A general catalytic allylation using allyltrimethoxysilane

Article abstract of DOI:10.1021/ja0262582

A general and mild catalytic allylation of carbonyl compounds, applicable to aldehydes, ketones, and imines is developed using allyltrimethoxysilane as the allylating reagent. The reaction proceeds smoothly with 1-10 mol % of CuCl and TBAT in THF at ambie

Full text of DOI:10.1021/ja0262582

Published on Web 05/15/2002
A General Catalytic Allylation Using Allyltrimethoxysilane
Shingo Yamasaki,† Kunihiko Fujii,† Reiko Wada,† Motomu Kanai,†,‡ and Masakatsu Shibasaki*,†
Graduate School of Pharmaceutical Sciences, The UniVersity of Tokyo, Tokyo 113-0033, Japan, and PREST, Japan
Science and Technology Corporation, Tokyo 113-0033, Japan
Received March 21, 2002
Table 1. Catalytic Allylation Using Allylsilanea
The allylation of carbonyl derivatives such as aldehydes, ketones,
and imines is one of the most important carbon-carbon bond-
forming reactions because of the versatility of homoallylic alcohols
and amines as synthetic intermediates.1 Among various allylmetal
reagents, allylsilanes and allylstannanes are very useful because of
their moderate reactivity, which can be increased by catalyst
activation and thus allows for application to catalytic enantio-
selective reactions. Allylsilanes are generally more desirable,
particularly for environmental reasons, because they are less toxic
and more stable than allylstannanes. The low reactivity of allyl-
silanes, however, has limited their synthetic utility.2 The present
contribution describes a new catalytic allylation methodology using
allylsilanes that can be applied to a broad range of carbonyl
derivatives including imines.
To overcome the low reactivity of allylsilanes, our initial idea
was to use the dual activation concept: if a catalyst activates both
an allylsilane and a carbonyl substrate simultaneously, high catalyst
activity is obtained.3 A fluoride anion can activate the silylated
nucleophiles.4 Therefore, we investigated the combination of a
fluoride anion and a Lewis acid metal to activate an allylsilane
and a substrate, respectively.5 Allyltrimethoxysilane should be more
suitable for the present purpose than allyltrimethylsilane because
it can form a pentacoordinate silicate6 more easily than allytri-
methylsilane. Thus, using tetrabutylammonium difluorotriphenyl-
silicate (TBAT; 10 mol %)7 and allyltrimethoxysilane8 (1.5 equiv),
the combination was screened with various Lewis acids for the
reaction of benzaldehyde as a substrate. Although combinations
with hard Lewis acids such as Et2AlCl, ZrCl4‚2THF, and HfCl4‚
2THF did not promote the reaction at all, a facile and clean
conversion was obtained using soft metals such as AgCl (60% yield)
and CuCl (100% yield; 1.5 h). Completely optimized reaction
conditions were determined (1 mol % of CuCl and TBAT in THF
at room temperature) after surveying the copper salts, the fluoride
anion sources, solvents, and the catalyst loading.9
As shown in Table 1, the reaction proved to be general and could
be applied to a broad range of aldehydes, ketones, and imines.
Specifically, the reaction proceeds smoothly under very mild
conditions at ambient temperature and neutral pH with a syntheti-
cally acceptable catalyst loading (1-10 mol %). The applicability
a For experimental procedures, see Supporting Information. b Isolated
to the relatively complex substrate 1h clearly demonstrates these
advantages (entry 8).10 The reaction was completely chemoselective,
and the lactone was tolerated. Enone 3h gave the 1,2-adduct 4h
with complete regioselectivity (entry 16).11 Imines, including
aliphatic imines and a ketoimine, gave the allylated compounds in
yield. c Desilylation was conducted with TBAF. d A mixture of diastereo-
t
mers was used. e 2 equiv of allylsilane was used. BuOH (1 equiv) was
slowly added for 5 h.
crotylsilanes, and the reactions with 1a proceeded with complete
γ-selectivity, giving syn-7 as the major isomer from both E- and
Z-crotylsilanes (Scheme 1). Therefore, this new allylation reaction
is general and synthetically useful.
t
high yield with slow addition (5 h) of a proton source, BuOH
(entries 17-20).12 The present reaction was also applicable to
To gain insight into the reaction mechanism, we traced the
reaction procedure observing 19F NMR in THF (Scheme 2).13 The
reaction of CuCl with TBAF (1:1) afforded a new 19F NMR peak
(-155.7 ppm, broad) that was assigned to be CuF on the basis of
* To whom correspondence should be addressed. E-mail: mshibasa@
mol.f.u-tokyo.ac.jp.
† The University of Tokyo.
‡ PRESTO.
9
6536
J. AM. CHEM. SOC. 2002, 124, 6536-6537
10.1021/ja0262582 CCC: $22.00 © 2002 American Chemical Society

C O M M U N I C A T I O N S
Scheme 1. Catalytic Crotylation
Acknowledgment. Financial support was provided by RFTF
of Japan Society for the Promotion of Science and PRESTO of
Japan Science and Technology Corporation (JST).
Supporting Information Available: Experimental procedures and
characterization of the products (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
Scheme 2. Proposed Scheme for Generation of Reactive Species
Scheme 3. Catalytic Enantioselective Allylation of Acetophenone
References
(1) Yamamoto, Y.; Asao, N. Chem. ReV. 1993, 93, 2207.
(2) Only a few examples of catalytic allylation of ketones using allylsilanes
have been reported: (a) Hosomi, A.; Shirahata, A.; Sakurai, H. Tetrahe-
dron Lett. 1978, 3043 (cat. TBAF, THF reflux). (b) Ishihara, K.; Hiraiwa,
Y.; Yamamoto, H. Synlett 2001, 12, 1851 (cat. HNTf2, room temperature).
To our knowledge, there is no synthetically useful catalytic allylation of
ketoimines and aliphatic aldimines using allylsilanes.
(3) For an extension of this concept to enantioselective catalysis, see:
Shibasaki, M.; Kanai, M. Chem. Pharm. Bull. 2001, 49, 511.
(4) For selected examples using activated nucleophiles with a catalytic amount
of fluorides, see: (a) Nakamura, E.; Shimizu, M.; Kuwajima, I.; Sakata,
J.; Yokoyama, K.; Noyori, R. J. Org. Chem. 1983, 48, 932. (b) Pagenkopf,
B. L.; Carreira, E. M. Chem. Eur. J. 1999, 5, 3437.
the chemical shift of the reported CuF‚3PPh3‚2EtOH14 (8). Upon
addition of allyltrimethoxysilane (1 equiv) to the mixture, the CuF
peak disappeared, and a new peak (-129.0 ppm) appeared, which
was assigned to be allylfluorodimethoxysilane 9 on the basis of
the chemical shift and the presence of a doublet (J ) 112 Hz)
satellite peak derived from coupling with 29Si. Therefore, CuOMe
should be generated at this stage. On one hand, the addition of
aldehyde 1a to this reaction mixture (CuOMe and 9) did not give
any product, at least in 3 h. On the other hand, further addition of
allyltrimethoxysilane (1 equiv) to the mixture produced 2a. Thus,
CuOMe and allyltrimethoxysilane are necessary to generate the
reactive species.
At this point, we hypothesized that the fluoride ion might only
work as an initiator to generate the copper alkoxide and that direct
entry into the catalytic cycle might be possible with the copper
alkoxide and allyltrimethoxysilane, deduced from the mechanism
of the catalytic enantioselective aldol reaction reported by Car-
reira.15,16 In the present system, however, this was not the case;
only a trace amount (<1%) of 2a was obtained from 1a in the
presence of CuOtBu17 (10 mol %) and allyltrimethoxysilane (1.5
equiv). When (EtO)3SiF (10 mol %) was added as an analogue of
9, however, the reaction proceeded, and 2a was obtained in 27%
yield (4 h).18 Because addition of Ph3SiF, instead of (EtO)3SiF,
produced no product, alkoxysilyl fluoride appears to be the required
structural element for promoting the reaction. Therefore, in the
present reaction, the fluoride anion is not only an initiator to produce
the actual catalytic species, but it also has an active role in the
catalytic cycle, and CuOMe, allyltrimethoxysilane, and silyl fluoride
9 are all essential for the facile catalytic promotion of the reaction.
Although elucidation of the precise reactive species is under
investigation, these three components should generate the reactive
species19 through dynamic ligand exchange.20
(5) For a similar strategy for the allylsilylation of aldehydes, see: Gauthier,
D. R., Jr.; Carreira, E. M. Angew. Chem., Int. Ed. Engl. 1996, 35, 2363.
(6) Review: (a) Chuit, C.; Corriu, R. J. P.; Reye, C.; Young, J. C. Chem.
ReV. 1993, 93, 1371. (b) Sakurai, H. Synlett 1989, 1.
(7) Pilcher, A. S.; Ammom, H. L.; Deshong, P. J. Am. Chem. Soc. 1995,
117, 5166.
(8) For a catalytic asymmetric allylation of aldehydes using allyltrimethoxy-
silane, see: Yanagisawa, A.; Kageyama, H.; Nakatsuka, Y.; Asakawa,
K.; Matsumoto, Y.; Yamamoto, H. Angew. Chem., Int. Ed. 1999, 38, 3701.
(9) See Supporting Information (SI) for details. Both a copper (I) salt and a
fluoride source are essential for the facile reaction. No reaction occurred
using allyltrimethylsilane as an allylating reagent.
(10) Due to the ability to form the linear triene compound through â-elimination
at the lactone moiety, 1h is not stable under acidic or basic conditions.
No target allylated product 2h was obtained through the Grignard reaction
or the conventional Lewis acid (Ti)-catalyzed allylation using allyltribu-
tyltin.
(11) This complete selectivity is different from either the reaction of allylcopper
(1,2-adduct:1,4-adduct ) 1:3.6) generated through transmetalation (Lip-
shutz, B. H.; Ellsworth, E. L.; Dimock, S. H.; Smith, R. A. J. Am. Chem.
Soc. 1990, 112, 4404) or the reaction of allyltrimethylsilane catalyzed by
TBAF (1,2-adduct:1,4-adduct ) 2:1; ref 2a).
(12) In the absence of the protic additive, the reaction was very sluggish. The
presumed role of the additive is to facilitate catalyst turnover through the
protonation of the allylated intermediate. For an example of a beneficial
effect of a protic additive, see: Takamura, M.; Hamashima, Y.; Usuda,
H.; Kanai, M.; Shibasaki, M. Angew. Chem., Int. Ed. 2000, 39, 1650.
(13) NMR studies were conducted using TBAF as a fluoride source for
simplification. 1H, 13C, and 29Si NMR did not provide useful information.
(14) Gulliner, D. J.; Levason, W.; Webster, M. Inorg. Chim. Acta 1981, 52,
153. The fact that CuF acts as a (pre)catalyst was also confirmed by the
reaction using 8 (1 mol %); 98% of 2a was obtained from 1a for 1.5 h.
(15) Pagenkopf, B. L.; Kru¨ger, J.; Stojanovic, A.; Carreira, E. M. Angew. Chem.,
Int. Ed. 1998, 37, 3124.
(16) For an example of a fluoride anion as an initiator, see: Wiedemann, J.;
Heiner, T.; Mloston, G.; Prakash, G. K.; Olah, G. A. Angew. Chem., Int.
Ed. 1998, 37, 820 (trifluoromethylation with TMSCF3).
(17) Tsuda, T.; Hashimoto, T.; Saegusa, T. J. Am. Chem. Soc. 1972, 94, 658.
(18) The low yield might be due to the instability of the active species generated
under these specific conditions.
Preliminary attempts to extend this reaction to a catalytic
enantioselective allylation of ketones were promising (Scheme 3).
Thus, 4a was obtained from 3a in 65% yield with 61% ee (4 °C)
using p-tol-BINAP-CuCl-TBAT (15 mol %). Although the enan-
tioselectivity was still moderate, this is the first example of catalytic
enantioselective allylation of ketones using allylsilane.21
In summary, we developed a general catalytic allylation of
carbonyl derivatives using allyltrimethoxysilane that can be applied
to aldehydes, ketones, and imines. The reaction is promoted by a
catalytic amount of CuCl and TBAT under very mild conditions.
Elucidation of the exact reactive species and improvement of the
enantioselectivity are in progress.
(19) Possible reactive species might be either an allylcopper or an exceptionally
reactive allyl silicate. The stereochemistry of the crotylation products
(Scheme 1) might suggest the latter mechanism. See Denmark, S. E.;
Almsted, N. G. In Modern Carbonyl Chemistry; Otera, J., Ed.; Wiley-
VCH: Weinheim, 2000; pp 298-401.
(20) The role of the alkoxysilyl fluoride might be as a mediator of the ligand
exchange through the formation of pentavalent silicate. See SI for details.
For an example of ligand exchange on silicon, see: Farnham, W. B.;
Harlow, R. L. J. Am. Chem. Soc. 1981, 103, 4608.
(21) Catalytic enantioselective allylation of ketones using allylstannanes: (a)
Casolani, S.; D’Addario, D.; Tagliavini, E. Org. Lett. 1999, 1, 1061. (b)
Cunningham, A.; Woodward, S. Synlett 2002, 43.
JA0262582
9
J. AM. CHEM. SOC. VOL. 124, NO. 23, 2002 6537