Table 2 Sigmatropic rearrangements of 5a under reductive conditions
We clearly observed a beneficial effect in yield of 7a from
reducing the Lewis acidity of the medium and so sought to
improve this further. Thus, dichloroketene was generated in situ
by the alternative procedure that employs Et3N with di-
chloroacetyl chloride (elimination conditions8). When this
procedure was applied to 5a, we were delighted to find a very
clean conversion to the ketene Claisen product 7a. None of the
[1,3] rearranged product was observed with this new protocol.
This procedure was applied to the other substrates (Scheme 3,
Table 3, entries 2–4 and 6) and in each case single products were
obtained in high yield and as single diastereoisomers. Re-
arrangement of 5d (entry 4) gave 7d bearing a quaternary centre
as a single diastereoisomer. In the case of 5e, the 84/16 ratio of
E/Z isomers rearranged to an inseparable 84/16 ratio of
diastereoisomers of 7e, thus demonstrating that the rearrange-
ment was stereospecific.
Interestingly, the E double bond of 7d isomerised to the Z
isomer during silica gel chromatography.10 Isomerisations of E
to Z double bonds has been previously reported in nine-
membered ring lactones11 and indeed it has been shown that the
Z isomer is thermodynamically more stable than the E isomer in
nine-membered carbocycles.12 In order to evaluate the stability
of the E/Z isomers of our ten-membered ring thiolactone 7d, a
Monte Carlo conformational search was conducted (Macro
Model V6.513) and indeed the Z isomer was found to be 2.5
kcal/mol more stable than the corresponding E isomer.
In summary, we have reported the first examples of the
ketene Claisen rearrangement of 1,3-oxathianes. Under elim-
ination conditions for the in situ generation of dichloroketene, a
clean and smooth transformation of easily accessible camphor-
derived oxathianes of a,b-unsaturated aldehydes occurred to
give diastereomerically and enantiomerically pure thiolactones.
In addition to achieving the stereocontrolled construction of
tertiary and quaternary stereogenic centres, the current process
also converts readily available six-membered rings into the
much more difficultly accessible ten-membered rings. The
process could therefore have applications beyond our original
goals.
Yield
Entry
T/°C
t/h
Additive
Yield 7aa (%) 9aa (%)
1
2
3
4
5
r.t.
278
0
r.t.
r.t.
3
2
1
4
1
—
—
—
POCl3
DME
—
—
—
—
48
37
21
32
40
8
a Isolated yield after silica gel chromatography (light petroleum).
zinc. However, rather than obtaining the product derived from a
[3,3]-sigmatropic rearrangement 7a, 9a was isolated instead, a
product formally derived from a [1,3] rearrangement. Indeed,
under a number of conditions, 9a was the main product with
little or none of the required product 7a (Table 2, entries 1–3).
The main by-product in the reactions was hydroxythiol 4 which
was presumably formed through ZnCl2 mediated hydrolysis.
POCl3 has been previously employed to neutralise the Lewis
acidity of the reaction mixture in related cyclobutanone
formation from unreactive olefins with excellent results.9
However, in this case, 9a was still the major product (entry 4).
Finally, using a bidentate ligand,9 DME (2 eq.), to sequester
ZnCl2, the cycloenlarged product 7a derived from the [3,3]-sig-
matropic rearrangement was obtained in moderate yield,
together with a small amount of 9a. Gratifyingly, 7a was formed
as a single diastereoisomer, whose relative stereochemistry was
determined by NOE (see ESI†). Furthermore, exclusive reac-
tion occurred at the sulfur atom to furnish the ten-membered
ring thiolactone 7a. The formation of the two products 7a and
9a as single diastereoisomers can be rationalised by considering
the structure of the intermediate betaine. Attack at the equatorial
lone pair at sulfur (treating related compounds with m-CPBA
provides only the equatorial sulfoxide6) gives betaine 6
(Scheme 2) which, in the absence of naked ZnCl2, rapidly
rearranges via a chair transition state to give 7a. As a single
diastereomer of the betaine is formed and because the thioacetal
cannot invert, only a single isomer is possible, i.e. the
rearrangement is stereospecific.
We thank Pfizer and Merck for generous support and Prof.
Placido Neri for helpful discussion on the Monte Carlo
conformational search.
Formation of 9a is more difficult to rationalise and pre-
sumably arises from stabilisation of the enolate by ZnCl2. This
slows down the rate of the [3,3]-sigmatropic rearrangement and
presumably allows time for the thioacetal to ring open and form
the oxonium ion 8 (Scheme 2). Subsequent reaction of the zinc
enolate with the oxonium ion 8 then furnishes 9a.
Notes and references
1 E. J. Corey and A. Guzman-Perez, Angew. Chem., Int. Ed., 1998, 37,
388.
2 (a) R. Malherbe and D. Bellus˘, Helv. Chim. Acta, 1978, 68, 3096; (b) R.
Malherbe, G. Rist and D. Bellus˘, J. Org. Chem., 1983, 48, 860.
3 (a) M. Ishida, H. Muramaru and S. Kato, Synthesis, 1989, 562; (b) E. D.
Edstrom, J. Am. Chem. Soc., 1991, 113, 6690; (c) R. Maruya, C. A.
Pittol, R. J. Pryce, S. M. Roberts, R. J. Thomas and J. O. Williams, J.
Chem. Soc., Perkin Trans. 1, 1992, 1617; (d) C. J. Deur, M. W. Miller
and L. S. Hedegus, J. Org. Chem., 1996, 61, 2871.
4 (a) D. W. C. MacMillan, T. P. Yoon and M. Dong, J. Am. Chem. Soc.,
1999, 121, 9726; (b) D. W. C. MacMillan and M. Dong, J. Am. Chem.
Soc., 2001, 123, 2448; (c) D. W. C. MacMillan and P. Yoon, J. Am.
Chem. Soc., 2001, 123, 2911.
5 R. Oehrlein, R. Jeschke, B. Ernst and D. Bellus˘, Tetrahedron Lett.,
1989, 30, 3517.
6 V. K. Aggarwal, J. G. Ford, S. Forquerna, H. Adams, R. V. H. Jones and
R. Fieldhouse, J. Am. Chem. Soc., 1998, 120, 8328.
7 R. Annunziata, M. Cinquini, F. Cozzi, L. Raimondi and S. Stefanelli,
Tetrahedron Lett., 1987, 28, 3139.
Scheme 3
Table 3 Ketene Claisen rearrangement of 5 under elimination conditions
Yield 7
(%)
Entry
R
R1
5
t/h
8 W. T. Brady, H. G. Liddel and W. Vaughn, J. Org. Chem., 1966, 31,
626.
9 B. D. Johnston, E. Czyzewska and A. C. Oehlschlager, J. Org. Chem.,
1987, 52, 3693.
10 E–Z isomerization of 7d products confirmed by 1H and 13C NMR (see
ESI†).
11 M. R. Kling, G. A. McNaughton-Smith and R. J. K. Taylor, J. Chem.
Soc., Chem. Commun.., 1993, 21, 1593.
12 A. C. Cope, P. T. Moore and W. R. Moore, J. Am. Chem. Soc., 1959, 81,
3153.
13 F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, C. Caufield,
G. Chang, T. Hendrickson and W. C. Still, J. Comput. Chem., 1990, 11,
440.
1
2
Ph
Me
H
H
5a
5b
1.8
1.5
95ab
67a
70a
96b
3
4
H
5c
1
Ph
Me
5d
2.5
5
6
Me
H
5e
5f
1
76b
91b
p-NO2-C6H4
1.5
a Isolated yield after recrystallization of the crude reaction mixture from
pentane/absolute ethanol at 220 °C. b Isolated yield after silica gel (pre-
treated with 1% of Et3N) chromatography (Et2O/light petroleum, 2/98).
CHEM. COMMUN., 2002, 2534–2535
2535