2
W. Ding et al. / Tetrahedron Letters xxx (2014) xxx–xxx
(A). Previsous work: stepwise route
With the optimal reaction conditions in hand, we then investi-
Oxone®, H2O2,
HNO2, et al.
step 1
OH
HCO2H, KOH or
gated the substrate scope of this sequential photocatalytic aerobic
oxidation and semipinacol rearrangement. As illustrated in Table 2,
a variety of structurally diverse 2-phenyl-3-benzyl indoles were
able to undergo this reaction smoothly. As for the effect of elec-
tronic properties of the substituents on the phenyl ring, the elec-
tro-rich substrates (1b and 1c) showed slightly higher reactivity
than electro-deficient one (1d). Importantly, the phenol group
(1e), which can be easily oxidized to quinone, was also well toler-
ated under the present oxidative condition, affording the desired
product in a moderate yield (2e: 45% yield). In addition, the naph-
thyl (1f) and heteroaryl (1g) substituted substrates participated in
the reaction well to give the corresponding products 2f and 2g in
47% and 51% yield, respectively. Notably, several indoles with alkyl
groups at 3-position, such as cyclohexyl (1h), iso-butyl (1i), and
allyl (1j), were also easily transformed into the corresponding
products 2h–2j in moderate to good yields.
R1
R2
heating conditions
step 2
N
R1
O
Semipinacol
R1
R2
R2
Rearrangement
N
H
N
O
H
H
R1
R2
O2
acid/base-free
& room temp.
N
(B) Our work: one-step route
• aerobic oxidation/semipinacol rearrangement sequence
(step-, atom- and redox-economy)
• visible light as driving force (photocatalysis)
• mild reacton conditions and green oxidizing agent (
O
)
2
Scheme 1. Construction of 2,2-disubstituted indolin-3-ones via semipinacol
rearrangement.
In addition to the phenyl group, variation of aryl substituents on
the 2-position of indole substrates were also successfully achieved.
As highlighted in Table 3, all these substrates can be transformed
into the corresponding products (2k–n) with moderate to excellent
yields. For example, highest isolated yield (94%) was obtained in
the case of substrate 1l with a methoxyl group on phenyl ring,
probably due to the higher susceptivity to the single electron oxi-
dation by the excited photocatalyst (please see the mechanism
analysis in Scheme 2). Moreover, we found that indole 1o with a
thienyl at 2-position was well compatible with the reaction condi-
tions and furnished the desired product 2o in 78% yield.
Finally, we turned our attention to the diversity of the benzene
ring of indole. As shown in Table 4, variation of the substation pat-
tern has no deleterious effect on the reaction efficiency (2p, 2q:
76%). Besides, various indoles bearing electron-rich (MeO) and -
deficient (F, Br) substituents at 5-position can react well to furnish
the desired products 2r–2t in moderate to good yields (52–84%).
The structure of representative product 2t was further confirmed
by X-ray crystallographic analysis.10
and document a new step-, atom- and redox-economic8 methodol-
ogy, that can be used for the preparation of a wide range of
2,2-disubstituted indolin-3-ones.
Initially, the feasibility of the proposed reaction was examined
by using 3-benzyl-2-phenyl indole 1a as the model substrate in
the presence of 5 mol % of Ir(ppy)2(dtb-bpy)PF6 under 3W white
LED irradiation. To our delight, the reaction indeed proceeded
smoothly in CHCl3 and afforded the desired product 2a in 64% yield
(Table 1, entry 1). Encouraged by this result, we continued to opti-
mize the condition by varying reaction parameters. It was found
that solvents had a dramatic impact on the reaction, and CHCl3
was identified as the best solvent of choice (entries 1–6). Subse-
quently, a series of iridium, ruthenium, and dye photocatalysts
were screened to further improve the reaction efficiency (entries
7–12). It turned out that the yield could be improved to 69% when
we applied Ru(bpy)3Cl2ꢀ6H2O as a photocatalyst, albeit with a
longer reaction time (entry 9). Notably, when the catalyst loading
was decreased to 3 mol %, the reaction could give the best result
(entry 13, 74% yield). Significantly, the direct application of air as
oxygen source also proved to be suitable for the reaction, while a
slight decrease in yield (entry 14, 56% yield) was observed. Finally,
control experiments confirmed that photocatalyst, visible light,
and oxygen were all critical for this process.9
Importantly, a gram scale reaction with 1a (4 mmol, 1.13 g) has
been carried out to prove the practicability of this sequential pro-
cess, providing the desired product 2a in 71% yield without affect-
ing the efficiency (Eq. (1)). More significantly, this sequential
photocatalytic aerobic oxidation/semipinacol rearrangement
Table 1
Optimization of reaction conditionsa
O
Bn
Photocatalyst (5 mol%)
Solvent, O2, 30 oC, 3W White LED
Bn
Ph
Ph
N
H
N
H
1a
2a
Entry
Photocatalyst
Solvent
Time (h)
Conversionb (%)
Yieldb (%)
1
2
3
4
5
6
7
8
9
10
11
12
13c
14c,e
Ir(ppy)2(dtb-bbpy)PF6
Ir(ppy)2(dtb-bbpy)PF6
Ir(ppy)2(dtb-bbpy)PF6
Ir(ppy)2(dtb-bbpy)PF6
Ir(ppy)2(dtb-bbpy)PF6
Ir(ppy)2(dtb-bbpy)PF6
Ir(ppy)2(bpy)PF6
Ir(ppy)3
CHCl3
CH2Cl2
DCE
CH3CN
DMF
DMSO
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
48
40
48
25
25
16
48
48
72
96
48
48
72
72
96
95
56
97
90
>99
97
97
90
80
93
99
95
81
64
29
32
20
17
18
60
33
69
Ru(bpy)3Cl2ꢀ6H2O
Ru(bpy)3(PF6)2
56
44
61
Ru(bpm)3(BArF)2
Eosin Y
Ru(bpy)3Cl2ꢀ6H2O
74(76)d
56
Ru(bpy)3Cl2ꢀ6H2O
a
Unless otherwise noted, the reaction was carried out with 1a (0.3 mmol), photocatalyst (5 mol %) in the solvent (3.0 mL) at 30 °C under 3W white LED irradiation.
Determined by GC.
b
c Using 3 mol % of photocatalyst.
d Isolated yield in parentheses.
e Performed under an air atmosphere.