Palladium-catalyzed coupling reaction of perfluoroarenes with diarylzinc compounds

Article abstract of DOI:10.1002/chem.201303451

This report describes the first Pd⁰-catalyzed cross-coupling of hexafluorobenzene (C₆F₆) with diarylzinc compounds to give a variety of pentafluorophenyl arenes. This reaction could be applied to other perfluoroarenes, such as octafluorotoluene, pentafluoropyridine, and perfluoronaphthalene, to give the corresponding polyfluorinated coupling products. The optimal ligand in this catalytic reaction was PCy₃, and lithium iodide was indispensable as an additive for the coupling reaction. One of the roles of lithium iodide in this catalytic reaction was to promote the oxidative addition of one Ci￡F bond of C₆F ₆ to palladium. Stoichiometric reactions revealed that an expected oxidative-addition product, trans-[Pd(C₆F₅)I(PCy ₃)₂], generated from the reaction of [Pd(PCy ₃)₂] with C₆F₆ in the presence of lithium iodide, was not involved in the catalytic cycle. Instead, a transient three-coordinate, monophosphine-ligated species, [Pd(C₆F ₅)I(PCy₃)], emerged as a potential intermediate in the catalytic cycle. Therefore, we isolated a novel Pd^II complex, [Pd(C₆F₅)I(PCy₃)(py)], in which pyridine (py) acted as a labile ligand to generate the transient species. In fact, in the presence of lithium iodide, this Pd^II complex was found to react smoothly with diphenylzinc to give the desired pentafluorophenyl benzene, whereas the same reaction conducted in the absence of lithium iodide resulted in a decreased yield of pentafluorophenyl benzene, which indicated that the other role of lithium iodide was to enhance the reactivity of the organozinc species during the transmetalation step.

Full text of DOI:10.1002/chem.201303451

DOI: 10.1002/chem.201303451
Full Paper
&
Synthetic Methods
Palladium-Catalyzed Coupling Reaction of Perfluoroarenes with
Diarylzinc Compounds
Masato Ohashi,*[a] Ryohei Doi,[a] and Sensuke Ogoshi*[a, b]
Abstract: This report describes the first Pd0-catalyzed cross-
coupling of hexafluorobenzene (C6F6) with diarylzinc com-
pounds to give a variety of pentafluorophenyl arenes. This
reaction could be applied to other perfluoroarenes, such as
octafluorotoluene, pentafluoropyridine, and perfluoronaph-
thalene, to give the corresponding polyfluorinated coupling
products. The optimal ligand in this catalytic reaction was
PCy3, and lithium iodide was indispensable as an additive for
the coupling reaction. One of the roles of lithium iodide in
this catalytic reaction was to promote the oxidative addition
of one CÀF bond of C6F6 to palladium. Stoichiometric reac-
tions revealed that an expected oxidative-addition product,
trans-[Pd(C6F5)I(PCy3)2], generated from the reaction of
[Pd(PCy3)2] with C6F6 in the presence of lithium iodide, was
not involved in the catalytic cycle. Instead, a transient three-
coordinate, monophosphine-ligated species, [Pd(C6F5)I-
(PCy3)], emerged as a potential intermediate in the catalytic
cycle. Therefore, we isolated a novel PdII complex, [Pd(C6F5)I-
(PCy3)(py)], in which pyridine (py) acted as a labile ligand to
generate the transient species. In fact, in the presence of
lithium iodide, this PdII complex was found to react smooth-
ly with diphenylzinc to give the desired pentafluorophenyl
benzene, whereas the same reaction conducted in the ab-
sence of lithium iodide resulted in a decreased yield of pen-
tafluorophenyl benzene, which indicated that the other role
of lithium iodide was to enhance the reactivity of the orga-
nozinc species during the transmetalation step.
compounds have been reported.[3–9] Radius et al. reported the
coupling reaction of octafluorotoluene (C7F8) and perfluorobi-
phenyl (C12F10) with aryl boronic acids in the presence of a cata-
lytic amount of NHC–nickel(0) catalyst (NHC=N-heterocyclic
carbene).[3] This group also demonstrated the usefulness of
a NHC–nickel(0) complex for CÀF bond activation of hexafluor-
obenzene (C6F6); indeed, the Ni0/NHC complex showed catalyt-
ic activity for the hydro-defluorination of C6F6.[10] However, effi-
cient catalytic transformation of C6F6 in which CÀC bond for-
mation is involved is very rare. To the best of our knowledge,
only two examples of transition-metal-catalyzed CÀC bond-
forming reactions of C6F6 to give biaryl compounds have been
reported to date.[4,5] We also demonstrated the reaction of C6F6
with an aryl boronate catalyzed by Radius’ complex.[6] These re-
ports suggest that nickel catalysts with strong electron-donat-
ing ligands are efficient for carbon–fluorine bond activa-
tion.[8,11,12] Yoshikai and Nakamura et al. reported the coupling
reaction of multifluorinated benzenes with arylzinc reagents
catalyzed by nickel ligated with alkoxydiphosphine; the same
group also achieved the selective activation of a CÀF bond.[5]
From a practical point-of-view, however, there is still no easily
accessible catalyst system applicable for the coupling reaction
that could introduce a perfluorinated aryl group to a specific
position of arene compounds. Recently, we reported the palla-
dium-catalyzed coupling reaction of tetrafluoroethylene with
diarylzinc reagents to give triflurorovinylarenes, in which lithi-
um iodide was crucial in the oxidative-addition step because
of the formation of a very strong LiÀF bond.[13] This reaction
suggested the possibility of cleavage of the unreactive
Introduction
Efficient methods have been developed for the synthesis of or-
ganofluorine compounds because functionalized fluorinated
organic compounds are indispensable in our daily life. These
methods are divided into two major approaches. The first is se-
lective monofluorination at a specific position of an organic
compound, which is an essential transformation for the effi-
cient synthesis of pharmaceuticals and agrochemicals. Thus, re-
actions and reagents for selective fluorination have been de-
veloped at a rapid rate.[1] The second approach is the selective
transformation of a specific carbon–fluorine bond of a per-
fluoro- or multifluoro-compound, which allows the preparation
of organic compounds that include a fluorinated functional
group.[2] In particular, the transformation of perfluoroarenes is
an efficient and economical method for the preparation of
highly functionalized organofluorine compounds and, there-
fore, a limited number of coupling reactions with aryl-metal
[a] Dr. M. Ohashi, R. Doi, Prof. Dr. S. Ogoshi
Department of Applied Chemistry, Faculty of Engineering
Osaka University, Suita, Osaka 565-0871 (Japan)
Fax: (+81)6-6879-7394
E-mail: ohashi@chem.eng.osaka-u.ac.jp
ogoshi@chem.eng.osaka-u.ac.jp
[b] Prof. Dr. S. Ogoshi
JST, Advanced Catalytic Transformation Program
for Carbon Utilization (ACT-C)
Suita, Osaka 565-0871 (Japan)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201303451.
Chem. Eur. J. 2014, 20, 2040 – 2048
2040
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
carbon–fluorine bond of C6F6 by the cooperation of palladi-
um(0) and lithium iodide. Herein, we report the coupling reac-
tion of perfluoroarenes (e.g., C6F6) with diarylzinc compounds,
catalyzed by palladium(0) in the presence of lithium iodide. In
addition, this report also introduces a possible reaction path-
way for transmetalation with aryl zinc reagents based on cer-
tain stoichiometric reactions and the robustness of trans-[Pd-
(C6F5)I(PCy3)2], formed by the oxidative addition of C6F6 to
[Pd(PCy3)2] in the presence of lithium iodide.
pling reaction (Table 1, entry 6). A mixture of [Pd2(dba)3] and
PCy3 (5 and 20 mol%, respectively) showed catalytic activity to
give 1 in 77% yield, whereas a greater palladium catalyst load-
ing (10 mol% [Pd2(dba)3]) was required for smooth progress in
the coupling reaction (Table 1, entry 7). When either DCPE (1,2-
dicyclohexylphosphinoethane) or DCPB (1,4-dicyclohexylphos-
phinobutane) were employed the coupling reaction was clearly
retarded (Table 1, entries 8 and 9).
The palladium-catalyzed coupling reaction of perfluoroar-
enes with a variety of Ar2Zn reagents in the presence of LiI
was carried out and the results are summarized in Scheme 1.
Results and Discussion
Optimization and substrate scope in the Pd0-catalyzed cou-
pling of perfluoroarenes with diarylzinc compounds
We first applied the reaction conditions of the coupling reac-
tion of TFE (tetrafluoroethylene) with Ar2Zn[13] to the coupling
reaction of C6F6 with Ar2Zn (Table 1). In the presence of tris(di-
benzylideneacetone)dipalladium(0) ([Pd2(dba)3], 5 mol% (com-
Table 1. Optimization of the catalytic reaction.
Entry
Catalyst ([mol%])
Additive
t [h]
Yield [%][a]
1
2
3
4
[Pd2(dba)3] (5)/PPh3 (20)
[Pd(PCy3)2] (5)
none
[Pd(PCy3)2] (5)
[Pd(PCy3)2] (5)
[Pd(OAc)2] (5)/PCy3 (10)
[Pd2(dba)3] (5)/PCy3 (20)
[Pd(OAc)2] (5)/DCPE (5)
[Pd(OAc)2] (5)/DCPB (5)
LiI (2.4 equiv)
LiI (2.4 equiv)
LiI (2.4 equiv)
LiI (3.6 equiv)
none
LiI (2.4 equiv)
LiI (3.6 equiv)
LiI (2.4 equiv)
LiI (2.4 equiv)
10
4
21
6
10
4
4
trace
70
–
75
5
65
77
13
5
6[b]
7
8[b]
9[b]
9
15
trace
[a] Tetradecane was used as an internal standard. [b] ZnPh2 (0.7 equiv)
was employed.
parable to 10 mol% Pd0 atom)), PPh3 (20 mol%), and LiI
(2.5 equiv) at 608C in THF, the reaction of C6F6 with Ph2Zn, pre-
pared in situ by treatment of ZnCl2 with PhMgBr (2 equiv),
gave a trace amount of pentafluorophenyl benzene (1) and
C6F6 remained intact (Table 1, entry 1). To promote the oxida-
tive addition of C6F6 to palladium, [Pd(PCy3)2] was examined as
a catalyst precursor for the coupling reaction. The desired
product 1 was obtained in 70% yield (Table 1, entry 2).[14,15] In
the absence of a palladium catalyst, no coupling product was
observed (Table 1, entry 3). An increase in the amount of LiI im-
proved the yield of 1 to 75% (Table 1, entry 4). In the absence
of LiI, 1 was obtained in 5% yield, even after a prolonged reac-
tion time (Table 1, entry 5), which indicates that the addition of
LiI is crucial for the occurrence of the coupling reaction. This
result contrasted with the reaction of TFE and Ph2Zn, generat-
ed from PhMgBr and ZnCl2, in which 44% of the coupling
product was obtained.[13] In the presence of PCy3, palladium(II)
acetate (Pd(OAc)2) was also an effective catalyst for the cou-
Scheme 1. Palladium(0)-catalyzed coupling reaction of perfluoroarenes with
diarylzinc compounds in the presence of LiI.
Both (4-MeC6H4)2Zn and (3-MeC6H4)2Zn reacted with C6F6 to
give coupling products 2 and 3 in 70 and 53% yield, respec-
tively. However, no coupling reaction product was obtained
from the reaction of C6F6 with (2-MeC6H4)2Zn. The aryl zinc
compounds with electron-donating groups such as (4-
Me2NC6H4)2Zn and (4-MeOC6H4)2Zn afforded the coupling com-
pounds 4 and 5 in 74 and 76% yield, respectively. The reac-
Chem. Eur. J. 2014, 20, 2040 – 2048
www.chemeurj.org
2041
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tions of aryl zinc reagents with electron-withdrawing groups,
(4-FC6H4)2Zn and (3,5-F2C6H3)2Zn, also yielded the correspond-
ing coupling products (6 and 7) in 66 and 49% yield, respec-
tively. The reaction of C6F6 with (2-C10H9)2Zn under the same re-
action conditions gave 2-pentafluorophenylnaphthalene (8) in
65% yield after 8 h. When a thienyl group was introduced, the
reaction gave 2-pentaphenylthiophene (9) in 55% yield. Other
functionalized aryl zinc species prepared according to Kno-
chel’s procedure,[16] LiCl·(p-EtCOOC6H4)ZnI and LiCl·(p-
NCC6H4)ZnI, were successfully applied to the coupling reaction
with C6F6 to give 10 and 11, respectively, in moderate isolated
yields.
Scheme 2. Stoichiometric reaction of C6F6 with Pd0/PCy3 in the presence of
LiI.
The reaction was applicable to other perfluoroarenes. The
coupling reaction of octafluorotoluene (C7F8) with Ph2Zn, (4-
MeOC6H4)2Zn, or (2-MeC6H4)2Zn took place at the 4-position of
C7F8 to give the corresponding products 12–14 in good-to-ex-
cellent yields. The reaction of C7F8 with (4-MeOC6H4)2Zn pro-
ceeded very smoothly, which allowed the confirmation of
a background reaction. In the absence of [Pd(PCy3)2], 13 was
obtained in 30% yield at 608C for 6 h, which indicates that the
palladium-catalyzed coupling reaction proceeds much faster
than the background reaction. Perfluoronaphthalene and per-
fluorobiphenyl reacted with (4-MeOC6H4)2Zn to give 2-(4-
MeOC6H4)C10F7 (15) and 4’-(4-MeOC6H4)C12F9 (16) in 53 and
32% yield, respectively. In contrast, the reaction of pentafluor-
opyridine (C5F5N) with Ph2Zn gave a mixture of tetrafluoro-4-
phenylpyridine (17) and tetrafluoro-2-phenylpyridine (17’) in
65 and 17% yield, respectively. Pentafluorobenzene also suc-
cessfully participated in the coupling reaction with Ph2Zn,
however, the reaction product was obtained as a mixture of
two regioisomers, 2,3,5,6-tetrafluorobiphenyl (18) and 2,3,4,5-
tetrafluorobiphenyl (18’), and the combined yield of the iso-
mers was only 38%.
not take place, which is consistent with the observation that
no reaction occurred in the presence of PPh3 (Table 1, entry 1).
The ORTEP representation of 19 unambiguously demon-
strates that the palladium center in 19 adopts a square-planar
coordination geometry and is coordinated by two PCy3 ligands
in a trans manner (Figure 1). A similar coordination geometry
was observed in structurally well-defined PdII complexes, such
as trans-[Pd(C6F5)Cl(PPh3)2] and trans-[Pd(C6F5)I(PCy2Fc)2] (Fc=
ferrocenyl).[18]
Oxidative addition of perfluoroarenes to Pd0 in the presence
of lithium iodide
To gain deeper insight into the reaction pathway, stoichiomet-
ric reactions of C6F6 with palladium(0) complexes were con-
ducted. In a previous report by Grushin et al., the reaction of
C6F6 with [Pd(PCy3)2] in THF at 708C for 24 h occurred very
slowly to give a perfluorophenylpalladium(II) fluoride, trans-
[Pd(C6F5)F(PCy3)2], in 3% yield.[17] On the other hand, in the
presence of lithium iodide the oxidative addition proceeded
much faster to give a perfluorophenylpalladium(II) iodide,
trans-[Pd(C6F5)I(PCy3)2] (19), which indicates that acceleration of
the oxidative addition is an important role of lithium iodide
(Scheme 2a). Although the catalytic reaction of C6F6 with
Ph2Zn occurred in the presence of [Pd2(dba)3] and PCy3
(4 equiv) to give 1 in 77% yield (Table 1, entry 7), oxidative ad-
dition did not occur in the presence of DBA in the stoichiomet-
ric reaction at 608C (Scheme 2b). This result might have been
due to inhibition of the coordination of C6F6 to palladium by
DBA under the stoichiometric reaction conditions and could in-
dicate why [Pd(PCy3)2] is a more efficient catalyst than the
[Pd2(dba)3]/PCy3 system. In contrast, even in the presence of
lithium iodide, the oxidative addition of C6F6 to [Pd(PPh3)4] did
Figure 1. ORTEP drawing of 19 with thermal ellipsoids at the 30% probabili-
ty level. Hydrogen atoms are omitted for clarity.
Similar oxidative-addition products, trans-[Pd(p-CF3C6F4)I-
(PCy3)2] (20) and trans-[Pd(2-C10F7)I(PCy3)2] (21), can be isolated
by treatment of either C7F8 or perfluoronaphthalene with
[Pd(PCy3)2] in the presence of lithium iodide (Scheme 3). In the
former reaction, the CÀF bond at the 4-position of perfluoroto-
luene was exclusively cleaved, whereas the CÀF bond at the 2-
position of perfluoronaphthalene was exclusively activated in
the latter reaction. These regioselectivities of CÀF bond activa-
tion were consistent with those observed in the corresponding
catalytic reaction (Scheme 1), as well as with those observed in
the reactivity of [{(NHC)2Ni}2(cod)] (cod=1,5-cyclooctadie-
ne).[3,10b] The ORTEP drawings of 20 and 21 are represented in
Figure 2, and definitely show that the palladium center in both
20 and 21 has the same coordination geometry as in 19.
Chem. Eur. J. 2014, 20, 2040 – 2048
www.chemeurj.org
2042
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 4. Reaction of 19 with diphenylzinc in the presence of LiI.
chiometric reaction conducted at 608C for 7 h in the presence
of an excess amount of lithium iodide (Scheme 4), whereas
1 was obtained in 69% yield under the catalytic reaction condi-
tions mentioned above (608C, 6 h; Scheme 1). This result
strongly indicates that 19 is unlikely to be a reaction inter-
mediate due to the steric hindrance around the palladium
center caused by the two bulky PCy3 ligands. Thus, we as-
sumed that oxidative addition of C6F6 to [Pd(PCy3)2] in the
presence of lithium iodide might involve dissociation of a PCy3
ligand to give a transient [Pd(C6F5)I(PCy3)] species. The resul-
tant three-coordinate transient intermediate would undergo
re-coordination of a PCy3 ligand in the absence of ZnPh2 to
yield the thermodynamically favored, and unreactive, species
19. On the other hand, in the presence of ZnPh2, transmetala-
tion between the transient iodopalladium(II) species and ZnPh2
would take place smoothly to give the coupling product 1.
These assumptions are consistent with the results from kinetic
studies performed by Hartwig and co-workers: the oxidative
addition of chlorobenzene to [Pd(PCy3)2], to give trans-[PdCl-
(PCy3)2(Ph)] involved the dissociation of a PCy3 ligand at the in-
itial stage of the reaction.[19] Unfortunately, any attempt to pre-
pare the transient intermediate failed due to its coordinative
unsaturation and, therefore, [Pd(C6F5)I(PCy3)(py)] (22), in which
pyridine acts as a labile ligand to generate a tentative three-
coordinate [Pd(C6F5)I(PCy3)] species, was prepared as an alter-
native catalytic precursor.
Scheme 3. Stoichiometric reaction of either C7F8 or C10F8 with [Pd(PCy3)2] in
the presence of LiI.
A synthetic route for preparation of 22 is summarized in
Scheme 5. We chose [Pd(C6F5)2(py)2] (23)[20] as a starting materi-
al. In accordance with the literature,[21] treatment of [PdCl2(py)2]
with C6F5Li (3 equiv), generated in situ by reaction of C6F5Cl
with nBuLi at À788C, gave 23 in 72% yield. The reaction of 23
with PdCl2 in acetone,[22] followed by treatment with PCy3 in
Figure 2. ORTEP drawings of 20 (top) and 21 (bottom) with thermal ellip-
soids at the 30% probability level. One of the two independent molecules
in a unit cell of 20 is depicted. Hydrogen atoms and solvate molecules
(hexane) are omitted for clarity.
Reactivity of 20 toward Ph2Zn and preparation of [Pd(C6F5)I-
(PCy3)(py)] (22)
To confirm whether or not 19 is an intermediate in the Pd0-cat-
alyzed cross-coupling reaction of C6F6 with diarylzinc com-
pounds, a stoichiometric reaction of 19 with diphenylzinc was
carried out. A yield of only 5% of 1 was obtained from a stoi-
Scheme 5. Preparation of 22–24.
Chem. Eur. J. 2014, 20, 2040 – 2048
www.chemeurj.org
2043
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
pyridine, resulted in the formation of [Pd(C6F5)Cl(PCy3)(py)] (24)
in 32% yield. Substitution of an iodide for the chloride ligand
in 24 was accomplished by treatment of a solution of 24 in
acetone with excess NaI to afford the desired palladium(II)
iodide 22 in 55% yield.
In the 19F NMR spectrum of the bis(pentafluorophenyl)palla-
dium complex 23, three resonances assigned to the ortho-,
meta-, and para-positions of the pentafluorophenyl rings ap-
peared with an integral ratio of 2:2:1, and the combustion
analysis of 23 agreed with the original literature. However, an
X-ray diffraction study of 23 revealed that the PdII center had
a square-planar geometry and was coordinated by two penta-
fluorophenyl ligands in the cis configuration, although a trans
configuration was proposed in the original literature
(Figure 3).[20] This result clearly demonstrates that, at least in
Figure 4. ORTEP drawings of 24 (top) and 22 (bottom) with thermal ellip-
soids at the 30% probability level. Hydrogen atoms and solvate molecules
(THF) in 22 are omitted for clarity.
Figure 3. ORTEP drawing of 23 with thermal ellipsoids at the 30% probabili-
ty level. Hydrogen atoms are omitted for clarity.
Reactivity of 22 toward diphenylzinc
the crystal lattice, 23 is the cis isomer. Both the pyridine and
pentafluorophenyl rings in 23 were tilted away from the palla-
dium coordination plane to reduce the steric repulsion (dihe-
dral angles=81.91(9) and 66.37(8)8 for the pyridine rings;
78.61(9) and 68.93(8)8 for the pentafluorophenyl rings).
We next evaluated the reactivity of 22 toward diphenylzinc in
the presence or absence of lithium iodide; the results are sum-
marized in Scheme 6. In the presence of lithium iodide
The novel pentafluorophenyl palladium(II) halides 22 and 24
were characterized by NMR spectroscopy and elemental analy-
1
sis, as well as by X-ray diffraction analysis. The H NMR spectra
of these complexes clearly showed that both the pyridine and
PCy3 ligands were coordinated to the PdII center in a ratio of
1:1. In addition, the ligands in 22 and 24 were situated in
a mutual cis position with a square-planar PdII geometry, as
shown by X-ray diffraction (Figure 4). The PdÀN bond lengths
of 2.032(10) ꢁ in 22 and 2.0407(13) ꢁ in 24 were slightly short-
er than those observed in 23 (2.093(2) and 2.105(3) ꢁ), which
reflects the difference in the trans influence between halides
and a pentafluorophenyl ligand. In contrast, the PdÀC6F5 bond
lengths showed no significant differences (2.015(5) ꢁ for 19,
2.069(12) ꢁ for 22, 2.001(3) and 2.025(3) ꢁ for 23, and
2.0519(16) ꢁ for 24). In addition, the bond lengths between
the palladium and phosphorus atoms in 22 and 24 (2.359(3)
and 2.3604(4) ꢁ, respectively) were close in value to those ob-
served in 19 (2.3691(16) and 2.3839(16) ꢁ).
Scheme 6. Reactions of 22 with ZnPh2 in the presence or absence of LiI.
(1.5 equiv), 22 reacted smoothly with ZnPh2, which was careful-
ly purified by sublimation prior to use, in THF at room temper-
ature to give the desired coupling product 1 as the sole prod-
uct in 63% yield. The addition of PCy3 to this reaction mixture
affected neither the yield nor the selectivity of the reaction
product. On the other hand, in the absence of lithium iodide,
the reaction of 22 with ZnPh2 under the same reaction condi-
tions afforded a pentafluorophenylzinc species, C6F5ZnX (X=I
or C6F5), as the major product (54%) and 1 as the minor prod-
Chem. Eur. J. 2014, 20, 2040 – 2048
www.chemeurj.org
2044
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
uct (27%). These observations suggest the following: 1) a tran-
sient [Pd(C6F5)I(PCy3)] species, generated by dissociation of the
labile pyridine ligand of 22, could be crucial for the smooth oc-
currence of transmetalation between the palladium(II) species
and ZnPh2, and 2) the existence of lithium iodide is essential
for selective transmetalation to generate 1.
Based on these results, a plausible reaction mechanism was
proposed (Scheme 7). In the presence of lithium iodide, oxida-
tive addition of a CÀF bond in C6F6 to Pd0 would occur by dis-
Conclusion
We have demonstrated the Pd0/PCy3-catalyzed cross-coupling
reaction of C6F6 with a variety of diarylzinc compounds to give
the corresponding pentafluorobiaryl compounds in good-to-
excellent yields. Stoichiometric reactions that employed model
complexes, trans-[Pd(C6F5)I(PCy3)2] and [Pd(C6F5)I(PCy3)(py)],
with diphenylzinc in the presence of lithium iodide revealed
both the catalytic reaction mechanism and the role of lithium
iodide in this catalytic reaction. The key intermediate in this
catalytic cycle is a transient, three-coordinated, monophos-
phine-ligated species, [Pd(C6F5)I(PCy3)], which was generated
by oxidative addition of the CÀF bond of C6F6 to [Pd(PCy3)2],
followed by dissociation of a PCy3 ligand. The role of lithium
iodide in this catalytic reaction was not only to accelerate the
oxidative addition step, but also to generate a reactive zincate,
such as Li[ArZnXI] (X=Ar or I), which would enable an efficient
transmetalation with the key intermediate. We have also dem-
onstrated that this catalytic reaction could be applied to other
monocyclic perfluorinated compounds, such as octafluoroto-
luene and pentafluoropyridine, as well as to polycyclic per-
fluorinated compounds, such as perfluoronaphthalene and
perfluorobiphenyl, to give the corresponding coupling prod-
ucts.
Experimental Section
Scheme 7. A plausible mechanism.
General
sociation of a PCy3 ligand to form a [Pd(C6F5)I(PCy3)] intermedi-
ate (A). We assume that the LiI-promoted CÀF bond activation
of C6F6 on [Pd(PCy3)2] would take place to give the cis-oxida-
tive-addition product, then the rapid dissociation of a PCy3
ligand might occur due to the steric hindrance of the two
bulky phosphine ligands. On the basis of theoretical and ex-
perimental studies, Radius et al. assumed that a related CÀF
bond activation of C6F6 on a Ni(NHC)2 fragment, to yield trans-
[Ni(NHC)2(C6F5)F], would proceed by the corresponding cis-oxi-
dative-addition product.[10b] Transmetalation between A and
the diarylzinc compound in the presence of lithium iodide
would take place to give a bisarylpalladium(II) intermediate (B).
This transmetalation step would progress in preference to the
re-coordination of a PCy3 ligand to give unreactive trans-[Pd-
(C6F5)I(PCy3)2] (A’). The role of lithium iodide in this reaction
step might be rationalized by the formation of a reactive zin-
cate, such as Li[ArZnXI] (X=Ar or I), which would enable the
efficient formation of B.[13,23] Then, reductive elimination from
B, followed by the re-coordination of a PCy3 ligand would yield
the coupling product, along with regeneration of the Pd0 spe-
cies.
All manipulations were conducted under a nitrogen atmosphere
by using standard Schlenk or drybox techniques. 1H, 19F, 31P, and
13C NMR spectra were recorded on a Bruker Avance III 400 spec-
1
trometer. The chemical shifts in the H and 13C NMR spectra were
recorded relative to residual protic solvent. The chemical shifts in
the 31P NMR spectra were recorded by using aqueous H3PO4 (85%)
as an external standard. The chemical shifts in the 19F NMR spectra
were recorded relative to a,a,a-trifluorotoluene (d=À62.8 ppm in
CDCl3). Recycling preparative HPLC was performed with a Japan
Analytical Industry LC9225NEXT instrument equipped with JAIGEL-
1H and JAIGEL-2H in-line columns. Elemental analyses were per-
formed at the Instrumental Analysis Center, Faculty of Engineering,
Osaka University. X-ray crystal data were collected with a Rigaku
RAXIS-RAPID Imaging Plate diffractometer or Mercury 375R/M CCD
(XtaL LAB mini) diffractometer.
Materials
The degassed and distilled solvents (toluene and hexane) used in
this work were commercially available. THF, [D8]THF, and C6D6 were
distilled from sodium benzophenone ketyl. All the Grignard re-
agents used in this work were purchased from Aldrich as THF
solutions and their concentrations were determined by titration
with
presence of 1,10-phenanthroline as an indicator. Trans-bis-
(pyridine)dichloropalladium(II),[21] [Pd(PCy3)2],[24]
solution of
a solution of sec-BuOH in absolute m-xylene in the
Another possible route for the coupling reaction might in-
volve the formation of a dimer intermediate, [Pd(C6F5)(m-
I)(PCy3)]2. Hor et al. argued for the possibility that both catalytic
pathways, via mononuclear cis/trans geometric isomers and via
a dinuclear iodide-bridged intermediate, may contribute to the
Pd0-catalyzed coupling reaction of pentafluorophenyl iodide
with phenylboronic acid.[18b]
a
LiCl·(p-cyanophenyl)zinc iodide in THF,[16] and LiCl·(p-EtCOOPh)zinc
iodide[16] were prepared by published procedures. Zinc chloride
(3N) was purchased from WAKO Pure Chemicals, and dried under
vacuum with heating until melting. Other commercially available
reagents were distilled and degassed prior to use.
Chem. Eur. J. 2014, 20, 2040 – 2048
www.chemeurj.org
2045
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
À788C. Then, [PdCl2(py)2] (670 mg, 2.0 mmol) was added to the so-
lution. The resultant yellow suspension was stirred at À788C for
1 h, then warmed to rt and stirred for 2 h. The resulting white sus-
pension was quenched with ether (5 mL) that contained a small
amount of water and evaporated to dryness. The residue was ex-
tracted with boiling acetone and the acetone solution was filtered
through a pad of Celite and dried out. Recrystallization from hot
acetone/ethanol at À308C overnight afforded 23 as white needles
General procedure for the optimization of the catalytic reac-
tion
In a drybox, ZnCl2 (9.54 mg, 0.07 mmol), PhMgBr (1m in THF,
140 mL, 0.14 mmol), LiI (32.1 mg, 0.24 mmol), and THF (160 mL)
were added to a vial equipped with a stirrer bar. A solution of
Pd(OAc)2 (1.1 mg, 0.005 mmol) in THF, PCy3 (2.8 mg, 0.010 mmol),
1 (11.5 mL, 0.1 mmol), and tetradecane (26 mL, 0.1 mmol), as an in-
ternal standard, were added to the mixture. The vial was sealed
and heated, with stirring, by means of a preheated sand bath.
After the reaction was complete, the solution was quenched with
methanol and analyzed by GC. The yield was estimated by compar-
ison of the peak areas of pentafluorobiphenyl and tetradecane
with a sensitivity ratio determined by the GC spectra of isolated
samples.
1
(862 mg, 72%). H NMR (400 MHz, CDCl3, rt): d=8.74 (m, 4H), 7.64
(tt, J=7.8, 1.5 Hz, 2H), 7.23 ppm (m, 4H); 19F NMR (376 MHz, CDCl3,
rt): d=À122.4 (m, 4F), À160.4 (t, J=19.5 Hz, 2F), À162.5 ppm (m,
4F); 13C NMR (100 MHz, CDCl3, rt): d=153.4, 137.8, 125.4 ppm (sig-
nals assigned to the C6F5 moiety could not be detected due to
multiple 13C–19F couplings), elemental analysis calcd (%) for
C22H10F10N2Pd: C 44.13, H 1.68, N 4.68; found: C 44.11, H 1.89, N
4.74; X-ray data for 23: Mr =598.72; colorless; monoclinic; P21/
c (no. 14); a=9.8891(10), b=16.9318(13), c=13.0111(12) ꢁ; b=
General procedure for the Pd0-catalyzed coupling of per-
fluoroarenes with diarylzinc compounds in the presence of
LiI
109.524(3)8; V=2053.3(3) ꢁ3; Z=4; 1calcd =1.937 gcmÀ3
À150(0)8C; R1 (wR2)=0.0355 (0.0786).
;
T=
In a drybox, a solution of arylmagnesium halide (1.2 mmol) in THF
and ZnCl2 (81.8 mg, 0.6 mmol) were added to a vial equipped with
a stirrer bar. The mixture was diluted with THF (total volume=
5 mL) and vigorously stirred until ZnCl2 dissolved completely.
[Pd(PCy3)2] (33.3 mg, 0.05 mmol), LiI (321 mg, 2.4 mmol), and per-
fluroarene (0.1 mmol) were added to the solution. The reaction
mixture was heated with stirring, then quenched with an aqueous
solution of 1m HCl (15 mL). The water layer was separated and ex-
tracted with ether (4ꢂ5 mL). The combined organic layers were
dried over MgSO4, filtered, and evaporated to dryness. The result-
ing solid was purified by flash column chromatography to give
pure product.
Preparation of [Pd(C6F5)Cl(py)2][22]
Compound 23 (599 mg, 1.0 mmol), PdCl2 (195 mg, 1.1 mmol), and
acetone (35 mL) were added to a round-bottomed flask equipped
with a stirrer bar. The resultant reddish-brown suspension was
heated at reflux for 3 h with vigorous stirring. After the reddish-
brown suspension of PdCl2 disappeared, pyridine (1 mL) was
added. After an additional 30 min of heating at reflux, the volatile
compounds were removed by evaporation. The resulting solid was
extracted with Et2O. The organic layer was evaporated to dryness
and recrystallization from acetone afforded [Pd(C6F5)Cl(py)2] as
1
white needles (541 mg, 58%). H NMR (400 MHz, C6D6, rt): d=8.60
(m, 4H), 6.44 (tt, J=7.8, 1.5 Hz, 2H), 6.13 ppm (m, 4H); 19F NMR
(376 MHz, C6D6, rt): d=À125.1 (m, 2F), À162.0 (t, J=20.2 Hz, 1F),
À164.9 ppm (m, 2F); 13C NMR (100 MHz, C6D6, rt): d=153.4, 137.6,
124.7 ppm (signals assigned to the C6F5 moiety could not be de-
tected due to multiple 13C–19F couplings); elemental analysis calcd
(%) for C16H10ClF5N2Pd: C 41.14, H 2.16, N 6.00; found: C 41.29, H
Isolation of compound 19
In a drybox, [Pd(PCy3)2] (202 mg, 0.3 mmol), LiI (41 mg, 0.3 mmol),
C6F6 (34.5 mL, 0.3 mmol), and THF (5 mL) were added. The reaction
mixture was stirred at 608C for 5 h in a metal bath. Volatile com-
pounds were removed under vacuum and the resultant solid was
extracted with Et2O, filtered, and dried under vacuum to afford 19
as a yellow solid (197 mg, 68%). Recrystallization from Et2O at
À358C afforded good crystals for analysis by X-ray diffraction.
1H NMR (400 MHz, C6D6, rt): d=2.4–1.0 ppm (m, 66H; Cy group);
19F NMR (376 MHz, C6D6, rt): d=À111.2 (d, J=27.4 Hz, 2F), À164.7
(t, J=20.1 Hz, 1F), À166.0 ppm (m, 2F); 31P NMR (162 MHz, C6D6,
rt): d=29.3 ppm (s); 13C NMR (100 MHz, C6D6, rt): d=37.1 (t, J=
9.7 Hz), 30.8, 28.0 (t, J=5.2 Hz), 26.7 ppm (signals assigned to the
C6F5 moiety could not be detected due to multiple 13C–19F cou-
plings); elemental analysis calcd (%) for C42H66F5IP2Pd: C 52.48, H
6.92; found: C 52.45, H 7.10; X-ray data for 19: Mr =961.19; color-
less; monoclinic; P21/c (no. 14); a=16.474(11), b=16.227(10), c=
1
2.38, N 6.09. In the H and 13C NMR spectra, the two pyridine rings
were observed equivalently, which indicated they occupy the trans
positions of the square-planar PdII geometry. The configuration of
the product, however, was not mentioned in the original litera-
ture.[22]
Preparation of compound 24
In a dry box, PCy3 (287 mg, 1.02 mmol) was added to a solution of
[Pd(C6F5)Cl(py)2] (434 mg, 0.93 mmol) in pyridine (7 mL). Hexane
was added to the resultant yellow solution to give a yellowish-
white precipitate. The suspension was filtered and washed with
hexane to give a yellowish-white powder. The crude material was
recrystallized from acetone by cooling to À358C to yield yellow
block crystals of 24·acetone (320 mg, 52%). 1H NMR (400 MHz,
[D8]THF, rt): d=8.88 (m, 2H), 7.85 (tt, J=7.6, 1.5 Hz, 1H), 7.45 (m,
2H), 2.1–1.0 ppm (m, 33H; Cy group); 19F NMR (376 MHz, [D8]THF,
rt): d=À127.5 (m, 2F), À169.3 (t, J=19.6 Hz, 1F), À170.5 ppm (m,
2F); 31P NMR (162 MHz, [D8]THF, rt): d=17.7 ppm (m); 13C NMR
(100 MHz, [D8]THF, rt): d=154.5, 139.2, 126.8, 33.6 (d, J(C,P)=
17 Hz), 30.4, 28.3 (d, J(C,P)=11 Hz), 27.0 ppm (signals assigned to
the C6F5 moiety could not be detected due to multiple 13C–19F cou-
plings); elemental analysis calcd (%) for C29H38ClF5NPPd·C3H6O: C
52.90, H 6.10, N 1.93; found: C 53.03, H 6.29, N 2.09; X-ray data for
24·acetone: Mr =726.50; yellow; monoclinic; P21/c (no. 14); a=
9.8563(4), b=16.1075(7), c=20.7981(10) ꢁ; b=100.527(2)8; V=
17.847(12) ꢁ; b=116.358(6)8; V=4275(5) ꢁ3; Z=4; 1calcd
1.493 gcmÀ3; T=À120(0)8C; R1 (wR2)=0.0551 (0.1086).
=
Preparation of compound 23[20]
Absolute ether (20 mL, dried over sodium benzophenone ketyl)
and chloropentafluorobenzene (740 mL, 6.0 mmol) were added to
a two-necked round-bottomed flask equipped with a stirrer bar.
The solution was cooled to À788C. A solution of nBuLi (1.6m,
3.8 mL, 6.0 mmol) in hexane was added dropwise with stirring
(Caution! (Pentafluorophenyl)lithium is very thermally unstable
and to avoid explosion it must be prepared and reacted at low
temperatures). The colorless solution was stirred for 30 min at
Chem. Eur. J. 2014, 20, 2040 – 2048
www.chemeurj.org
2046
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3246.3(3) ꢁ3; Z=4; 1calcd =1.486 gcmÀ3; T=À120(0)8C; R1 (wR2)=
Angew. Chem. Int. Ed. 2011, 50, 9081–9084; n) P. Tang, T. Ritter, Tetrahe-
dron 2011, 67, 4449–4454; o) P. S. Fier, J. F. Hartwig, J. Am. Chem. Soc.
2012, 134, 10795–10798; p) Y. Ye, M. S. Sanford, J. Am. Chem. Soc. 2013,
135, 4648–4651.
0.0241 (0.0281).
Preparation of compound 22
[2] For recent reviews, see: a) M. F. Kuehnel, D. Lentz, T. Braun, Angew.
Chem. 2013, 125, 3412–3433; Angew. Chem. Int. Ed. 2013, 52, 3328–
3348; b) M. Klahn, U. Rosenthal, Organometallics 2012, 31, 1235–1244;
c) A. Nova, R. Mas-Ballestꢃ, A. Lledꢄs, Organometallics 2012, 31, 1245–
1256; d) J.-F. Paquin, Synlett 2011, 289–293; e) E. Clot, O. Eisenstein, N.
Jasim, S. A. Macgregor, J. E. McGrady, R. N. Perutz, Acc. Chem. Res. 2011,
44, 333–348; f) T. Braun, F. Wehmeier, Eur. J. Inorg. Chem. 2011, 613–
625; g) A. D. Sun, J. A. Love, Dalton Trans. 2010, 39, 10362–10374; h) G.
Meier, T. Braun, Angew. Chem. 2009, 121, 1575–1577; Angew. Chem. Int.
Ed. 2009, 48, 1546–1548; i) H. Amii, K. Uneyama, Chem. Rev. 2009, 109,
2119–2183; j) R. N. Perutz, Science 2008, 321, 1168–1169.
[3] T. Schaub, M. Backes, U. Radius, J. Am. Chem. Soc. 2006, 128, 15964–
15965.
[4] H. Guo, F. Kong, K.-i. Kanno, J. He, K. Nakajima, T. Takahashi, Organome-
tallics 2006, 25, 2045–2048.
[5] Y. Nakamura, N. Yoshikai, L. Ilies, E. Nakamura, Org. Lett. 2012, 14, 3316–
3319.
[6] M. Ohashi, H. Saijo, M. Shibata, S. Ogoshi, Eur. J. Org. Chem. 2013, 443–
447.
[7] a) M. R. Cargill, G. Sandford, A. J. Tadeusiak, D. S. Yufit, J. A. K. Howard, P.
Kilickiran, G. Nelles, J. Org. Chem. 2010, 75, 5860–5866; b) M. R. Cargill,
G. Sandford, P. Kilickiran, G. Nelles, Tetrahedron 2013, 69, 512–516.
[8] For Ni-catalyzed coupling reactions of poly- or monofluoroarenes by CÀ
F bond cleavage, see: a) K. Inamoto, J.-i. Kuroda, T. Sakamoto, K. Hiroya,
Synthesis 2007, 2853–2861; b) V. P. W. Bçhm, C. W. K. Gstçttmayr, T. We-
skamp, W. A. Herrmann, Angew. Chem. 2001, 113, 3500–3503; Angew.
Chem. Int. Ed. 2001, 40, 3387–3389; c) L. Ackermann, R. Born, J. H.
Spatz, D. Meyer, Angew. Chem. 2005, 117, 7382–7386; Angew. Chem. Int.
Ed. 2005, 44, 7216–7219; d) M. Tobisu, T. Xu, T. Shimasaki, N. Chatani, J.
Am. Chem. Soc. 2011, 133, 19505–19511; e) A. Steffen, M. I. Sladek, T.
Braun, B. Neumann, H. G. Stammler, Organometallics 2005, 24, 4057–
4064; f) T. Saeki, Y. Takashima, K. Tamao, Synlett 2005, 1771–1774;
g) A. D. Sun, J. A. Love, Org. Lett. 2011, 13, 2750–2753; h) T. Braun, R. N.
Perutz, M. I. Sladek, Chem. Commun. 2001, 2254–2255; i) Y. Kiso, K.
Tamao, M. Kumada, J. Organomet. Chem. 1973, 50, C12–C14; j) K.
Tamao, K. Sumitani, Y. Kiso, M. Zembayashi, A. Fujioka, S.-i. Kodama, I.
Nakajima, A. Minato, M. Kumada, Bull. Chem. Soc. Jpn. 1976, 49, 1958–
1969.
In a drybox, NaI (300 mg, 2.0 mmol) was added to a solution of 24
(145 mg, 0.2 mmol) in acetone (10 mL). The resultant orange solu-
tion was stirred for 3 h. The solution turned to an orange suspen-
sion. Toluene (30 mL) was added and the resultant precipitates
were removed by filtration. All volatile compounds were removed
under vacuum and the resultant solid was taken out of the drybox.
The solid was washed with ethanol until no yellow color was ob-
served in the wash liquor, then washed with small amount of
water and ethanol. The resultant solid was dissolved in acetone
and dried under vacuum to give 22 as a yellow powder (83 mg,
55%). The complex was recrystallized from THF/hexane. 1H NMR
(400 MHz, [D6]acetone, rt): d=8.86 (d, J=5.2 Hz, 2H), 7.98 (tt, J=
7.6, 1.5 Hz, 1H), 7.62 (m, 2H), 2.1–1.0 ppm (m, 33H; Cy group);
19F NMR (376 MHz, [D6]acetone, rt): d=À123.2 (m, 2F), À167.3 (t,
J=18.6 Hz, 1F), À168.7 ppm (m, 2F); 31P NMR (162 MHz,
[D6]acetone, rt): d=21.0 ppm (m); 13C NMR (100 MHz, [D6]acetone,
rt): d=154.0, 140.0, 127.5, 35.1 (d, J(C,P)=18 Hz), 31.1, 28.5 (d,
J(C,P)=10 Hz), 27.1 ppm (signals assigned to the C6F5 moiety could
not be detected due to multiple 13C–19F couplings); elemental anal-
ysis calcd (%) for C29H38F5INPPd: C 45.84, H 5.04, N 1.84; found: C
45.92, H 5.65, N 2.39; X-ray data for 22·C4H8O: Mr =831.98; color-
less; monoclinic; P21/n (no. 14); a=9.9348(4), b=16.3295(7), c=
21.2257(9) ꢁ; b=105.0560(10)8; V=3325.2(2) ꢁ3; Z=4; 1calcd
1.646 gcmÀ3; T=À150(0)8C; R1 (wR2)=0.1348 (0.3479).
=
CCDC-958538 (19), -958539 (20·C6H14), -958540 (21·0.5(C6H14)),
-958541 (22·C4H8O), -958542 (23), and -958543 (24·C3H6O) contain
the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
[9] For transition-metal-catalyzed (except for Ni) coupling reactions of poly-
or monofluoroarenes by CÀF bond cleavage, see: a) R. Wilhelm, D. A.
Widdowson, J. Chem. Soc. Perkin Trans. 1 2000, 3808–3813; b) D. A.
Widdowson, R. Wilhelm, Chem. Commun. 2003, 578–579; c) K. Manabe,
S. Ishikawa, Synthesis 2008, 2645–2649; d) T. J. Korn, M. A. Schade, S.
Wirth, P. Knochel, Org. Lett. 2006, 8, 725–728; e) T. Wang, B. J. Alfonso,
J. A. Love, Org. Lett. 2007, 9, 5629–5631.
[10] a) T. Schaub, U. Radius, Chem. Eur. J. 2005, 11, 5024–5030; b) T. Schaub,
P. Fischer, A. Steffen, T. Braun, U. Radius, A. Mix, J. Am. Chem. Soc. 2008,
130, 9304–9317; c) P. Fischer, K. Gçtz, A. Eichhorn, U. Radius, Organome-
tallics 2012, 31, 1374–1383.
[11] For other examples for CÀF bond activation on nickel(0), see: a) D. R.
Fahey, J. E. Mahan, J. Am. Chem. Soc. 1977, 99, 2501–2508; b) I. Bach,
K.-R. Pçrschke, R. Goddard, C. Kopiske, C. Krꢅger, A. Rufinska, K. Seevo-
gel, Organometallics 1996, 15, 4959–4966; c) T. Yamamoto, M. Abla, J.
Organomet. Chem. 1997, 535, 209–211; d) T. Braun, S. P. Foxon, R. N.
Perutz, P. H. Walton, Angew. Chem. 1999, 111, 3543–3545; Angew. Chem.
Int. Ed. 1999, 38, 3326–3329; e) T. Braun, L. Cronin, C. L. Higgitt, J. E.
McGrady, R. N. Perutz, M. Reinhold, New J. Chem. 2001, 25, 19–21; f) S.
Kuhl, R. Schneider, Y. Fort, Adv. Synth. Catal. 2003, 345, 341–344; g) S. A.
Johnson, C. W. Huff, F. Mustafa, M. Saliba, J. Am. Chem. Soc. 2008, 130,
17278–17280.
[12] For examples of structurally-characterized NiII–F complexes as a result
of CÀF bond activation on nickel(0), see: a) L. Cronin, C. L. Higgitt, R.
Karch, R. N. Perutz, Organometallics 1997, 16, 4920–4928; b) M. I.
Sladek, T. Braun, B. Neumann, H.-G. Stammler, J. Chem. Soc. Dalton
Trans. 2002, 297–299; c) S. Burling, P. I. Elliott, N. A. Jasim, R. J. Lindup,
J. McKenna, R. N. Perutz, S. J. Archibald, A. C. Whitwood, Dalton Trans.
2005, 3686–3695; d) S. A. Johnson, E. T. Taylor, S. J. Cruise, Organome-
This work was supported by a Grant-in-Aid for Scientific Re-
search (A) (No. 21245028), a Grant-in-Aid for Young Scientists
(A) (No. 25708018), and a Grant-in-Aid for Scientific Research
on Innovative Areas “Molecular Activation Directed toward
Straightforward Synthesis” (No. 23105546) from MEXT. M.O.
also acknowledges The Noguchi Institute.
Keywords: CÀF activation
· cross-coupling · palladium ·
perfluoroarenes · reaction mechanisms
[1] For selected examples of fluorination reactions, see: a) G. Balz, G. Schie-
mann, Ber. Dtsch. Chem. Ges. 1927, 60, 1186–1190; b) W. E. Barnette, J.
Am. Chem. Soc. 1984, 106, 452–454; c) D. J. Adams, H. Clark, Chem. Soc.
Rev. 1999, 28, 225–231; d) H. Sun, S. G. DiMagno, Angew. Chem. 2006,
118, 2786–2791; Angew. Chem. Int. Ed. 2006, 45, 2720–2725; e) G. Sand-
ford, J. Fluorine Chem. 2007, 128, 90–104; f) P. Tang, T. Furuya, T. Ritter,
J. Am. Chem. Soc. 2010, 132, 12150–12154; g) K. L. Hull, W. Q. Anani,
M. S. Sanford, J. Am. Chem. Soc. 2006, 128, 7134–7135; h) D. A. Watson,
M. Su, G. Teverovskiy, Y. Zhang, J. Garcia-Fortanet, T. Kinzel, S. L. Buch-
wald, Science 2009, 325, 1661–1664; i) T. Furuya, T. Ritter, Org. Lett.
2009, 11, 2860–2863; j) T. Furuya, A. E. Strom, T. Ritter, J. Am. Chem. Soc.
2009, 131, 1662–1663; k) S. Yamada, A. Gavryushin, P. Knochel, Angew.
Chem. 2010, 122, 2261–2264; Angew. Chem. Int. Ed. 2010, 49, 2215–
2218; l) P. Anbarasan, H. Neumannm, M. Beller, Angew. Chem. 2010, 122,
2265–2268; Angew. Chem. Int. Ed. 2010, 49, 2219–2222; m) C. S. L.
Chan, M. Wasa, X. Wang, J.-Q. Yu, Angew. Chem. 2011, 123, 9247–9250;
Chem. Eur. J. 2014, 20, 2040 – 2048
www.chemeurj.org
2047
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tallics 2009, 28, 3842–3855; e) M. E. Doster, S. A. Johnson, Angew.
Chem. 2009, 121, 2219–2221; Angew. Chem. Int. Ed. 2009, 48, 2185–
2187.
[18] a) T. Kashiwabara, M. Tanaka, Organometallics 2006, 25, 4648–4652;
b) S. Teo, Z. Weng, T. S. A. Hor, Organometallics 2006, 25, 1199–1205.
[19] F. Barrios-Landeros, B. P. Carrow, J. F. Hartwig, J. Am. Chem. Soc. 2009,
131, 8141–8154.
[20] R. Uson, J. Fornies, J. Gimeno, P. Espinet, R. Navarro, J. Organomet.
Chem. 1974, 81, 115–122.
[13] M. Ohashi, T. Kambara, T. Hatanaka, H. Saijo, R. Doi, S. Ogoshi, J. Am.
Chem. Soc. 2011, 133, 3256–3259.
[14] When isolated ZnPh2 (purchased from Strem) was employed in the cou-
pling reaction in place of ZnPh2 generated in situ from ZnCl2 and
[21] X. Chen, X. Luo, Y. Song, S. Zhou, L. Zhu, Polyhedron 1998, 17, 2271–
2278.
PhMgBr,
1 was obtained in 63% yield (conditions: [Pd(PCy3)2]
(10 mol%), isolated ZnPh2 (0.6 equiv), LiI (2.0 equiv), [D8]THF, 608C, 4 h).
[15] Based on the fact that the catalytic reaction with pentafluoroiodoben-
zene (C6F5I) as a substrate hardly gave 1 (<5% yield, 70% starting ma-
terial recovered), we ruled out the possibility that C6F5I could be gener-
ated as a result of nucleophilic attack of an iodide anion on C6F6, and it
is subsequently the species that undergoes oxidative addition at palla-
dium(0).
[22] R. Usꢄn, J. Fornies, R. Navarro, Inorg. Chim. Acta 1979, 33, 69–75.
[23] G. T. Achonduh, N. Hadei, C. Valente, S. Avola, C. J. O’Brien, M. G. Organ,
Chem. Commun. 2010, 46, 4109–4111.
[24] T. Yoshida, S. Otsuka, D. G. Jones, J. L. Spencer, P. Binger, A. Brinkmann,
P. Wedemann, Inorg. Synth. 1974, 19, 101–107.
[16] A. Krasovskiy, V. Malakhov, A. Gavryushin, P. Knochel, Angew. Chem.
2006, 118, 6186–6190; Angew. Chem. Int. Ed. 2006, 45, 6040–6044.
[17] S. A. Macgregor, D. C. Roe, W. J. Marshall, K. M. Bloch, V. I. Bakhmutov,
V. V. Grushin, J. Am. Chem. Soc. 2005, 127, 15304–15321.
Received: September 3, 2013
Published online on January 15, 2014
Chem. Eur. J. 2014, 20, 2040 – 2048
www.chemeurj.org
2048
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim