A new radical-ionic allylation sequence

Article abstract of DOI:10.1055/s-2004-837191

Fluoride induced elimination of β-xanthylsilanes, formed by the radical addition of xanthates to allylsilanes, leads to the formation of diversely substituted allyl derivatives.

Full text of DOI:10.1055/s-2004-837191

334
LETTER
A New Radical-Ionic Allylation Sequence
A
New Radical-Ionic
A
i
llylati
c
on
S
equenc
h
e
ael E. Briggs, Samir Z. Zard*
Laboratoire de Synthèse Organique Associé au CNRS, Ecole Polytechnique, 91128 Palaiseau, France
Fax +33(1)69333851; E-mail: zard@poly.polytechnique.fr
Received 29 September 2004
Our approach is summarised in Scheme 1. Radical addi-
tion of a xanthate such as 1 to allyl silane 2 gives the cor-
responding adduct 3 by a radical chain process that can be
initiated using a small amount of peroxide.7 We hoped
that the action of the fluoride anion would take place at the
silicon atom rather than at the xanthate group and that the
latter would in fact simply act as a nucleofuge in the pro-
cess. Variously substituted a-xanthyl ketones and amides
1a–e underwent lauroyl peroxide (DLP) initiated radical
addition to allyltrimethylsilane 2a to give b-xanthylsi-
lanes 3a–e as shown by the examples compiled in Table 1.
Indeed, exposure to tetrabutylammonium fluoride indu-
ced clean elimination to afford the terminal olefins 4a–e
as the sole products. Typically short reaction times (<2 h)
were observed for both steps and the products were easily
isolated by chromatography on silica gel. The unopti-
mised yields were generally good (>70%) and a number
of useful functional groups were tolerated. The direct in-
troduction of the very useful Weinreb amide8 moiety in 3a
is noteworthy. It was possible to carry out the allylation in
a ‘one-pot’ procedure (67% yield from 1a to 4a) by re-
moval of the solvent and excess silane in vacuo prior to
the addition of TBAF, although the combined yield over
two steps was slightly higher (76%). The two-step proce-
dure has the added advantage of forming what is essential-
ly a masked olefin as an intermediate, which could be
retained in the molecule while further manipulations are
carried out.
Abstract: Fluoride induced elimination of b-xanthylsilanes,
formed by the radical addition of xanthates to allylsilanes, leads to
the formation of diversely substituted allyl derivatives.
Key words: radical addition, xanthates, silanes, olefins, elimina-
tion
Radical allylation reactions have become an important
and useful tool for organic chemists. The majority of these
reactions involve the use of either allyl stannanes1 or an
allyl phenyl sulfide2 or sulfone3 and a stoichiometric
amount of a tin-based mediator. With the exception of
Keck’s allyl sulfone, which was used to good effect in the
synthesis of (+)-pseudomic acid, it has only been possible
to prepare unsubstituted or simple, 2-substituted allyl
moieties. In the past we have demonstrated that it is pos-
sible to accomplish a radical allylation of iodides and
dithiocarbonates by exploiting the extrusion of sulfur di-
oxide from alkyl sulfonyl radicals generated upon addi-
tion–fragmentation of allyl alkyl sulfones.4 This allows
allylation, and indeed vinylation,5 procedures that do not
involve tin or other heavy metals. Porter et al. have also
reported the Lewis acid-promoted free radical allylation
of alkyl bromides with 3-silyl-1-butenes.6 We herein re-
port a new radical allylation sequence starting from easily
prepared xanthates and commercially available allyltri-
methylsilane. The use of readily accessible substituted
allyl silanes provides a route to more complex structures.
As the reaction with unsubstituted allylsilane proved suc-
cessful, we examined the possibility of extending the se-
quence to include substituted allylsilanes (Scheme 2). b-
Hydroxy-allylsilanes, prepared essentially as single dia-
stereoisomers according to the Yamamoto–Reetz one-pot
procedure, were chosen as substrates.9 Lauroyl peroxide
initiated radical addition of the xanthates to the substituted
allylsilanes afforded, as expected, the b-xanthylsilanes as
a mixture of two diastereoisomers. The ratio of products
was deduced by integration of the two TMS peaks ob-
served in the 1H NMR spectra. The stereochemistry of the
major product was assumed to be syn by analogy with the
work of Renaud et al. for a similar radical addition to allyl
silanes: the resulting radicals were then trapped with sul-
fonyl azides.10 The assumption is supported by the fact
that the diastereoselectivity increased with the size of the
xanthate and the silane.
S
S
OEt
SiMe3
S
2
O
SiMe3
O
S
OEt
(lauroyl peroxide)
1,2-dichloroethane
reflux
R
R
3
1
Bu4NF
THF, r.t.
O
R
4
Scheme 1
It is well documented that fluoride induced eliminations
of b-silyl halides and more recently b-silyl azides11
proceed via a concerted E2 mechanism. We hoped there-
fore that the diastereoselectivity observed in the radical
SYNLETT 2005, No. 2, pp 0334–0336
Advanced online publication: 17.12.2004
DOI: 10.1055/s-2004-837191; Art ID: D30104ST
© Georg Thieme Verlag Stuttgart · New York
0
1.
0
2.
2
0
0
5

LETTER
A New Radical-Ionic Allylation Sequence
335
S
O
SiMe3
OH
Table 1 Radical Addition/Fluoride Induced Elimination (S =
-SCSOEt)
O
R'
S
OEt
R
Xanthate 1
Adduct 3 (yield, %)
Product 4 (yield, %)
R
EtOCSS
6a–f syn
(lauroyl peroxide)
1a,d,f
+
+
O
SiMe3
Σ
O
O
1,2-dichloroethane
reflux
SiMe3
Σ
R'
O
R
R'
N
N
Me
OMe
Me
OMe
EtOCSS
HO
OH
SiMe3
6a–f anti
N
OH
4a (80)
1a
Me
OMe
5a,b
Bu4NF
THF, r.t.
3a (95)
O
Σ
O
R'
Σ
O
O
O
N
N
R'
Me
Ph
Me
+
R
R
SiMe3
N
4b (90)
Me
Ph
OH
7a–f E
7a–f Z
3b (72)
1b
O
O
Σ
Σ
SiMe3 OH
O
SiMe3 OH
O
1f
N
N
Ph
Ph
O
O
SiMe3
N
5c
6g 25%
O
O
EtOCSS
O
1c
4c (75)
Bu4NF
THF, r.t.
O
O
OH
Ph
3c (82)
O
Σ
Σ
O
Σ
Σ
7g 73%
O
Ph
Ph
1d
4d (88)
Scheme 2
SiMe3
SiMe3
Ph
3d (95)
O
O
in 73% yield. The radical addition in the synthesis of 6g
proceeded poorly because the olefinic partner 5c could not
be obtained pure from the reaction of the allyltrimethyl-
silane anion with the epoxide of allylbenzene, and a crude
sample had to be used.
O
F
F
We attempted to improve the diastereoselectivity of the
radical addition of the Weinreb amide 1a to the allyl-
silanes 2b,c by lowering the reaction temperature using
BEt3 as the radical initiator. Although an increase in the
diastereoselectivity of the addition was observed (up to
9:1) the yield of the products was reduced.
1e
4e (55)
F
3e (82)
addition to the allylsilanes would be reflected in the E/Z
ratio of the newly formed olefins. The syn isomer should
afford the Z-olefin and the anti isomer the E-olefin. How-
ever, the fluoride induced elimination in fact proved
problematic. The E/Z ratio failed to mirror the syn:anti ra-
tio (Table 2) and several by-products were observed. The
unexpected E/Z ratio may be attributable to isomerisation
of the less stable Z-isomer to the E-isomer by sulfur con-
taining by-products formed in the elimination step: thiols
are known to isomerise olefins through a reversible radi-
cal addition to the unsaturated bond.12
Table 2 Addition–Elimination to Substituted Allylsilanes
R in 1
R¢ in 5
Yielda
(%)
Syn/antib Yieldc
Z/E
(%)
CH3O(CH3)N, 1a Et, 5a
6a, 65
6b, 47
6c, 79
6d, 34
6e, 27
6f, 68
2:1
3:1
2:1
4:1
5:1
3:1
7a, 68
7b, 29
7c, 32
7d, 63
7e, 17
7f, 64
1:1
1:1
1:2
3:2
2:1
2:1
Ph, 1d
Et, 5a
Et, 5a
CH3, 1f
The yield in the elimination step was generally lower than
in the simple allylation. The by-products obtained were
found to contain olefin functionality but were lacking hy-
droxyl groups, suggesting that a competing reaction path
exists involving the vicinal alcohol function. The problem
may indeed be due to a competition between the xanthate
and the hydroxy as leaving groups or, perhaps more like-
ly, a competing Peterson olefination reaction. In line with
this reasoning is the observation that analogue 6g, which
possessed an extra methylene unit between the silyl group
and the alcohol, underwent smooth elimination to give 7g
CH3O(CH3)N, 1a Ph, 5b
Ph, 1d
Ph, 5b
CH3, 1f
Ph, 5b
a Combined yield of syn/anti diastereomers.
b Product ratio calculated from 1H NMR.
c Combined yield of Z/E isomers.
Synlett 2005, No. 2, 334–336 © Thieme Stuttgart · New York

336
M. E. Briggs, S. Z. Zard
LETTER
(8) Nahm, S.; Weinreb, S. Tetrahedron Lett. 1981, 39, 3815.
(9) (a) Reetz, M.; Steinbach, R.; Peter, R.; Wenderoth, B. Chem.
Ber. 1985, 118, 118. (b) Ikeda, Y.; Yamamoto, H. Bull.
Chem. Soc. Jpn. 1986, 59, 657.
(10) Chabaud, L.; Landais, Y.; Renaud, P. Org. Lett. 2002, 4,
425.
(11) Chabaud, L.; Landais, Y. Tetrahedron Lett. 2003, 44, 6995.
(12) Thalman, A.; Oertle, K.; Gerlach, H. Org. Synth. 1984, 63,
193.
In conclusion we have developed a new two-step method
for radical allylation which avoids the use of toxic tin
containing compounds.13 Of particular note is the ability
to introduce functionalised allyl moieties, which has pre-
viously been difficult to achieve. Further studies to better
delineate the scope of the process and to improve the
selectivity of the elimination step are nevertheless still
needed.
(13) Typical Experimental Procedures:
To a solution of xanthate 1c (0.63 g, 2.53 mmol, 1 equiv) and
trimethylallylsilane 2a (1.39 mL, 8.75 mmol, 3 equiv) in
refluxing degassed 1,2-dichloroethane (5.8 mL) was added
lauroyl peroxide (DLP) (0.050 g, 0.13 mmol, 0.05 equiv)
under N2 atmosphere. DLP (0.03 equiv) was added every 2
h until complete consumption of the starting material. 5%
DLP was needed to complete this reaction. The reaction was
allowed to cool to r.t. and the solvent removed in vacuo.
Purification by flash chromatography (EtOAc–petroleum
ether 3:17) gave xanthate 3c (0.75 g, 82%) as a light yellow
oil. 1H NMR (400 MHz, CDCl3): d = 4.58 (2 H, m,
OCH2CH3), 4.47 (2 H, t, J = 8.1 Hz, OCH2CH2N), 3.97 (2 H,
t, J = 8.1 Hz, OCH2CH2N), 3.86 [1 H, m, CHS(CS)OEt],
3.07 [1 H, m, N(CO)CHaHb], 2.93 [1 H, m, N(CO)CHaHb],
2.14 [1 H, m, N(CO)CH2CHaHb], 1.89 [1 H, m,
Acknowledgment
We thank Drs Fabien Gagosz and Gilles Ouvry for stimulating dis-
cussion and the Ecole Polytechnique for a grant (MEB). This paper
is dedicated with respect to the memory of Jean Mathieu.
References
(1) (a) Keck, G.; Yates, B. J. Am. Chem. Soc. 1982, 104, 5829.
(b) Enholm, E.; Kachensky, D.; Keck, G. Tetrahedron Lett.
1984, 25, 1867. (c) Enholm, E.; Keck, G.; Wiley, M.; Yates,
J. Tetrahedron 1985, 41, 4079. (d) Grignon, J.; Pereyre, M.
J. Organomet. Chem. 1973, 61, C33. (e) Kosugi, M.;
Kurino, K.; Takayama, K.; Migita, T. J. Organomet. Chem.
1973, 56, C11.
(2) Keck, G.; Byers, J. J. Org. Chem. 1986, 50, 5442.
(3) Keck, G.; Tafesh, A. J. Org. Chem. 1989, 54, 5845.
(4) (a) Quiclet-Sire, B.; Seguin, S.; Zard, S. Z. Angew. Chem.
Int. Ed. 1998, 37, 2864. (b) Quiclet-Sire, B.; Zard, S. Z. J.
Am. Chem. Soc. 1996, 118, 1209. (c) Le Guyader, F.;
Quiclet-Sire, B.; Seguin, S.; Zard, S. Z. J. Am. Chem. Soc.
1997, 119, 7410. (d) Bertrand, F.; Quiclet-Sire, B.; Zard, S.
Z. Angew. Chem. Int. Ed. 1999, 38, 1943. (e) For related
work, see: Xiang, J.; Fuchs, P. L. J. Am. Chem. Soc. 1996,
118, 11986. (f) Xiang, J.; Jiang, W.; Gong, J.; Fuchs, P. L. J.
Am. Chem. Soc. 1997, 119, 4123. (g) Gong, J.; Fuchs, P. L.
J. Am. Chem. Soc. 1996, 118, 4486. (h) Xiang, J.; Jiang, W.;
Fuchs, P. L. Tetrahedron Lett. 1997, 38, 6635.
N(CO)CH2CHaHb], 1.47 (3 H, t, J = 7.1 Hz, OCH2CH3),
1.08 (1 H, dd, 2J = 14.9 Hz and J = 7.2 Hz, CHaHbSiMe3),
1.01 (1 H, dd, 2J = 14.9 Hz and J = 8.1 Hz, CHaHbSiMe3),
0.03 (9 H, s, SiMe3). 13C NMR (400 MHz, CDCl3): d =
214.1, 172.3, 153.2, 69.4, 61.8, 47.6, 42.2, 32.1, 31.1, 22.9,
13.4, –1.1. MS (CI): m/z (%) = 242 (100) [M – S(CS)OEt].
To a solution of b-silylxanthate 3c (0.11 g, 0.33 mmol, 1
equiv) in THF (1.1 mL) at r.t. under a N2 atmosphere was
added a solution of TBAF (0.65 mL, 1 M in THF, 0.65
mmol, 2 equiv). After 3 h the solvent was removed in vacuo
and the residue was purified by column chromatography
(EtOAc–petroleum ether 2:8) to afford olefin 4c (0.042 g,
75%) as a light yellow oil. 1H NMR (400 MHz, CDCl3): d =
5.88–5.78 (1 H, m, CH=CHtHc), 5.06 (1 H, dq, J = 17.1 Hz,
2J and 4 J = 1.6 Hz, CH=CHtHc), 4.99 (1 H, dq, J = 10.2 Hz,
2J and 4 J = 1.4 Hz, CH=CHtHc), 4.40 (2 H, t, J = 8.2 Hz,
OCH2CH2N), 4.00 (2 H, t, J = 8.2 Hz, OCH2CH2N), 3.01
[2 H, t, J = 7.4 Hz, N(CO)CH2], 2.40 [2 H, m,
(5) Bertrand, F.; Quiclet-Sire, B.; Zard, S. Z. Angew. Chem. Int.
Ed. 1999, 38, 1943.
(6) Porter, N.; Zhang, G.; Reed, A. Tetrahedron Lett. 2000, 41,
5773.
N(CO)CH2CH2]. 13C NMR (400 MHz, CDCl3): d = 172.3,
153.2, 136.4, 115.3, 61.8, 42.1, 34.1, 27.8. MS (CI):
(7) For reviews on xanthate transfer additions, see: (a) Zard, S.
Z. Angew. Chem., Int. Ed. Engl. 1997, 36, 672. (b) Quiclet-
Sire, B.; Zard, S. Z. Phosphorus, Sulfur Silicon Relat. Elem.
1999, 153-154, 137. (c) Zard, S. Z. In Radicals in Organic
Synthesis, Vol. 1; Renaud, P.; Sibi, M. P., Eds.; Wiley-VCH:
Weinheim, 2001, 90.
+
m/z (%) = 187 (100) [MNH4 ], 170 (30) [M + H].
Synlett 2005, No. 2, 334–336 © Thieme Stuttgart · New York