Organic Letters
Letter
The irradiation of 4-tert-butylanisole (1a) with blue light in
the presence of K2S2O8, pyridine, and Ru(phen)3(PF6)2 (cat.) in
CH3CN/H2O (1:1) and subsequent treatment with piperidine
gave the desired primary aromatic amine 2a in an excellent yield
(Table 1, entry 1). Indeed, irradiation was essential for the
Table 1. Optimization of Photocatalytic C−H Amination*
a
yield
entry
1
photocatalyst
oxidant
K2S2O8
(%)
Ru(phen)3(PF6)2
96
b
[Ered(RuIII/RuII) = +1.37 V]
c
2
Ru(phen)3(PF6)2
K2S2O8
K2S2O8
0
0
3
4
5
Ir[dF(CF3)ppy]2(dtbpy)
b
[Ered(IrIV/IrIII) = +1.80 V]
Ru(bpz)3(PF6)2
K2S2O8
K2S2O8
40
47
b
b
[Ered(RuIII/RuII) = +1.79 V]
Ru(dfmb)3(PF6)2
[Ered(RuIII/RuII) = +2.05 V]
6
7
8
9
none
K2S2O8
0
98
99
0
Ru(phen)3(PF6)2
Ru(phen)3(PF6)2
Ru(phen)3(PF6)2
Ru(phen)3(PF6)2
Na2S2O8
(NH4)2S2O8
DDQ
10
CBrCl3
0
*
4-tert-Butylanisole (1a) (0.2 mmol), pyridine (2 mmol), oxidant
(0.4 mmol), and Ru(phen)3(PF6)2 (5 mol %) were stirred at 23 °C in
CH3CN/H2O (1/1, 5 mL) for 3 h with blue light irradiation, and the
resulting reaction mixture was treated with piperidine (10 mmol) at
a
b
70 °C for 12 h. Determined by 1H NMR analysis [Ered(Mn+1/Mn)]
is a reduction potential of a high oxidation state of photocatalysts vs
SCE. Reaction was carried out in the dark.
Figure 2. Generation of radical cations of aromatic compounds by
photoredox catalysts. (a) Previous C−H functionalization of aromatics
through their radical cations. (b) Redox potential range of applicable
substrates in previous studies. (c) This work: the photocatalytic C−H
amination of aromatics overcoming the redox potential limitations. (d)
Redox potential range of applicable substrates in the present system.
c
reaction to take place (entry 2). Interestingly, the choice of
catalyst was important; Ir[dF(CF3)ppy]2(dtbpy),10 Ru-
(bpz)3(PF6)2,11 and Ru(dfmb)3(PF6)2,12 which have reduction
potentials at high oxidation states higher than those of
Ru(phen)3(PF6)2 were not suitable (entries 3−5). Clearly, the
oxidizing power of the photocatalyst was not correlated with the
yield of product obtained. No product was obtained in the
absence of the photocatalyst (entry 6). Furthermore, the choice
of oxidant was also important, where persulfates were found to
be suitable oxidants, irrespective of which countercation was
employed (entries 7 and 8). In contrast, no desired amine was
obtained when the common organic oxidant DDQ was utilized
(entry 9). In addition, the use of CBrCl3, which is a typical
oxidant for photocatalytic reactions,13 failed to yield the desired
product (entry 10).
aromatic compounds with oxidation potentials lower than the
reduction potential of the photocatalyst excited state (Figure
2b). Although it is known that thermodynamic favorability is not
a prerequisite for successful electron transfer,7 it is a de facto
requirement for the photocatalytic C−H functionalization of
aromatics through their radical cations. For example, C−H
phosphonylation using Ru(bpz)3(PF6)2 [E*red(PC3) = +1.45
V] is only applicable to highly electron-rich aromatics such as di-
or trimethoxy-substituted benzenes but not to anisole [Eox
=
+1.76 V].6g
We herein report the photocatalytic C−H amination8,9 of
aromatics overcoming redox potential limitations. Radical
cations of aromatic compounds were generated photocatalyti-
cally using Ru(phen)3(PF6)2, which has a reduction potential at
a high oxidation state (Ered(RuIII/RuII) = +1.37 V vs SCE) lower
than the oxidation potentials of the aromatic substrates
examined herein (Eox = +1.65 to +2.27 V vs SCE) (Figure
2c). The generated radical cations are trapped with pyridine to
give N-arylpyridinium ions, which are subsequently converted to
aromatic amines. In this reaction, electron transfer from the
aromatic substrate to photocatalytically generated Ru(III) is
thermodynamically disfavored (Figure 2d); this disfavorability is
up to +22.5 kcal/mol.
The scope of the present photocatalytic amination was then
examined (Table 2), and a remarkable functional group
compatibility was revealed. More specifically, anisoles bearing
iodo, bromo, benzoyl, cyano, trifluoromethyl, and nitro groups
gave the corresponding primary aromatic amines in good to
excellent yields (entries 2−7). Significantly, 4-nitroanisole (1g)
gave the corresponding amine 2g in 83% yield, despite electron
transfer from 1g (Eox = +2.27 V) to Ru(III) [Ered(RuIII/RuII) =
+1.37 V] being disfavored by 22.5 kcal/mol, as calculated from
derivation). In contrast, aniline was not obtained from benzene
(Eox = +2.54 V), presumably due to the fact that the electron
transfer from benzene to Ru(III), which is disfavored by 28.7
B
Org. Lett. XXXX, XXX, XXX−XXX