Stereoselective intermolecular carboazidation of chiral allylsilanes

Article abstract of DOI:10.1021/ol026824y

(equation presented) Easily available chiral allylsilanes were used as substrate for carboazidation reactions. For the first time, a substantial control of the stereochemistry of the azidation of acyclic nonconjugated radicals was achieved.

Full text of DOI:10.1021/ol026824y

ORGANIC
LETTERS
2002
Vol. 4, No. 24
4257-4260
Stereoselective Intermolecular
Carboazidation of Chiral Allylsilanes
,†
,‡
Laurent Chabaud,† Yannick Landais,* and Philippe Renaud*
UniVersite´ de Bordeaux 1, Laboratoire de Chimie Organique et Organome´tallique,
351, Cours de la Libe´ration, F-33405 Talence Cedex, France,
and UniVersity of Berne, Department of Chemistry and Biochemistry, Freiestrasse 3,
CH-3000 Berne 9, Switzerland
philippe.renaud@ioc.unibe.ch
Received August 30, 2002
ABSTRACT
Easily available chiral allylsilanes were used as substrate for carboazidation reactions. For the first time, a substantial control of the
stereochemistry of the azidation of acyclic nonconjugated radicals was achieved.
The use of free radical reactions in multistep synthesis has
steadily increased over the last years, mainly because of their
compatibility with a large number of functional groups and
their high potential to perform sequential transformations.1
Recently, we have developed a novel method allowing the
efficient formation of carbon-nitrogen bonds by reaction
of radicals with sulfonyl azides.2 This process has been
extended successfully to intra- and intermolecular carboazi-
dation reactions.3,4 This last reaction has led to a very short
and efficient two-step procedure for the preparation of the
backbone of pyrrolizidine and indolizidine alkaloids (see
Scheme 1 for a typical example). However, the control of
Scheme 1. Two-Step Preparation of 3-Methylpyrrolizidinone
by Carboazidation of 5-Bromopent-1-enea
* Email for additional corresponding author: y.landais@lcoo.u-bor-
deaux.fr.
† University of Bordeaux.
‡ University of Berne.
a (a) PhSO2N3 (3 equiv), (Bu3Sn)2 (1.5 equiv), tBuONdNOtBu
(3 mol %), refluxing benzene. (b) In/NH4Cl, refluxing EtOH; then
Et3N (5 equiv)/refluxing EtOH.
(1) For a general review on radical reactions, see: Radicals in Organic
Synthesis; Renaud, P., Sibi, M., Eds; Wiley-VCH: Weinheim, 2001; Vols.
1 and 2. Giese, B. Radicals in Organic Synthesis: Formation of Carbon-
Carbon Bonds; Pergamon: Oxford, 1988. Curran, D. P. In ComprehensiVe
Organic Synthesis; Trost, B. M., Fleming, I., Eds; Pergamon: Oxford,
1991; Vol. 4, p 715. Motherwell, W. B.; Crich, D. Free Radical Chain
Reactions in Organic Synthesis; Academic Press: London, 1992. Fossey,
J.; Lefort, D.; Sorba, J. Free Radicals in Organic Synthesis; Wiley:
Chichester, 1995. Chatgilialoglu, C.; Renaud, P. In General Aspects of
the Chemistry of Radicals; Alfassi, Z. B., Ed.; Wiley: Chichester, 1999; p
501.
the stereochemistry of the intermolecular carboazidation of
acyclic olefins was not possible in the systems examined so
far.
The control of the stereoselectivity of radical reactions has
reached a level that was unexpected a decade ago.5 However,
the stereoselectivity of reactions of acyclic alkyl radicals that
are not substituted by strongly stabilizing moieties such as
(2) Ollivier, C.; Renaud, P. J. Am. Chem. Soc. 2000, 122, 6496.
(3) Ollivier, C.; Renaud, P. J. Am. Chem. Soc. 2001, 123, 4717.
(4) Renaud, P.; Ollivier, C.; Panchaud, P. Angew. Chem., Int. Ed. 2002,
41, 3460.
10.1021/ol026824y CCC: $22.00 © 2002 American Chemical Society
Published on Web 10/30/2002

esters, alkoxy, amino, and aryl groups is usually not
controlled.6 Recently, Porter has observed a very interesting
E/Z stereocontrol in Lewis acid promoted radical addition
to branched allylsilanes.7 This result was interpreted as a
consequence of a diastereoselective radical bromine atom
transfer. We anticipated that such stereocontrol should also
be operating with other electrophilic radical traps such as
sulfonyl azides.8 We report here our first attempt to control
the stereochemistry of the azidation step by using chiral
allylsilanes as substrates for the carboazidation process.9
Chiral substituted allylsilanes are easily available by well-
established protocols (Scheme 2). For instance, 1 and 2 were
phenylsilane, a low yield of anti-6 was obtained along with
a large amount of the γ-alkylation product (not shown). With
allyltriphenylsilane, the coupling occurred mainly at the
R-position and provides anti- and syn-7 in good yield. The
diastereomers were easily separated by chromatography.13
Allylsilanes 1-7 were then submitted to carboazidation
conditions (eq 1). A mixture of allylsilane (2 equiv) and
ethyl[(ethoxycarbonothioyl)thio]acetate (1 equiv) was heated
under reflux in benzene in the presence of PhSO2N3 (3
equiv), (Bu3Sn)2 (1.5 equiv), and a substoichiometric amount
of di-tert-butylhyponitrite (1,2-di-tert-butoxydiazene) as
initiator. The results are presented in Table 1. â-Azidosilanes
Scheme 2. Chiral Allylsilanes and Their Preparation
Table 1. Carboazidation of Chiral Allylsilanes 1-7 According
to Eq 114
entry
silane
product
syn/antia,b
yield (%)
1
2
3
4
5
6
7
1
2
3
4
5
6
8
9
73:27
80:20
63:37
>90:10
90:10
64
55
71
47
29c
70
30c
10
11
12
13
14
80:20
g80:20d
anti-7
a See below for the attribution of the relative stereochemistry. b Ratio
estimated from 1H NMR of the crude reaction mixture. c Yield of isolated
major diastereomer after several purifications by flash chromatography.d Not
determined with precision because of purification problems.
prepared in 70% yield through silyl-cupration of the corre-
sponding allylic chlorides.10 â-Hydroxysilanes 3-5 were
obtained in diastereomerically pure form (anti, >98%) from
the corresponding allylsilanes following Yamamoto-Reetz
one-pot protocol involving deprotonation of the allylsilane
and then transmetalation with Ti(Oi-Pr)4 followed by cou-
pling with suitable aldehydes.11 Finally, γ-hydroxysilanes
anti-6 and anti/syn-7 were prepared by metalation of the
allyl(dimethyl)phenylsilane and allyltriphenylsilane followed
by reaction with styrene oxide.12 With allyl(dimethyl)-
8-14 were obtained in moderate to good yields and
diastereoselectivities. The preliminary investigations carried
out on simple allylsilanes 1 and 2 led to reasonable levels
of diastereocontrol. The syn â-azidosilanes are formed as
the major isomer (vide infra). Diastereomeric ratios were
easily estimated using 1H NMR, and the major isomers were
isolated by chromatography. However, we were unable to
isolate the minor isomer in diastereomerically pure form. The
best levels of diastereocontrol were obtained with allylsilanes
4 and 5 having a â-hydroxyl group. Moving the hydroxy
group one carbon away, as in γ-hydroxysilane 6, had no
deleterious effect on the diastereocontrol since â-azidosilane
13 was obtained as a 80:20 syn/anti mixture of diastereomer
in 70% yield (entry 6). In the last case (entry 7), it was not
possible to determine precisely the diastereoselectivity from
1H NMR of the crude product but flash chromatography
afforded only one diastereoisomer of 14 in 30% yield.
(5) For a general review, see: Curran, D. P.; Porter, N. A.; Giese, B.
Stereochemistry of Radical Reactions; VCH: Weinheim, 1995.
(6) For a review on acyclic radical, see: Giese, B. In Radicals in Organic
Synthesis; Renaud, P., Sibi, M. P., Eds; Wiley-VCH: Weinheim, 2001;
Vol. 1, p 381.
(7) Porter, N. A.; Zhang, G.; Reed, A. D. Tetrahedron Lett. 2000, 41,
5773.
(8) The silyl group is expected to stabilize polarized transition states
bearing a partially positive charge in â-position. Therefore, transition state
conformations allowing such a stabilization should be favored. See refs 20
and 21.
(9) Diasteroselective azidation of â-silyl Barton ester are reported by
Masterson and Porter in the preceding paper: Porter, N. A.; Masterson, D.
S. Org. Lett. 2002, 4, 4253-4256.
(10) Laycock, B.; Kitching, W.; Wickman, G. Tetrahedron Lett. 1983,
24, 5785.
The relative configuration of our products was determined
through a chemoselective â-elimination of â-silyl azides
(Scheme 3).15 Treatment of major azide 10 with a 1 M
(11) Reetz, M. T.; Steinbach, R.; Westermann, J.; Peter, R. Wenderoth,
B. Chem. Ber. 1985, 118, 1441. Ikeda, Y.; Yamamoto, H. Bull. Chem. Soc.
Jpn. 1986, 59, 657.
(13) The relative stereochemistry of 6 and 7 was assigned by analogy to
the reaction with propylene oxide where product stereochemistry was proven
by X-ray crystallography (results to be published).
(12) Schaumann, E.; Kirschning, A. Tetrahedron Lett. 1988, 29, 4281.
4258
Org. Lett., Vol. 4, No. 24, 2002

a partial positive charge would appear at the carbon atom as
a result of the high electrophilicity of the radical trap (see
Figure 1), and this partial positive charge is best stabilized
Scheme 3. Determination of Relative Configuration of
2-Azidosilanesa
a syn and anti refer to the relative configuration at the two
adjacent centers bearing the silyl and the azido groups.
solution of TBAF at room temperature led stereospecifically
to the corresponding (Z)-allylic alcohol 15 in 40% yield.16
Interestingly, under these reaction conditions, the azido group
fragments faster than the hydroxy group. Similarly, the azide
11 afforded the (Z)-allylic alcohol 16 in 83% yield. Assuming
the anti-stereospecificity for this fluoride-mediated Peterson-
like elimination, it is possible to correlate the (Z)-configured
allylic alcohols 15 and 16 with syn-10 and syn-11 (Scheme
3).17
With the relative configuration of our â-azidosilanes in
hand, we then attempted to rationalize the stereochemistry
of the process. Since the intermediate radicals are noncon-
jugated, none of the classical model to rationalize the
stereochemistry of radical reactions can be applied.18 How-
ever, Curran and Giese have reported one example of high
1,2-asymmetric induction in nonconjugated acyclic radicals.19
The model they propose to rationalize the stereochemical
outcome is not applicable to our system because it requires
a bulky substituent at the radical center to achieve a high
stereocontrol. However, their model is based on the need to
pyramidalize the transition state into roughly staggered
conformation to avoid gauche interactions between large
groups. The same principle should apply to our system
together with the well-established behavior of silanes that
should favor an approach of electrophilic reagents anti to
silyl groups.20 Indeed, it was anticipated that, as bonding
between the radical center and the sulfonyl azide develops,
Figure 1. Rationalization of the 1,2-asymmetric induction observed
during the carboazidation of chiral allylsilanes.
by a coplanar electron-rich C-Si bond (silicon â-effect).21
On the basis of these two assumptions (pyramidalized
staggered transition state and silicon â-effect) we propose
the models A and B depicted in Figure 1 to explain the
stereochemical outcome of the carboazidation of chiral
allylsilanes. These two models are characterized by (1) a
quasi staggered transition state; (2) the orthogonal relation-
ship between the bulky silyl group and the CH2CH2CO2Et
substituent at the radical center; and (3) the formation of a
C-N bond nearly anti to the silyl group. Model A, leading
to the major syn product, is favored relative to model B by
the absence of steric interactions between the incoming
sulfonyl azide and the CH2CH2CO2Et substituent.22 This
model fits well with the experimental results: increasing the
size of the silyl group (entries 3/5, Table 1) and the size of
the R′ group (entries 3/4, Table 1) are both leading to an
enhancement of the diastereoselectivity.
We have demonstrated that the carboazidation of chiral
allylsilanes enables the stereoselective construction of poly-
functional â-silyl azides having up to three contiguous
stereogenic centers. Further elaboration of these substrates
is at hand. For instance, oxidation with retention of config-
uration of the C-Si bond should allow for the introduction
of an additional hydroxy group.23 Moreover, the results
presented here extend considerably the scope of the car-
boazidation process for the asymmetric synthesis of alkaloids
such as pyrrolizidinones and indolizidinones. Finally, this
work demonstrates further the unique role of silyl group in
the control of the stereochemistry of nonconjugated acyclic
radicals. Further applications of this concept in radical-
mediated asymmetric synthesis are anticipated.
(14) Experimental Procedure. Caution: as sulfonyl azides are capable
of exploding, it is strongly recommended to apply standard safety rules
and to use a safety shield. A solution of ethyl [(ethoxycarbonothioyl)thio]-
acetate (0.05 mL, 0.39 mmol), allylsilane (0.39 mmol), benzenesulfonyl
azide (192 mg, 1.05 mmol), (Bu3Sn)2 (0.27 mL, 0.52 mmol), and di-tert-
butylhyponitrite (2.5 mg, 0.015 mmol) in benzene (1.5 mL) was heated
under reflux. The reaction was monitored by TLC and further portions of
di-tert-butylhyponitrite (2.5 mg, 0.015 mmol) were added every 1.5 h. Upon
completion, the solvent was removed under reduced pressure, the crude
product was filtered through silica gel (hexane, then hexane/EtOAc 9:1),
and the combined fractions were concentrated in vacuo. The residue was
purified by chromatography (hexane/EtOAc).
(15) Arimoto, M.; Yamaguchi, H.; Fujita, E.; Nagao, Y.; Ochiai, M.
Chem. Pharm. Bull. 1989, 37, 12, 3221. Bernhard, W.; Fleming, I. J.
Organomet. Chem. 1984, 271, 281.
(16) The coupling constant J ) 9 Hz between the two vicinal olefinic
protons and NOESY experiments allowed us to assign unambiguously the
(Z)-stereochemistry for the double bond of 15 and 16.
(17) A detailed study of the stereospecificity of this elimination is cur-
rently underway in Bordeaux. Chabaud, L.; Landais, Y. Unpublished results.
(18) The stereochemical outcome of radical addition to branched
allylsilane was rationalized by a Felkin-Anh transition state (ref 7).
(19) Thoma, G.; Curran, D. P.; Geib, S.; Giese, B.; Damm, W.; Wetterich,
F. J. Am. Chem. Soc. 1993, 115, 8585.
(21) A radical â to a SiR3 group is known to be stabilized by ∼3 kcal/
mol: Auner, N.; Walsh, R.; Westrup, J. J. Chem. Soc., Chem. Commun.
1986, 207.
(22) For a related example, see: Angelaud, R.; Landais, Y. Tetrahedron
Lett. 1997, 38, 233.
(23) Fleming, I. Chemtracts: Org. Chem. 1996, 9, 1. Landais, Y.; Jones,
G. Tetrahedron 1996, 52, 7599. Tamao, K. In AdVances in Silicon
Chemistry; Jai Press Inc.: Greenwich, CT 1996; Vol. 3, p 1.
(20) Fleming, I.; Dunogue`s, J.; Smithers, R. Org. React. 1989, 37, 57.
Fleming, I.; Barbero, A.; Walter, D. Chem. ReV. 1997, 97, 2063.
Org. Lett., Vol. 4, No. 24, 2002
4259

Acknowledgment. This work was supported in France
by CNRS and the Re´gion Aquitaine and in Switzerland
by the National Science Foundation and the Federal Office
for Science and Education (OFES/BBW). We are very
grateful to Professor Ned A. Porter for helpful discussions.
as well as for informing us about his results prior to
publication.
Supporting Information Available: Representative ex-
perimental procedures including preparation and character-
ization of allylsilanes 1-7, their carboazidation, and the anti
elimination of 10 and 11. This material is available free of
charge via the Internet at http://pubs.acs.org.
OL026824Y
4260
Org. Lett., Vol. 4, No. 24, 2002