Reducing the catalyst loading to 1 mol % led to a remarkable
decrease in the yield (entries 5), but a higher catalyst
loading did not improve the yield (entry 4). Besides the
desired product, some undesired byproducts were also
generated.10 Lewis acids were added to accelerate the
Povarov reaction between styrene and glycine imine.11
Figure 1. Radical cation salt initiated CꢀH functionalization.
FeCl3 and BF3 Et2O were ineffective, and mostly starting
3
material was recovered in both cases, despite the side
reaction being inhibited. The reaction was enhanced by
the addition of 10% InCl3, and the yield was increased to
90%. To confirm the participation of O2, the reaction was
performed under argon, and only a trace amount of
chemistry,7 we found that a commercially available persis-
tent radical cation salt, tris(4-bromophenyl)aminium hexa-
chloroantimonate (TBPAþ•), is stable in the solid state but
will be decomposed after 3 h in CH3CN under an O2
atmosphere.8 This finding implies the coupling between O2
and TBPAþ• would be possible. We wondered if such a
peroxyl radical cation could react with a substrate to
achieve CꢀH functionalization reactions, especially in a
catalytic way (Figure 1). If our hypothesis is feasible, we
will find a new class of catalyst to achieve CDC reactions.
With this idea in mind, we decided to test the reaction
between glycine derivative 1a and styrene 2a, which has
1
product was detected by H NMR analysis of the crude
reaction mixture. Interestingly, after the solution was
stirred for 24 h under argon and then exposed to air, the
reaction occurred, giving the desired product in 82% yield,
which suggests that O2 is crucial in this reaction.
We subsequently studied the scope under optimal reac-
tion conditions (Table 2). The experiments showed that
electron-rich styrenes gave higher yields of the desired
quinolines than electron-deficient styrenes. For example,
1a reacted smoothly with both 4-methylstyrene and
4-methoxystyrene to give the quinoline products in 93%
isolated yield in both cases, respectively (entries 2ꢀ3),
while the use of 4-fluoro- or 4-chlorostyrene provided the
corresponding quinolines in 67 and 79% isolated yield,
respectively (entries 6ꢀ7). Steric hindrance also has a
deleterious effect on reaction efficiency, as β- methylstyrene
reacted to form the desired product in 50% yield (entry 4).
Alkynes can also serve as dienophiles to participate in this
reaction, but the yields are lower (entries 1ꢀ4). When
indene was used in this reaction, a polycyclic quinoline
was obtained in 82% yield (entry 8). Next, a variety of
substituted glycines 1 was applied in the reaction with 2a.
The reactions of electron-rich glycine derivatives occurred
smoothly, but electron-withdrawing groups make the re-
action slower even in an O2 atmosphere due to the lower
stability of the iminium ion intermediate (entries 9ꢀ14).
Free hydroxyl groups were also well tolerated, affording
the desired product in 71% yield (entry 14).
~
been achieved by the Mancheno and Hu group using an
excess amount of TEMPO oxoammonium salt and di-tert-
butylperoxide asoxidants, respectively.9 Wefirst examined
the reaction of 1a and 2a using 1 equiv of TBPAþ•. To our
delight, 1awas consumed only after 6 h, and the product3a
was produced in 47% yield (Table 1, entry 1). A compa-
rable yield was obtained (entry 2), when 5 mol % TBPAþ•
was used, which implied catalytic oxidation prompted by a
radical cation salt is feasible. These encouraging results
prompted us to optimize the reaction conditions further
(see Supporting Information).
Table 1. Optimization of Reaction Conditions
We next turned our focus toward testing the intramolecu-
lar variant of this reaction. To our delight, the intramolecular
reaction occurred smoothly even in the absence of InCl3,
TBPAþ•
time
(h)
yield
(%)a
entry
(mmol %)
additives
1
2
3
4
5
6
7
8
9
100
5
none
none
none
none
none
6
47
6
40
(6) (a) Wurster, C.; Sendtner, R. Ber. 1879, 12, 1803. (b) Wurster, C.
Ber. 1879, 12, 2071. (c) Wieland, H. Ber. 1907, 40, 426.
10
20
1
6
64
(7) (a) Jia, X.-D.; Wang, X.-E.; Yang, C.-X.; Huo, C.-D.; Wang, W.-J.;
Ren, Y.; Wang, X.-C. Org. Lett. 2010, 12, 732. (b) Jia, X.-D.; Ren, Y.;
Huo, C.-D.; Wang, W.-J.; Chen, X.-N.; Wang, X.-C. Chin. Chem. Lett.
2011, 22, 671. (c) Jia, X.-D.; Ren, Y.; Huo, C.-D.; Wang, W.-J.; Chen,
X.-N.; Xu, X.-L.; Wang, X.-C. Tetrahedron Lett. 2010, 51, 6779. (d) Jia,
X.-D.; Han, B.; Zhang, W.; Jin, X.; Yang, L.; Liu, Z.-L. Synthesis 2006,
2831. (e) Jia, X.-D.; Lin, H.-C.; Huo, C.-D.; Wei, Z.; Lu, J.-M.; Yang, L.;
Zhao, G.-Y.; Liu, Z.-L. Synlett 2003, 1707.
6
61
12
6
21
b
10
10
10
10
FeCl3
47
BF3 Et2Ob
6
42
3
b
InCl3
6
90
<5%c
b
InCl3
24
(8) The characteristic blue color of TBPAþ• will fade after 1 h, giving
a brown solution. After 3 h, it will turn to a bright yellow solution.
a Determined 1H NMR analysis of the crude reaction mixture.
b 10 mmol % Lewis acid added. c After stirring under argon for 24 h,
the reaction solution was exposed to open air and an 82% yield of the
desired product was obtained.
~
(9) (a) Richter, H.; Mancheno, O. G. Org. Lett. 2011, 13, 6066. (b)
Liu, P.; Wang, Z.; Lin, J.; Hu, X. Eur. J. Org. Chem. 2012, 1583.
(10) We found that, in the absence of styrene, the same byproducts
were also produced, but if 10 mol % InCl3 was added, these byproducts
could not be detected. So InCl3 may be just an inhibitor of the side
reaction. Further investigation is underway in our laboratory.
(11) Review of Povarov reaction catalyzed by Lewis acid in synthesis
of quinoline: Kouznetsov, V. V. Tetrahedron 2009, 65, 2721.
We first evaluated the catalyst loading. The best yield
was obtained in the presence of 10 mol % TBPAþ• (entry 3).
Org. Lett., Vol. 14, No. 15, 2012
4031
Jia, Xiaodong
