Dye-sensitized intrazeolite photooxygenation of 4-substituted cyclohexenes. Remote substituent effects in regioselectivity and diastereoselectivity

Article abstract of DOI:10.1016/j.tetlet.2004.05.050

The product distribution in the dye-sensitized photooxygenation of α-terpinyl acetate and terpinen-4-ol is quite similar in solution, however, by zeolite Y confinement, is substantially influenced by the position of the remote polar substituents relative

Full text of DOI:10.1016/j.tetlet.2004.05.050

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 5433–5436
Dye-sensitized intrazeolite photooxygenation of 4-substituted
cyclohexenes. Remote substituent effects in regioselectivity
and diastereoselectivity
Manolis Stratakis,* Nikoletta Sofikiti, Constantinos Baskakis and Christos Raptis
Department of Chemistry, University of Crete, 71409 Iraklion, Greece
Received 4 March 2004; revised 7 May 2004; accepted 11 May 2004
Abstract—The product distribution in the dye-sensitized photooxygenation of a-terpinyl acetate and terpinen-4-ol is quite similar in
solution, however, by zeolite Y confinement, is substantially influenced by the position of the remote polar substituents relative to
the reacting double bond. The intrazeolite results were rationalized in terms of Naþ-substrate and Naþ-singlet oxygen interactions.
Ó 2004 Elsevier Ltd. All rights reserved.
The reaction of singlet oxygen (1O2) with alkenes with
zeolite Y confinement is a promising methodology for
their selective oxyfunctionalization.1 It has been shown
that the intrazeolite photooxygenation of trisubstituted
alkenes is regioselective with exclusive or predominant
formation of the secondary allylic hydroperoxides.2 In
addition, enhanced twin regioselectivity was found for
the case of gem-dimethyl trisubstituted alkenes.3 Alk-
enylarenes afford within NaY the ene products in a
highly chemoselective manner,4 while for chiral alkenes
bearing a phenyl group on the stereogenic carbon atom,
the diastereoselectivity trend changes from threo in
solution to erythro within NaY.5
cycloalkenes,7 however, with negligible diastereoselec-
tivity (Scheme 1). We examined the effect of zeolite Y
confinement in the photooxygenation of the natural
products terpinen-4-ol (3) and a-terpinyl acetate (4).
Compounds 3–4 have the carbon skeleton of p-menth-1-
ene (2), and possess remote polar substituents relative to
the reacting double bond. Photooxygenation of 3 and 4
in solution gives quite reproducible ratios of the three
regoisomeric allylic hydroperoxides, taking into account
the product distribution from the photooxygenation of
limonene.8 The allylic hydroperoxides 3a–f and 4a–f
(Table 1) were isolated by flash column chromatography
using chloroform/diethyl ether ¼ 5:1 as eluent.9 The
presence of the remote polar functionalities (–OH and
–OAc, respectively) do not affect significantly, as
expected, the product selectivity. However, in their intra-
zeolite photooxygenation, quite significant changes were
In this letter we report that polar substituents at a re-
mote position with respect to the reacting double bond
can substantially influence the photooxygenation prod-
ucts of 4-substituted-1-methyl-1-cyclohexenes, on going
from the reaction in solution to the confined environ-
ment of zeolite Y. The thionin-sensitized intrazeolite
photooxygenation of limonene (1) and p-menth-1-ene
(2) has been reported by Ramamurthy and co-workers6
to give exclusively the exocyclic ene allylic hydroperox-
ides 1a and 2a, respectively, in accordance with the
general regioselectivity trend found in 1-methyl-
OOH
1O2
H2C
H3C
H2C
H3C
CH3
CH2
Na-Y
1a
1
>97% (d.e. ~5%)
OOH
CH2
1O2
H3C
H3C
H3C
H3C
CH3
Na-Y
2a
2
Keywords: Singlet oxygen; Hydroperoxides; Diastereoselection; Zeolite
NaY; Terpenes.
>97% (d.e. not reported)
Scheme 1. Intrazeolite photooxygenation of limonene (1) and
p-menth-1-ene (2).
* Corresponding author. Tel.: +30-2810-393687; fax: +30-2810-3936-
01; e-mail: stratakis@chemistry.uoc.gr
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.05.050

5434
M. Stratakis et al. / Tetrahedron Letters 45 (2004) 5433–5436
Table 1. Photooxygenation of terpinen-4-ol (3) and a-terpinyl acetate (4) in solutiona and within NaYb
H3C
CH2
CH2
OOH
HOO
H3C
HOO
H3C
+
+
+
CH3
OH
OH
OH
H3C
CH3
CH3
3a
CH3
3b
1O2
3c
CH3
CH3
H3C
OH
OOH
+
H3C
HOO
H3C
HOO
CH3
+
3
OH
CH3
3e
OH
CH3
3f
OH
H3C
H3C
CH3
3d
CH2
HOO
CH2
CH3
HOO
HOO
+
+
+
CH3
H3C
CH3
OAc
4b
H3C
CH3
OAc
4c
H3C
CH3
1O2
OAc
4a
H3C
CH3
OOH
CH3
H3C
CH3
OAc
HOO
+
HOO
4
+
H3C
CH3
OAc
H3C
CH3
H3C
CH3
OAc
OAc
4f
4d
4e
Substrate
a
b
c
d
e + fc
3
4
67 (21)
5 (27)
5 (15)
4 (17)
9 (44)
2 (7)
16 (8)
87 (40)
3 (12)
2 (9)
a The values in parentheses indicate the product distribution for the photooxygenation in solution (dichloromethane/methylene blue).
b The intrazeolite photooxygenation experiments were carried out as described in Ref. 3b.
c The relative stereochemistry for the minor hydroperoxides 3–4e and 3–4f was not assigned.
found, not only between them, but also compared to the
intrazeolite photooxygenation of limonene (1) and p-
menth-1-ene (2). The results are summarized in Table 1.
The stereochemical assignment of the main products
was performed as follows. DFTcalculations revealed
that for 3b, the allylic hydrogen on the carbon atom
bearing the hydroperoxy functionality forms, with the
adjacent methylenic hydrogens, dihedral angles of 63°
and 52°, respectively. Therefore, small coupling con-
stants are expected, and this hydrogen appears as a
broad triplet at 4.60 ppm with J ¼ 3:5 Hz. For the dia-
stereomer 3a, which is more polar compared to 3b, the
analogous dihedral angles were calculated to be 55° and
172°, respectively. Thus, a large and a small coupling
constant are expected, and indeed this allylic hydrogen
exclusive formation of the ene adducts resulting from
allylic hydrogen atom abstraction from the less substi-
tuted side of the endocyclic double bond, by placing an
acetate functionality at the remote 8-position of the
menthene carbon skeleton (a-terpinyl acetate, 4), the
reaction gives mainly one regioisomeric adduct (4d) in
>97% diastereomeric excess, in which allylic hydrogen
atom abstraction has occurred from the more substi-
tuted side of the cycloalkene. The substantial formation
of the tertiary hydroperoxide 4d for the photooxygen-
ation of 4 within NaY is unique for an intrazeolite
reaction. Generally, for the case of trisubstituted alk-
enes, the secondary allylic hydroperoxides are produced
predominantly.12 On the other hand, for 3, the exocyclic
ene allylic hydroperoxide is formed substantially in
agreement to the intrazeolite photooxygenation results
of 1 and 2, however, in a highly diastereoselective
manner (93% de).
appears as
a doublet of doublets at 4.76 ppm
(J1 ¼ 11:5 Hz, J2 ¼ 4:0 Hz). Compound 3c was reduced
by PPh3 to the corresponding known cis-1,4-diol.10
Finally, hydroperoxide 4c reacted with LiAlH4 in
diethyl ether to afford, quantitatively, the known trans-
p-menth-2-ene-1,8-diol.11
The highly regioselective and diastereoselective intrazeo-
lite photooxygenation of 4 can be explained considering
the acetate–Naþ interaction within the zeolite cavities,
which directs singlet oxygen via an electrostatic inter-
action towards the more substituted side of the alkene
While the intrazeolite photooxygenation of limonene (1)
and p-menth-1-ene (2) is highly regioselective, with

M. Stratakis et al. / Tetrahedron Letters 45 (2004) 5433–5436
5435
tent with cation binding to the ester or hydroxy func-
tionality, which directs 1O2 attack towards the more
substituted side of the double bond, as proposed in TS1
of Scheme 2 to form the tertiary allylic hydroperoxide.
The moderate selectivity in the case of 6 (59%), might be
attributed to a less efficient cation–1O2 interaction
within NaY, since the binding site of Naþ to a-terpineol
is different compared to 4 or 5.
δ−
O
δ+
O
H3C
AcO
H3C
OOH
CH3
AcO
H3C
H
CH3
3
4d
H3C
H
TS1
δ−
O
δ+
CH3
In conclusion, we have shown that remote polar sub-
stituents with respect to the reacting double bond can
substantially affect the product selectivity in the int-
razeolite photooxygenation of p-menth-1-ene type
compounds. In addition, we have presented the first
examples of the predominant formation of tertiary
allylic hydroperoxides in the photooxygenation of tri-
substituted alkenes by zeolite confinement.
O
H3C
HO
H
H3C
H3C
OOH
3a
O
O
TS2
H
H
δ−
O
H3C
H3C
δ+
O
CH3
OOH
H
HO
CH3
H3C
H3C
3
3d
TS3
Acknowledgements
Scheme 2. Naþ-directing regioselectivity and diastereoselection in the
intrazeolite photooxygenation of a-terpinyl acetate and terpinen-4-ol.
This work was supported by the Greek Secretariat of
Research and Technology. We thank Professor G. E.
Froudakis and Mr. A. Mavrandonakis for performing
the DFTcalculations.
(TS1 in Scheme 2), where it abstracts the axially oriented
allylic hydrogen atom at C-3. For terpinen-4-ol (3), we
propose that binding of the Naþ to the hydroxy func-
tionality places the cation on the one diastereotopic face
of the double bond, thus shielding oxygen attack from
References and notes
1
that face (Scheme 2). Therefore, O2 attacks from the
1. (a) Ramamurthy, V.; Lakshminarasimhan, P.; Grey, C. P.;
Johnston, L. J. Chem. Commun. 1998, 2411–2418; (b)
Pace, A.; Clennan, E. L. J. Am. Chem. Soc. 2002, 124,
11236–11237; (c) Stratakis, M. Curr. Org. Synth. 2004, 1,
in press.
opposite face, mainly oriented towards the less substi-
tuted side of the alkene (TS2), to give the allylic
hydroperoxide 3a in 67% relative yield. A similar dia-
stereoselectivity trend has been reported by us, in the
intrazeolite photooxygenation of (R)-())-a-phelland-
rene.13 This ‘cation-shielding’ assumption for the int-
razeolite photooxygenation of 3 is supported by the fact
that the tertiary hydroperoxide 3d (TS3, Scheme 2) is
formed predominantly over its diastereomer 3c. By
contrast, the predominant formation of 3c over 3d in
2. Robbins, R. J.; Ramamurthy, V. Chem. Commun. 1997,
1071–1072.
3. (a) Clennan, E. L.; Sram, J. P. Tetrahedron 2000, 56, 6945–
6950; (b) Stratakis, M.; Nencka, R.; Rabalakos, C.;
Adam, W.; Krebs, O. J. Org. Chem. 2002, 67, 8758–8763.
4. Stratakis, M.; Rabalakos, C.; Mpourmpakis, G.; Frouda-
kis, G. E. J. Org. Chem. 2003, 68, 2839–2843.
5. Stratakis, M.; Kalaitzakis, D.; Stavroulakis, D.; Kosmas,
G.; Tsangarakis, C. Org. Lett. 2003, 5, 3471–3474.
6. Shailaja, J.; Sivaguru, J.; Robbins, J.; Ramamurthy, V.;
Sunoj, R. B.; Chandrasekhar, J. Tetrahedron 2000, 56,
6927–6943.
1
solution could be attributed to a O2–hydroxyl steering
effect.14
Photooxygenation of substrates 5 and 6 (a-terpineol)15
supported the intrazeolite regioselectivity results for the
case of 4 (Scheme 3). While in solution, the typical
regiochemical outcome was found as in the case of 1–4,
by zeolite confinement the tertiary allylic hydroperox-
ides were formed in 77% and 59% relative yield,
respectively. This regioselectivity trend is again consis-
7. Li, X.; Ramamurthy, V. J. Am. Chem. Soc. 1996, 118,
10666–10667.
8. Gollnick, K.; Schade, G. Tetrahedron Lett. 1973, 11, 857–
860.
9. Selected NMR data for the main allylic hydroperoxides in
1
CDCl3. Compound 3a H NMR: 4.95 (br s, 1H), 4.87 (br
s, 1H), 4.76 (dd, 1H, J1 ¼ 11:5 Hz, J2 ¼ 4:0 Hz), 2.45 (dt,
1H, J1 ¼ 13:0 Hz, J2 ¼ 4:0 Hz), 2.27 (td, 1H, J1 ¼ 13:0 Hz,
J2 ¼ 4:0 Hz), 2.13 (ddd, 1H, J1 ¼ 10 Hz, J2 ¼ 5:0 Hz,
J3 ¼ 2:5 Hz), 1.65 (m, 2H), 1.48 (m, 2H), 0.94 (d, 3H,
J ¼ 7:0 Hz), 0.93 (d, 3H, J ¼ 7:0 Hz). 13C NMR: 146.92,
105.13, 82.82, 75.56, 39.57, 38.28, 35.11, 29.57, 16.78,
16.76. Compound 3b 1H NMR: 5.07 (br s, 1H), 5.04 (br s,
1H), 4.60 (br t, 1H, J ¼ 3:5 Hz), 2.61 (m, 1H), 2.20 (m,
2H), 1.46–1.85 (m, 4H), 0.93 (d, 3H, J ¼ 7:0 Hz), 0.92 (d,
3H, J ¼ 7:0 Hz). 13C NMR: 144.96, 114.59, 86.18, 72.84,
10 (<1)
CH3 54 (41)
12 (7)
CH3
H3C
HO
H3C
H3C
MeO
46 (16)
36 (59)
42 (77)
O
6
5
Scheme 3. Regioselectivity in the photooxygenation of 5 and 6 (the
values in parentheses indicate the relative reactivity for allylic hydro-
gen atom abstraction within NaY).
1
37.83, 37.34, 35.32, 25.85, 16.92, 16.79. Compound 3c H
NMR: 5.82 (d, 1H, J ¼ 10:0 Hz), 5.67 (d, 1H,

5436
M. Stratakis et al. / Tetrahedron Letters 45 (2004) 5433–5436
J ¼ 10:0 Hz), 2.09 (td, 1H, J1 ¼ 13:5 Hz, J2 ¼ 3:0 Hz), 1.96
11. Clark, B. C.; Powell, C. C.; Radford, T. Tetrahedron 1977,
33, 2187–2191. 1H NMR (CDCl3): 5.88 (d, 1H,
J ¼ 10:0 Hz), 5.77 (dd, 1H, J1 ¼ 10:0 Hz, J2 ¼ 2:0 Hz),
2.06 (m, 1H), 1.50–1.85 (m, 4H), 1.29 (s, 3H), 1.25 (s, 3H),
1.19 (s, 3H). 13C NMR: 134.93, 130.08, 72.51, 66.85, 47.10,
37.03, 29.65, 28.08, 26.22, 20.48. For comparison pur-
poses, the NMR data of the diastereomeric cis-p-menth-2-
ene-1,8-diol were kindly provided by Dr. R. K. Razdan
(Organix Inc.).
12. Clennan, E. L.; Sram, J. P. Tetrahedron Lett. 1999, 40,
5275–5278.
13. Stratakis, M.; Sofikiti, N. J. Chem. Res. (S) 2002, 374–
375.
(dt, 1H, J1 ¼ 13:5 Hz, J2 ¼ 3:0 Hz), 1.69–1.79 (m, 2H),
1.51 (td, 1H, J1 ¼ 13:5 Hz, J2 ¼ 3:0 Hz), 1.37 (s, 3H), 0.97
(d, 3H, J ¼ 7:0 Hz), 0.91 (d, 3H, J ¼ 7:0 Hz). 13C NMR:
137.12, 130.86, 78.86, 71.74, 37.18, 28.24, 27.22, 24.42,
17.35, 16.26. Compound 3d 1H NMR: 5.77 (d, 1H,
J ¼ 10:5 Hz), 5.74 (d, 1H, J ¼ 10:5 Hz), 2.25 (m, 2H),
1.60–1.75 (m, 3H), 1.26 (s, 3H), 0.97 (d, 3H, J ¼ 7:0 Hz),
0.90 (d, 3H, J ¼ 7:0 Hz). 13C NMR: 134.96, 132.96, 81.48,
71.61, 36.97, 28.45, 28.20, 22.65, 17.41, 16.30. Compound
4d 1H NMR: 5.92 (d, 1H, J ¼ 10:0 Hz), 5.66 (dd, 1H,
J1 ¼ 10:0 Hz, J2 ¼ 2:0 Hz), 2.81 (m, 1H), 2.20 (td, 1H,
J1 ¼ 13:0 Hz, J2 ¼ 3:5 Hz), 2.00 (s, 3H), 1.97 (m, 1H),
1.55–1.62 (m, 2H), 1.45 (s, 3H), 1.40 (s, 3H), 1.34 (s, 3H).
13C NMR: 170.51, 133.74, 129.89, 84.40, 78.62, 44.36,
31.67, 24.81, 23.68, 23.06, 22.47, 20.09.
14. Prein, M.; Adam, W. Angew. Chem., Int. Ed. Engl. 1996,
108, 519–538.
15. Substrate 5 was prepared as a mixture of the diastereomers
R,R/R,S ꢀ 2:1 Guth, H. Helv. Chim. Acta 1996, 79, 1559–
1571. Compound 6 is commercially available.
10. Suzuki, M.; Ohtake, H.; Kameya, Y.; Hamanaka, N.;
Noyori, R. J. Org. Chem. 1989, 54, 5292–5302.