-C double bond.
On the basis of the experimental results and the theoretical
calculations, we propose in Scheme 4 a model that can
requirements of a perepoxide as is evident from the lack of
a primary isotope effect in the photooxygenation of cis-
tetramethylethylene-d6.6 For trisubstituted alkene such as 1,
the alkali metal cation binds to the double bond, forming
two species either to the more or to the less substituted side
of the alkene. The cation-double bond interaction is strong,14
and it is likely to play a significant role in the stability of
the transition states of the hydroperoxidation reaction.
Recently, it was calculated that in the gas phase the Na+-
ethylene interaction is exothermic by 12.7 kcal/mol, and the
cation sits in the middle of the double bond.15
Scheme 4a
The incoming oxygen can form the four possible inter-
mediates in which the singlet oxygen and cation can be
placed in the same or at different sides of the double bond.
Considering the electrostatic interaction of the negatively
charged pendant oxygen of the perepoxide to the cation,6
we assume that the intermediates in which the oxygen is
placed on the same side of the alkene as the cation are the
most favorable. In the transition state for the formation of
perepoxide PEI, oxygen interacts simultaneously with one
allylic hydrogen (D in the present case) and the positively
charged cation. Therefore, it gains significant stabilization.
It is likely that in the transition state leading to PEI, the cation
has moved from its original position in the complex with
the olefin, resulting in a more efficient interaction with the
negatively charged oxygen atom. On the other hand, when
the cation binds on the more substituted side of the alkene,
steric factors place the alkyl group in a conformation away
from the cation as has been proposed by Ramamurthy.2 Thus,
in the transition state leading to PEII, oxygen interacts also
with the cation and one allylic hydrogen (from the methyl
group), because methylene hydrogens have an unfavorable
conformation. PEI leads to D abstraction from the anti allylic
methyl group, while PEII leads to the ene product formed
by hydrogen abstraction only from the syn allylic methyl
group. On the other hand, in the absence of the alkali metal
cation, a transition state leading to the formation of the
perepoxide on the more substituted side of the alkene is more
a Cation is omitted in the structure of PEII.
explain the regioselectivity in the intrazeolite photooxygen-
ation of the geminal dimethyl trisubstituted alkenes. We view
the reaction as proceeding through the intermediacy of an
irreversible perepoxide or an exciplex with the structural
(8) For the calculation of the loading levels, see: Zhou, W.; Clennan,
E. L. J. Am. Chem. Soc. 1999, 121, 2915-2916.
(9) Jayathirma Rao, V.; Perlstein, D. L.; Robbins, R. J.; Lakshminarasim-
han, P. H.; Kao, H.-M.; Grey, C. P.; Ramamurthy, V. J. Chem. Soc., Chem.
Commun. 1998, 269-270. We used limonene rearrangements as a standard
(see ref 5).
(10) Gaussian 94, ReVision D.4; Frisch, M. J.; Trucks, G. W.; Schlegel,
H. B.; Gill, P. M. W.; Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.;
Keith, T.; Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.;
Al-Laham, M. A.; Zakrewski , V. G.; Ortiz, J. V.; Foresman, J. B.;
Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng,
C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J. L.; Replogle, R.;
Gomperts, R.; Martin, R. L.; Fox, D. L.; Binkley, J. S.; Defrees, D. J.;
Baker, J.; Stewart, J. P.; Head-Gordon, M.; Gonzalez, C.; Pople, J. A.
Gaussian, Inc.: Pittsburgh, PA, 1995.
(11) Becke, D. J. Chem. Phys. 1993, 98, 5648.
(14) For a review on cation-π interactions, see: Ma, J. C.; Dougherty,
D. A. Chem. ReV. 1997, 97, 1303-1325.
(15) Hoyau, S.; Norrman, K.; McMahon, T. B.; Ohanessian, G. J. Am.
Chem. Soc. 1999, 121, 8864-8875.
(12) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(13) Barich, D. H.; Xu, T.; Zhang, J.; Haw, J. F. Angew. Chem, Int. Ed.
1998, 37, 2530-2531.
Org. Lett., Vol. 2, No. 10, 2000
1371

favorable, because singlet oxygen has the ability to interact
simultaneously with two allylic hydrogens and leads to the
expected “cis effect” selectivity.16 Within the alkali metal
exchanged zeolite, however, the transition states leading to
PEI and PEII appear to have similar stabilizing interactions;
thus they are close in energy.
the alkene.17 The model we proposed in this Letter is a
working hypothesis; however, so far we cannot find a better
alternative that fits the experimental results.18
Theoretical calculations are currently underway to under-
stand the nature of the cation-π interaction for several
nonsymmetrical alkenes and to locate the possible transition
1
For alkene 3, theoretical calculations show that the phenyl
group and the double bond are strongly bound to the cation
in such a conformation that the benzylic hydrogens have an
unfavorable geometry for interaction with the incoming
oxygen, by not being perpendicular to the olefinic plane (see
figure in Supporting Information). Furthermore, the cation
sits almost on top of the disubstituted olefinic carbon and
cannot interact efficiently with the oxygen placed either on
the more or on the less substituted side of 3. Thus, in the
transition state leading to the formation of the intermediate
on the more substituted side of 3, oxygen interacts with only
one allylic methyl hydrogen. Therefore, the perepoxide on
the less substituted side of the alkene, where again oxygen
interacts with one allylic hydrogen from the methyl group,
is formed substantially. Similar anti “cis effect” selectivity
is observed for trisubstituted alkenes that do not have the
capability for a simultaneous interaction of singlet oxygen
with two allylic hydrogens on the more substituted side of
states of perepoxide formation in the reaction of O2 with
trimethylethylene bound to a Li+.
Acknowledgment. We thank Professor C. R. Theocharis
(University of Cyprus) for a generous donation of zeolite
Na-Y, Professor M. Orfanopoulos for valuable comments,
and Mr. Giannis Lykakis for assistance in the preparation
of the substrates.
Supporting Information Available: Four 1H NRM
spectra of the photooxygenation reactions and two figures
showing the Li+ interactions with 3. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL000021Y
(17) Stratakis, M.; Orfanopoulos, M. Tetrahedron Lett. 1995, 36, 4291-
4294.
(18) Note added in proof: After this manuscript had been submitted
we were informed by professor E. L. Clennan of some results similar to
those reported here. We thank him for a prepublication copy of his
manuscript. Clennan, E. L.; Sram, J. P. Tetrahedron, in press.
(16) Stratakis, M.; Orfanopoulos, M. Tetrahedron 2000, 64, 1595-1615.
1372
Org. Lett., Vol. 2, No. 10, 2000