[1,5]-anion relay/[2,3]-wittig rearrangement of 3,3-bis(silyl) allyl enol ethers: Synthesis of useful vinyl bis(silane) species

Article abstract of DOI:10.1021/ol300004b

The [1,5]-anion relay/[2,3]-Wittig rearrangement of 3,3-bis(silyl) enol allyl ethers has been developed. This reaction provides an efficient method to synthesize versatile vinyl bissilanes, which can be transformed into trisubstituted vinylsilanes through

Full text of DOI:10.1021/ol300004b

ORGANIC
LETTERS
2012
Vol. 14, No. 4
1094–1097
[1,5]-Anion Relay/[2,3]-Wittig
Rearrangement of 3,3-Bis(silyl) Allyl Enol
Ethers: Synthesis of Useful Vinyl
Bis(silane) Species
Xianwei Sun,†,§ Jian Lei,†,§ Changzhen Sun,† Zhenlei Song,*,†,‡ and Linjie Yan†
Key Laboratory of Drug-Targeting of Education Ministry and Department of Medicinal
Chemistry, West China School of Pharmacy, State Key Laboratory of Biotherapy,
West China Hospital, Sichuan University, Chengdu 610041, P. R. China
zhenleisong@scu.edu.cn
Received January 3, 2012
ABSTRACT
The [1,5]-anion relay/[2,3]-Wittig rearrangement of 3,3-bis(silyl) enol allyl ethers has been developed. This reaction provides an efficient method to
synthesize versatile vinyl bissilanes, which can be transformed into trisubstituted vinylsilanes through a [1,4]-Brook rearrangement/alkylation
protocol using a wide range of electrophiles.
Bis(silyl) compounds1 are a special type of organosilane2
in organic synthesis. Although their great potential for
bifunctional reactivity makes them attractive synthons,3
they have been poorly investigated because of the lack
of suitable methods for their synthesis. Recently, we
reported a facile retro-[1,4]-Brook rearrangement of
3-silyl allyloxy silanes 1 that generated efficiently a
variety of Z-3,3-bis(silyl) enol derivatives 2 (Scheme 1).4
One focus of our studies on these compounds was the
generation of bis(silyl) allyl anion 3, which could lead to
transformations into other useful bis(silyl) building
blocks. Because initial attempts based on directed metala-
tion5 proved unsuccessful, we proposed a conceptually
new strategy (Scheme 2).
† Key Laboratory of Drug-Targeting of Education Ministry and Depart-
ment of Medicinal Chemistry, West China School of Pharmacy.
‡ State Key Laboratory of Biotherapy, West China Hospital.
§ These authors contributed equally.
(1) For studies on geminal bis(silyl) species, see: (a) Fleming, I.;
Floyd, C. D. J. Chem. Soc., Perkin Trans. 1 1981, 969. (b) Brook, A. G.;
Chrusciel, J. J. Organometallics 1984, 3, 1317. (c) Klumpp, G. W.;
Mierop, A. J. C.; Vrielink, J. J.; Brugman, A.; Schakel, M. J. Am. Chem.
Soc. 1985, 107, 6740. (d) Lautens, M.; Ben, R. N.; Delanghe, P. H. M.
Angew. Chem., Int. Ed. Engl. 1994, 33, 2448. (e) Princet, B.; Gariglio,
H. G.; Pornet, J. J. Organomet. Chem. 2000, 604, 186. (f) Hodgson,
D. M.; Barker, S. F.; Mace, L. H.; Moran, J. R. Chem. Commun. 2001,
Scheme 1. Initial Attempts To Synthesize Bis(silyl) Allyl Anion
by Directed Metalation
ꢀ
153. (g) Williams, D. R.; Morales-Ramos, A. I.; Williams, C. M. Org.
Lett. 2006, 8, 4393.
(2) For selective reviews on organosilanes, see: (a) Overman, L. E.;
Blumenkopf, T. A. Chem. Rev. 1986, 86, 857. (b) Panek, M.; Masse, C. E.
Chem. Rev. 1995, 95, 1293. (c) Langkopf, E.; Schinzer, D. Chem. Rev. 1995,
95, 1375. (d) Fleming, I.; Barbero, A.; Walter, D. Chem. Rev. 1997, 97, 2063.
(3) For reviews on geminal bimetallic species, see: (a) Marek, I.;
Normant, J. F. Chem. Rev. 1996, 96, 3241. (b) Marshall, J. A. Chem. Rev.
1996, 96, 31. (c) Marek, I. Chem. Rev. 2000, 100, 2887.
(4) (a) Song, Z. L.; Lei, Z.; Gao, L.; Wu, X.; Li, L. J. Org. Lett. 2010,
12, 5298. (b) Gao, L.; Lin, X. L.; Lei, J.; Song, Z. L.; Lin, Z. Org. Lett.
2012, 14, 158.
The key to this strategy lies in using O-allylated 2 as the
substrate. Deprotonation of 2 should occur regioselec-
tively on the allyl group, in which the proton in blue is
(5) For a review on directed metalation, see: Snieckus, V. Chem. Rev.
1990, 90, 879.
r
10.1021/ol300004b
Published on Web 01/25/2012
2012 American Chemical Society

containing a smaller bis(Me3Si)2CH group (2b), larger
bis(t-BuMe2Si)2CH group (2c), or electronically different
bis(PhMe2Si)2CH group (2d) all showed poor reactivities.
Scheme 2. [1,5]-Anion Relay/[2,3]-Wittig Rearrangement of
Bis(silyl) Allyl Enol Ether
Scheme 3. Reaction of Deuterium-Substituted 2a-D
While the desired product was obtained as expected,
some key mechanistic issues remained to be elucidated.
First, did the reaction proceed by a [1,5]-anion relay or by
some other pathways, such as directed metalation and
intermolecular proton shift? Second, did product 8a form
by [1,2]- or [2,3]-Wittig rearrangement, since either path-
way could yield the same result in this case? To address
these questions, the reaction of deuterium-substituted 2a-
D was performed and 8a-D was obtained as an E/Z = 1:1
mixture in 80% yield (Scheme 3). This result unambigu-
ously confirmed that the reaction proceeds by a [1,5]-anion
relay/[2,3]-Wittig rearrangement. Further support for the
[1,5]-anion relay was obtained when the reaction of 2k-E-
E,10 having the enol double bond as an E-configuration,
failed to give the desired product 8k. Apparently, the
oxy allyl anion 4 generated from 2k-E-E cannot adopt a
6-membered ring transition state.
To gain further mechanistic insight, the reaction of 2a
was quenched with D2O at 15 and 30 min to afford three
products 2a, 2a-D, and 8a (Scheme 4). No formation of 9
and 10 implies that the [1,5]-anion relay occurs immedi-
ately once the oxy allyl anion 4 forms by the initial
deprotonation. The quenching yield of 2a-D and 8a at
30 min implies that the [1,5]-anion relay should be faster
than the [2,3]-Wittig rearrangement. In addition, forma-
tion of 2a-D and not 11 suggests that, in the bis(silyl) allyl
anion, the carbon center attached to the silyl group has a
much more sterically accessible than the one in red adja-
cent to the bis(silyl) group, to generate the oxy allyl anion
4. Given that silicon can stabilize the R-carbanion through
a pÀd π-bonding interaction,6 a [1,5]-anion relay7,8 of 4 via
a 6-membered ring transition state may proceed to gen-
erate the thermodynamically more stable bis(silyl) allyl
anion 5. The process could be further driven by a [2,3]-
Wittig rearrangement9 of the resonance structure 6 to
generate allyloxy lithium 7. Here we report detailed studies
applying this novel approach.
The feasibility of the proposed reaction was quickly
established using 3,3-bis(triethylsilyl) allyl enol ether 2a4a
as the substrate and t-BuLi as the base, with 3.0 equiv of
HMPA in THF at À78 °C for 2 h. The desired bis(silyl)
allylic alcohol 8a was obtained in 83% yield (Table 1,
entry 3). HMPA as a cosolvent proved to be crucial for
good efficiency; using no or smaller amounts of HMPA led
to either no reaction or to only a moderate yield (entries 1
and 2). Inaddition, the reaction isalsohighlydependent on
the bis(silyl) group. As shown in entries 4À6, enol ethers
Table 1. Screening of Reactions Conditions
(8) Anion relay chemistry (ARC), first introduced by Smith III a
decade ago, features a negative charge migration in a “through-space”
fashion involving the shift of a silyl group from a carbon to an oxygen
atom. For reviews, see: (a) Smith, A. B., III; Adams, C. M. Acc. Chem.
Res. 2004, 37, 365. (b) Smith, A. B., III; Wuest, W. M. Chem. Commun.
2008, 5883. For the latest advances from this group, see: (c) Smith, A. B.,
III; Kim, W. S. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 6787. (d) Smith,
A. B., III; Tong, R. B.; Kim, W. S.; Maio, W. A. Angew. Chem., Int. Ed.
2011, 50, 8904. (e) Smith, A. B., III; Han, H.; Kim, W. S. Org. Lett. 2011,
13, 3328.
HMPA
entry
substrate (Si)a
(equiv)
product
yieldb
1
2
3
4
5
6
2a (Et3Si)
À
À
NR
2a (Et3Si)
1.5
3.0
3.0
3.0
3.0
8a
8a
8b
8c
8d
50%
83%
<5%
<5%
<5%
2a (Et3Si)
2b (Me3Si)
2c (t-BuMe2Si)
2d (PhMe2Si)
(9) For a review of the [2,3]-Wittig rearrangement, see: (a) Nakai, T.;
Mikami, K. Chem. Rev. 1986, 86, 855. For the latest advances, see: (b)
Druais, V.; Hall, M. J.; Corsi, C.; Wendeborn, S. V.; Meyer, C.; Cossy, J.
Org. Lett. 2009, 11, 935. (c) Shapland, P. D. P.; Thomas, E. J. Tetra-
hedron 2009, 65, 4201. (d) Sasaki, M.; Ikemoto, H.; Kawahata, M.;
Yamaguchi, K.; Takeda, K. Chem.;Eur. J. 2009, 15, 4663. (e)
Kitamura, M.; Hirokawa, Y.; Maezaki, N. Chem.;Eur. J. 2009, 15,
9911. (f) Anderson, J. C.; Davies, E. A. Tetrahedron 2010, 66, 6300. (g)
Ikemoto, H.; Sasaki, M.; Takeda, K. Eur. J. Org. Chem. 2010, 34, 6643.
(h) Hirokawa, Y.; Kitamura, M.; Kato, C.; Kurata, Y.; Maezaki, N.
Tetrahedron Lett. 2011, 52, 581. (i) Chandrasekhar, B.; Prasada, R. J.;
Venkateswara, R. B.; Naresh, P. Tetrahedron Lett. 2011, 52, 5921.
(10) 2k-E-E with the enol double bond as an E-configuration was
obtained in 15% yield by reacting 1a with E-crotyl tosylate for 15 h.
a Reaction conditions: 0.25 mmol of 2, 0.75 mmol of t-BuLi, and
0.75 mmol of HMPA in THF (2.5 mL) at À78 °C. b Isolated yields after
purification by silica gel column chromatography.
(6) (a) Brinkman, E. A.; Berger, S.; Brauman, J. I. J. Am. Chem. Soc.
1994, 116, 8304. (b) Chan, T. H.; Wang, D. Chem. Rev. 1995, 95, 1279.
(7) Reports of a [1,5]-anion relay via an intramolecular proton shift
are rare in the literature: (a) Maercker, A.; Eckers, M.; Passlack, M.
J. Organomet. Chem. 1980, 186, 193. (b) Maercker, A.; Stoetzel, R.
Chem. Ber. 1987, 120, 1695.
Org. Lett., Vol. 14, No. 4, 2012
1095

higher electron density than that adjacent to an anion-
destabilizing OR group. This may be the reason why the
[2,3]-Wittig rearrangement proceeds slowly.
Table 2. Scope of the [1,5]-Anion Relay/[2,3]-Wittig
Rearrangement Process
Scheme 4. Reaction of 2a Quenched with D2O
The scope of this reaction was then examined (Table 2).
The reaction is generally suitable for enol ethers with
substitution at the 2-position of the allyl chain, giving
goods yields (entries 1À6). With one substituent at the
3-position, two diastereomers were obtained, with the syn-
isomer being a major product (entries 7À10).11 The geo-
metry of the allylic double bond appears to have some
impact on the stereochemical outcome, since the reaction
of 2k-E gave a higher diastereoselectivity than that of 2k-
Z.12 Interestingly, when enol allyl ethers 2n and 2o sub-
stituted at both the 2- and 3-positions were used, the anti-
isomers were formed predominantly (entries 11 and 12).11
It is also noteworthy that when the allyl chain contained a
carbanion-stabilizing phenyl or trimethylsilyl group, 2p
failed to undergo the desired reaction (entry 13).13 Prob-
ably, the greater stability of the initially formed allyl anion
raises the energy barrier for formation of the 6-membered
ring transition state, making the subsequent [1,5]-anion
relay much more difficult.
a Reaction conditions: 2 (1.0 equiv), t-BuLi (3.0 equiv), and HMPA
(3.0 equiv) in THF at À78 °C. b Diastereoselectivity was determined by
1H NMR spectroscopy. c Isolated yields after purification by silica gel
column chromatography. d The calculated diastereoselectivity of the
reaction from pure 2k-Z was syn/anti = 67:33.
We rationalized the diastereoselectivity of the [2,3]-
Wittig rearrangement is based on Nakai’s model,9a which
features a folded-envelope transition state (Scheme 5).
Generally, for E-alkenes, the G group attached to the
carbanion prefers to adopt an equatorial orientation to
Scheme 5. Model Analysis of Diastereoselectivity in the [2,3]-
Wittig Rearrangement
(11) The stereochemistry of the syn-isomer of 8k was assigned by
transformation of 8k to the known vinyliodide compound ( Fuwa, H.;
Suzuki, T.; Kubo, H.; Yamori, T.; Sasaki, M. Chem.;Eur. J. 2011, 17,
2678. See Supporting Information for details). The stereochemistry was
also determined based on Greeves’ observations: In similar structures to
ours, Ha in the syn-isomers is generally 0.1À0.3 ppm downfield of the
equivalent atom in the anti-isomers; vicinal coupling constants between
Ha and Hb in the anti-isomers are 1.0À1.5 Hz higher than those in the
syn-isomers (Greeves, N.; Lee, W. M. Tetrahedron Lett. 1997, 38, 6445).
This method was also applied to the stereochemical assignment of the
syn-isomers of 8l and 8m and the anti-isomers of 8n and 8o.
(12) While no isomerization of the allyl chain in 2k-E was observed,
such isomerization occurred partially (from 94:6 to 73:27) via a [1,5]-anion
relay in the reaction of 2k-Z-Et with an Et substituted at the 3-position.
(Experiments were performed with 2k-E and 2k-Z-Et by quenching the
reactions with D2O at 40 and 30 min, respectively. See Supporting
Information for details.) Thus, the real diastereoselectivity obtained from
pure 2k-Z should be slightly lower than the calculated ratio 67:33.
(13) The reaction of 2p-SiMe3-D, which is deuterium-substituted at
the same position as 2a-D, gave 3,3-bis(triethylsilyl) aldehyde in 70%
yield and left the deuterium untouched. This result implies that the
initially formed allyl anion 4 undergoescleavage of the CÀO bond rather
than a [1,5]-anion relay.
avoid a pseudodiaxial interaction with the R group at
the 2-position, giving the anti-isomer as the predominant
product. The opposite syn-selectivity for 2k-E (R = H)
may be due to increased gauche repulsion with the 3-Me
group, which forces the bulky G group (;CHdC(SiEt3)2)
to adopt an axial orientation as in TS-E-2. However, when
the 2-position contains an additional substituent, such as
in 2n, the diaxial interaction between G and 2-Me groups
again dominates, making TS-E-1 more favorable and
1096
Org. Lett., Vol. 14, No. 4, 2012

Table 3. Scope of the [1,4]-Brook Rearrangement/Alkylation
Reaction
Scheme 6. One-Pot Transformation of 1a into 12a
bulky G group appears to suffer the effects of gauche and
diaxial interactions with the 3-H and 2-H groups, making
TS-Z-1 slightly favored and giving rise to the observed
poor syn-selectivity.
Elaboration of the bis(silyl) allylic alcohols to function-
allyincreased trisubstituted vinylsilaneswas thenexplored.
The process features a sequential [1,4]-Brook rearrange-
ment14/alkylation reaction based on a procedure mod-
ified from Takeda.15 As summarized in Table 3, the
reaction is suitable not only for a wide range of allylic
electrophiles (entries 2À7, 10, and 11) but also for
propargyl bromide (entry 8) and phenyl disulfide
(entry 9). Moreover, the leaving group of electrophiles
can include several choices such as halides and tosylate
(entry 1), with the latter one having not been reported in
previous work.
Combining the processes to transform 1a into 2a, with
subsequent generation of 8a and 12a, appeared to be
feasible in a one-pot procedure, which would be of sub-
stantial interest as a synthetic approach. Therefore we
tested this process and obtained the efficient one-pot
transformation of 1a directly into 12a with overall 33%
yield (Scheme 6).16
Here we have described a [1,5]-anion relay/[2,3]-
Wittig rearrangement of 3,3-bis(silyl) allyl enol ethers
to generate versatile 3,3-bis(silyl) allyl alcohols. We
have also demonstrated the synthetic value of this
approach by efficiently preparing a wide range of tri-
substituted vinylsilanes using a sequential [1,4]-Brook/
alkylation protocol. We hope this conceptually new
strategy can be applied as a general method in related
systems. Further investigations in this regard are cur-
rently underway.
a Reaction conditions: 8(1.0 equiv), electrophile (2.0 equiv), t-BuOLi (3.0
equiv), and CuCN (3.0 equiv) in DMF/THF = 5:3 as cosolvent at 25 °C.
b Isolated yields after purification by silica gel column chromatography.
yielding the anti-isomer as the major product. In contrast,
in the transition states TS-Z-1 and TS-Z-2 for 2k-Z, the
(14) For reviews on Brook rearrangement, see: (a) Brook, A. G. Acc.
Chem. Res. 1974, 7, 77. (b) Moser, W. H. Tetrahedron 2001, 57, 2065.
For the latest advances, see: (c) Schmitt, D. C.; Johnson, J. S. Org. Lett.
2010, 12, 944. (d) Sasaki, M.; Oyamada, K.; Takeda, K. J. Org. Chem. 2010,
75, 3941. (e) Boyce, G. R.; Johnson, J. S. Angew. Chem., Int. Ed. 2010, 49,
8930. (f) Song, Z. L; Kui, L. Z; Sun, X. W.; Li, L. J. Org. Lett. 2011, 13,
1440. (g) Hayashi, M.; Nakamura, S. Angew. Chem., Int. Ed. 2011, 50, 2249.
(h) Li, H.; Liu, L. T.; Wang, Z. T; Zhao, F.; Zhang, S. G; Zhang, W. X; Xi,
Z. F. Chem.;Eur. J. 2011, 17, 7399. (i) Sasaki, M.; Kondo, Y.; Kawahata,
M.; Yamaguchi, K.; Takeda, K. Angew. Chem., Int. Ed. 2011, 50, 6375. (j)
Martin, D. B. C.; Vanderwal, C. D. Chem. Sci. 2011, 2, 649.
Acknowledgment. This research was supported by the
NSFC (21172150, 21021001), the National Basic Research
Program of China (973 Program, 2010CB833200), and
Sichuan University (2011SCU04A18).
(15) (a) Taguchi, H.; Ghoroku, K.; Tadaki, M.; Tsubouchi, A.; Takeda,
T. Org. Lett. 2001, 3, 3811. (b) Taguchi, H.; Takami, K.; Tsubouchi, A.;
Takeda, T. Tetrahedron Lett. 2004, 45, 429. (c) Tsubouchi, A.; Enatsu, S.;
Kanno, R.; Takeda, T. Angew. Chem., Int. Ed. 2010, 49, 7089. (d) Smith,
A. B., III; Kim, W. S.; Tong, R. B. Org. Lett. 2010, 12, 588.
(16) It is necessary to remove the excess allyl bromide under reduced
pressure after the first step to avoid undesired side reactions in the next
two steps.
Supporting Information Available. Experimental pro-
cedures and spectral data for products (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 4, 2012
1097