C O M M U N I C A T I O N S
Scheme 1. Proposed Catalytic Cycle
Scheme 2. Application to the Catalytic Enantioselective Reaction
In summary, we developed a new methodology for the catalytic
aldol reaction to ketones. The success of the reaction depended on
a unique, dynamic ligand exchange between silicon and copper
atoms. The method was applied to a catalytic enantioselective
reaction. Detailed mechanistic studies, improvement of the enan-
tioselectivity, and application of the present method to other
important carbon-carbon bond-forming reactions are now in
progress.
intermediary of 5 was also supported by the fact that the reaction
between 1a and triethoxysilyl enolate of ethyl acetate 6a proceeded
smoothly (room temperature, 25 min, 80% yield) in the presence
of CuF‚3PPh3‚2EtOH (2.5 mol %) and in the absence of (EtO)3SiF.
Moreover, the 1H and 19F NMR spectra of a mixture of CuF‚3PPh3,
(EtO)3SiF, and 2a (1:3:1) were almost identical to those of CuF‚
3PPh3 and 6a (1:1), except for the presence of TMSF and excess
(EtO)3SiF peaks. Thus, the copper silicate (4)-mediated conversion
of trimethylsilyl enolate 2 to silicate 5 is the first key step in the
catalytic cycle.
Acknowledgment. Financial support was provided by PRESTO
of the Japan Science and Technology Corp. (JST).
Supporting Information Available: Experimental procedures and
characterization of the products (PDF). This material is available free
References
(1) Carreira, E. M. In ComprehensiVe Asymmetric Catalysis; Jacobsen, E.
N., Pfaltz, A., Yamamoto, H., Eds.; Springer-Verlag: Heidelberg, 1999;
Vol. III, Chapter 29.1.
(2) For example, see: (a) Kobayashi, S.; Matsui, S.; Mukaiyama, T. Chem.
Lett. 1988, 1491. (b) Chen, J.; Sakamoto, K.; Orita, A.; Otera, J. J. Org.
Chem. 1998, 63, 9739.
(3) Denmark, S. E.; Fan, Y. J. Am. Chem. Soc. 2002, 124, 4233.
(4) Yamasaki, S.; Fujii, K.; Wada, R.; Kanai, M.; Shibasaki, M. J. Am. Chem.
Soc. 2002, 124, 6536.
(5) For a copper fluoride-catalyzed asymmetric aldol reaction of a silyl
dienolate to aromatic and R,â-unsaturated aldehydes, see: (a) Kru¨ger, J.;
Carreira, E. M. J. Am. Chem. Soc. 1998, 120, 837. (b) Pagenkopf, B. L.;
Kru¨ger, J.; Stojanovic, A.; Carreira, E. M. Angew. Chem., Int. Ed. 1998,
37, 3124. Aliphatic substrates gave poor yields in those cases.
(6) Gulliver, D. J.; Levason, W.; Webster, M. Inorg. Chim. Acta 1981, 52,
153.
We next performed kinetic studies to address the question of
whether the actual nucleophile was silicate 5 or copper enolate 7,
which was produced through further ligand exchange between
silicon and copper atoms.11 The order dependency of the initial
reaction rate on [CuF‚3PPh3] and [6a] was 1.5 and -0.8, respec-
tively.12 Although not conclusive, the results suggest that the actual
nucleophile was copper enolate 7.13 The generation of highly
nucleophilic copper enolate would be the second key for the
catalytic cycle. When 7 reacts with a ketone, copper alkoxide 8
should be produced. To gain insight into the catalyst turnover step
from 8, we attempted the reaction of 1a and 2a using CuOtBu14 as
the catalyst (10 mol %) in the presence of (EtO)3SiF (1.2 equiv).
Unexpectedly, the reaction did not proceed well, and the aldol
product was obtained in only 29% yield after 20 h. When the same
reaction was conducted in the presence of 30 mol % PPh3, however,
the reaction proceeded smoothly in 1.5 h, giving the product in
92% yield. These results indicated that the phosphine ligand was
essential for the catalyst turnover from the copper alkoxide. Thus,
the catalytic cycle for the present reaction is proposed in Scheme
1 as a working hypothesis. First, a dynamic ligand exchange
between CuF‚3PPh3, (EtO)3SiF, and 2 produces silicate 5 and
copper enolate 7, which reacts with a ketone to give copper alkoxide
8. 8 then reacts with (EtO)3SiF to produce silicate 9. Finally, fluoride
exchange between 9 and (EtO)3SiF produces the triethoxysilyl-
protected aldol product 10 and active silicate 4 to complete the
catalytic cycle.15
(7) For example, the reaction of aliphatic ketone 1g produced multiple
products, and 3ga was obtained in only 28% yield after 74 h in the absence
of (EtO)3SiF. See the Supporting Information (SI) for details.
(8) To avoid the undesired solvolysis of the active species, EtOH was
completely eliminated by azeotropic evaporation with toluene in the NMR
studies. EtOH-free catalyst contained the same activity as CuF‚3PPh3‚
2EtOH.
(9) There have been no reports on the 19F NMR chemical shift values of
trialkoxydifluoro silicates. When TBAT was mixed with (EtO)3SiF, a peak
at -130 ppm (with a doublet satellite peak, J ) 156 Hz) was observed,
which supports the assignment of 4.
(10) Peaks of 5 and 7 were not observed on 19F and 1H NMR, possibly due to
the fast ligand exchange between silicon and copper atoms.
(11) React IR did not provide any clear information in our case. A very small
amount of (EtO)3SiF was observed on the 19F NMR of a CuF‚PPh3-6a
mixture, suggesting that 5 and 7 were in equilibrium.
(12) 6a was used for the kinetic studies to eliminate the effect of the complex
initial ligand exchange process. The fractional order dependency might
suggest that the overall reaction contains several rate-determining steps.
The ligand-exchange step to generate the active nucleophile, the aldol
addition step, and the catalyst turnover step are possible candidates. The
inhibitory feature of the ketene silyl acetal (minus order dependency) might
be due to the nonproductive fluoride exchange between silicon atoms
(inhibition of the active nucleophile generation step).
This new methodology could, in principle, be extended to the
catalytic enantioselective reaction, and the preliminary results are
shown in Scheme 2. Thus, in the presence of 2.5 mol % CuF‚
3PPh3‚2EtOH-(S)-p-tol-BINAP complex, the aldol reaction between
ketone 1l and silyl enolate 2g gave the product with up to 82% ee.
As expected, the sense of enantioselectivity was almost completely
controlled by the asymmetric catalyst, and not by the geometry of
the initial ketene silyl acetal.
(13) The possibility that Cu+ acts as a Lewis acid cannot be completely
excluded.
(14) Tsuda, T.; Hashimoto, T.; Saegusa, T. J. Am. Chem. Soc. 1972, 94, 658.
(15) The proposed mechanism is distinct from the silyl Lewis acid-catalyzed
aldol reaction (Ishihara, K.; Hiraiwa, Y.; Yamamoto, H. Chem. Commun.
2002, 1564) or the tin(II) triflate- and tin fluoride-mediated asymmetric
aldol reaction (Kobayashi, S.; Uchiro, H.; Fujishita, Y.; Shiina, I.;
Mukaiyama, T. J. Am. Chem. Soc. 1991, 113, 4247).
JA034993N
9
J. AM. CHEM. SOC. VOL. 125, NO. 19, 2003 5645