Effect of tether length on Ti(III)-mediated cyclization of epoxyalkenes and unsaturated epoxyketones

Article abstract of DOI:10.1055/s-2004-834824

The chemo- and regioselectivity of the radical cyclization induced by titanocene chloride of a series of epoxyalkenes and another of unsaturated epoxyketones are investigated. 5-Exo and 6-exo cyclizations are the main processes with epoxyalkenes. 3-Exo and 4-exo cyclizations onto the carbonyl group and 6-exo and 8-endo onto the carbon-carbon double bond are the preferred processes with epoxyenones. A tandem reaction, 6-exo onto C=C, followed by a 3-exo onto C=O, and a disfavored 5-endo cyclization onto C=C are reported.

Full text of DOI:10.1055/s-2004-834824

LETTER
2553
Effect of Tether Length on Ti(III)-Mediated Cyclization of Epoxyalkenes and
Unsaturated Epoxyketones
Effect of
T
ethe
r
Le
ngthonTi(I
F
II)-Mediated
C
e
yclization
r
of
E
poxy
n
ketones ández-Mateos,*a L. Mateos Burón,a E. M. Martín de la Nava,a R. Rabanedo Clemente,a R. Rubio González,a
F. Sanz Gonzálezb
a
Departamento de Química Orgánica, Facultad de Ciencias Químicas, Universidad de Salamanca, Plaza de los Caídos 1–5,
37008 Salamanca, Spain
E-mail: afmateos@usal.es
b
General X-Ray Diffraction Service, Universidad de Salamanca, Plaza de los Caídos 1-5, 37008 Salamanca, Spain
Received 12 July 2004
(k25 = 5 × 103 s–1) is slightly faster than the 7-endo process
Abstract: The chemo- and regioselectivity of the radical cycliza-
(k25 = 9 × 102 s–1).4 Also, in this case a competitive 1,5-
tion induced by titanocene chloride of a series of epoxyalkenes and
hydrogen transfer may occur.1 For radicals A, B and D,
powerful accelerating effects will be required for synthet-
another of unsaturated epoxyketones are investigated. 5-Exo and 6-
exo cyclizations are the main processes with epoxyalkenes. 3-Exo
and 4-exo cyclizations onto the carbonyl group and 6-exo and 8- ic applications (gem-dimethyl substituents, heteroatoms
endo onto the carbon–carbon double bond are the preferred process-
es with epoxyenones. A tandem reaction, 6-exo onto C=C, followed
by a 3-exo onto C=O, and a disfavored 5-endo cyclization onto C=C
instead of C-3, exo carbonyl group, etc.) and also rapid
trapping of the new radical formed.
Radical cyclization reactions are not restricted to addi-
tions onto carbon–carbon double bonds, and other accep-
tors, such as carbon-oxygen double bonds, can also be
employed. The available kinetic data for the cyclization of
the 5-oxo-pentyl G (k25 = 1.5 × 105 s–1) and the 6-oxo-
hexyl radical H (k25 = 5.5 × 103 s–1) show that both pro-
cesses are similar to those corresponding to the pentenyl
C and hexenyl D radicals. The drawback is that the
are reported.
Key words: radical cyclization, titanocene chloride, unsaturated
epoxyketones, cyclobutanodiol, cyclooctanodiol
Radical cyclization is a valuable tool for the construction
of carbocyclic compounds. Hexenyl radical C cyclization
is the best understood and the most representative parent,
and mainly gives cyclopentane (k25 = 2 × 105 s–1), instead
of cyclohexanes (k25 = 4 × 103 s–1), in an irreversible
way.1 Other cyclizations, such as that of butenyl A, pente-
nyl B and heptenyl D radicals, have unique features, and
are less well known (Figure 1).2
reverse cleavages of the acyclic alkoxy radical G (–k80
=
5 × 108 s–1) and H (–k80 = 1 × 107 s–1) are faster than the
cyclization (Figure 2).1
E n = 1
F n = 2
O
O
k
H
G n = 3
H n = 4
H
n
n
-k
A n = 1
k
k
B n = 2
C n = 3
D n = 4
n
Figure 2 Radical addition onto C=O double bond
-k
-k
n
n
Figure 1 Radical addition onto C=C double bond
We have recently reported a series of radical cyclizations
on carbonyl groups of aldehydes and ketones to afford, in
good yields, compounds ranging from cyclopropanols to
cyclohexanols.5 The n-exo process is based on the ho-
molytic cleavage of an epoxide with titanocene chloride
and further intramolecular addition to a carbonyl group.
The resulting products after hydrolysis are 1,3-cyclic
diols. Success in the latter cyclizations was achieved by
trapping the cyclic alkoxy radicals with Ti(III)
(Scheme 1).
The latter radicals have not found much use in synthesis
due to their reluctance to cyclize. Thus, the butenyl radical
A does not give cyclization in 4-endo mode and under-
goes 3-exo cyclization at a very slow rate (k25 = 2 × 103
s–1). The process is reversible (–k25 = 1 × 108 s–1) and, due
to ring strain, the equilibrium lies to the side of the open
radical.1 The behavior of the pentenyl radical B is very
similar to that of the previous butenyl A, although several
examples of a priori stereoelectronically disfavored 5-
endo processes have been reported.3
In this context, here we investigated the behavior of two
series of systems. The first to be studied were alkenyl rad-
icals generated by homolytic cleavage of epoxyalkenes
with titanocene chloride (Scheme 2). This reagent was
discovered for organic synthesis by W. A. Nugent and T.
V. RajanBabu,6 and the scope and applications have been
the object of several reviews by A. Gansäuer.7a–c To our
knowledge, the radical cyclization accomplished by
this method gives cyclopentane6,7d,e or cyclohexane
Heptenyl radical D cyclizations are generally irreversible,
and the rate of ring closure is slower than that of the hex-
enyl radical C for entropic reasons; 6-exo cyclization
SYNLETT 2004, No. 14, pp 2553–2557
2
2
.1
1
.2
0
0
4
Advanced online publication: 20.10.2004
DOI: 10.1055/s-2004-834824; Art ID: G26704ST
© Georg Thieme Verlag Stuttgart · New York

2554
A. Fernández-Mateos et al.
LETTER
TiO
TiO
stead, we observed radical reduction followed by Cp2TiH
elimination. The same happened with the epoxyalkene 2:
the only product obtained was the dienol 2a. The acceler-
ating effect of the exo carbonyl group on the double bond
of 1 was tested with epoxy ionone 5, which quantitatively
afforded the 3-exo cyclization product 5a, (Scheme 3). A
similar case has been described recently by A. Gansäuer
for an unsaturated epoxyamide obtained from cyclo-
citral.7b
O
Ti (III)
R
R
R
n
n
n
OTi
O
O
HO
H3O+
R
n
OH
Scheme 1 1,3-Cycloalkanediols from epoxyketones
The expected 5-exo cyclization of the radical generated
from epoxyalkene 3 afforded cis-indanol (3a) and trans-
indanol 3b at a ratio of 3:1, respectively.11 The heptenyl
radical from 4 gave the trans-decalin 4a by the 6-exo
cyclization process. No 7-endo products were found, but
dienol 4b appeared after the reduction and elimination of
Cp2TiH.
compounds.8 No attempts have been made to obtain other
types of cyclic compounds, such as cyclopropane, without
any type of activation.
O
?
Ti (III)
n
n
HO
OTi
O
n = 1–4
O
1. Ti (III)
O
O
O
2. H3O+
H
?
Ti (III)
O
n
n
5
5a
OTi
n = 1–4
Scheme 3 Reaction product from the epoxy ionone 5 with Cp2TiCl
Scheme 2 Series of epoxyalkenes and unsaturated epoxyketones in
radical reaction with Cp2TiCl
The last series of test substrates, 6–9, was also prepared
from a-cyclocitral.9b
The next series studied involved radicals tethered to a,b-
unsaturated ketones, generated by the cleavage of epoxy
alkenones with Ti(III). The former systems can be consid-
ered as a model to compare with the latter systems in
which an extra endo carbonyl group is added. Our interest
in this area of chemistry was sparked by our investigation
of the chemo- and regioselectivity of the cyclization of
this kind of chemical system in order to obtain new and
selective methods for synthesis.
As is known, radical cyclization starts by homolytic
cleavage of the oxirane ring induced by titanocene chlo-
ride. However, other alternative independent reactions
could take place with the enone system, as reported by E.
Doris; i.e., the reduction of the conjugated double bond.
Fortunately for us, the Doris reduction was slower and did
not compete with our cyclization.12
Treatment of unsaturated epoxyketone 6 (Table 2) with
titanocene chloride gave the bicyclic diol 6a as the main
product.13 In both diastereomers, 6 and 6¢, the 3-exo
cyclization onto carbonyl group is the preferred process.
Neither 4-exo nor 5-endo cyclization onto C=C occurred,
although as could be expected the endo carbonyl group
would enhance the 5-endo-trig process, as discussed
below.
The former series of test substrates, 1–4, were convenient-
ly prepared from a-cyclocitral and homologues com-
pound by standard procedures.9a
Treatment of epoxyalkene 1 (Table 1) with titanocene
chloride in THF afforded only the dienol 1a in good
yield.10 Neither 3-exo nor 4-endo cyclization occurred; in-
Table 1 Reaction Products from Epoxyalkenes 1–4 with Cp2TiCl
The epoxyketone 7 afforded a mixture of two bicyc-
lo[4.2.0]octane diols, 7a and 7b (1:2), the trans-indenone
7c, and the cis and trans decalones 7d and 7e (1:1). These
results point to several interesting findings: (i) the priority
of the 4-exo process onto carbonyl group over the 5-exo
onto C=C, which means that the former is a fast process;
(ii) the occurrence of the 6-endo process accelerated by
the endo carbonyl group, which did not occur with the
deoxo epoxide 3; and (iii) the stereoselectivity in the 4-
exo process, which afforded only trans fused bicyc-
lo[4.2.0]octanes. This remarkable stereoselectivity in the
formation of cyclobutane is probably one of the first
examples found in radical chemistry and in other modes
of cyclization.
Epoxides
Products (yield, %)
O
HO
R
HO
R
R
n
n
n
H
H
1 n = 0, R = Et
2 n = 1, R = H
3 n = 2, R = H
–
1a (90)
2a (74)
–
–
3a (48)
3b (16)
4 n = 3, R = H
4a (69)
4b (8)
Synlett 2004, No. 14, 2553–2557 © Thieme Stuttgart · New York

LETTER
Effect of Tether Length on Ti(III)-Mediated Cyclization of Epoxyketones
2555
Table 2 Radical Reaction of the Epoxy Enones 6–9 with Cp2TiCl
Epoxides
Products (yield, %)
O
HO
OH
HO
HO
HO
HO
OH
O
O
n
n
O
O
n
n
n
H
H
H
H
6 n = 0
7 n = 1
8 n = 2
9 n = 3
6a (68)
–
–
–
6b (13)
7a, 7b (40)
7c (17)
–
–
7d, 7e (21)
8b (26)
–
–
–
–
–
8a (51)
9a (22)
–
9b (47)
The radical arising from epoxyketone 8 afforded two Finally, an epoxy enone 10, different from those men-
products, 8a and 8b, at a ratio of 2:1. The tricyclic diol 8a tioned above, with the carbonyl group exo to the double
is the result of a cyclization cascade: the first step is a 6- bond C=C was tested against Ti(III). In this case, two
exo cyclization onto C=C, followed by a very rapid 3-exo products, 10a and 10b, were obtained at a ratio of 2:1, re-
cyclization onto carbonyl. The global process is complete- spectively (Scheme 5). The major product was the 5-
ly stereoselective.11 The minor product from 8 was the re- endo-trig cyclization product, indenone 10a. This is one
sult of a stereoselective 7-endo cyclization to give the of the few examples reported for a 5-endo cyclization pro-
trans bicyclo hydroxyketone 8b.14 It is interesting to note cess in the literature without a heteroatom in the newly
from the kinetic data that although the potential of 5-exo formed cyclopentane.3 It is worth noting that an exo
cyclization onto carbonyl (k25 = 1.5 × 105 s–1) must be carbonyl group promotes the 5-endo cyclization, as in
faster than the 6-exo (k25 = 5.5 × 103 s–1) or 7-endo compound 10, while an endo carbonyl group, as in
cyclization onto C=C, no product resulting from the 5-exo compound 7, does not.
process was obtained. The 7-endo process is undoubtedly
activated by the 7-endo carbonyl group, since the deoxo
HO
compound 4 did not give the corresponding cycloheptane
(Scheme 4).
O
O
H
1. Ti (III)
[Ti]O
10a (48%)
[Ti]O
2. H3O+
O
HO
Ti (III)
OH
8
O
O
O
10
8b
8a
H
H
10b (24%)
[Ti]O
[Ti]O
Scheme 5 Radical reaction of unsaturated epoxyketone 10 with
O
Cp2TiCl
H
H
The minor product was the 5-exo-trig cyclization product,
indenodiol 10b. The ring fusions of 10a and 10b are in
both cases trans.11
Scheme 4 Proposed mechanism to explain the formation of 8a
and 8b
The chemo- and regioselectivity of the radicals arising
from epoxyunsaturated ketones 6 to 9 depend of the dis-
tance from the radical to the carbonyl group: when the dis-
tance is short (compounds 6 and 7) cyclization onto the
carbonyl is the preferred mode. For longer distances
(compounds 8 and 9) the cyclization onto carbonyl does
not occur, and cyclization onto C=C is preferred.
The last radical in this series, obtained from epoxyketone
9 and titanocene chloride, gave two products, 9a and 9b,
at a ratio of 1:2, respectively. The minor product, the hy-
droxyketone 9a, corresponds to the 7-exo process and the
major product, the hydroxyketone 9b, corresponds to the
8-endo cyclization mode. No 6-exo product was found al-
though, again, the kinetic data (see above) indicate that The results indicate that the alkenyl radicals, not activated
this process would clearly be faster than the 7-exo or the by carbonyl groups generated from epoxyalkanes 1–4,
8-endo cyclization onto C=C, which are at the lower limit give the 5-exo and 6-exo cyclization products. Alkenyl
of synthetic utility.15 Clearly, the last process is accelerat- radicals from 5 and 10 activated with an exo carbonyl
ed by the endo carbonyl group.
group with respect to the C=C double bond, preferentially
give 3-exo and 5-endo cyclizations onto C=C, respec-
Synlett 2004, No. 14, 2553–2557 © Thieme Stuttgart · New York

2556
A. Fernández-Mateos et al.
LETTER
aldehydes I (n = 0, 1, 2, 3) have been reported in the
tively. The alkenyl radicals from 6–9 activated with an
endo carbonyl group with respect to C=C afford 3-exo and
4-exo onto C=O, and 6-exo and 8-endo onto C=C cycliza-
tion compounds, respectively, as the major products. In
summary, we report new knowledge about radical cy-
clization selectivity and new procedures in the synthesis
of cyclic compounds, ranging from cyclopropanes to
cyclooctanes.
literature. (i) For aldehydes I, n = 0, see: Geyde, R. N.; Aura,
P. C.; Deck, K. Can. J. Chem. 1971, 49, 1764. (ii) For
aldehyde I, n = 1, see: Fernández-Mateos, A.; Lopéz Barba,
A. J. Org. Chem. 1995, 60, 3580. (iii) For aldehyde I, n = 2,
see: Fernández-Mateos, A.; Pascual Coca, G.; Rubio
González, R.; Tapia Hernández, C. J. Org. Chem. 1996, 61,
9097. (iv) The aldehyde I, n = 3, was obtained from I, n = 2,
by Wittig reaction with Ph3P=CHOCH3, followed by
treatment with HClO4 (60%) in THF. All compounds
synthesized are racemic, although only one enantiomer is
depicted (Figure 3). (b) The epoxyketones 6–9 were
obtained by a three-step sequence from the aldehydes I,
n = 0, 1, 2, 3: i) Grignard reaction with vinyl magnesium
bromide, ii) oxidation with Dess–Martin reagent, and iii)
epoxidation with mCPBA. (c) The structure of
Acknowledgment
Financial support for this work from the Ministerio de Ciencia y
Tecnología of Spain (PPQ2002-00290) is gratefully acknowledged.
We also thank the Junta de Castilla y León for the fellowship to
L.M.B., and the Ministerio de Ciencia y Tecnología for the fel-
lowships to E.M.M.N. and R.R.C.
epoxyketones 6–10, in which the oxiranic oxygen and the
side chain is cis, is based on spectroscopic data and
comparison with the epoxy compound described by K. Mori
et al., whose structure was determined by X-ray. See: Mori,
K.; Aki, S.; Kido, M. Liebigs Ann. Chem. 1993, 83.
References
(1) (a) Curran, D. P. In Comprehensive Organic Synthesis, Vol.
4; Trost, B. M.; Fleming, I.; Paquette, L. A., Eds.; Pergamon:
Oxford, 1991, 815. (b) Giese, B.; Kopping, B.; Göbel, T.;
Thoma, G.; Dickhaut, J.; Kulicke, K. J.; Trach, F. In Organic
Reactions, Vol. 48; Paquette, L. A., Ed.; Wiley: New York,
1996, 308. (c) Beckwith, A. L. J.; Hay, B. D. J. Am. Chem.
Soc. 1989, 111, 230. (d) Beckwith, A. L. J.; Hay, B. D. J.
Am. Chem. Soc. 1989, 111, 2674. (e) Beckwith, A. L. J.;
Raner, K. D. J. Org. Chem. 1992, 57, 4954. (f) Beckwith,
A. L. J. Tetrahedron 1981, 37, 3073.
(2) Srikrishna, A. In Radicals in Organic Synthesis, Vol. 2;
Renaud, P.; Sibi, M. P., Eds.; Wiley-VCH: Weinheim, 2001,
151.
(3) (a) Ishibashi, H.; Sato, T.; Ikeda, M. Synthesis 2002, 695.
(b) Chatgilialoglu, C.; Ferreri, C.; Guerra, M. J. Am. Chem.
Soc. 2002, 124, 10765.
CHO
n
I , n = 0, 1, 2, 3
Figure 3
(10) General Procedure. A mixture of Cp2TiCl2 (2.2 mmol) and
Zn (3.0 mmol) in strictly deoxygenated THF (4 mL) was
stirred at r.t. until the red solution turned green. In a separate
flask, the epoxy compound (1.0 mmol) was dissolved in
strictly deoxygenated THF (10 mL). The green Ti(III)
solution was slowly added via cannula to the epoxide
solution. After 30 min, an excess of sat. NaH2PO3 was
added, and the mixture was stirred for 20 min. The product
was extracted into Et2O and washed with sat. NaHCO3 and
H2O. After removal of the solvent, the crude product was
purified by flash chromatography. All homolytic cleavages
were absolutely selective and always afforded the tertiary
radical.
(4) Bailey, W. F.; Longstaff, S. C. Org. Lett. 2002, 3, 2217.
(5) Fernández-Mateos, A.; Martín de la Nava, E.; Pascual Coca,
G.; Ramos Silvo, A.; Rubio González, R. Org. Lett. 1999, 1,
607.
(11) The relative configuration of the newly created stereocenters
has been assigned by spectroscopic data and H-C
(6) (a) RajanBabu, T. V.; Nugent, W. A. J. Am. Chem. Soc.
1994, 116, 986. (b) Nugent, W. A.; RajanBabu, T. V. J. Am.
Chem. Soc. 1988, 110, 8561.
correlation, except for structures 7b, 8a, and 10b, whose
crystallographic data have been deposited with the
Cambridge Crystallographic Data Centre as supplementary
publication numbers CCDC 234516, 2345517 and 234515,
respectively. Spectroscopic data of three selected
(7) (a) Gansäuer, A.; Bluhm, H. Chem. Rev. 2000, 100, 2771.
(b) Gansäuer, A.; Lauterbach, T.; Narayan, S. Angew. Chem.
Int. Ed. 2003, 42, 5556. (c) Gansäuer, A.; Rinker, B.;
Pierobon, M.; Grimme, S.; Gerenkamp, M.; Mück-
Lichtenfeld, C. Angew. Chem. Int. Ed. 2003, 42, 3687.
(d) Gansäuer, A.; Pierobon, M.; Bluhm, H. Synthesis 2001,
2500. (e) Gansäuer, A.; Bluhm, H.; Pierobon, M. J. Am.
Chem. Soc. 1998, 120, 12849.
(8) (a) Yamada, H.; Hasegawa, T.; Tanaka, H.; Takahashi, T.
Synlett 2001, 1935. (b) Fuse, S.; Hanochi, M.; Doi, T.;
Takahashi, T. Tetrahedron Lett. 2004, 45, 1961.
(c) Barrero, A. F.; Cuerva, J. M.; Herrador, M. M.; Valdivia,
M. V. J. Org. Chem. 2001, 66, 4074. (d) Barrero, A. F.;
Oltra, J. E.; Cuerva, J. M.; Rosales, A. J. Org. Chem. 2002,
67, 2566. (e) Barrero, A. F.; Rosales, A.; Cuerva, J. M.;
Oltra, J. E. Org. Lett. 2003, 5, 1935. (f) Haïdour, A.; Oltra,
J. E.; Barrero, A. F.; Cardenas, D. J.; Cuerva, J. M. Chem.–
Eur. J. 2004, 10, 1778. (g) Ruano, G.; Grande, M.; Anaya, J.
J. Org. Chem. 2002, 67, 8243.
compounds:
1-(2-Hydroxy-1,5,5-trimethylbicyclo[4.1.0]hept-7-yl)-
propan-2-one (5a). IR (film) n = 3430, 2902, 1715, 1040
cm–1. 1H NMR (CDCl3) d = 0.39 (1 H, d, J = 5.9 Hz), 0.87 (3
H, s), 0.90 (1 H, m), 1.00 (3 H, s), 1.01 (3 H, s), 0.90–1.30 (4
H, m), 2.13 (3 H, s), 2.13 (1 H, m), 2.79 (1 H, dd, J = 4.2,
J¢ = 18.8 Hz), 3.81 (1 H, t, J = 6.2 Hz) ppm. 13C NMR
(CDCl3) d = 16.29 (CH), 19.69 (CH3), 26.83 (C), 27.01
(CH2), 27.93 (C), 28.38 (CH3), 29.48 (CH3), 31.38 (CH3),
33.39 (CH2), 39.54 (CH), 42.91 (CH2), 71.96 (CH-O),
210.06 (C=O) ppm. MS (EI): m/z (%) = 210 (19) [M+], 177
(9), 153 (53), 97 (100), 71 (95). HRMS (IE): m/z calcd for
C13H22O2 [M+]: 210.1619. Found: 210.1620.
7-Hydroxy-1,4,4,7a-tetramethyloctahydroinden-2-one
(7c). IR (film) n = 3453, 2955, 1732, 1051 cm–1. 1H NMR
(CDCl3) d = 0.81 (3 H, s), 0.90 (3 H, s), 0.94 (3 H, s), 1.09
(3 H, d, J = 6.9 Hz), 1.20–1.70 (4 H, m), 1.52 (1 H, dd,
J = 7.7 Hz, J¢ = 14 Hz), 2.05 (2 H, m), 2.18 (1 H, dd, J = 7.7
Hz, J¢ = 18 Hz), 3.60 (1 H, dd, J = 5.2 Hz, J¢ = 10.3 Hz)
(9) (a) The epoxyalkenes 1–4 were obtained by a Wittig reaction
of the aldehydes I (n = 0, 1, 2, 3) followed by selective
epoxidation with mCPBA in CH2Cl2 at –30 °C. The starting
Synlett 2004, No. 14, 2553–2557 © Thieme Stuttgart · New York

LETTER
Effect of Tether Length on Ti(III)-Mediated Cyclization of Epoxyketones
2557
ppm. 13C NMR (CDCl3) d = 8.58 (CH3), 9.54 (CH3), 20.62
(13), 67 (41), 55 (100),(54), 55 (100). HRMS (IE): m/z calcd
for C17HO3 [M+]: 280.2038. Found: 280.2033.
(12) Moisan, L.; Hardouin, C.; Rousseau, B.; Doris, E.
Tetrahedron Lett. 2001, 43, 2013.
(CH3), 29.32 (CH2), 31.48 (C), 32.75 (CH3), 35.17 (CH2),
39.79 (CH2), 46.75 (C), 52.01 (CH), 59.23 (CH), 79.91 (CH-
O), 217.59 (C=O) ppm. MS (EI): m/z (%) = 210 (41) [M+],
177 (10), 139 (67), 110 (54), 95 (100), 69 (48), 55 (51).
HRMS (IE): m/z calcd for C13H22O2 [M+]: 210.1620. Found:
210.1623.
(13) The diastereomer the 6 afforded exclusively bicyclic diol 6a¢
in 70% yield (Scheme 6).
HO
O
Acetic Acid 4,4,10a-Trimethyl-8-oxododecahydrobenzo-
cycloocten-1-yl Ester (9b). IR (film) n = 2953, 1734, 1697,
1244 cm–1. 1H NMR (CDCl3) d = 0.80 (3 H, s), 0.83 (3 H, s),
0.92 (3 H, s), 1.20–1.70 (11 H, m), 2.04 (3 H, s), 2.25 (2 H,
m), 2.66 (2 H, m), 4.78 (1 H, dd, J = 5.6 Hz, J¢ = 10 Hz) ppm.
13C NMR (CDCl3) d = 17.60 (CH3), 21.12 (CH3), 21.27
(CH3), 24.10 (CH2), 25.43 (CH2), 31.25 (CH2), 31.70 (CH2),
31.85 (CH3), 34.81 (C), 38.72 (CH2), 40.41 (CH2), 40.56
(CH2), 41.15 (C), 49.74 (CH), 74.22 (CH-O), 170.30 (C=O),
216.62 (C=O) ppm. MS (EI): m/z (%) = 220 (9) [M+ – 60],
205 (16), 187 (9), 150 (17), 135 (11), 109 (25), 95 (30), 81
OH
H
6a'
O
6'
Scheme 6
(14) Yet, L. Tetrahedron 1999, 55, 9349.
(15) Beckwith, A. L.; Schiesser, C. H. Tetrahedron 1985, 41,
3925.
Synlett 2004, No. 14, 2553–2557 © Thieme Stuttgart · New York