Table 1 Group selective cyclisation of 11 to 12 and 13
xylene), and found to give a mixture of sulfoxides 12 and 13
in an identical ratio to that derived from thermolysis of 11.
The reversible nature of the sulfoxide elimination–sulfenic
acid addition reaction means that the product sulfoxides 12
and 13 can also eliminate to reform the sulfenic acid inter-
mediate under the reaction conditions – presumably the
temperature required for elimination of a sulfenic acid from
12 or 13 is similar or less than that from tert-butyl sulfoxide
11.15
In conclusion the stereoselective synthesis of cis-fused per-
hydrobenzothiophenes related to the breynolide ring system
can be controlled by a combination of the stereoelectronic
requirements for the addition of sulfenic acids to alkenes, and
the preference of an oxygen substituent other than hydroxyl to
occupy a position on the newly formed thiolane ring cis to the
adjacent bridgehead methyl and the lone pair on sulfur. The
overall process generates four contiguous stereocentres starting
from a simple sulfoxide starting material.
Entry
R
Time/h
Yield (%)a
Ratio 12 : 13
1
2
3
4
5
6
7
H
Me
Bn
Ac
Bz
TIPS
TBDMS
3.5
3
1
4
2.5
2
41b
65
83
80
74
45
59
1 : 1c
4 : 1d
3.6 : 1d
2 : 1c
2 : 1d
4.3 : 1d
4.9 : 1c
2.5
a Combined isolated yield of 12 and 13. b 14% recovered starting
material 11a. c Ratio of separated cycloadducts. d Ratio determined by
integration in 1H NMR of combined cycloadducts.
Acknowledgements
We thank the EPSRC for funding this work (studentship to
P. T., grant reference number GR/R20465/01).
Notes and references
§ Although the relative stereochemistry between the alcohol and the
sulfoxide in 11 is unimportant in determining the ratio of diastereomers
12 : 13 (since the sulfoxide stereocentre is destroyed in the process of
elimination to the sulfenic acid intermediate), the fact that the sulfoxide
stereocentre can be used to set the alcohol stereochemistry means that
this approach can be used to access enantiomerically pure compounds
via a chiral relay starting from enantiomerically pure tert-butyl methyl
sulfoxide.
¶ Crystal data for 12a: C9H15O2.50S, M = 195.27, colourless prism,
0.80 × 0.25 × 0.20 mm3, triclinic, space group P-1 (No. 2), a = 7.1304(3),
b = 10.9979(4), c = 13.4699(6) Å, α = 109.079(2), β = 92.0540(10),
γ = 100.681(2)Њ, V = 975.68(7) Å3, Z = 4, Dc = 1.329 g cmϪ3, F000 = 420,
Nonius KappaCCD diffractometer, MoKα radiation, λ = 0.71073 Å,
T = 120(2)K, 2θmax = 54.9Њ, 3808 reflections collected, 2637 unique
(Rint = 0.0534). Final GooF = 1.047, R1 = 0.0347, wR2 = 0.0832,
R indices based on 2203 reflections with I > 2σ(I ) (refinement on F 2),
239 parameters, 0 restraints. Lp and absorption corrections applied, µ =
0.298 mmϪ1. CCDC reference number 218530. See http://www.rsc.org/
suppdata/ob/b3/b314176d/ for crystallographic data in.cif or other
electronic format.
Crystal data for 13a: C9 H14 O2 S, M = 186.26, 0.20 × 0.15 ×
0.10 mm3, monoclinic, space group P21/n (No. 14), a = 7.5780(3),
b = 9.7836(4), c = 12.0885(5) Å, β = 92.028(2)Њ, V = 895.68(6) Å3, Z = 4,
Dc = 1.381 g cmϪ3, F000 = 400, Nonius KappaCCD diffractometer,
MoKα radiation, λ = 0.71073 Å, T = 120(2)K, 2θmax = 54.9Њ, 2782
reflections collected, 1640 unique (Rint = 0.0388). Final GooF = 1.069,
R1 = 0.0361, wR2 = 0.0779, R indices based on 1265 reflections with
I > 2σ(I ) (refinement on F 2), 115 parameters, 0 restraints. Lp and
absorption corrections applied, µ = 0.317 mmϪ1. CCDC reference
crystallographic data in.cif or other electronic format.
Scheme 3 Reagents and conditions: (i) Na, NH3, THF, t-BuOH then
piperylene, MeI, 93%; (ii) t-BuS(O)Me, LDA, THF, Ϫ78 ЊC, 76%;
(iii) DIBAL-H, THF, Ϫ78 ЊC, 96% (> 95% de); (iv) xylene, reflux; (v) b:
NaH, MeI, THF, 66%; c: NaH, BnBr, THF, 74%; d: Ac2O, cat. DMAP,
65 ЊC, 80%; e: BuLi, BzCl, THF, Ϫ78 ЊC, 54%; f: TIPSOTf, 2,6-lutidine,
CH2Cl2, 56%; g: TBDMSCl, imidazole, DMF, 49%.
case the stereochemistry of the major and minor products was
determined by comparison with alcohols 12a and 13a of known
configuration, either by O-alkylation of 12a and 13a (in the
case of R = Me and Bn), or deprotection of 12d–g and 13d–g
(TBAF in the case of silyl protecting groups, basic hydrolysis in
the case of ester protecting groups) to yield 12a and 13a.
In all cases the reaction gave rise to a major isomer 12b–g
with the bridgehead methyl and adjacent substituted oxygen cis
to one another. However, the diastereomeric ratio is apparently
not simply a reflection of the relative size of the group attached
to oxygen.14 Alkyl substituents gave rise to similar levels of
selectivity (entries 2 and 3). Lower and identical selectivity
is observed in the case of ester groups (entries 4 and 5). The
best selectivity is seen in the case of a silyl ether (entries
6 and 7), with the TBDMS group proving optimum in terms
of diastereomeric ratio, product stability, and ease of separ-
ation.
1 (a) M. Maier, in Organic Synthesis Highlights II, ed. H. Waldmann,
VCH, New York, 1995, p. 203; (b) C. S. Poss and S. L. Schreiber, Acc.
Chem. Res., 1994, 27, 9; (c) S. R. Magnuson, Tetrahedron, 1995, 51,
2167.
2 M. C. Willis, J. Chem. Soc., Perkin. Trans. 1, 1999, 1765.
3 Representative examples involving diastereotopic alkene and alkyne
groups: (a) S. F. Martin, S. K. Davidsen and T. A. Puckette, J. Org.
Chem., 1987, 52, 1962; (b) S. F. Martin and C. L. Campbell, J. Org.
Chem., 1988, 53, 3184; (c) P. Wipf and Y. Kim, Tetrahedron Lett.,
1992, 33, 5477; (d ) P. Wipf, S. R. Rector and H. Takahashi,
J. Am. Chem. Soc., 2002, 124, 14848; (e) D. Bland, D. J. Hart
and S. Lacoutiere, Tetrahedron, 1997, 53, 8871; ( f ) D. Bland,
G. Chambournier, V. Dragan and D. J. Hart, Tetrahedron, 1999, 55,
8953; (g) H. Fujioka, S. Kitagaki, N. Ohno, H. Kitagawa, Y. Kita
and K. Matsumoto, Tetrahedron: Asymmetry, 1994, 5, 333; (h) M. C.
Carreño, M. P. González, M. Ribagorda and K. N. Houk, J. Org.
Chem., 1998, 63, 3687; (i) M. C. Carreño, M. P. González and
K. N. Houk, J. Org. Chem., 1997, 62, 9128; (j) M. Suginome,
Y. Yamamoto, K. Fujii and Y. Ito, J. Am. Chem. Soc., 1995, 117,
The origin of the selectivity presented in Table 1 has yet to be
determined, but we believe it represents a thermodynamic
rather than kinetic preference. A number of separated sulf-
oxides 12 and 13 (R = H, Me, Ac, Bz, TBDMS) have been
independently resubjected to the reaction conditions (refluxing
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 1 5 1 – 1 5 3
152