Organic Letters
Letter
(3b−3g), generating the dihydropyridinone products with
excellent diastereoselectivities and high enantioselectivities.
The absolute configurations of the dihydropyridinone products
were confirmed by analogy to product 3c, whose structure was
unambiguously assigned by X-ray crystallography (CCDC
1993830).20 Next, we investigated the scope of the
substitutions at the 3-position of allylic alcohols. Encourag-
ingly, a variety of substituted phenyl groups were amenable
with the standard conditions (3h−3l), as well as a thienyl
substituent (3m). Finally, the scope of the 4-subsitutions of
oxazolone 2 was also explored, and a range of alkyl groups
were well compatible with the optimal conditions (3n−3o), as
well as the benzyl (3p) and allyl groups (3q). In addition, the
functional group-containing substituents were also well
tolerated, producing the products with high stereoselectivities
(3r and 3s).
were performed (Scheme 3). Treatment of oxazolone 2a with
the β,β-disubstituted α,β-unsaturated N-tosyl ketimine 4a9 did
Scheme 3. Control Experiments and Mechanistic Studies
To further extend the scope of these reactions, nPr/Me-
disubstituted 3-amido allylic alcohol 1t was examined under
the optimal conditions, however, which provided the
dihydropyridinone product with poor enantioselectivity.20
Encouragingly, after brief optimizations of the CPA catalysts
(see Table S2 in SI), the 2,4,6-trimethylphenyl substituted
SPINOL-derived phosphoric acid (R)-A9 was determined to
be the optimal catalyst for this reaction, which generated the
product in 70% yield with 98:2 dr and 98:2 er (Scheme 2, 3t).
Scheme 2. Scope for Enantioselective Synthesis of
Dihydropyridinones from Dialkyl-Substituted 3-Amido
Allylic Alcohols
not afford any desired dihydropyridinone product under the
standard conditions (Scheme 3a). In an attempt to prepare the
β,β-disubstituted α,β-unsaturated N-benzoyl imine 5a via
condensation9 between the corresponding enone and
BzNH2, only the dienamide product 6a was obtained.
Theoretically, this dienamide could still undergo isomerization
to give the α,β-unsaturated imine 5a under CPA catalysis
conditions; however, no dihydropyridinone product was
detected in the reaction between dienamide 6a and oxazolone
2a under the standard conditions, as well as no oxazolone ring-
opening addition product (Scheme 3b). Treatment of the
oxazolone ring-opening addition product 3a′ with the standard
conditions could not provide any dihydropyridinone product,
which indicated 3a′ was probably not an intermediate in these
cycloaddition reactions (Scheme 3c). The absence of the N-1
Bz group in the products is a key feature of these reactions,
which is also very important to understand the reaction
mechanism. Selective benzoylation of the dihydropyridinone
product 3a gave the N-1 benzoylated product 7a, whose
structure was confirmed by 2D NMR analysis. Treatment of 7a
with CPA catalyst A7 and H2O did not give any
debenzoylation product, even at higher reaction temperature,
which suggested that 7a was also not an intermediate of these
reactions21 (Scheme 3d). All these results clearly suggested
that the α,β-unsaturated N-Bz ketimine 5a is probably not the
intermediate in these reactions. Accordingly, we envisioned
that the relatively rare α,β-unsaturated N−H ketimine may be
the key intermediate of these cycloaddition reactions. To prove
this hypothesis, irradiation of the mixture of α- and γ-allyl
azides22 (8a and 8a′) by ruthenium catalysis (9a) under
fluorescent light in THF led to the formation of α,β-
unsaturated N−H ketimine (10a) in 41% conversion, which
existed as an E/Z mixture.23 After removal of the solvent, the
obtained crude mixture was directly subjected into the
standard reaction conditions with oxazolone 2a, which finally
provided the desired dihydropyridinone 3a in 39% yield with
a
a
Reactions were performed with 1 (0.15 mmol), 2a (0.1 mmol), (R)-
A9 cat (0.01 mmol), CCl4 (1 mL), AW-300 MS (10 mg) at 18 °C for
40 h. Yields were isolated yields. Dr and er values were determined by
HPLC analysis on a chiral stationary phase.
Subsequently, a series of alkyl groups were examined at the 1-
position of the allylic alcohols, which provided the
dihydropyridinones with high diastereo- and enantioselectiv-
ities (3u and 3v), including the aryl- and olefine-containing
substituents (3w and 3x). It is worth mentioning that even the
Et/Me-disubstituted 3-amido allylic alcohol could afford the
product with excellent stereoselectivity under the standard
conditions without exception (3y).
To shed light on the reaction mechanism and demonstrate
the uniqueness of these reactions, some control experiments
C
Org. Lett. XXXX, XXX, XXX−XXX