Organic Letters
Letter
5
α-amino acids under mild conditions. This strategy has been
extended to the synthesis of β-phenyldimethylsilyl α-amino
a
Conditions
acids by the addition of Suginome’s reagent (PhMe Si-Bpin)
2
to dehydroalanine under copper(I)-catalyzed conditions
6
(
Scheme 1b). The asymmetric synthesis of N-Boc-(R)-
silaproline was also achieved by the Rh-catalyzed intra7-
molecular hydrosilylation of dehydroalanine (Scheme 1c).
Nonetheless, these methods are far from being ideal. The
expensive and toxic transition-metal catalysts are still required
in the process. Furthermore, only a few examples were
demonstrated, accompanied by moderate reaction efficiency.
Herein we report a visible-light-mediated hydrosilylation
approach merging organophotoredox and hydrogen atom
transfer (HAT) catalysis (Scheme 1d). The protocol features
mild, green, and metal-free reaction conditions and utilizes
readily accessible and structurally diverse hydrosilanes as the
silicon pronucleophile, allowing the facile transformation of
dehydroalanine ester and analogues into a variety of β-silyl α-
amino acids, especially those with an additional alkyl or aryl
group, in moderate to high yields. Notably, the use of chiral
methyleneoxazolidinone as the substrate enabled the highly
stereoselective synthesis. The reaction could also be performed
in a continuous-flow fashion and scaled up to the gram scale.
Furthermore, the preparative power of this method was further
highlighted by its application in the direct silylation of
dehydropeptide and the synthesis of silaproline.
b
entry
substrate
HAT cat. (x mol %)
time (h)
yield (%)
c
1
2
3
4
5
6
7
8
9
2a1
2a2
2a3
2a3
2a3
2a3
2a3
2a3
2a3
2a3
2a4
2a3
2a3
2a3
2a3
H1 (10 mol %)
H1 (10 mol %)
H1 (10 mol %)
H2 (10 mol %)
H3 (10 mol %)
H4 (10 mol %)
H5 (10 mol %)
H4 (20 mol %)
H4 (50 mol %)
H4 (20 mol %)
H4 (20 mol %)
H4 (20 mol %)
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
n.d.
n.d.
c
82
40
n.d.
c
86
84
90
64
d
1
1
1
1
1
1
0
1
2
3
4
5
75
80
Recently, visible-light photoredox catalysis has emerged as a
8
c
c
c
,
e
f
n.d.
n.d.
n.d.
powerful tool for C−Si bond formation. Notably, a visible-
light-mediated hydrosilylation of alkenes using hydrosilanes
has been developed by the synergistic combination of
organophotoredox and HAT catalysis, which allows the facile
,
H4 (20 mol %)
H4 (20 mol %)
g
trace
3
a
construction of a Csp −Si bond under mild and metal-free
Reaction conditions: 24 W blue LEDs, 1a (0.3 mmol, 1.5 equiv), 2a
0.2 mmol, 1 equiv), 4CzIPN (0.006 mmol, 3 mol %), HAT cat. (0.02
to 0.1 mmol, 10−50 mol %), MeCN (1.5 mL), N , rt, unless
8
d
(
conditions. Drawing from the mechanistic evidence amassed
in this study and in connection to our continued interest in the
synthesis of medicinally valuable silicon-containing functiona-
2
b
1
otherwise noted. Yields were determined by H NMR using
c
d
8
b
mesitylene as the internal standard. Not detected. 1.1 equiv of 1a
was employed. Performed in darkness. Performed in the absence of
lities, we envisioned that this strategy could be extended to
the synthesis of β-silyl α-amino acids via the radical addition of
hydrosilanes to dehydroalanine ester.
e f
g
4
CzIPN. Under air.
The initial investigation of the proposed reaction started
with subjecting the readily accessible N-Boc-protected
dehydroalanine methyl ester 2a1 and phenyldimethylhydrosi-
lane 1a to the reaction conditions established in previous work
using 4CzIPN (3 mol %) and quinuclidine H1 (10 mol %) as
the catalysts in the presence of 24 W blue-light emitting diodes
structure failed to give the product (entry 5, Table 1). The
introduction of an acetoxyl group at the C3 position of
quinuclidine led to a slightly increased yield (entry 6, 86%
yield, Table 1), which might be attributed to the improved
solubility of the protonated HAT catalyst in CH CN.
3
8
d
(
(
LEDs). Unfortunately, the desired product was not detected
entry 1, Table 1). The introduction of an extra methyl group
However, the replacement of the acetoxyl group with a more
lipophilic benzoyloxy group could not further improve the
yield (entry 7, 84% yield, Table 1). Quinuclidin-3-yl acetate
H4 was then identified as the optimal HAT catalyst for further
screening, and 20 mol % was proved to be the optimal amount
(entries 8 and 9, 64−90% yields, Table 1). It should be noted
that the reduction of the amount of 1a to 1.1 equiv led to
inferior reaction efficiency (entry 10, 75% yield, Table 1).
(Table S2) identified the optimal conditions as follows: In the
presence of 24 W blue LEDs, 3 mol % of 4CzIPN, and 20 mol
% of quinuclidin-3-yl acetate H4, the reaction of 2a3 with 1.5
equiv of 1a was conducted in MeCN for 12 h at room
at the nitrogen of 2a1 proved to be ineffective (entry 2, Table
). However, the reaction of bis-N-Boc-protected dehydroala-
1
nine methyl ester 2a3 proceeded well, providing the product in
good yield and with excellent regioselectivity (entry 3, 82%
yield, Table 1). The formation of a distinct orthogonal
conformation enabled by the introduction of two bulky Boc
groups could weaken the conjugation effect of the nitrogen
lone pair into the adjoining π system and significantly decrease
the electronic density of the olefin, which might contribute to
5b
explain the observed reactivity. The screening of photo-
(Table S1). Surprisingly, hexamethylenetetramine H2, which is
temperature under N . Under these reaction conditions, 3a3
2
broadly used in polymer chemistry, organic synthesis, and
various biomedical applications, could also promote the
reaction as an alternative HAT catalyst, albeit with inferior
reaction efficiency (entry 4, 40% yield, Table 1). However,
triethylenediamine (DABCO, H3) with a similar rigid
was produced in excellent yield (entry 8, 90% yield, Table 1).
N-Phthaloyl-dehydroalanine methyl ester 2a4 also reacted well
under the optimized reaction conditions (entry 11, 80% yield,
Table 1). Control experiments established the importance of
the visible light, photocatalyst, and HAT catalyst, as no
9
1
407
Org. Lett. 2021, 23, 1406−1410
Zhang, Yongqiang
