Enantioselective alkenylation and phenylation catalyzed by a chiral CuF complex

Article abstract of DOI:10.1021/ja0507362

A new method for CuF-catalyzed alkenylation and phenylation of aldehydes and an activated ketone using air- and moisture-stable alkenylsilanes and phenylsilane as a nucleophile is described. This methodology was extended to highly enantioselective catalytic alkenylation and phenylation using DTBM-SEGPHOS as a chiral ligand. Substrate generality is broad, and an alkenylsilane with a long alkyl chain and an internal alkenylsilane can be also used as a nucleophile. The key to success partly involves the accelerated regeneration of reactive alkenylcopper and phenylcopper through transmetalation from the silylated nucleophiles, and stabilization of the reactive copper reagents, both of which are effected by the diphosphine ligands. Copyright

Full text of DOI:10.1021/ja0507362

Published on Web 03/05/2005
Enantioselective Alkenylation and Phenylation Catalyzed by a Chiral CuF
Complex
Daisuke Tomita,† Reiko Wada,† Motomu Kanai,*,†,‡ and Masakatsu Shibasaki*,†
Graduate School of Pharmaceutical Sciences, The UniVersity of Tokyo, Tokyo 113-0033, Japan, and PRESTO,
Japan Science and Technology Agency, Japan
Received February 4, 2005; E-mail: mshibasa@mol.f.u-tokyo.ac.jp
Table 1. Optimization of the Vinylation Reaction: Ligand Effects
Chiral allylic alcohols are among the most versatile synthetic
intermediates. Catalytic enantioselective synthesis of these chiral
building blocks currently relies on two main methods; kinetic
resolution using the Sharpless epoxidation1 and asymmetric addition
of alkenylzinc reagents to carbonyl compounds.2 These ground-
breaking reactions are reliable and have established new concepts
in the history of organic synthesis. Because of the recent demands
for safe and sustainable organic synthesis, however, resolution
methodology or the use of highly sensitive reagents is not desirable,
especially for large-scale synthesis.3 In this communication, we
report a new catalytic method for chiral allylic alcohol synthesis
using air-stable and commercially available vinylsilanes as nucleo-
philes. In addition, the method also allows for a highly enantiose-
lective phenylation of aldehydes, producing another group of
valuable chiral building blocks, diarylmethanols.
We previously reported that silylated nucleophiles such as
allyltrimethoxysilanes and ketene silyl acetals are activated by a
catalytic amount of CuF, producing highly nucleophilic allylcoppers
or copper enolates through transmetalation.4,5 These species are
sufficiently reactive to perform catalytic enantioselective allylation
and aldol reactions to ketones, when the CuF is modified by chiral
phosphine ligands. On the basis of these findings, we planned to
use trimethoxyvinylsilane as a nucleophile. Previous reports on
CuCl-promoted homo-coupling of alkenylfluorosilanes,6 as well as
stoichiometric Cu(I)-induced protodesilylation7 and allylation8 of
alkenylsilanes support our expectation that the generation of
alkenylcopper through transmetalation between copper and silicon
atoms should be possible. On the other hand, the reactivity of the
thus-generated vinylcopper to carbonyl compounds and the pos-
sibility of catalyst turnover could not be predicted.9
Our initial studies focused on the determination of vinylation
conditions for aldehydes using achiral CuF complexes, and we first
found a dramatic difference in reactivity between allylation and
vinylation reactions. When the optimized conditions for catalytic
allylation4a were applied to catalytic vinylation of benzaldehyde
(1a) using trimethoxyvinylsilane (2a), the reaction did not proceed
(Table 1, entries 1 and 2). The allylation was completed in 1 min
under the same conditions. During the reaction, the solution color
changed from pale yellow (derived from CuF) to dark brown, which
might indicate the generation of a vinylcopper via transmetalation.
To enhance the reactivity of the generated vinylcopper species,
ligand-acceleration effects were investigated (entries 3-7). After
systematic screening of diphosphine ligands regarding the electronic
and steric factors, dppf was determined to be the optimum achiral
ligand (entry 7).10 The reaction rate was faster in DMF than in
THF, and the reaction was completed in 3 h using 5 mol % catalyst
(entry 8).11 To our knowledge, this is the first example of a catalytic
entry
catalyst
ligand
time (h)
yield (%)
1
2
3
4
5
6
7
8d
CuCl-TBATa
none
none
dppe
24
24
24
24
24
24
4
0
0
61
47
57
89
100
100
CuF‚3PPh3‚2EtOH
CuF‚3PPh3‚2EtOH
CuF‚3PPh3‚EtOH
CuF‚3PPh3‚2EtOH
CuF‚3PPh3‚2EtOH
CuF‚3PPh3‚2EtOH
CuF‚3PPh3‚2EtOH
d(p-Cl)ppeb
d(p-MeO)ppec
dppp
dppf
dppf
3
a Tetrabutylammonium difluorotriphenylsilicate. b 1,2-Bis(di-p-chloro-
phenylphosphino)ethane. c 1,2-Bis(di-p-methoxyphenylphosphino)ethane.
d 5 mol % catalyst and 7 mol % dppf were used. Solvent ) DMF.
intermolecular vinylation of aldehydes using weakly nucleophilic
simple vinylsilane as a nucleophile.12
Having established the new catalytic vinylation conditions,
extension to a catalytic enantioselective reaction was investigated
using chiral diphosphines. Screening available chiral diphosphines
led us to determine that DTBM-SEGPHOS (4) was the best
ligand;13 reaction of 1a was completed in 0.5 h using 3 mol %
catalyst (generated via in situ reduction of CuF2‚2H2O by excess
chiral phosphine), and the product was obtained in 99% yield with
94% ee (Table 2, entry 1).
The substrate generality of this reaction is summarized in Table
2. Excellent enantioselectivity was generally produced from a wide
range of aldehydes. As for enolizable aliphatic aldehyde 1h, the
typical procedure afforded mainly undesired product via self-aldol
condensation.14 In this case, however, simple tuning of the reaction
conditions using dimethoxymethylvinylsilane (2b) as a nucleophile
in toluene solvent gave the vinylation product in high chemical
yield (entry 8). Alkenylsilane 2c with a longer alkyl chain, which
was conveniently synthesized through olefin cross metathesis,15 can
be utilized as a nucleophile (entries 9 and 10). The reaction
proceeded even using an internal alkenylsilane 2d, which was
synthesized through Trost’s ruthenium-catalyzed regioselective
hydrosilylation of the corresponding alkyne16 (entry 11). Although
yield and enantioselectivity require further improvement in this case,
the result is noteworthy because products containing 1,1′-disubsti-
tuted alkenes are not accessible by reactions using alkenylzinc,
which is normally generated through a terminal position-selective
hydroboration-transmetalation sequence.2 The reaction proceeded
chemoselectively with an activated ketone 1i, giving the chiral
tertiary allylic alcohol with high enantioselectivity (entry 12).17
Moreover, this methodology can be extended to a catalytic
enantioselective phenylation18 using dimethoxydiphenylsilane (2e)
as a nucleophile (entries 13 and 14).
† The University of Tokyo.
‡ PRESTO.
9
4138
J. AM. CHEM. SOC. 2005, 127, 4138-4139
10.1021/ja0507362 CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Catalytic Enantioselective Vinylation and Phenylation
Johnson, R. A.; Sharpless, K. B. In Catalytic Asymmetric Synthesis; Ojima,
I., Ed.; Wiley-VCH: New York, 2000; p 231.
(2) For selected examples, see: (a) Oppolzer, W.; Radinov, R. N. HelV. Chim.
Acta 1992, 75, 170. (b) Wipf, P.; Ribe, S. J. Org. Chem. 1998, 63, 6454.
(c) Chen, Y. K.; Lurain, A. E.; Walsh, P. J. J. Am. Chem. Soc. 2002, 124,
12225. (d) Li, H.; Walsh, P. J. J. Am. Chem. Soc. 2004, 126, 6538.
For general reviews, see; (e) Pu, L.; Yu, H.-B. Chem. ReV. 2001, 101,
757. (f) Noyori, R.; Kitamura, M. Angew. Chem., Int. Ed. Engl. 1991,
30, 49.
(3) Recent contributions from Jamison’s group (reductive coupling of alkynes
and aldehydes) are consistent with these demands. Substrate generality,
however, remains to be improved. Miller, K. M.; Huang, W.-S.; Jamison,
T. F. J. Am. Chem. Soc. 2003, 125, 3442.
(4) (a) Yamasaki, S.; Fujii, K.; Wada, R.; Kanai, M.; Shibasaki, M. J. Am.
Chem. Soc. 2002, 124, 6536. (b) Oisaki, K.; Suto, Y.; Kanai, M.; Shibasaki,
M. J. Am. Chem. Soc. 2003, 125, 5644. (c) Wada, R.; Oisaki, K.; Kanai,
M.; Shibasaki, M. J. Am. Chem. Soc. 2004, 126, 8910.
(5) For leading references of metal fluoride activation of silylated nucleophiles
through transmetalation, see: (a) Pagenkopf, B. L.; Kru¨ger, J.; Stojanovic,
A.; Carreira, E. M. Angew. Chem., Int. Ed. 1998, 37, 3124 (CuF activation
of silyl dienolate). (b) Yanagisawa, A.; Kageyama, H.; Nakatsuka, Y.;
Asakawa, K.; Matsumoto, Y.; Yamamoto, H. Angew. Chem., Int. Ed. 1999,
38, 3701 (AgF activation of allyltrimethoxysilane).
(6) Using a stoichiometric amount of Cu(I) salt: (a) Yoshida, J.-i.; Tamao,
K.; Kakui, T.; Kumada, M. Tetrahedron Lett. 1979, 13, 1141. (b)
Ikegashira, K.; Nishihara, Y.; Hirabayashi, K.; Mori, A.; Hiyama, T. Chem.
Commun. 1997, 1039. Using a catalytic amount of CuCl: (c) Nishihara,
Y.; Ikegashira, K.; Toriyama, F.; Mori, A.; Hiyama, T. Bull. Chem. Soc.
Jpn. 2000, 73, 985.
(7) Trost, B. M.; Ball, Z. T.; Jo¨ge, T. J. Am. Chem. Soc. 2002, 124, 7922.
(8) Taguchi, H.; Ghoroku, K.; Tadaki, M.; Tsubouchi, A.; Takeda, T. Org.
Lett. 2001, 3, 3811.
(9) The allylcopper species generated through transmetalation from allylsilane
demonstrated completely different reactivity and stability from that
prepared through a conventional method (transmetalation from allyllithium
or allylmagnesium reagents). For example, CuF-catalyzed allylation
proceeded with complete 1,2-selectivity to enones at room temperature
(ref 4a).
a Isolated yield. b Determined by chiral HPLC. c The reaction was
performed at room temperature in toluene. d The reaction was performed
at 60 °C in the presence of 10 mol % TBAT as an additive. In the absence
of TBAT, the reaction did not proceed. e The reaction was performed at
room temperature. f The absolute configuration was determined to be (S).
(10) Steric congestion around the copper atom appeared to be more important
for the ligand-acceleration effects than the electronic factors (for electronic
effects, see Table 1, entries 3-5; for steric effects, see Table 1, entries 3,
6, and 7 and reactivity comparison between Et- and iPr-DuPHOS described
in ref 13), and steric effects). The difference between DTBM-SEGPHOS
(4) and DMM-SEGPHOS (5) under optimized conditions for enantiose-
lective reaction is also consistent with this tendency: CuF-5 complex (3
mol % in DMF) gave 3aa in only 31% yield for 3 h, while the reaction
was completed in 0.5 h using 4 (Table 2, entry 1). This acceleration effect
might be due to stabilization of a hypothetical monomeric, active copper
species and/or acceleration of the rate-determining ligand exchange (see
text) to regenerate the active vinylcopper. For examples of acceleration
effects by sterically hindered ligands, see: (a) Littke, A. F.; Schwarz, L.;
Fu, G. C. J. Am. Chem. Soc. 2002, 124, 6343. (b) Strieter, E. R.;
Blackmond, D. G.; Buchwald, S. L. J. Am. Chem. Soc. 2003, 125, 13978.
(c) Yamasaki, S.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2001, 123,
1256.
To gain insight into the reaction mechanism, several observations
were made. First, generation of an alkenylcopper was strongly
supported by the observation of an 19F NMR signal corresponding
to (MeO)3SiF, when the CuF-dppf complex was mixed with 2a
in a 1:3 ratio.11,19 Second, enantioselectivity was not affected by
the substituents of the silicon atom of alkenylsilane.11 Thus, the
silicon species is not relevant to the enantiodifferentiating addition
step. Third, the catalytic cycle can start from a copper alkoxide-
DTBM-SEGPHOS complex with the same enantioselectivity as
that of the chiral CuF complex.11 These results, combined with
kinetic studies,20 suggested two key factors: (1) an alkenylcopper
(or a phenylcopper) generated through transmetalation works as
an active nucleophile, and (2) the diphosphine ligands facilitate
the rate-determining catalyst turnover step (regeneration of the
alkenylcopper from an intermediate copper alkoxide).
(11) For the substrate scope using the CuF-dppf catalyst, results of mechanistic
studies, and the proposed catalytic cycle, see Supporting Information for
details.
(12) For a comprehensive review on the reactions using alkenylsilanes, see:
Fleming, I.; Dunogue´s, J.; Smithers, R. Org. React.1989, 37, 57.
(13) See Supporting Information for the effects of chiral ligands and the catalyst
preparation method. Briefly, p-tol-BINAP, Et-DuPHOS, and iPr-DuPHOS-
complexes (10 mol %) gave 3aa in 47% with 61% ee (24 h), in 51%
with 38% ee (24 h), and in 87% with 64% ee (5 h), respectively.
(14) The self-aldol reaction might be promoted by CuOMe, which is generated
through competitive methoxy ligand transfer from silicon to copper, instead
of the desired vinyl transfer. For a use of copper alkoxide as a Brønsted
base catalyst for direct aldol reaction, see: Suto, Y.; Kumagai, N.;
Matsunaga, S.; Kanai, M.; Shibasaki, M. Org. Lett. 2003, 5, 3147.
In conclusion, we developed a new catalytic enantioselective
method for chiral allylic alcohol and diarylmethanol synthesis using
air- and moisture-stable alkenylsilanes and phenylsilane as nucleo-
philes. Detailed mechanistic studies are in progress.
(15) Pietraszuk, C.; Fischer, H.; Kujiwa, M.; Marciniec, B. Tetrahedron Lett.
2001, 42, 1175.
(16) Trost, B. M.; Ball, Z. T. J. Am. Chem. Soc. 2001, 123, 12726.
(17) Simple ketones did not give the addition product at the current stage.
Acknowledgment. Financial support was provided by PRESTO
of Japan Science and Technology Agency (JST). R.W. thanks to
JSPS for a research fellowship.
(18) For selected examples of catalytic enantioselective phenylation of alde-
hydes using phenylzinc species, see: (a) Dosa, P. I.; Ruble, J. C.; Fu, G.
C. J. Org. Chem. 1997, 62, 444. (b) Bolm, C.; Rudolph, J. J. Am. Chem.
Soc. 2002, 124, 14850. For a review, see: (c) Bolm, C.; Hildebrand, J.
P.; Mun˜iz, K.; Hermanns, N. Angew. Chem., Int. Ed. 2001, 40, 3284 and
ref 2e.
Supporting Information Available: Experimental procedures and
characterization of the products. This material is available free of charge
via the Internet at http://pubs.acs.org.
(19) Interestingly, no such signal was observed using CuF‚3PPh3 or TBAT as
a fluoride source, which demonstrated no catalyst activity.
(20) The rate-determining step was identified on the basis of kinetic studies.
The order dependencies of the initial reaction rate were 1, 0, and 0.5
regarding CuF, aldehyde, and vinylsilane, respectively.
References
(1) (a) Gao, Y.; Klunder, J. M.; Hanson, R. M.; Masamune, H.; Ko, A. Y.;
Sharpless, K. B. J. Am. Chem. Soc. 1987, 109, 5765. For review, see; (b)
JA0507362
9
J. AM. CHEM. SOC. VOL. 127, NO. 12, 2005 4139