124
Z. Fu et al. / Tetrahedron Letters 56 (2015) 123–126
Cheng24a and Fairlamb24b reported direct arylation of benzoxazole
C–H bonds with aryliodine(III) diacetate. With particular interest
in the transformation reaction of electron-deficient arene C–H
bond, we describe that such an attempt to use ArI(OAc)2 reagents
in an oxidative electron-deficient polyfluoroarene C–H bond
transformation resulted in the direct arylation reaction of elec-
tron-deficient polyfluoroarene with Pd/Ag bimetallic system, and
this observation represents a beneficial complement to the well
documented versions of direct arylation reaction of electron-
deficient polyfluoroarenes.
(entry 14). The effect of different loading of Ag2CO3 was the next
variable evaluated, decreasing the amount of Ag2CO3 to 1.5 equiv
gave comparable yield (75%), however, further reducing Ag2CO3
loading to 1 equiv led to a slightly lower yield (70%) (entries 15
and 16). Varying the ratio of DMSO to DMF in the mixed solvents
system led to change in the yield, as a result, replacing 5%
DMSO–DMF with 2.5% DMSO–DMF depressed the yield (entry
17). The influence of reaction temperature on the catalytic process
was also investigated, studies showed that lowering reaction tem-
perature to 110 °C did not affect the efficiency (74%), whereas
unsatisfied yield (64%) was obtained when reaction conducted at
90 °C (entries 18–19). Reaction time also had significant effect on
this reaction, and a shorter reaction time did not contribute to
increasing the yield (entry 20).
After identifying the factors influencing the reaction outcome,
the scope of this protocol with respect to polyfluoroarene was
explored by employing the conditions of entry 18 in Table 1. As
shown in Table 2, beside the pentafluorobenzene, a broad array
of polyfluoroarenes, tetrafluoro-, trifluoro-, and even some poly-
chloroarenes were chemoselectively arylated with (diacetoxy-
iodo)benzene 2a under the standard conditions to furnish
corresponding products in moderate to good yields. Electron-rich
substituents such as methyl and methoxy, electron-deficient
substituents such as bromo, trifluoromethyl, cyano, and aldehyde
groups all could be quite well tolerated, and it was observed that
the reaction was facilitated when a substituent at the para-position
of tetrafluorobenzene derivative has electron-donating effect or
p–pi conjugation effect (3a–d), on the contrary, the reaction was
undermined when an electron-withdrawing group at the
para-position of tetrafluorobenzene derivative (3e–g). 1,2,3,5-tet-
rafluorobenzene and 1,2,4,5-tetrafluorobenzene, each of which
possesses two potential reaction sites, generated mixtures of
mono- and di-substituted arylation products (3i and 3i0, 3j and
3j0) in 62% and 57% overall yield, respectively. Interestingly, aryla-
tion of pentachlorobenzene was also achieved (3h), the low yield
suggested that a C–H bond flanked with chlorine atoms was less
We initiated our investigation by taking reaction of pentaflu-
orobenzene 1a with iodobenzene diacetate 2a as the model
reaction for the optimization studies. Table 1 presents some
selected results from these optimization studies that showed the
effects of the solvent, base, and other factors on the reaction out-
come. At first, using a stoichiometric K3PO4 as base in the model
reaction in DMF solvent, it was found that the catalyst greatly
affects reaction efficiency, the reaction of 1a with 2a conducted
in the presence of 10 mol % Pd(OAc)2 as catalyst under nitrogen
atmosphere furnished desired product 2,3,4,5,6-pentafluorobiphe-
nyl 3a in 54% isolated yield in DMF at 130 °C (entry 3, Table 1),
however, either inferior yield (23%) displayed when 20 mol % CuI
instead of 10 mol % Pd(OAc)2 or no product was obtained when
Pd catalyst was absent (entries 1–2), the results indicated that Pd
catalyst has an essential role in the catalytic reaction. Subse-
quently, a brief survey of the solvents under otherwise identical
conditions, including polar and less polar solvents, revealed that
better yields could be obtained in polar solvents than that in less
polar solvent such as toluene (entries 4–9). Considering DMSO
could play an important role and function as a ligand to activate
the Pd catalyst and prevent the formation of palladium
black,11c,24b,25 thus, different bases were screened in the model
reaction in 5% DMSO/DMF (v/v) mixed solvents (entries 10–14).
Among them, K3PO4, KOAc, KOtBu, K2CO3, LiOtBu, and Ag2CO3
appeared to be the best choice and had a significant influence on
the reaction, delivering 3a in synthetically useful levels (76% yield)
Table 1
Selected results of screening the optimal conditionsa
F
F
F
F
F
F
F
F
catalyst
base, ligand, additive
F
PhI(OAc)2
+
F
temp, solvent
20 h
1a
2a
3a
Entry
Catalyst (mol %)
Base (equiv)
Temp. (°C)
Solvent
Isolated yield (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20b
—
K3PO4 (2)
K3PO4 (2)
K3PO4 (2)
K3PO4 (2)
K3PO4 (2)
K3PO4 (2)
K3PO4 (2)
K3PO4 (2)
K3PO4 (2)
KOAc (2)
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
130
110
90
DMF
DMF
DMF
DMSO
NMP
CH3CN
THF
Toluene
5% DMSO–DMF
5% DMSO–DMF
5% DMSO–DMF
5% DMSO–DMF
5% DMSO–DMF
5% DMSO–DMF
5% DMSO–DMF
5% DMSO–DMF
2.5% DMSO–DMF
5% DMSO–DMF
5% DMSO–DMF
5% DMSO–DMF
0
23
54
37
36
24
20
<5
39
43
40
<5
35
76
75
70
69
74
64
66
CuI (20)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
Pd(OAc)2 (10)
KOtBu (2)
K2CO3 (2)
LiOtBu (2)
Ag2CO3 (2)
Ag2CO3 (1.5)
Ag2CO3 (1)
Ag2CO3 (1.5)
Ag2CO3 (1.5)
Ag2CO3 (1.5)
Ag2CO3 (1.5)
110
a
Conditions: 1a (0.2 mmol), 2a (1.2 equiv), solvent (2 mL), 20 h.
Reaction run for 10 h.
b