Development of a formal catalytic asymmetric [4 + 2] addition of ethyl-2,3-butadienoate with acyclic enones

Article abstract of DOI:10.1021/ol202301f

Allene esters are unique not only as excellent electrophiles but also because of their ability for subsequent reactivity after the initial nucleophilic attack. A mechanistically inspired cyclization using ethyl-2,3-butadienoate and acyclic enones to provi

Full text of DOI:10.1021/ol202301f

ORGANIC
LETTERS
2011
Vol. 13, No. 21
5732–5735
Development of a Formal Catalytic
Asymmetric [4 þ 2] Addition of Ethyl-2,3-
butadienoate with Acyclic Enones
Kumar Dilip Ashtekar, Richard J. Staples, and Babak Borhan*
Department of Chemistry, Michigan State University, East Lansing, Michigan 48824,
United States
babak@chemistry.msu.edu
Received August 10, 2011
ABSTRACT
Allene esters are unique not only as excellent electrophiles but also because of their ability for subsequent reactivity after the initial nucleophilic
attack. A mechanistically inspired cyclization using ethyl-2,3-butadienoate and acyclic enones to provide dihydropyrans in excellent yields and
enantioselectivity under solvent-free conditions at room temperature is reported.
The BaylisꢀHillman reaction has been extensively studied
for its utility to forge CꢀC bonds catalyzed by N-based Lewis
bases such as 1,4-diazabicyclo[2,2,2]octane (DABCO).1
Recent advancements include the development of catalytic
asymmetric variants.3 The use of allene esters as primary
electrophiles for the BaylisꢀHillman reaction expands the
repertoire of products, as subsequent reactions add to the
complexity of the final structures.3c In the latter context,
the use of chiral N- or P-based Lewis bases has been
reported with various secondary electrophiles;2,4 however
development of a reaction with acyclicenonesassecondary
electrophiles has not been explored.6 Considering our
endeavor in developing synthetic routes to heterocyclic
nuclei,5 our interest was piqued by the possibility of employ-
ing acyclic enones as secondary electrophiles toward the
preparation of a library of complex dihydropyrans as key
intermediates for constructing complex motifs.7
Figure 1 illustrates the divergence in products obtained
from the reaction of 1 with allenoate 2, catalyzed with
either phosphines2a or amines.2b Two main factors seem to
contribute to the formation of cyclic products in the
phosphine catalyzed pathway: (a) the presence of ‘d’
orbitals on phosphorus that support an expanded valence
shell enables the reaction of the transient enolates 6a,c
in the manner depicted to generate ylides 6b,d; (b) a
rapid proton transfer in 6b,d initiates catalyst turnover.
Yet, nitrogen cannot exhibit a similar genre of reactivity,
as the lack of ‘d’ orbitals does not indulge ylide forma-
tion under these conditions. Thus, proton transfer leading
to the illustrated elimination (Figure 1, 7a,b), albeit slowly,
results in the formation of R-substituted allenes.
(1) (a) Basavaiah, D.; Rao, A. J.; Satyanarayana, T. Chem. Rev. 2003,
103, 811. (b)Baylis, A. B. H., M. E. D. InGerman Patent 2 1972; Vol. 155. (c)
Morita, K.; Suzuki, Z.; Hirose, H. Bull. Chem. Soc. Jpn. 1968, 41, 2815.
(2) (a) Du, Y. S.; Lu, X. Y.; Yu, Y. H. J. Org. Chem. 2002, 67, 8901.
(b) Evans, C. A.; Miller, S. J. J. Am. Chem. Soc. 2003, 125, 12394.
(3) (a) Iwabuchi, Y.; Nakatani, M.; Yokoyama, N.; Hatakeyama, S.
J. Am. Chem. Soc. 1999, 121, 10219. (b) Langer, P. Angew. Chem., Int.
Ed. 2000, 39, 3049. (c) Chen, X. Y.; Wen, M. W.; Ye, S.; Wang, Z. X.
Org. Lett. 2011, 13, 1138.
(4) (a) Cowen, B. J.; Saunders, L. B.; Miller, S. J. J. Am. Chem. Soc.
2009, 131, 6105. (b) Guan, X. Y.; Wei, Y.; Shi, M. J. Org. Chem. 2009, 74,
6343. (c) Meng, X. T.; Huang, Y.; Chen, R. Y. Org. Lett. 2009, 11, 137.
(d) Voituriez, A.; Panossian, A.; Fleury-Bregeot, N.; Retailleau, P.;
Marinetti, A. J. Am. Chem. Soc. 2008, 130, 14030. (e) Wallace, D. J.;
Sidda, R. L.; Reamer, R. A. J. Org. Chem. 2007, 72, 1051. (f) Zhao,
G. L.; Huang, J. W.; Shi, M. Org. Lett. 2003, 5, 4737.
(5) (a) Narayan, R. S.; Borhan, B. J. Org. Chem. 2006, 71, 1416.
(b) Schomaker, J. M.; Borhan, B. J. Am. Chem. Soc. 2008, 130, 12228.
(c) Zheng, T.; Narayan, R. S.; Schomaker, J. M.; Borhan, B. J. Am.
Chem. Soc. 2005, 127, 6946.
(6) During the preparation of this manuscript, Tong and coworkers
demonstrated the feasibility of this reaction using activated chalcones as
secondary electrophiles and benzyl-2,3-butadienoate as a primary elec-
trophile in toluene as a solvent at ꢀ30 °C; Wang, X; Fang, T.; Tong, X.
Angew. Chem., Int. Ed. 2011, 50, 5361.
(7) (a) Boivin, T. L. B. Tetrahedron 1987, 43, 3309. (b) Connor, D. T.;
Young, P. A.; Strandtmann, M. V. J. Org. Chem. 1977, 42, 1364.
(c) Coppi, L.; Ricci, A.; Taddei, M. J. Org. Chem. 1988, 53, 911. (d) Gerth,
K.; Washausen, P.; Hofle, G.; Irschik, H.; Reichenbach, H. J. Antibiot.
1996, 49, 71. (e) Hayakawa, H.; Miyashita, M. Tetrahedron Lett. 2000,
41, 707. (f) Schmidt, B.; Westhus, M. Tetrahedron 2000, 56, 2421.
r
10.1021/ol202301f
Published on Web 10/04/2011
2011 American Chemical Society

Our venture into the use of acyclic enones was based on the
assumption that the increased conformational flexibility of
the enolate intermediate could lead to a facile ring closure in
preference to the slow proton transfer. As depicted in Figure
2, cyclization of the hard oxyanion onto the hard enamine, as
opposed to that observed with phosphine catalysis (cycliza-
tion of the softer carbon onto the softer vinylphosphonium;
see Figure 1), would yield the desired dihydropyran product.
Consequently, a variety of enone electrophiles were screened
for reactivity with allene ester 2. Allene ester 8, incapable of
proton transfer, was also chosen to further facilitate the
cyclization of 9. In the event, treatment of acyclic enones
(see Table 1) with ethyl-2-methyl-2,3-butadienoate (8) in the
presence of 20 mol % DABCO provided no product and the
secondary electrophile was recovered unreacted. The inert-
ness of 8in this reaction may be attributed to sterics as a result
of the R-methyl group substitution, rendering the intermedi-
ate enolate incapable of attacking the secondary electrophile
(enone). In contrast, ethyl-2,3-butadienoate (2) provided
good yields of the formal [4 þ 2] adducts, albeit via the
unanticipated attack of the γ-enolate derived from the
activation of 2 with DABCO (Figure 3, enolate 2a). In fact,
previous studies with 2 only report products that arise from
R-substitution during the BaylisꢀHillman reaction.2b,4a
Figure 3 illustrates our proposed mechanism leading to the
observed products in Table 1, highlighting the γ-substitution
of the allene ester (2af2b) and subsequent oxygen trap
(2bf2c) to yield the dihydropyran without proton transfer.
Although product 10a, formed by the γ-attack of the
allene predominates, a minor fraction of R-substituted
allene product 10b was observed (Table 1, entry 1). Opti-
mization of reaction conditions revealed that the presence
of adventitious water leads to the formation of 10b. The
Figure 2. Hypothetical [4 þ 2] addition reaction.
˚
same reaction charged with 4 A molecular sieves under
argon atmosphere yields 10a exclusively. As shown in
Table 1, several secondary electrophiles with varying sub-
stitution patterns wereemployed toinvestigate the scopeof
this reaction. Enones 11 and 12 provide the corresponding
dihydropyran product as anticipated (Table 1, entries 2
and 3). Dypnone (13), a β,β-disubstituted enone (entry 4),
did not yield product, presumably indicating the intoler-
ance of the reaction to increased sterics at the β-position of
the enone. It is noteworthy that the isolated mass balance
of the latter reactions was the unreacted enone. The allene
ester 2 does decompose at rt regardless of the absence or
presence of the secondary electrophile, an observation that
was helpful in further optimization of this reaction, leading
to high yields of products as will be described below.
We next explored the possibility of enantiocontrol at C4
by employing cinchona alkaloids (and their derivatives) as
chiral amine catalysts for the BaylisꢀHillman reaction.
Chalcone 10 was chosen as the model substrate in the reac-
tion of 2 as the primary electrophile using 10 mol % of the
chiral amine for initial screening efforts; the reactions were
performed in toluene at rt. Preliminary results were encoura-
ging since every catalyst that furnished the desired product
displayed enantioselectivity, with most surpassing 90% ee.
Not surprisingly, the monohydrochloride salts of cinchonine
(G) and cinchonidine (H) did not yield product, suggesting
that the quinuclidine nitrogen is needed for catalysis.
Although the initial screening delivered the desired
products in good enantiomeric excess, the low yields
(10ꢀ20%) were clearly a problem. Surprisingly, increasing
the catalyst loading up to 30 mol % did not make any
quantifiable difference in theisolated yields. Hatekeyama’s
catalyst (E), which reportedly enhances the rate of reaction
through H-bonding with secondary electrophiles,3a mar-
ginally improved the yield (30%), although the ee suffered
in the process (59%). Any attempt to externally activate
the secondary electrophile by addition of acidic or basic
additives led to faster decomposition of 2. A screen of
different solvents with a large range of polarities was not
conclusive, with comparable efficiencies for bothpolar and
nonpolar solvents (see Supporting Information (SI)).
We next resorted to a concentration study, mindful of the
tendency for cinchona alkaloids to aggregate at high con-
centrations (which often leads to deterioration of their
catalytic and stereoinductive ability).8 Gratifyingly, the high-
est yields were obtained under neat reaction conditions (see
SI), providing the products in synthetically useful quantities,
and also maintaining high enantiomeric excess. Since, the
catalyst and chalcone are both solids, the loading of 2 up to
Figure 1. Phosphine and quinuclidine2 catalytic pathway for
allene ester mediated addition reaction.
Org. Lett., Vol. 13, No. 21, 2011
5733

Table 1. Initial Results for [4 þ 2] Addition Reaction
Figure 3. Proposed mechanism for the [4 þ 2] addition.
Moreover, we were able to access both enantiomers by a
simple switch of the pseudoenantiomeric catalyst. Regard-
less of electronic and steric factors, the enantioselectivity of
the reaction was not greatly influenced by the substitution
pattern on R1 or R2. Figure 3 depicts a putative mechanism
for the formation of the dihydropyran products in Tables 1
and 2. Evidence for this mechanism was obtained from
ESI-MS analyses of reaction intermediates (see SI).
a Isolated yields after column chromatography. Due to thermal
sensitivity of the products, solvents were removed under vacuum with-
b
out heating. Reaction was performed in presence of 4 A MS. c Not
˚
observed under anhydrous conditions.
To further probe the basis for stereoinduction, an exhaus-
tive DFT calculation at the B3LYP/6-31G* level using
toluene as solvent was performed. A large number of possible
reaction trajectories (>20) for the approach of chalcone
relative to the adduct of catalyst B and 2 were examined
(see SI for details). The results revealed that the difference in
3ꢀ4 equiv was necessary to provide a reaction medium with
efficient mixing.9 The increased concentration along with the
higher equivalence of 2 leads to a faster reaction rate prior to
its degradation via nonproductive pathways. It is also
noteworthy that no significant deleterious effects result from
the self-aggregation of cinchona alkaloids or their deri-
vatives, most probably because stereochemical induc-
tion results after the addition of the catalyst to the
allenoate (2). Aggregation of the zwitterionic intermedi-
ate is less likely as compared to the neutral catalyst.
The scope of the reaction was tested with a number of
enones as secondary electrophiles, employing the best four
catalysts from Figure 4 (Table 2 lists only three catalysts;
complete data for all four catalysts are given in the SI). It is
evident from the results that electron donation through R1
does not favor the formation of the transient oxyanion 2b
(Figure 3) and furnishes a low product yield (Table 2, entries
2 and 9). Although, electron-withdrawing R2 groups gave
better yields (entries 4, 5, and 8), the yields are not affected
dramatically by electron-donating groups (entries 6, 11, and
16). Aliphatic enones provided the desired products in
lower yields (Table 2, entries 13-15); presumably, the
slower reaction rate leads to more reactant degradation.
Aromatic and heteroaromatic enones were stable under
the reaction conditions and furnished excellent yields of the
desired products with excellent enantioselectivity.
(8) (a) Rho, H. S.; Oh, S. H.; Lee, J. W.; Lee, J. Y.; Chin, J.; Song, C. E.
Chem. Commun. 2008, 1208. (b) Williams, T.; Pitcher, R. G.; Bommer, P.;
Gutzwill, J.; Uskokovi, M. J. Am. Chem. Soc. 1969, 91, 1871.
(9) Ethyl-2,3-butadienoate (2) was mostly decomposed after 48 h at rt
even when stored under an inert atmosphere. This rate of decomposition was
one of the reasons that compelled its use in superstoichiometric amounts.
Figure 4. Development of an asymmetric protocol.
Org. Lett., Vol. 13, No. 21, 2011
5734

Table 2. Substrate Scope of the Asymmetric [4 þ 2] Addition
Reactiona
entry
1
R1
R2
catalyst product yield % ee
Ph
p-OMe-C6H4 Ph
p-NO2-C6H4 R-naphthyl
Ph
A
D
A
D
Bb
D
A
D
A
D
A
D
A
D
A
D
A
D
A
D
A
D
A
D
Bb
D
A
D
Bb
D
A
D
A
D
10a-S
10a-R 89% 88
14a-S 39% 96
14a-R 42% 82
15a-S >99% 97
15a-R 94% 90
93% 97
2
3
4
5
6
7
8
9
Figure 5. Origin of enantioselectivity (diastereomeric transition
states TS1 and TS2 determined at B3LYP/6-31G* level). The
gauche interactions (highlighted in red bonds) make TS2 en-
ergetically less favored than TS1.
Ph
Ph
Ph
p-CN-C6H4
p-Br-C6H4
16a-S
16a-R 55% 89
17a-S
92% 96e
17a-R 82% 91e
18a-S
60% 96e
18a-R 52% 83e
19a-S
58% 95e
19a-R 49% 84e
20a-S
66% 90e
20a-R 50% 92e
21a-S 16% 95
81% 97
p-OMe-C6H4
p-NO2-C6H4 p-OMe-C6H4
p-CH3-C6H4 p-Cl-m-NO2-C6H3
p-MeO-C6H4 p-Br-C6H4
Scheme 1. Rh(II) Mediated Cyclopropanation of 24a-S
21a-R 15% 85
22a-S >99% 97e
22a-R 63% 81e
10 o-MeO-C6H4 p-F-C6H4
11 o-Br-C6H4 2-furanyl
12 m-Br-C6H4 p-Ph-C6H4
23a-S
23a-R 52% 79
24a-S
98% 98c
24a-R 70% 79
61% 96
13 CH3
n-C5H11
Ph
25a-S
25a-R
12a-S
10% 87
11% 77
45% 95
14
15
H
H
mode; moreover, upon electrophilic functionalization at
C3, the resulting oxacarbenium can undergo attack by
nucleophiles, also in a stereoselective manner. As a demon-
stration of its applicability, Rh2(OAc)4 mediated cyclopro-
panation of 24a-(S) provided product 24b in 74% isolated
yield as a single isomer by NMR (Scheme 1). The crystal
structure of 24b provides the absolute stereochemistry
of the product, suggesting that the C4 substituent is the
stereochemical driver in this reaction.
12a-R 30% 80
n-C3H7
26a-S
18% N.D.d
26a-R 12% N.D.d
27a-S >99% 97e
27a-R 58% 77e
16 o-Cl-C6H4
17 p-I-C6H4
p-OMe-C6H4
p-Br-C6H4
28a-S
28a-R 36% 87e
51% 95e
a Ratios were determined by chiral HPLC. b Catalyst loading was
20 mol %. c Reaction was performed on 1 g scale of 24. d Enantiomers
could not be resolved by HPLC analysis. e Reactions were performed
using 4 equiv of 2.
In summary, we have exploited the difference in phos-
phine and amine catalysis for the construction of novel
dihydropyrans. The employment of cinchona alkaloid
catalyzed formal [4 þ 2] addition of acyclic enones and
allenoate 2 creates an asymmetric center at C4 with
efficient stereocontrol, providing a handle for the stereo-
chemical functionalizations of these novel dihydropyrans.
The ability to access cyclic products by intermediacy of a
hard nucleophile unlocks opportunities toward assembly
of complex and novel heterocyclic products.
energy for the two diastereomeric transition states is 2.5 kcal/
mol in favor of the observed (S)-enantiomer (Figure 5). This
is in excellent agreement with the experimentally observed
selectivity of 98:2 er (see SI, Table S1, entry 14). The two
transition states in Figure 5 orient the reacting molecules such
that a close proximity of the counterions (electrostatic
stabilization) is achieved. The gauche interaction encoun-
tered in TS2 (highlighted bonds in red, Figure 5) makes this
transition state energetically more demanding than the or-
ientation suggested in TS1.
The synthetic utility of this transformation is dictated
by its ability to access both enantiomers with excellent
selectivity and its tolerance to various functional groups
under solvent-free conditions at rt. Interestingly, by ex-
ploiting the stereocenter and the rigid framework of these
molecules one can imagine a plethora of electrophiles
reacting at the nucleophilic ‘enol ether’ in a stereoselective
Acknowledgment. We acknowledge the NIH (R01-
GM082961) for generous funding. The authors thank
Dr. Daniel Whitehead (Clemson University) for illuminat-
ing discussions and review of this manuscript.
Supporting Information Available. Experimental pro-
cedures and DFT computational data. This material is
available free of charge via the Internet at http://pubs.
acs.org.
Org. Lett., Vol. 13, No. 21, 2011
5735