On the stereochemistry of β-elimination of β-silyl azides

Article abstract of DOI:10.1016/S0040-4039(03)01800-8

Fluoride-mediated elimination of syn and anti β-silyl azides was shown to afford the corresponding (Z)- and (E)-olefins, respectively, demonstrating that β-elimination of β-silyl azides is stereospecifically anti.

Full text of DOI:10.1016/S0040-4039(03)01800-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 6995–6998
On the stereochemistry of b-elimination of b-silyl azides
Laurent Chabaud and Yannick Landais*
University Bordeaux-I, Laboratoire de Chimie Organique et Organome´tallique, 351, Cours de la Libe´ration,
33405 Talence Cedex, France
Received 3 June 2003; revised 18 July 2003; accepted 23 July 2003
Abstract—Fluoride-mediated elimination of syn and anti b-silyl azides was shown to afford the corresponding (Z)- and
(E)-olefins, respectively, demonstrating that b-elimination of b-silyl azides is stereospecifically anti.
© 2003 Elsevier Ltd. All rights reserved.
b-Silyl azides are useful intermediates for organic syn-
thesis which have yet received little attention.1 We
recently reported a stereocontrolled carbo-azidation of
chiral allylsilanes such as 1a–b which led to the forma-
tion of b-silyl azides 2a–b having up to three contiguous
stereogenic centers (Scheme 1).2 While the C1ꢀC2 rela-
tive configuration of 2a–b could be assigned without
ambiguities, the relation between C2 and C3 centers
mer, suggesting that b-silyl azides 2a–b had the
C2ꢀC3ꢀsyn relative configuration.5
In order to formerly establish the stereochemistry of
this reaction,3 we initiated a study on the fluoride-medi-
ated elimination of b-silyl azides of known relative
configuration. We report here our conclusions regard-
ing the stereochemistry of the b-elimination of b-silyl
azides.
1
was found to be more difficult to determine. H NMR
studies provided no conclusive information regarding
this relative configuration and we were unable to pro-
duce crystals for X-ray diffraction studies. It was thus
decided to convert 2a–b, using a transformation of
known stereochemical course, into compounds which
As mentioned above, there are few efficient methods to
prepare configurationally defined b-silylazides.1 One of
the most straightforward methods relies on the elec-
trophilic azidation of b-silylester enolates using trisyl
azide (triisopropylsulfonyl azide).6 As shown by the
pioneering studies of Panek, this transformation pro-
duce anti-b-silyl azides with high stereocontrol.1a,b,7 b-
silylesters 4a and 4b were thus prepared through
silyl-cupration of the corresponding unsaturated esters,8
and their enolate (generated using LDA) treated with
trisyl azide, eventually leading to b-silyl azides 5a–b in
reasonable yield as a unique diastereomer in each case
(Scheme 2).
1
structures would be more amenable to simple H NMR
structure determination. We thus envisioned carrying
out a fluoride-ion mediated b-elimination of the silicon
and the azido group.3 It was anticipated, based on the
known stereochemistry of b-silyl halide elimination in
similar conditions,4 that this transformation would pro-
ceed in an anti fashion. Treatment of the major
diastereomers 2a–b using tetrabutylammonium fluoride
(TBAF) led to (Z)-alkenes 3a–b, as a unique stereoiso-
Scheme 1.
Keywords: b-elimination reaction; silicon and compounds; azides.
* Corresponding author. Tel.: +33 (0)5 40 00 22 89; fax: +33 (0)5 40 00 62 86; e-mail: y.landais@lcoo.u-bordeaux1.fr
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)01800-8

6996
L. Chabaud, Y. Landais / Tetrahedron Letters 44 (2003) 6995–6998
Scheme 2.
Scheme 3.
sponding olefin 10c with the same ratio. Similarly, a
54/46 mixture of anti and syn-b-silyl azides 7a and 8a
led, in the same conditions, to the formation of a 55:45
mixture of (E)/(Z)-b-bromostyrenes 10c–d (entry 6).
These few examples demonstrate that, syn- and anti-
isomers affording (Z)- and (E)-olefins respectively, the
i-elimination of i-silyl azides is an anti-stereospecific
process (Scheme 5).
In order to obtain unambiguous answer regarding the
stereospecificity of the process, it was essential to carry
out the b-elimination, independently, on pure syn and
anti-b-silyl azides. As no method was available to
access rapidly to the syn-isomers of 5a–b, we turned
our attention to b-silyl azides such as 7a and 8a–b, the
preparation of which has recently been described and
their relative configurations well secured.9 Addition of
BrN3 onto E-vinylsilanes10 6a was thus shown to afford
the corresponding syn and anti-b-silyl azides 7a and 8a,
depending on the method used to generate BrN3. Treat-
ment of 6a using method A (Br2, NaN3, CH2Cl2, HCl)
led to the anti-isomer 7a with a 99:1 ratio (GC), while
treatment using method B (NBS, NaN3, dioxane–H2O)
produced the syn-isomer 8a with a >98:<2 ratio (1H
NMR) (Scheme 3). With non-styrenic substrates, only
the syn-isomer was produced as shown by the transfor-
mation of 6b into 8b (method B).
However, although all these data point towards an anti
process, b-elimination of diastereomerically pure anti-b-
silyl azide 5a unexpectedly produced a 90:10 E/Z mix-
ture of the corresponding olefin 10a (entry 1). This
indicates that when the silicon group is located on a
benzylic centre, the reaction probably follows a differ-
ent pathway from that observed for alkyl analogues.
Similar observations have been made recently by
Porter3 and earlier on by Fleming on related com-
pounds.4a A E1CB mechanism was proposed instead,
which may be rationalized by the stabilization of a
carbanion thus generated at the benzylic position. Par-
tial configurational inversion of this benzylic anion
(R1=Ph in A, Scheme 5), followed by azide elimination
would then overall result in a syn-elimination, thus
The above anti- and syn-b-silyl azides were then sub-
mitted to fluoride-mediated elimination and led to the
corresponding olefins 10a–e in high yield and in most
cases, in a stereospecific fashion (Scheme 4, Table 1).
As most olefins were quite volatile, the isolated yields
did not always match those estimated through GC
analysis, which therefore provide a more accurate pic-
ture about the efficiency of the elimination process.
Our results are summarized in Table 1. Some interest-
ing features emerge from these data. First, configura-
tionally pure anti-b-silyl azide 5b produces exclusively
the corresponding (E)-olefin 10b (entry 2), while syn-b-
silyl azides 8a–b furnished (Z)-olefins 10d–e (entries
4–5). As shown in entry 3, starting from a 99:1 anti/syn
isomeric mixture of b-silyl azide 7a led to the corre-
Scheme 4.

L. Chabaud, Y. Landais / Tetrahedron Letters 44 (2003) 6995–6998
Table 1. b-Elimination of b-silyl azides (Scheme 4).
6997
Entry
b-Silyl azides
anti/syn
Olefins
E/Z (%)
Yield (%)
1
2
3
4
5
6
5a
5b
7a
8a
8b
7a/8a
>98:<2a
>98:<2a
99/1b
10a
10b
10c
10d
10e
10c/10d
90:10a
88c
64d
97d
95d
99d
94d
>98:<2a
98:2b
<2:>98a
<2:>98a
54:46b
<2:>98b
<2:>98b
55:45b
a Estimated ratio by 1H NMR.
b Estimated ratio by GC analysis.
c Isolated yield.
d Estimated yield by GC analysis.
Scheme 5.
explaining the minor amount of (Z)-olefin formed dur-
ing elimination of 5a.11 Such a stabilization is not
present with analogue 5b, and the elimination in this
case follows a pure E2 pathway as indicated by the
stereospecificity of the process.
gratefully thanked for providing a copy of his results
prior to publication.
References
In conclusion, except in the particular case of benzylic
b-silyl azides (i.e. 5a), elimination of b-silyl azides
induced by fluorine-ion proceeds in an anti fashion. The
formation of less stable (Z)-olefins and the precedent in
the literature3,4 for anti-elimination of b-silyl acetates
and halides provide a strong support for this anti-
stereospecificity. This study finally confirms our
assumption that b-silyl azides 2a–b, prepared through
carbo-azidation of allylsilanes 1a–b, have the syn-
configuration.2 This result should be helpful for the
unambiguous determination of the relative configura-
1. (a) Panek, J. S.; Zhang, J. J. Org. Chem. 1993, 58,
294–296; (b) Sparks, M. A.; Panek, J. S. J. Org. Chem.
1991, 56, 3431–3438; (c) Grayson, E. J.; Whitham, G. H.
Tetrahedron 1988, 44, 4087–4094; (d) Duboudin, F.;
Laporte, O. J. Organomet. Chem. 1979, 174, C18–C20.
2. Chabaud, L.; Landais, Y.; Renaud, P. Org. Lett. 2002, 4,
4257–4260.
3. Conversion of b-silyl azides into olefins using fluoride-ion
catalysis has been described, but the stereochemistry of
the process has never been firmly established, see: (a)
Arimoto, M.; Yamaguchi, H.; Fujita, E.; Nagao, Y.;
Ochiai, M. Chem. Pharm. Bull. 1989, 37, 3221–3224; (b)
Materson, D. S.; Porter, N. A. Org. Lett. 2002, 4, 4253–
4256.
1
tion of b-silyl azides, through simple H NMR determi-
nation of the geometry of the olefins issued from their
b-elimination.
4. (a) Bernhard, W.; Fleming, I. J. Organomet. Chem. 1984,
271, 281–288; (b) Miller, R. B.; McGarvey, G. J. Org.
Chem. 1978, 43, 4424–4431; (c) Carey, F. A.; Toler, J. R.
J. Org. Chem. 1976, 41, 1966–1971; (d) Lambert, J. B.;
Finzel, R. B. J. Am. Chem. Soc. 1982, 104, 2020–2022.
5. The (Z)-stereochemistry of olefins 3a–b was determined
using NOE experiments and was confirmed by the easy
preparation of the seven-membered ring lactone 11 from
3b.
Acknowledgements
The authors thank the Region Aquitaine, the CNRS
and the Institut Universitaire de France for financial
support. L.C. thanks the Ministere de la Recherche for
a fellowship. Professor N. Porter (Duke University) is

6998
L. Chabaud, Y. Landais / Tetrahedron Letters 44 (2003) 6995–6998
P. G. J. Chem. Soc., Perkin Trans. 1 2000, 1173–1180.
6. Evans, D. A.; Britton, T. C.; Ellman, J. A.; Dorow, R. L.
10. (a) Koumaglo, K.; Chan, T. H. Tetrahedron Lett. 1984, 25,
717–720; (b) Jeffery, T. Tetrahedron Lett. 1999, 40, 1673–
1676.
J. Am. Chem. Soc. 1990, 112, 4011–4030.
7. Panek, J. S.; Beresis, R.; Xu, F.; Yang, M. J. Org. Chem.
1991, 56, 7341–7344.
11. This experiment does not constitute a formal proof for an
E1cb mechanism. Other mechanisms, although unlikely,
such as a syn-elimination may also be operative.
8. Fleming, I.; Waterson, D. J. Chem. Soc., Perkin Trans. 1
1984, 1809–1813.
9. Bassindale, A. R.; Kyle, P. A.; Soobramanien, M.; Taylor,