Organic Letters
Letter
the aminium radical cation to alkenes to give the α-amino
oxime derivatives. The photochemical behaviors of N-nitroso-
amines have been studied in detail,10 and the triplet and singlet
excited-state energy transitions of N-nitrosoamines were
identified as S0 → S2 (πN → π*NO), S0 → S1 (nN → π*NO),
and S0 → T1 (πO → π*NO).11 The burning question of whether
visible-light excitation of S0 → T1 (πO → π*NO) at 453 nm
could induce the homolytic cleavage of N-nitrosoamines to
generate the aminium radical cations arises, given that the S0
→ S2 (πN → π*NO) and S0 → S1 (nN → π*NO) transitions are
related to the homolytic N−NO bond cleavage at 230 and 360
nm, respectively.11a The implication of such visible-light-
induced aminium radical cation formation will be significant in
understanding of photochemical pathways of N-nitrosoamines
as well as the synthetic application of in situ-generated
aminium radical cations. Herein, we disclose the generation of
aminium radical cations from N-nitrosoamines under visible-
light irradiation for the first time, and the subsequent reaction
with alkenes to give the α-amino oxime derivatives (Scheme
1c).
The photochemical N−NO bond cleavage of N-nitroso-
amines was investigated in the presence of indene 2a (Table
1). Thus, the irradiation of N-nitrosopiperidine 1a and 2a
using blue LED light did not promote the formation of the
aminium radical cation (entry 1). The protonated form of 1a
with HCl failed to undergo the desired N−NO bond cleavage
in the absence of a light source (entry 2). The use of visible
light with blue LED light at 470 nm drastically changed the
outcome of the reaction, where protonated 1a smoothly
underwent the photoinduced N−NO bond cleavage, and
subsequently reacted with alkene 2a to give the corresponding
α-amino oxime 3a as respective (Z) and (E) isomers in a
combined yield of 93% (entry 3). The acidity−reactivity
relationship of protonated 1a was further examined using
different acids (entries 4−10), and the employment of acids
with pKa values of no less than −0.44 was beneficial to the
current visible-light-promoted photoaddition of N-nitrosopi-
peridine 1a to alkene 2a (TfOH pKa = −14, Tf2NH pKa =
−11.9, HCl pKa = −8.0, MsOH pKa = −2.6, TsOH pKa =
−0.51, HBF4 pKa = −0.44, HCOOH pKa = 3.77, and PhCO2H
pKa = 4.2). Solvent screening also revealed that the current
photoaddition reaction was not sensitive to the nature of
solvents, providing good to excellent yields of 3a in various
organic solvents (entries 11−17). The wavelength of the
applied light source significantly influenced the reaction rates
(entries 18−21). Thus, the reaction under the green LEDs
gave only 24% conversion within 48 h (entry 18), whereas the
reactions under purple light and UV light significantly
decreased the reaction time to 6 and 2 h, respectively (entries
19 and 20, respectively). The reaction could be also performed
under white light without much change in the reaction time or
chemical yield (entry 21). The use of 1 equiv of TsOH·H2O or
1a was tested, and the slightly reduced yields of 3a were
observed to be 77−78% (entries 22 and 23). As a result of the
optimization process, the use of TsOH·H2O was chosen to
broaden the substrate scope due to the easy handling of the
solid acid. Also, the current photoaddition reaction could be
equally performed in laboratory grade EtOH, implying the
beneficial effect of a nontoxic solvent.12 The reaction of N-
nitrosodiethylamine 1b resulted in the formation of 3b in 62%
yield under purple LED light in 48 h (entry 24). The use of N-
nitrosopyrrolidine 1c and N-nitroso 4-phenylpiperidine 1d also
provided the desired α-amino oximes 3c and 3d, respectively,
Table 1. Optimization of Photoaddition of N-Nitrosoamines
to Indene
a
b
entry
1/acid
1a/−
1a/HCl
1a/HCl
hυ
solvent
yield (%)
1
2
3
4
5
6
7
8
Blue LED’s
−
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
EtOH
i-PrOH
MeCN
acetone
EtOAc
DMF
DMSO
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
EtOH
0
0
93
95
84
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
blue LEDs
green LEDs
purple LEDs
UV-A LEDs
white LEDs
white LEDs
white LEDs
purple LEDs
purple LEDs
purple LEDs
1a/TfOH
1a/HNTf2
1a/MsOH
1a/TsOH
1a/HBF4
1a/HCOOH
1a/PhCO2H
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1a/TsOH
1b/TsOH
1c/TsOH
1d/TsOH
96
93 (92)
82
9
7
0
88
80
39
80
49
77
89
24
89
79
93
77
78
62
10
11
12
13
14
15
16
17
18
19
20
21
c
d
e
f
g
22
g
c
c
c
,h
23
24
25
26
EtOH
EtOH
66
72
a
Reaction conditions: 1 (0.6 mmol), 2a (0.3 mmol), and acid (0.6
mmol) in degassed solvent (0.1 M) under LED light at 23 °C for 16
b
h. Yield determined by 1H NMR using an internal standard (isolated
c
d
e
yield in parentheses). Reaction for 48 h. Reaction for 6 h. Reaction
f
g
for 2 h. Reaction for 15 h. Reaction with 1 equiv of TsOH.
Reaction with 1 equiv of 1a.
h
in 66−72% yields (entries 25 and 26, respectively). Thus, the
conformation of N-nitrosoamines greatly influenced the
photoaddition reactions of N-nitrosoamines.13 It is noteworthy
that the addition of aminium radical cations to 2a was highly
selective, resulting in the regioselective formation of α-amino
oximes from the more stable benzylic radical intermediate
species. This observation concurred with Chow’s results.9
While the visible-light-promoted photoaddition of N-nitro-
soamines to alkenes possesses high synthetic potential in the
preparation of 1,2-diamine derivatives,14 the toxicity associated
with the N-nitrosoamines prompted the in situ generation of
N-nitrosoamines through the aerobic oxidation of amines
(Scheme 2). Thus, piperidine was aerobically oxidized to N-
nitrosopiperidine 1a in the presence of an o-naphthoquinone
(o-NQ) catalyst.15 The subsequent one-pot photoaddition
reaction did not occur under white LED light (400−700 nm),
although the use of UV-A LED light (365 nm) provided
desired product 3a in 60% yield (Scheme 2a). Because the
3106
Org. Lett. 2021, 23, 3105−3109