Confined space-controlled hydroperoxidation of trisubstituted alkenes adsorbed on pentasil zeolites

Article abstract of DOI:10.1021/jo050081n

Photosensitized oxidation of trialkylalkenes 2-methyl-2-pentene (1), 1-methylcyclohexene (2), trans-3-methyl-2-pentene (3), cis-3-methyl-2-pentene (4), and 2-methyl-2-butene (5) included in the internal framework of Na-ZSM-5 zeolites was investigated. The zeolite samples having adsorbed the alkenes were suspended in isooctane, and the sensitizer, tetraphenylporphyrin (TPP), was dissolved in the solution. Singlet oxygen produced in the solution diffused into the internal framework of the zeolites and reacted with alkenes. For all the substrates studied, the ene-type allylic hydroperoxides were obtained in a highly regioselective manner. The regiochemistry for 1-4 in favor of the allylic hydrogen abstraction from the largest substituents is in contrast to their photooxidation within the dye-supported zeolite Na-Y, where the secondary hydroperoxides are preferentially produced. The tight confinement of the alkenes within the narrow channels of the ZSM-5 zeolites is likely to be responsible for this selectivity.

Full text of DOI:10.1021/jo050081n

Confined Space-Controlled Hydroperoxidation of Trisubstituted
Alkenes Adsorbed on Pentasil Zeolites
Yu-Zhe Chen, Li-Zhu Wu,* Li-Ping Zhang, and Chen-Ho Tung*
Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100101, China, and
The Graduate School of the Chinese Academy of Sciences, Beijing 100049, China
chtung@mail.ipc.ac.cn; lzwu@mail.ipc.ac.cn
Received January 14, 2005
Photosensitized oxidation of trialkylalkenes 2-methyl-2-pentene (1), 1-methylcyclohexene (2), trans-
3-methyl-2-pentene (3), cis-3-methyl-2-pentene (4), and 2-methyl-2-butene (5) included in the
internal framework of Na-ZSM-5 zeolites was investigated. The zeolite samples having adsorbed
the alkenes were suspended in isooctane, and the sensitizer, tetraphenylporphyrin (TPP), was
dissolved in the solution. Singlet oxygen produced in the solution diffused into the internal
framework of the zeolites and reacted with alkenes. For all the substrates studied, the ene-type
allylic hydroperoxides were obtained in a highly regioselective manner. The regiochemistry for
1-4 in favor of the allylic hydrogen abstraction from the largest substituents is in contrast to their
photooxidation within the dye-supported zeolite Na-Y, where the secondary hydroperoxides are
preferentially produced. The tight confinement of the alkenes within the narrow channels of the
ZSM-5 zeolites is likely to be responsible for this selectivity.
Introduction
reaction is suppressed by the fact that reaction with
substrates containing several distinct allylic hydrogens
generally produces several distinct products, and the
selectivity of the reaction is poor. To gain the selectivity
in the reaction, various efforts have been made in the
past decades, and remarkable regio- and stereocontrol
has been obtained.9-12 For example, Adam and co-
workers9 successfully utilized hydrogen bonding to con-
trol the facial diastereoselectivity of the ene reactions in
a series of allylic alcohols and amines. On the other hand,
The reaction of singlet oxygen (1O2) with alkenes
containing allylic hydrogens has attracted much attention
because the allylic hydroperoxide products are important
building blocks for synthetic organic chemistry.1-7 This
reaction was first introduced by Schenck8 in the late
1940s, and is often referred to as the “ene” or Schenck
reaction.6,7 A large number of experimental and theoreti-
cal studies have established a mechanism for this reac-
tion involving the perepoxide intermediate, and the
effects of the substituents in alkenes on the direction of
the approach of singlet oxygen to the stereogenic faces
of the substrate molecules have been thoroughly inves-
tigated.6,7 However, the potential synthetic use of this
(9) (a) Adam, W.; Nestler, B. J. Am. Chem. Soc. 1992, 114, 6549-
6550. (b) Adam, W.; Nestler, B. J. Am. Chem. Soc. 1993, 115, 5041-
5049. (c) Adam, W.; Bru¨nker, H. G. J. Am. Chem. Soc. 1993, 115, 3008-
3009. (d) Bru¨nker, H.-G.; Adam, W. J. Am. Chem. Soc. 1995, 117,
3976-3982.
(10) (a) Li, X.; Ramamurthy, V. J. Am. Chem. Soc. 1996, 118,
10666-10667.(b) Robbins, R. J.; Ramamurthy, V. J. Chem. Soc., Chem.
Commun. 1997, 1071-1072. (c) Ramamurthy, V.; Lakshminarasimhan,
P.; Grey, C. P.; Johnston, L. J. J. Chem. Soc., Chem. Commun. 1998,
2411-2424. (d) Shailaja, J.; Sivaguru, J.; Robbins, R. J.; Ramamurthy,
V.; Sunoj, R. B.; Chandrasekhar, J. Tetrahedron 2000, 56, 6927-6943.
(e) Kaanumalle L. S.; Shailaja J.; Robbins, R. J.; Ramamurthy, V. J.
Photochem. Photobiol. A 2002, 153, 55-65. (f) Shailaja, J.; Ponchot,
K. J.; Ramamurthy, V. Org. Lett. 2000, 2, 937-940. (g) Joy, A.;
Scheffer, J. R.; Ramamurthy, V. Org. Lett. 2000, 2, 119-121. (h) Joy,
A.; Robbins, R. J.; Pitchumani, K.; Ramamurthy, V. Tetrahedron Lett.
1997, 38, 8825-8828. (i) Ramamurthy, V.; Shailaja, J.; Kaanumalle,
L. S.; Sunoj, R. B.; Chandrasekhar, J. Chem. Commun. 2003, 1987-
1999. (j) Ramamurthy, V. J. Photochem. Photobiol. C 2000, 1, 145-
166.
(1) Wasserman, H. H.; Murray, R. W. Singlet Oxygen; Academic
Press: New York, 1979.
(2) Frimmer, A. A.; Stephenson, L. M. In Singlet Oxygen. Reactions,
Modes and Products; Frimer, A. A., Ed.; CRC Press: Boca Raton, FL,
1985.
(3) Foote, C. S.; Clennan, E. L. In Active Oxygen in Chemistry; Foote,
C. S., Valentine, J. S., Greenberg, A., Liebman, J. F., Eds.; Chapman
& Hall: London, 1995; pp 105-140.
(4) Adam, W.; Prein, M. Acc. Chem. Res. 1996, 29, 275-283.
(5) Adam, W.; Wirth, T. Acc. Chem. Res. 1999, 32, 703-710.
(6) Clennan, E. L. Tetrahedron 2000, 56, 9151-9179.
(7) Stratakis, M.; Orfanopoulos, M. Tetrahedron 2000, 56, 1595-
1615.
(8) Schenck, G. O. Naturwissenschaften 1948, 35, 28-29.
10.1021/jo050081n CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/10/2005
4676
J. Org. Chem. 2005, 70, 4676-4681

Space-Controlled Hydroperoxidation of Trisubstituted Alkenes
SCHEME 1. Schematic Presentation of the
Reaction System
several research groups have employed organized and
constrained media to enhance the selectivity in the
hydroperoxidation of alkenes.10-12 In particular, Rama-
murthy,10 Clennan,11 Stratakis,12 and their co-workers
have used the dye-supported Y-type zeolites as media for
selective oxidation of alkenes. They demonstrated that
for trisubstituted alkenes the ene reaction within these
intrazeolites is regioselective with preferential double-
bond formation at the more substituted carbon of the
double bond and the “cis effect” selectivity in solution does
not operate within zeolites. For the photoxygenation of
1-methylcycloalkenes,10c in contrast to the reaction in
solution, even an anti “cis effect” selectivity was found
in the intrazeolites. Ramamurthy10 and Clennan11 pro-
posed, respectively, that the metal cation comlexation to
the alkene and the electrostatic interaction of the cation
with the pendant oxygen in the perepoxide intermediate
are responsible to the regiochemistry of the intrazeolite
photooxidation.
Although the photooxidation of the alkenes within
Y-zeolites has been extensively investigated, curiously
the ene reaction conducted within the internal framework
of pentasil zeolites was scarcely studied.13,14 Here, we
report the photooxidation of trisubstituted alkenes 2-meth-
yl-2-pentene (1), 1-methylcyclohexene (2), trans-3-methyl-
2-pentene (3), cis-3-methyl-2-pentene (4), and 2-methyl-
2-butene (5) within ZSM-5 zeolite, a member of the
pentasil zeolite family. The internal surface of ZSM-5
consists of two types of pore systems (channels):15 one is
sinusoidal with a near-circular cross-section of ca. 5.5 Å,
and the other is straight and perpendicular to the
sinusoidal channels. The straight channels are roughly
elliptical with dimensions of ca. 5.2 × 5.8 Å. These
channels of ZSM-5 can allow the absorption of benzene
and other molecules of similar molecular size but prevent
molecules that possess a larger size/shape from being
sorbed into the internal framework. In the present work,
tetraphenylporphyrin (TPP) and isooctane were selected
as the photosensitizer and solvent, respectively. We
trapped the alkenes in the channels of ZSM-5 zeolite and
dissolved the sensitizer in the solvent. The zeolite sample
was suspended in the solution (Scheme 1). The choice of
the sensitizer and solvent was motivated by the desire
that they were prevented from being sorbed into the
ZSM-5 channels due to their size and shape character-
istics.16 Thus, the internal framework of the zeolite was
“dry,” and the substrate was protected from being
extracted to the solution during photolysis. Singlet
oxygen was generated in the solution and diffused into
the zeolite channels to react with the alkene. We found
that in addition to the cation complexation to the alkene
and to the pendant oxygen of the perepoxide, the narrow
internal confined space of ZSM-5 channels also play a
crucial role in the control of the direction of the pendant
oxygen, and a remarkable selectivity of the hydroperoxi-
dation is observed. In many cases, the ene product
distribution is significantly different both from those in
Na-Y zeolites and in solutions. In this sense, the
photosensitized oxidation of alkenes within ZSM-5 is
complementary with that in Y-type zeolites. Therefore,
one can choose Y-type or ZSM-5 zeolites as reaction
media to direct the selectivity in alkene photooxidation
toward the desired products.
Results and Discussion
General Methods. We used the ZSM-5 with a differ-
ent Si/Al ratio as the reactive media: a “low” Al content
(Si/Al ) 55) and a “high” Al content (Si/Al ) 25). All of
the zeolites were sodium cation exchanged forms. Inclu-
sion of 1-5 within the zeolites was achieved by using
dichloromethane as the solvent. The powder of ZSM-5
having adsorbed the substrate was collected by filtration
of the solvent, dried under nitrogen, and then washed
with isooctane to remove the substrate adsorbed on the
external surface of ZSM-5. The loading level was kept at
ca. 17 mg of substrate on 1 g of ZSM-5. The sample
prepared above was suspended in isooctane. The sensi-
tizer, TPP, was dissolved in the solution. The suspension
was saturated with oxygen by bubbling the gas during
photolysis. A 450-W medium-press Hanovia Hg lamp was
used as the light source, and a glass filter was employed
to cut off the light with wavelength < 400 nm. Generally,
after 30 min of photolysis the conversion of the starting
material was near 100%. Similar conversion was obtained
in isooctane in the absence of ZSM-5 zeolite after the
same irradiation period. After photolysis, the ZSM-5
powder was separated from the solvent by filtration. The
hydroperoxide products were extracted with dichlo-
romethane. For ease of analysis, the produced hydro-
peroxides were converted to the corresponding alcohols
(11) (a) Pace A.; Clennan, E. L. J. Am. Chem. Soc. 2002, 124, 11236-
11237. (b) Clennan, E. L.; Sram, J. P. Tetrahedron Lett. 1999, 40,
5275-5278. (c) Clennan, E. L.; Sram, J. P. Tetrahedron 2000, 56,
6945-6950. (c) Clennan, E. L.; Sram, J. P.; Pace, A.; Vincer, K.; White,
S. J. Org. Chem. 2002, 67, 3975-3978.
(12) (a) Stratakis, M.; Froudakis, G. Org. Lett. 2000, 2, 1369-1372.
(b) Stratakis, M.; Rabalakos C. Tetrahedron Lett. 2001, 42, 4545-4547.
(c) Stratakis, M.; Nencka, R.; Rabalakos, C.; Adam, W.; Krebs, O. J.
Org. Chem. 2002, 67, 8758-8763. (d) Stratakis, M.; Rabalakos, C.;
Mpourmpakis, G.; Froudakis, G. E. J. Org. Chem. 2003, 68, 2839-
2843. (e) Stratakis, M.; Kalaitzakis, D.; Stavroulakis, D.; Kosmas, G.;
Tsangarakis, C. Org. Lett. 2003, 5, 3471-3474. (f) Stratakis, M.;
Sofikiti, N.; Baskakis, C.; Raptis, C. Tetrahedron Lett. 2004, 45, 5433-
5436.
(13) Tung, C. H.; Wang, H. W.; Ying, Y. M. J. Am. Chem. Soc. 1998,
120, 5179-5186.
(14) Baldov´ı, M. V.; Corma, A.; Garc´ıa, H.; Mart´ı, V. Tetrahedron
Lett. 1994, 35, 9447.
(15) (a) Olson, D. H.; Haag, W. O.; Lago, R. M. J. Catal. 1980, 61,
390. (b) Jacobs, P. A.; Beyer, H. K.; Valyon, J. Zeolites 1981, 1, 161. (c)
Meier, W. M.; Olson, D. H. Atlas of Zeolite Structure Types; Structure
Commission of IZA; Polycrystal Book Service: Pittsburgh, 1978.
(16) The kinetic diameter of isooctane is >6.2 Å based on the fact
that for neopentane the kinetic diameter is 6.2 Å. Thus, it cannot be
sorbed into the ZSM-5 internal framework; see: Turro, N. J.; Wan, P.
J. Am. Chem. Soc. 1985, 107, 678-682. TPP has a greater kinetic
diameter compared with isooctane.
J. Org. Chem, Vol. 70, No. 12, 2005 4677

Chen et al.
SCHEME 2. Product Distributions in
Photosensitized Oxidation of Alkenesa
FIGURE 1. Perepoxide intermediate and Ramamurthy’s
model for oxidation of alkenes within Na-Y zeolite.
FIGURE 2. Clennan model for oxidation of alkenes within
Na-Y zeolite.
selectivity to complexation of the alkene to a sodium
cation in the interior of the zeolite (Figure 1b). This
complexation would compel the sterically demanding
allylic methyl group to rotate to the face approached by
singlet oxygen, thereby preventing access to the allylic
hydrogens on the methylene carbon. As a result, no
tertiary hydroperoxide could be generated. In 1999,
Clennan and co-workers11b proposed that sodium cation
complexed to the pendant oxygen in the perepoxide
rather than to the double bond (Figure 2). Markovnikov
directing effects (as a result of greater charge buildup
on the carbon framework) and the preferential formation
of the perepoxide with the pendant oxygen on the least
substituted side of the double bond (Figure 2, perepoxide
a) were responsible to the observed regiochemistry.
Seemly, a combination of ideas proposed by Ramamurthy
and Clennan can account for the regioselectivity during
the oxidation within dye-supported Na-Y zeolite.6,17
Surprisingly, the regioselectivity in the photooxidation
of 1 within ZSM-5 is in dramatic contrast to its photo-
oxidation within dye-supported Na-Y. The isolated yield
of 6 + 7 was greater than 95% based on the consumption
of the starting material, and the main product is the
tertiary (7) rather than the secondary (6) hydroperoxide.
To ascertain that the selectivity is not due to any
experimental artifacts, we performed the following con-
trol experiment. First, we carried out the photooxidation
of 1 in isooctane by using TPP as the sensitizer. Then
we added ZSM-5 to the solution. Analysis of the hydro-
peroxides in the solution showed that almost all of the
hydroperoxides were adsorbed into the zeolite. After
separation of the ZSM-5 powder from the solvent by
filtration, the hydroperoxides were extracted again by
dichloromethane, and analysis (after reduction by tri-
phenyl phosphine) by gas chromatography revealed that
they could be quantatively recovered. Thus, the hydro-
peroxides were stable and had no unusual affinity for
zeolite interior.
a Key: *the product distributions for the reactions within Na-Y
are quoted from refs 10a-c and 11b,c.
by reaction with triphenylphosphine. The oxidation
products were analyzed by gas chromatography and
identified by 1HNMR and mass spectroscopies. Material
balance in general was greater than 95%.
Photosensitized Oxidation of 1. Reaction of 1 with
singlet oxygen gives secondary (6) and tertiary (7) allylic
hydroperoxides as shown in Scheme 2. The product
distributions for the reactions carried out in isooctane
solution, within ZSM-5 zeolite and within thionin-sup-
ported Na-Y zeolite are also shown in Scheme 2. The
hydroperoxidation of 1 in solution and within dye-doped
Na-Y has been extensively examined.6,7 In solution,
nearly equal amounts of secondary and tertiary hydro-
peroxides were obtained. This product distribution has
been rationalized on the basis of the fact that singlet
oxygen attacks the alkene from the top and doubly
substituted side of the double bond (cis effect). In such
an approach, the formed perepoxide is stabilized by
secondary interactions between the pendant oxygen and
allylic hydrogens, which are situated parallel to the
π-orbital (Figure 1a). Thus, both the methyl and meth-
ylene hydrogens can be placed in an appropriate geom-
etry for abstraction, and 6 and 7 are formed in a nearly
equal amount. Within thionin-supported Na-Y, 6 was
exclusively produced. In 1996, Ramamurthy and co-
workers10a attributed this dramatically enhanced regio-
We tentatively attribute the change in regioselectivity,
as the photooxidation is moved from the interior of Na-Y
to the ZSM-5, to the tight confinement of 1 in the ZSM-5
channels. As mentioned above, within Na-Y the sodium
cation complexes with alkene and the pendant oxygen
(17) Clennan, E. L. In Molecular and Supramolecular Photochem-
istry; Ramamurthy, V., Schanze, K., Eds.; Marcel Dekker: New York;
Vol. 9, 2003.
4678 J. Org. Chem., Vol. 70, No. 12, 2005

Space-Controlled Hydroperoxidation of Trisubstituted Alkenes
supported Na-Y zeolite (Scheme 2). In isooctane solution,
the oxidation afforded the endocyclic (8 and 9) and
exocyclic (10) ene allylic hydroperoxides with nearly
equal amounts. However, within Na-Y, the photooxy-
genation is highly regioselective with almost exclusive
formation of 10.10a This regiochemistry has been inter-
preted by Ramamurthy,10 Clennan,11 and Stratakis12 on
the basis of cation complexing to the alkene and the
pendant oxygen on the least substituted side of the
endocyclic double bond (Figure 4, perepoxide a). In
contrast, within the interior of ZSM-5, the endocyclic ene
allylic hydroperoxides (8 + 9) are the predominant
products. Mass balance in this reaction was excellent
(greater than 95%). This observation is again consistent
with the proposal that the confined space of ZSM-5
channels forces the perepoxide to orient with the pendant
oxygen directed toward the largest group substituted side
of the endocyclic double bond (Figure 4, perepoxide b).
As in the case of 1, we assume that the perepoxide will
orient with the double bond aligned with the zeolite
channel due to the interaction of the cation on the
channel wall with the alkene and the pendant oxygen.
Calculation showed that indeed perepoxide b has smaller
width than perepoxide a (Table 1), and can fit into the
ZSM-5 channels, while perepoxide a cannot be accom-
modated into the internal surface of the zeolite.
Photosensitized Oxidation of 3 and 4. The evidence
that the confined space of the ZSM-5 channels compels
the pendant oxygen in the perepoxide to direct toward
the largest group substituted side of the alkene is further
strengthened by the intrazeolite photooxidation of 3 and
4. The oxidation of 3 in isooctane almost exclusively
resulted in 11 and 12, and only a trace of 13 was
produced (Scheme 2). Again, this observation can be well-
rationalized in terms of cis-effect. In the intrazeolite
photooxygenations, the cis-effect is significantly dimin-
ished. Within Na-Y, although the main products were
originated from the hydrogen abstraction from the two
cis-methyl (11 and 12), considerable amount of 13 was
generated.11c Within ZSM-5, the change of the regiose-
lectivity is even more significant, and 13 becomes the
main product. Evidently, the photooxygenation both
within Na-Y and ZSM-5 favors the allylic hydrogen
abstraction from the ethyl group. For this substrate, the
notation that the pendant oxygen in the perepoxide
intermediate directs to the least substituted side of the
double bond is the same as the presentation that the
pendant oxygen is on the largest group substituted side.
Thus, the direction of the effects of the Na-Y and ZSM-5
media on the regioselectivity is the same. The calculated
molecular dimensions for the perepoxide with pendant
oxygen on the largest group substituted side of the double
bond (perepoxide b) and for that with pendant oxygen
on the cis-dimethyl side (perepoxide a) are given in Table
1. Clearly, only perepoxide b can be formed within ZSM-5
channels.
FIGURE 3. Na+- and zeolite-directing regioselectivity in the
photooxidation of 2-methyl-2-pentene.
in the perepoxide, and the pendant oxygen is preferen-
tially on the least substituted side of the double bond
(Figure 3, perepoxide a). Thus, 6 is exclusively produced.
In contrast, within ZSM-5 the narrow space of the
channel would force the perepoxide to orient with the
pendant oxygen on the largest group (ethyl) substituted
side of the double bond (perepoxide b), because the
perepoxide in such conformation is more compact com-
pared with perepoxide a. As a result, the abstraction of
the allylic hydrogen from the ethyl is possible. We have
estimated the molecular dimensions of perepoxides a and
b18 and compared them with the size of Na-ZSM-5
channel. First, we used a molecular dynamics annealing
process to obtain a rough lowest energy conformation for
perepoxide a (or perepoxide b), which was followed by a
molecular mechanics energy minimization. Then we
defined a cylinder that can accommodate the perepoxide
in such conformation and has minimum dimensions. The
length of the smallest cylinder is assigned to the length
of the longest axis of the perepoxide. We believe that the
perepoxide will orient with the long axis aligned with the
ZSM-5 channel, so that the double bond and the pendant
oxygen can interact with the sodium cation on the zeolite
channel wall. The dimensions of the perepoxides were
determined based on the assumption that the internulear
length in the molecule is equivalent.19 The calculated
width pluses 1 Å to account for the size of the exterior
hydrogen atom.20 The obtained widths and lengths for
perepoxides a and b are given in Table 1. Evidently, the
data determined in this way are somewhat arbitrary.
However, they give a relative ranking that is adequate
for the present purposes. The kinetic width of the
channels in Na-ZSM-5 are estimated to be 5.5 Å.21
Obviously, while perepoxide b can enter the internal
surface of the zeolite, perepoxide a is too large to fit into
the channels.
Photosensitized Oxidation of 2. As observed in the
case of 1, the photosensitized oxidation of 2 differed
significantly when included in ZSM-5 zeolite compared
to those in homogeneous solutions and within dye-
In contrast, the effects of the Na-Y and ZSM-5 media
on the regioselectivity in the photooxygenation of 4, the
isomer of 3, are significantly different. While the photo-
oxidation within Na-Y afforded the hydroperoxide gen-
erated from the abstraction of hydrogen in the trans-
(18) Calculations were performed using ChemBats3D-Pro, Cam-
bridgesoft Corp., Cambridge, MA. Details are given in the Experimen-
tal Section.
(19) Morkin, T. L.; Turro, N. J.; Kleinman, M. H.; Brindle, C. S.;
Kramer, W. H.; Gould, I. R. J. Am. Chem. Soc. 2003, 125, 14917-
14924.
(20) 1 Å was chosen as a rough compromise between the hydrogen
covalent (0.4) and van der Waals radii (1.1-1.3 Å). See: Cotton, F. A.;
Wilkinson, G. Advanced Inorganic Chemistry 3rd ed.; Interscience:
New York, 1972; pp 116-119.
(21) Olson, D. H.; Kokotallo, G. T.; Lawton, S. L.; Meier, W. M. J.
Phys. Chem. 1981, 85, 2238-2243.
J. Org. Chem, Vol. 70, No. 12, 2005 4679

Chen et al.
TABLE 1. Dimensions of the Perepoxide Intermediates
substrate
1
2
3
4
5
perepoxide
a
b
a
b
a
b
a
b
a
b
dimensions
(length (Å) × width (Å))
7.4 × 5.6 8.0 × 4.8 6.6 × 6.1 6.8 × 5.3 6.7 × 6.7 6.1 × 5.4 6.6 × 6.3 6.0 × 5.4 6.2 × 4.9 6.1 × 5.1
π-electron density of the substrate and the cation of
the zeolite.10,23 Based on the unit cell composition
(NaxSi96-xAlxO192) of Na-ZSM-5 zeolite,13,22b the concen-
tration of Al is calculated to be ca. 6.3 × 10-4 and 2.3 ×
10-4 M per gram of zeolite with a Si/Al value of 25 and
55, respectively. The loading level of the substrates
employed in the present study is ca. 2.1 × 10-4 M per
gram of zeolite. Thus, both for high and low Al content
samples every substrate molecule may associate with a
cation near an Al atom and the complexation of the cation
to the alkene and to the pendant oxygen can operate in
the manners described above. Thus, the Al content shows
no effect on the product distributions.
Conclusions
FIGURE 4. Na+- and zeolite-directing regioselectivity in the
photooxidation of 1-methylcyclohexene.
The product distributions for the photosensitized oxi-
dation of 1-4 adsorbed within ZSM-5 zeolites are sig-
nificantly different from those in homogeneous solutions
and within Na-Y zeolites. Within dye-supported Na-Y
zeolites, the “cis effect” selectivity in solution does not
operate, and for the oxidation of 1-methylcycloalkenes
even an anti “cis effect” selectivity was observed. This
has been interpreted in terms of the complexation of the
cation to the alkene and to the pendant oxygen of the
perepoxide and the orientation of the pendant oxygen to
the least substituted side of the double bond. In contrast,
within ZSM-5 zeolites the constrained space of the zeolite
channels compels the pendant oxygen to direct toward
the largest group substituted side of the double bond,
thus forming a more compact intermediate. As a result,
the oxidation proceeds with preferential formation of the
products derived from the hydrogen abstraction from the
largest group. The oxidation within ZSM-5 zeolites is
complementary with that in Y-type zeolites. Thus, one
can choose either ZSM-5 or Y-type zeolites as reaction
media to direct the selectivity in the photosensitized
oxidation of alkene toward the desired products.
methyl (12) as the main product,10c the internal surface
of ZSM-5 enhances the formation of 13, and 11 + 13
dominate the products. Again, these regioselectivities
clearly show that within Na-Y the pendant oxygen of
the perepoxide directs toward the least substituted side
of the double bond (perepoxide a), while in ZSM-5 the
pendant oxygen is on the largest group substituted side
of the alkene (perepoxide b). Indeed, the calculated
dimensions of the perepoxides (Table 1) indicate that only
perepoxide b can be accommodated into the ZSM-5
channels.
Photosensitized Oxidation of 5. The above argu-
ment on the regiochemistry of the photooxidation of
alkenes within the interior of ZSM-5 suggests that such
regioselectivity for the substrates without a large sub-
stituent would be absent. Experiments show that this is
indeed the case for 5. The photooxidation of 5 in isooctane
gave 14 and 15 with equal amounts, while the product
ratios both within Na-Y10b and ZSM-5 were in favor of
the secondary hydroperoxide, 15. For this alkene, the
calculated dimensions of the perepoxides are given in
Table 1. Evidently, all the two diastereomers of the
perepoxide can be accommodated within the ZSM-5
channels; thus, the character of photooxidation is almost
identical with that in Na-Y zeolite.
Effects of Al Contents on Product Distribution.
As mentioned above, we used the ZSM-5 zeolites with a
different Si/Al ratio as the reaction media: low Al content
(Si/Al ) 55) and high Al content (Si/Al ) 25). We found
that for all the substrates at the loading levels used in
the present work the products and their distributions are
independent of the Al content. This is readily understood
by consideration of the character of ZSM-5 zeolite. It has
been established that there exist hydrophilic centers in
the internal surface of pentasils, the most common of
which are hydroxyl groups or cations associated with a
tetrahedrally coordinated aluminum, and the remaining
regions in the internal surface are hydrophobic.22 Within
a zeolite alkene molecules are adsorbed to the surface
near Al atoms due to the interaction between the
Experimental Section
General Procedure for the Intrazeolite Photooxygen-
ations. Alkenes 1-5 and the solvents isooctane and dichlo-
romethane were spectra grade. ZSM-5 zeolites with Si/Al )
25 and 55 were sodium-exchanged forms and were baked at
550 °C for 10 h prior to use. For a typical photooxygenation
experiment, an activated zeolite sample was added to the
substrate solution in dichloromethane and stirred at room
temperature for 1 h. The zeolite sample having absorbed the
substrate was collected by filtration of the solvent, dried under
nitrogen, and washed with isooctane to remove the substrate
absorbed on the external surface of the zeolite. The sample
(22) (a) Turro, N. J.; Cheng, C. C.; Abrams, L.; Corbin, D. R. J. Am.
Chem. Soc. 1987, 109, 2449-2456. (b) Hill, S. G.; Seddon, D. Zeolites
1985, 5, 173. (c) Chen, N. Y. J. Phys. Chem. 1976, 80, 60-64. (d) Weisz,
P. B. Ind. Eng. Chem. Fundam. 1986, 25, 53. (e) Corbin, D. R.; Seidel,
W. C.; Abrams, L.; Herron, N.; Stucky, G. D.; Tolman, C. A. Inorg.
Chem. 1985, 24, 1800-1803.
(23) (a) Hepp, M. A.; Ramamurthy, V.; Corbin, C. R.; Dybowski, C.
J. Phys. Chem. 1992, 96, 2629-2632. (b) Staley, R. H.; Beauchamp, J.
L. J. Am. Chem. Soc. 1975, 97, 5920-5921.
4680 J. Org. Chem., Vol. 70, No. 12, 2005

Space-Controlled Hydroperoxidation of Trisubstituted Alkenes
was subsequently suspended in isooctane in a Pyrex tube. To
the tube was added a known amount of the sensitizer (TPP)
solution in isooctane. This suspension was bubbled by a
constant stream of oxygen during irradiation. A 450-W medium-
press Hanovia Hg lamp was used as the light source, and a
glass filter was employed to cut off the light with wavelength
<400 nm. The filter thus ensured the absence of direct
excitation of the alkene substrate. After irradiation, the zeolite
was collected by filtration and stirred in each 20 mL of
dichloromethane for three times. The extracted products were
analyzed by their 1HNMR spectra. The products were then
converted into the corresponding alcohols by the reaction with
triphenylphosphine and analyzed by gas chromatography-
mass spectroscopy on a GC-MS spectrometer and by com-
parison with authentic samples. For preparative samples, the
corresponding alcohols obtained from the reaction with tri-
phenylphosphine were isolated by column chromatography on
silica eluting with dichloromethane and identified by their
spectral properties.
12. 1HNMR (400 MHz, CD3CN) δ: 9.23 (s, 1H), 5.00 (s, 1H),
4.91 (s, 1H), 4.41 (q, 1H), 0.80-1.56 (m, 8H). MS of its
corresponding alcohol: m/e 43, 55, 67, 71, 85.
13. 1HNMR (400 MHz, CD3CN) δ: 9.09 (s, 1H), 5.48 (q, 1H),
4.30 (q, 1H), 0.80-1.56 (m, 9H). MS of its corresponding
alcohol: m/e 43, 55, 67, 71, 85, 100 (M+).
14. 1HNMR (400 MHz, CD3CN) δ: 8.88 (s, 1H), 5.97 (q, 1H),
5.19 (d, 1H), 5.11 (q, 1H), 1.26 (s, 6H). MS of its corresponding
alcohol: m/e 43, 59, 71, 86 (M+),
15. 1HNMR (400 MHz, CD3CN) δ: 9.25 (s, 1H), 4.92 (d, 2H),
4.39 (q, 1H), 1.72 (s, 3H), 1.18 (d, 3H). MS of its corresponding
alcohol: m/e 43, 71, 86 (M+).
Calculation of the Dimensions of the Perepoxides. The
dimensions of the smallest cylinder that can accommodate the
most reasonable conformation of a perepoxide were estimated
by first performing a molecular dynamics annealing calculation
on the perepoxide using MM2 in program ChemBats3D-Pro,
from CambridgeSoft. The cooling rate was 1 kcal/atom/ps, and
the final temperature was 300 K. An additional molecular
mechanics optimization was performed to a minimum rms
gradient of 0.1. The relevant axes were then determined by
visual inspection. Equivalent atomic center-to-center distances
were measured graphically, and 1.0 Å was added to this
distance to take into account the dimensions of the exterior
hydrogen atoms. Repeated calculations of this type gave
dimensions that were reproducible to ca. 0.15 Å.
6. 1HNMR (400 MHz, CD3CN) δ: 9.23 (s, 1H), 4.95 (s, 1H),
4.92 (s, 1H), 4.14 (t, 1H), 1.68 (s, 3H), 1.54 (m, 2H), 0.86 (t,
3H). MS of its corresponding alcohol: m/e 43, 71, 85, 100 (M+).
7. 1HNMR (400 MHz, CD3CN) δ: 8.74 (s, 1H), 5.66 (m, 1H),
5.56 (d, 1H), 1.69 (d, 3H), 1.24 (s, 6H). MS of its corresponding
alcohol: m/e 43, 67, 85, 100 (M+).
1
8. HNMR (400 MHz, CD3CN) δ: 9.35 (s, 1H), 5.68 (br s,
1H), 4.18 (br s, 1H), 1.73 (s, 3H), 1.20-1.95 (m, 6H). MS of its
corresponding alcohol: m/e 55, 69, 84, 97, 112 (M+).
9. 1HNMR (400 MHz, CD3CN) δ: 8.76 (s, 1H), 5.88 (m, 1H),
5.53 (d, 1H), 1.23 (s, 3H), 1.20-1.95 (m, 6H). MS of its
corresponding alcohol: m/e 55, 79, 84, 97, 112 (M+).
10. 1HNMR (400 MHz, CD3CN) δ: 9.28 (s, 1H), 4.85 (d, 2H),
4.29 (t, 1H), 1.2-1.95 (m, 8H). MS of its corresponding
alcohol: m/e 55, 69, 79, 84, 97, 112 (M+), 113 (M + 1).
11. 1HNMR (400 MHz, CD3CN) δ: 8.76 (s, 1H), 5.90 (q, 1H),
5.17 (d, 1H), 5.14 (d, 1H), 0.80-1.56 (m, 8H). MS of its
corresponding alcohol: m/e 43, 55, 67, 71, 85.
Acknowledgment. We thank the National Sci-
ence Foundation of China (Nos. 20125207, 20272066,
20333080, 20332040, 20202013) and the Ministry of Sci-
ence and Technology of China (Grant Nos. G2000078104,
G2000077502, and 2003CB716802) for financial support.
JO050081N
J. Org. Chem, Vol. 70, No. 12, 2005 4681