olefination of ketenes,6b and the chiral guanidine-catalyzed
isomerization of 3-alkynoates exemplify elegant transforma-
tions with these reagents.6c However, few reports have
described the catalytic asymmetric synthesis of chiral
R-substituted 2,3-allenoic esters.3b,6d
Table 1. Optimization of Reaction Conditionsa
Very recently, we established a series of chiral Brønsted
acid-catalyzed 1,3-dipolar cycloaddition reactions of azome-
thine ylides.7 In particular, the protocol is applicable to 2,3-
allenoates, yielding 3-methylenepyrrolidine derivatives with
high enantiopurity.7d Interestingly, when 2 equiv of racemic
4-ethyl buta-2,3-dienoates were reacted with the azomethine
ylide derived from 4-bromobenzaldehyde, a moderate enan-
tiomeric excess was observed for the recovered 4-ethyl buta-
2,3-dienoates. In view of the widespread applications of
enantiomerically enriched 2,3-allenoates, we were greatly
interested in the kinetic resolution8 of racemic 2,3-allenoates.
Although there have been examples describing the prepara-
tion of allenes in optically pure form through kinetic
resolution,3b organocatalytic kinetic variants have not been
available. Herein, we will report an alternative to the known
methods to access highly optically active R-substituted 2,3-
allenoic esters by kinetic resolution of racemic 2,3-allenoates
2 via a chiral Brønsted acid-catalyzed asymmetric 1,3-dipolar
cycloaddition reaction (eq 1).
5a
6
entry
x
3
2a: 3: 4 yieldb (%) eec (%) yieldb (%) eec (%)
1
2
3
4
5
6
10
10
10
10
3a 1:1.2:1
3b 1:1.2:1
3c 1:1.2:1
3d 1:1.2:1
47
43
45
48
47
46
47
50
45
65
51
96
94
99
98
99
98
99
90
>99
24
33
50
49
46
51
52
53
52
48
48
19
26
96
95
93
96
94
92
94
96
94
92
92
7.5 3d 1:1.2:1
5 3d 1:1.2:1
7d
8e
9
7.5 3d 1:1.2:1
7.5 3d 1:1.2:1
7.5 3d 1:1.1:1
7.5 3d 1:1.1:1
7.5 3d 1:1.1:1
10f
11g
a Unless indicated otherwise, the reaction was carried out on a 0.1 mmol
scale in PhCH3 (1 mL) with 3 Å MS (100 mg) for 3 days. b Isolated yield.
c Determined by HPLC. d 4 Å MS was used. e 5 Å MS was used. f CH2Cl2
was used as solvent. g CHCl3 was used as solvent.
The variation of molecular sieves afforded no positive
effects on the stereoselectivity (entries 7 and 8). Tuning the
stoichiometric ratio of 4-bromobenzaldehyde had little effect
on the stereoselectivity of 6d and the recovered 5a, and thus,
addition of only 1.1 equiv of 4-bromobenzaldehyde to 2a
was sufficient for the reaction, generating 5a with excellent
enantioselectivity of >99% ee (entry 9). A comparison of
solvents indicated that the reaction proceeded more slowly
in halogenated media than in toluene. Additionally, much
lower ee values were obtained for the recovered 5a, when
the reaction was conducted in either chloroform or dichlo-
romethane (entries 10 and 11).
With the optimized conditions in hand, the scope of 2,3-
allenoates was then examined (Table 2). The protocol
tolerated a wide range of allenylic R-substituents on the
allenoate 2, including benzyl, alkyl, and allyl moieties.
Generally, the reactions provided cycloaddition products 6
in 39-57% yields and 64-94% ee accompanied by yields
of 35-48% for the recovered 2,3-allenoates 5 in 85-99%
ee. The substituents at C4 of the 2,3-allenoate (R1 group)
had little effect on the enantioselectivity of recovered 2,3-
allenoates 5. The ee of the products eroded to a small degree
when the ethyl group was replaced with a benzyl group at
C4 of the allenoates (Table 2, entries 1 vs 11 and 2 vs 10).
Variation of the substituent at C2 of the allenoates showed
that larger substituents were beneficial to the enantiose-
lectivity of products while high ee values were observed
for the recovered allenoates (entries 1-5 and 9-12). The
6-chlorohexa-2,3-dienoate derivatives proved to be more
reactive than other alleonates toward azomethine ylides
in the 1,3-dipolar cycloaddition, giving products 6 in
higher yields under the similar conditions, whereas the
To establish the suitable aldehyde substrate for the efficient
kinetic resolution, we investigated the 1,3-dipolar cycload-
dition reaction of the racemic 9-anthracenylmethyl 2-ben-
zylhexa-2,3-dienoate (2a) with azomethine ylides derived
from diethyl 2-aminomalonate (4) and various aromatic
aldehydes in the presence of the bisphosphoric acid 1 (Table
1). The substituent on the aromatic aldehyde proved to have
considerable influence on the kinetic resolution, and the use
of 4-bromobenzaldehyde gave 6d in 51% yield and with 96%
ee, while remarkably, 5a was recovered in 48% yield and
with 98% ee (Table 1, entries 1-4). Reducing the catalyst
loading from 10 mol % to 7.5 mol % led to a subtle
enhancement in the enantioselectivity of 5a (entry 5).
However, when catalyst loading was reduced to 5 mol %,
slightly diminished ee values were observed for both the
recovered substrate and the product (entry 6).
(7) (a) Chen, X.-H.; Zhang, W.-Q.; Gong, L.-Z. J. Am. Chem. Soc. 2008,
130, 5652. (b) Liu, W.-J.; Chen, X.-H.; Gong, L.-Z. Org. Lett. 2008, 10,
5357. (c) Chen, X.-H.; Wei, Q.; Luo, S.-W.; Xiao, H.; Gong, L.-Z. J. Am.
Chem. Soc. 2009, 131, 13819. (d) Yu, J.; He, L.; Chen, X.-H.; Song, J.;
Chen, W.-J.; Gong, L.-Z. Org. Lett. 2009, 11, 4946.
(8) For some reviews of the kinetic resolution, see: (a) Kagan, H. B.;
Fiaud, J. C. Top. Stereochem. 1988, 18, 249. (b) Keith, J. M.; Larrow, J. F.;
Jacobsen, E. N. AdV. Synth. Catal. 2001, 343, 5. (c) Vedejs, E.; Jure, M.
Angew. Chem., Int. Ed. 2005, 44, 3974, and references cited therein. For
selected examples of the kinetic resolution of 2,3-allenoic acid, see: (d)
Ma, S.; Wu, S. Chem. Commun. 2001, 441.
Org. Lett., Vol. 12, No. 18, 2010
4051
Yu, Jie
