Organic Letters
Letter
were afforded when using CH3OH, THF, DCE, toluene, and
DMSO as the solvents (entries 12−16, respectively). More-
over, a control experiment showed that light irradiation was
critical to this transformation and the reaction did not proceed
in the dark (entry 17). The yield of product 4a decreases
without RB (entry 18).
Scheme 1. Visible Light-Induced Cascade Selenylation/
Cyclization of Unsaturated Compounds
After the determination of the optimal cascade cyclization
reaction conditions (Table 1, entry 7), different 3-amino-
indazoles were used to confirm the general adaptability of our
strategy (Scheme 2). Cyclization product 4a was analyzed by
a
Scheme 2. Substrate Scope of 3-Aminoindazoles
a
Table 1. Optimization of Reaction Conditions
b
entry
photocatalyst
additive
solvent
yield (%)
1
2
3
4
5
6
7
8
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
eosin Y
rhodamine 6G
fluorescein
Ir(ppy)3
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
Rose Bengal
−
DABCO
Cs2CO3
AcOH
TFA
ZnCl2
AlCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
FeCl3
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3OH
THF
not detected
not detected
13
18
35
66
80
75
73
65
70
53
60
66
50
45
a
Conditions: 1 (0.3 mmol), 2a (0.3 mmol), 3a (0.3 mmol), RB (3
mol %), FeCl3 (0.3 mmol), MeCN (3 mL), and irradiation with a 20
W blue LED for 12 h. Isolated yield.
9
single-crystal X-ray diffraction to ensure the accuracy of the
product structure (CDCC 2046543). Various substituted 3-
aminoindazoles reacted well with ynal 2a and diphenyl
diselenide 3a, resulting in the corresponding products 4b−4f
in 52−75% yields. Fortunately, when pyrazolo[3,4-b]-pyridines
were used in the reaction, we also obtained target products 4g
and 4h in good yields. With an electron-withdrawing group
(-CF3) at position C-4 of 3-aminoindazole, selenide product 4i
was not found.
10
11
12
13
14
15
16
DCE
toluene
DMSO
CH3CN
CH3CN
c
17
18
not detected
18
Then, the scope of ynals was investigated to test the
generality of the functional group for the reaction under the
standard conditions (Scheme 3). Ynals bearing a range of
substituents at the para position, such as -Me, -phenyl, -F, -Cl,
-Br, -CN, and -OCH3, provided corresponding products 5a−
5g in 62−75% yields. In addition, meta- and ortho-substituted
aryl ynals could work smoothly in this system, as well (5h−5j).
Likewise, a disubstituted ynal, 3-(3,4-dimethylphenyl)-
propiolaldehyde, was demonstrated to be a viable substrate,
supplying desired product 5k in satisfactory yield. Notably,
when naphthyl and thienyl substituents were applied to ynals,
the yield was still maintained at a good level (71% for 5l and
73% for 5m). Meanwhile, linear aliphatic ynals were also good
candidates for the cascade transformation and afforded
selenide product 5n in 65% yield under optimal conditions.
Next, we explored the substrate scope of the visible light-
induced cascade cyclization reaction with diverse chalcogens
(Scheme 4). A series of diselenides participating in the reaction
gave medium to high yields under the optimal conditions. For
example, the yields of products 6a−6e did not decrease
significantly when the corresponding substrates were employed
a
Conditions: 1a (0.3 mmol), 2a (0.3 mmol), 3a (0.3 mmol),
photocatalyst (3 mol %), additive (0.3 mmol), solvent (3 mL), and
irradiation with a 20 W blue LED for 12 h. Determined by GC
analysis. In the dark.
b
c
first, no desired selenide pyrimido[1,2-b]-indazoles 4a formed
when we selected bases such as DABCO and Cs2CO3 as
additives under 20 W blue LED irradiation using Rose Bengal
(RB) as the photosensitizer (entries 1 and 2, respectively).
Using organic acids in this reaction, we found only a small
amount of product (entries 3 and 4). To our delight, Lewis
acids were demonstrated to be efficient additives (entries 5−
7), in which FeCl3 was the best, giving the expected product in
80% yield (entry 7). To further optimize the condition, we
next examined the effect of photosensitizers, and the results
revealed that the efficiencies of eosin Y, rhodamine 6G,
fluorescein, and Ir(ppy)3 were lower than those of RB (65−
75% yields, entries 8−11, respectively). After screening a range
of solvents for this light-induced cyclization, we found CH3CN
to be the optimal solvent, and medium yields of product 4a
2755
Org. Lett. 2021, 23, 2754−2759