J. Am. Chem. Soc. 2001, 123, 12907-12908
12907
Table 1. Probe Reactions of 2-Methyl-2-vinyloxirane with Ethyl
Atom Economic Asymmetric Creation of Quaternary
Carbon: Regio- and Enantioselective Reactions of a
Vinylepoxide with a Carbon Nucleophile
Acetoacetatea
rxn time
(min)
ratioc
3:4:5
yieldd eee
3
entry ligand additiveb solvent temp
3
Barry M. Trost* and Chunhui Jiang
1
2
3
4
5
6
7
(()-L1
(()-L1
none
none
(S,S)-L1 none
CH2Cl2 rt
CH2Cl2 0 °C
CH2Cl2 rt
10
30
25
15
30
<5
70
120
79:16:5
64
n.a.f
54:22:24 n.d.g n.a.f
Department of Chemistry, Stanford UniVersity
63:24:14 45
73:16:11 56
60:30:10 48
70:17:13 57
93
93
96
97
96
94
Stanford, CA 94305-5080
(S,S)-L1 TBAT PhH
(S,S)-L2 none CH2Cl2 rt
rt
ReceiVed September 4, 2001
(S,S)-L2 TBAT PhH
(S,S)-L3 TBAT PhH
rt
40 °C
40 °C
79:17:4
70
68
The metal-catalyzed additions of pronucleophiles to vinyl-
epoxides with achiral phosphine ligands have a strong bias to
lead to 1,4-addition (eq 1, path a), due to the electronic effect of
the epoxide oxygen.1,2 On the other hand, the 1,2-adducts are
8h (S,S)-L3 TBAT PhH
n.d.g
a All reactions were conducted on 0.5 mmol scale in 5.0 mL of
solvent using 1.0 equiv of 1, 1.1 equiv of 2, 1 mol % of Pd2dba3‚CHCl3,
and 3 mol % of ligand unless indicated otherwise. b When added, 1
mol % employed. c Determined by 1H NMR spectroscopy on the crude
reaction mixture. d Isolated yield of pure 3. e Determined after dehydra-
tion of initial adduct. f n.a. ) not applicable. g n.d. ) not determined.
h Rection performed with 0.5 mol % of catalyst used at 0.2 M
concentration.
mixture of the 1,2 (i.e., 3)- and 1,4 (i.e., 4 and 5)-adducts were
obtained from which the 1,2-adduct 3a6 was isolated pure in 64%
yield. Lowering the temperature decreased the selectivity (entry
2). Interestingly, using enantiopure ligand S,S-L17 also reduced
the selectivity (entry 3), which allowed isolation of pure 3 in only
45% yield. Gratifyingly, the enantioselectivity was excellent (93%
ee, determined after conversion to the corresponding dihydrofuran
6). The reduced regioselectivity with the enantiopure ligand
compared to that of the racemic ligand was attributed to a kinetic
discrimination in the initial ionization with racemic epoxide and
racemic ligand to favor formation of enantiomers of the same
diastereomer I-1 (Figure 1) of the intermediate π-allylpalladium
species which had an intrinsic higher preference for addition at
the more substituted allyl terminus. With the enantiopure ligand
and racemic epoxide, two different diastereomeric π-allylpalla-
dium intermediates I-1 and I-2 are formed. While I-1 favors
formation of the branched product, its diastereomer I-2 favors
formation of the linear product.
It was hypothesized that increasing the rate of interconversion
of diastereomeric π-allylpalladium complexes would increase the
regioselectivity and maintain the ee.8 Indeed, adding 1 mol %
TBAT (tetra-n-butylammonium triphenyldifluorosilicate) did just
that (entry 4). A similar trend was observed with the more
sterically congested ligand L24c (entries 5 and 6). The more
flexible ligand L37 gave the best results, wherein a 70% isolated
yield of 3 of 96% ee was obtained (entry 7). Reducing the catalyst
load and increasing the concentration, each by a factor of 2,
showed little change (entry 8).
potentially valuable chiral building blocks as a result of the diverse
functionality present that permits chemoselective differentiation.
While the use of heteroatom pronucleophiles such as alcohols,
amines, and imides has been successfully employed,3-5 the use
of carbon-centered pronucleophiles has not been employed yet
represents perhaps the most significant type since it builds the
basic carbon framework. Particularly significant is the case of
R ) alkyl since it entails creating a quaternary carbon enantio-
selectively. The success of heteroatom pronucleophiles has been
attributed to the ability of such functionality to hydrogen bond
or in some other way coordinate to the epoxide oxygen to help
deliver the nucleophile to the adjacent carbon. The failure of
carbon-centered pronucleophiles to participate in such hydrogen
bonding would seem to make any extrapolation of results from
heteroatom to carbon pronucleophiles unlikely. Despite this
expectation, we explored â-ketoesters in such reactions to
determine the feasibility of eq 1, path b, dominating over path a.
As a probe of the effect of chiral ligands on regioselectivity,
isoprene monoepoxide 1 was reacted with ethyl acetoacetate 2
in the presence of 1 mol % Pd2(dba)3‚CHCl3 (Pd(0)) and 3 mol
% of racemic ligand (()-L1 (eq 2, and Table 1, entry 1). A
A range of â-ketoesters were examined as summarized in Table
2. Varying the size of the ester group R′ had little effect on the
selectivity. On the other hand, increasing the size of R does
increase the regioselectivity. In all cases, the enantioselectivity
(3) Trost, B. M.; Angle, S. R. J. Am. Chem. Soc. 1985, 107, 6123; Trost,
B. M.; Sudhakar, A. R. J. Am. Chem. Soc. 1988, 110, 7933; Trost, B. M.;
Tenaglia, A. Tetrahedron Lett. 1988, 29, 2931; Trost, B. M.; Hurnaus, R.
Tetrahedron Lett. 1989, 30, 3893.
(4) (a) Trost, B. M.; McEachern, E. J.; Toste, F. D. J. Am. Chem. Soc.
1998, 120, 12702. (b) Trost, B. M.; McEachern, E. J. J. Am. Chem. Soc. 1999,
121, 8649. (c) Trost, B. M.; Bunt, R. C.; Lemoine, R.; Calkins, T. L. J. Am.
Chem. Soc. 2000, 122, 5968.
(5) For related reactions of vinylthiiranes and vinylaziridines, see: Larksarp,
C.; Selllier, O.; Alper, H. J. Org. Chem. 2001, 66, 3502. For related reactions
of vinyloxetane, see: Larksarp, C.; Alper, H. J. Org. Chem. 1999, 64, 4152.
(6) This compound has been characterized spectroscopically and elemental
composition established by combustion analysis and high-resolution mass
spectrometry.
(1) Trost, B. M.; Molander, G. A., J. Am. Chem. Soc. 1981, 103, 5969;
Trost, B. M.; Warner, R. W. J. Am. Chem. Soc. 1982, 104, 6112; Trost, B.
M.; Cossy, J. J. Am. Chem. Soc. 1982, 104, 6881; Trost, B. M.; Granja, J. R.
J. Am. Chem. Soc. 1991, 113, 1044; Trost, B. M.; Granja, J. R. Tetrahedron
Lett. 1991, 32, 2193; Trost, B. M.; Ito, N.; Greenspan, P. D. Tetrahedron
Lett. 1993, 34, 1421; Trost, B. M.; Ceschi, M. A.; Ko¨nig, B. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1486; Trost, B. M.; Schroeder, G. M. J. Am. Chem.
Soc. 2000, 122, 3785.
(7) Trost, B. M.; Van Vranken, D. L.; Bingel, C. J. Am. Chem. Soc. 1992,
114, 9327.
(8) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 1999, 121, 4545. For a
review, see: Trost, B. M.; Van Vranken, D. L. Chem. ReV. 1996, 96, 395.
(2) Tsuji, J.; Kataoka, H.; Kobayashi, Y. Tetrahedron Lett. 1981, 22, 2575.
10.1021/ja012104v CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/28/2001