Organic Letters
Letter
a
Scheme 1. Intra- and Intermolecular Radical Addition to
Cyanamides
Table 1. Optimization of Reaction Conditions
b
entry
variation from general conditions
none
[Ir(dtbbpy)(ppy)2]PF6
fac-Ir(ppy)3
Eosin Y−H2
Eosin Y−Na2
yield (%)
1
2
3
4
5
6
7
74
70
62
c
68
c
70
c
Methylene blue
Riboflavin
68
c
68
8
9
THF (0.25 mL) in DCE (0.25 mL)
THF (0.25 mL) in EtOAc (0.25 mL)
TBHP
62
65
64
54
68
81
75
85
10
11
12
13
14
15
16
17
TBPB
1.5 equiv of DTBP
2.0 equiv of DTBP
2.5 equiv of DTBP
for 14 h
w/o photocatalyst
in dark
trace
N.R.
a
General conditions: 1a (0.1 mmol), 2a (0.5 mL), DTBP (3.0 equiv),
Ir[(dFppy)2(dtbbpy)]PF6 ([Ir−F]) (1 mol %), r.t., 25 W blue LED
b
c
(460−470 nm), 12 h. Isolated yields. Photocatalyst (3 mol %).
DTBP = di-tert-butyl peroxide, DCE = 1,2-dichloroethane, TBHP =
tert-butyl hydroperoxide (70% solution in water), TBPB = tert-butyl
peroxyl benzoate, N. R. = no reaction.
benzimidazole (1a) and tetrahydrofuran (2a) as model
substrates using 3.0 equiv of di-tert-butyl peroxide (DTBP)
as the oxidant for photoredox transformation under irradiation
by 25 W blue light LED lamp. An array of photocatalysts,
including the extensively studied Ir-/Ru-based complexes and
organocatalysts, were screened. All the photocatalysts tested
except phenothiazine could promote this photochemical
reaction (see the ESI, Table S1). The catalysts [Ir-
(dFppy)2(dtbbpy)]PF6 ([Ir−F]), Ir(dtbbpy)(ppy)2]PF6, and
fac-Ir(ppy)3 (1 mol % loading) afforded good yields of 62−
74% (Table 1, entries 1−3). Likewise, the organocatalysts, like
Eosin Y series, methylene blue, and riboflavin, achieved
comparable efficacies with 3 mol % being used (Table 1,
entries 4−7). The use of THF mixed, respectively, with DCE
and EtOAc in 1:1 volume ratio, led to slight decrease in the
yields (Table 1, entries 8 and 9). It is worth noting that the
type of oxidant exerted a significant impact on the reaction
efficiency. The peroxides, TBHP, TBPB, and DTBP, could
trigger the radical cascade cyclization, with the former being
less effective (Table 1, entries 10 and 11). The amount of
oxidant was optimized to be 2.0 equiv, and the time of
irradiation was kept for 14 h (Table 1, entries 12−15).
Notably, the blue light with the wavelength ranging from 420
to 470 nm could efficiently facilitate the transformation (see
the ESI, Table S5). Under the optimal conditions, 85% of
compound 3 was obtained. Besides, without either photo-
catalyst or light irradiation, only trace or none of 3 was
detected by GC-MS.
cyanobenzimidazoles tested could smoothly undergo the
visible-light initiated oxidative radical cascade cyclization,
providing 6-functionalized tetracyclic benzo[4,5]imidazo[1,2-
c]quinazolines in moderate to good yields. The substituents on
the C-2 aryl rings of benzimidazoles 1, -Me, -OMe, -F, -Cl, -Br,
and -CF3, could be well tolerated in this photochemical
oxidative reaction. It is worthwhile to note that the electron-
rich groups favored the reaction while the electron-deficient
ones diminished the reaction yields. Moreover, it was found
that the functional groups located at the ortho-, meta-, and
para-position of the benzimidazolyl C-2 aryl ring afforded
comparable yields (Scheme 2, for -Me, compounds 4, 7, and
10; for -Cl, compounds 6, 8, and 14), indicating that the steric
effect imposed little influence on the reaction efficiency.
Notably, this cascade cyclization possessed an excellent
regioselectivity, and the in situ generated iminyl radicals
preferentially attacked the aryl rings at the para- position of the
substituents with the meta-substituted subtrates (Scheme 2,
compounds 7−9). Benzimidazoles 1, with 2,4-dimethyl and
3,5-dimethyl attached at the C-2 aryl ring, reacted well under
the standard conditions to furnish compounds 17 and 18 in
82% and 83% yields, respectively. Similarly, compounds 19 and
20 bearing dichloro groups were obtained in 65% and 67%
yields, respectively. Furthermore, the introduction of sub-
stituents to the fused benzene ring of benzimidazoles enriched
the substrate scope with respect to benzimidazoles 1, with high
reaction efficiencies (Scheme 2, compounds 21−23).
Next, other ethers 2 such as 1,3-dioxolane, 1,4-dioxane, and
2-methyltetrahydrofuran, as radical precursors were also
explored. To our delight, these substrates delivered their
individual cyclization products 24−26 in 35−72% yields under
Having established the optimal reaction conditions, we
turned our attention to exploring the generality of this
photoredox process. As is shown in Scheme 2, all the N-
693
Org. Lett. 2021, 23, 692−696