156
V. K. Yadav et al. / Tetrahedron Letters 57 (2016) 155–158
Table 1
The literature records only a few reports on photocatalytic
Optimization of reaction conditionsa
cyclizationreactionsinvolvingthesulfurradical(Scheme1aandb).14
In view of the above discussion and our continued efforts for devel-
oping efficient heterocyclization reactions,13,15 we envisaged the
present eosin Y catalyzed visible-light-mediated aerobic oxidative
cyclodesulfurization of o-phenolic thioureas to the corresponding
2-aminobenzoxazoles (Scheme 1).
H
N
H
N
photocatalyst
base, solvent
N
N
S
OH
1a
O
visible light, air
rt, time (h)
2a
In order to realize the feasibility of our envisaged preparation of
2-aminobenzoxazoles, a model reaction was carried out by stirring
a mixture of N-substituted-2-hydroxyphenylthiourea 1a (1 mmol)
with Cs2CO3 (1.2 mmol) as a base and eosin Y as a photocatalyst in
DMF under an air atmosphere (without bubbling air) and irradia-
tion with visible light (green light emitting diodes (LEDs, 4.45 W,
kmax = 535 nm) at rt. To our delight, the desired product
2-aminobenzoxazole 2a was obtained in 90% yield after 2 h of irradi-
ation. Then, a series of control experiments were carried out, which
indicated that in the absence of any one of the reagents, catalyst/
reaction parameters, the present photo-oxidative cyclodesulfuriza-
tion could not take place (Table 1). Initially, several bases, namely
Na2CO3, K2CO3, Cs2CO3, DBU (1,8-diazabicycloundec-7-ene) and
DMAP (4-dimethylaminopyridine) were tested and Cs2CO3 was
found to work most efficiently in terms of the yield and reaction
time (Table 1, entry 1 vs 3–6). Inorganic bases are more efficient
than the amine bases probably because the latter are also involved
in the single electron transfer (SET) with the catalyst eosin Y, which
reduces the yield considerably. The bases are required to promote
the formation of a more electron donor thiolate anion from thiour-
eas. It was noted that the product 2a was also formed in the
absence of a base, but relatively a longer reaction time was
required (Table 1, entry 7). A decrease in loading of the catalyst
eosin Y from 1 mol % to 0.2 mol % resulted in lower yield of the pro-
duct (Table 1, entry 1 vs 8), and on increasing the amount of eosin
Y from 1 mol % to 2 mol % there was no effect on the yield of the
product (Table 1, entry 1 vs 9). It was noted that the product 2a
was not formed in the absence of eosin Y (Table 1, entry 10). More-
over, when the reaction was performed under a nitrogen atmo-
sphere, the desired product 2a was not formed (Table 1, entry
11). This shows that the presence of O2 is essential for the forma-
tion of 2a. However, the use of an O2 balloon instead of an air
atmosphere did not enhance the yield of 2a (Table 1, entry 1 vs
12). Similarly, when the reaction was conducted in the dark, there
was no conversion of 1a to 2a (Table 1, entry 13). These results
suggest that light, eosin Y, and O2 are essential requirements for
the reaction.
Entry Solvent Air Photocatalyst
Base (1.2 mmol) Time (h) Yieldb (%)
1
2
3
4
5
6
7
8
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
+
+
+
+
+
+
+
+
+
+
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (0.2 mol %)
EosinY (2 mol %)
—
Cs2CO3
Cs2CO3
K2CO3
Na2CO3
DBU
2
4
4
4
90
90
78
63
4
58
DMAP
—
4
30
2
n.d.
76
40
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
9
10
11
2
90
12
12
2
12
2
2
2
2
2
n.d.
n.d.
90
n.d.
69
75
67
72
36
N2 EosinY (1 mol %)
12c DMF
13d DMF
+
+
+
+
+
+
+
+
+
+
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
EosinY (1 mol %)
14
15
16
17
DCM
CH3OH
THF
CH3CN
18e DMF
19f
DMF
2
2
2
13
n.d.
88
20g DMF
21h DMF
Ru(bpy)3Cl2 (1 mol %) Cs2CO3
a
All reactions were run with 1a (1 mmol), base (1.2 mmol), and solvent (3 mL),
open to air (without bubbling air), irradiation through the flask’s bottom side using
Luxeon Rebel power green LEDs [2.50 W, kmax = 535 nm] at rt.
b
Isolated yield of 2a; n.d. = not detected.
O2 balloon was used.
Reaction was carried out in the dark.
18 W CFL.
c
d
e
f
Daylight.
g
The reaction was quenched with TEMPO (2 equiv).
Luxeon Rebel power blue LEDs [4.45 W, kmax = 447.5 nm] used for irradiation.
h
of high intensity green light. In a control study using 2 equiv of
TEMPO (2,2,6,6-tetramethyl-1-piperidinyloxyl), a well known rad-
ical-trapping reagent, no 2-aminobenzoxazole 2a was formed,
indicating that the reaction presumably involves a radical interme-
diate (Table 1, entry 20). Moreover, the reaction was also success-
ful on using Ru(bpy)3Cl2 as a photoredox catalyst under irradiation
with blue LEDs (Table 1, entry 21), but we opted to use eosin Y as
the organophotoredox catalyst in view of our goal to develop a
metal-free protocol.
Encouraged by the above studies, we investigated the substrate
scope under the optimized reaction conditions and results are
summarized in Table 2. o-Phenolic thioureas bearing electron-
donating or electron-withdrawing substituents generally afforded
2-aminobenzoxazoles in good to excellent yields (81–93%). How-
ever, phenolic thioureas with an electron-withdrawing group
afforded slightly higher yields (Table 2, entries 2d–2h, 2m, and
2n) as compared to those bearing an electron-donating group
(Table 2, entries 2b, 2c, and 2j). Interestingly, thioureas with vari-
ous functionalities such as CH3, C2H5, OCH3, Br, Cl, F, CF3, and NO2
were well tolerated to give aminobenzoxazoles 2 in excellent
yields and high purity.
With the aforementioned results, we screened several solvents,
DMF, DCM, CH3OH, THF, and CH3CN. DMF was found to be the best
solvent (Table 1, entry 1 vs 14–17). However, when the reaction
was carried out in 18 W CFL (Compact Fluorescent Light) the yield
was decreased to 36%. Similarly, the reaction also did not proceed
satisfactorily in daylight (Table 1, entry entries 18 and 19), which
shows the higher photocatalytic activity of eosin Y in the presence
Previous works
H
Ru(bpy)3Cl2
R
N
N
S
visible-light
open to air
(a)14a,b
R
S
R
eosin Y
S
N
visible-light
(b)14c
2
N
R
NH2
open to air
R
S
In view of an easy and high yielding conversion of o-aminophe-
nols into the corresponding phenolic thioureas with aryl isothio-
cyanates,12e we proceeded for one-pot synthesis of
2-aminobenzoxazole 2a starting directly from commercially
available phenyl isothiocynate and 2-aminophenol (Scheme 2).
Thus, we stirred phenyl isothiocynate (1 mmol) and
Present work
H
H
eosin Y
N
N
N
O
visible-light
H
N
R2
R1
R1
(c)
R2
S
open to air
OH
Scheme 1. Aerobic oxidative cyclization via sulfur radicals.