Journal of the American Chemical Society
Article
Table 5. Catalytic Asymmetric Allylation of Aldehydes with
Alkenes
Scheme 1. Mechanistic Studies
a
a
1
Yield and dr were determined by H NMR analysis of the crude
mixture using 1,1,2,2-tetrachloroethane as an internal standard.
a
General reaction conditions: 5 (0.25 mmol), 1 (5.0 mmol), CrCl2
(0.0125 mmol), ligand 8 (0.0125 mmol), HAT1 (0.025 mmol), PC1
(0.0125 mmol), and K2HPO4 (0.025 mmol) were reacted in DCM
(2.5 mL) at room temperature under blue LED irradiation for 48 h.
Yield is the isolated yield. The enantiomeric excess was determined by
chiral stationary HPLC analysis after isolation. 1 mL of 1c as a liquid
was used. ee = enantiomeric excess.
position of 1c was the exclusive pathway in the presence of
potentially reactive C−H bonds in the aldehyde substrates
(e.g., highlighted C−H bonds in 7ab and 7ac, the formyl C−H
bonds of aldehydes 5, and the allylic and benzylic C−H bonds
of products 7). Increasing the reaction scale to 1 mmol did not
affect the results (7q).
Next, we studied the reactions between more functionalized
aldehydes and 1c (Table 4). The natural product citronellal
was a competent substrate (7ad). An indometacin-conjugated
aldehyde provided product 7ae in 89% yield with high
diastereoselectivity. A cholesterol derivative containing allylic
C−H bonds also tolerated the reaction conditions (7af). The
free hydroxy group in propranolol did not affect the reaction
progress (7ag). A dipeptide and a nucleic acid derivative
containing coordinating polar functional groups (i.e., amides
and an adenine heterocycle) did not interfere with the reaction
(7ah and 7ai). Therefore, the allylation of aldehydes mediated
by ternary hybrid catalysis realized high functional group
compatibility and chemoselectivity using easily available
feedstock alkenes.
Catalytic Asymmetric Allylation. We applied this
method to a catalytic asymmetric allylation of aldehydes
(Table 5). The optimized chiral ligand was identified as
Indane-BOX ligand 8.10b,30,31 A series of alkenes afforded the
products in up to 88% ee ((S,S)-7a, 7b, 7f, (S,R)-7k). Both
aromatic and aliphatic aldehydes were also applicable ((S,S)-
7aj). In all the cases, regioselectivity was perfectly controlled to
favor branch selectivity (>20/1), although the diastereoselec-
tivity varied depending on the alkenes.32 To demonstrate the
b
7h). The C3 alkene feedstock, propene, was much less reactive
than other alkenes under the current conditions, however,
probably because of higher BDE of its allylic C−H bond (3%
yield from 5a).20,28
The aldehyde scope was then investigated by fixing the
alkene substrate to 1c (Table 3). A series of aromatic
aldehydes bearing halogens (7o−7q), electron-withdrawing
groups (7r), and a boronate ester (7s) reacted with 1c,
affording the corresponding products with high diastereose-
lectivity. A methyl substituent at the ortho, meta, and para
positions of the aromatic aldehydes did not affect the results
(7t−7v). The reactions of aliphatic aldehydes, including those
containing potentially sensitive functional groups, also
proceeded with high diastereoselectivity (7w−7y). Tertiary
amines (e.g., 7z) are generally susceptible to oxidative
conditions, providing N-cation radical or α-amino C-radical
species.29 Addition of 1 equiv of MsOH to the reaction
mixture, however, was effective for efficient reaction progress,
affording amino alcohol 7z. Sulfide moieties are also vulnerable
to various oxidative conditions; in this case, however, the
reaction proceeded in high yield without oxidizing the sulfide
moiety (7aa). When more than one C−H bond was
comparably reactive, the site selectivity was affected by the
relative concentration of the substrates: reaction at the allylic
E
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX