Dinuclear Systems in the Efficient Nickel-Catalyzed Kumada-Tamao-Corriu Cross-Coupling of Aryl Halides

Article abstract of DOI:10.1021/acs.organomet.6b00451

Highly active dinuclear nickel(I) complexes bearing bulky N-heterocyclic carbene ligands have been shown to be involved in the catalytic cycle of the Kumada-Tamao-Corriu cross-coupling reaction of aryl halides. The results of several stoichiometric reactions and kinetics experiments have revealed that monovalent and divalent dinickel species are the active species in the highly efficient, nickel-catalyzed, Kumada coupling reactions of aryl halides.

Full text of DOI:10.1021/acs.organomet.6b00451

Article
pubs.acs.org/Organometallics
Dinuclear Systems in the Efficient Nickel-Catalyzed Kumada−
Tamao−Corriu Cross-Coupling of Aryl Halides
Kouki Matsubara,*,† Hitomi Yamamoto,† Satoshi Miyazaki,† Takahiro Inatomi,† Keita Nonaka,†
Yuji Koga,† Yuji Yamada,† Luis F. Veiros,‡ and Karl Kirchner§
†Department of Chemistry, Fukuoka University, 8-19-1 Nanakuma, Jonan-ku, Fukuoka 814-0180, Japan
‡Centro de Química Estrutural, Instituto Superior Tec
Portugal
́ ́
nico, Universidade Tecnica de Lisboa, Av. Rovisco Pais No. 1, 1049-001 Lisboa,
§Institut fur Angewandte Synthesechemie, Technische Universitat Wien, Getreidemarkt 9, Wien 1060, Austria
̈
̈
S
* Supporting Information
ABSTRACT: Highly active dinuclear nickel(I) complexes
bearing bulky N-heterocyclic carbene ligands have been shown
to be involved in the catalytic cycle of the Kumada−Tamao−
Corriu cross-coupling reaction of aryl halides. The results of
several stoichiometric reactions and kinetics experiments have
revealed that monovalent and divalent dinickel species are the
active species in the highly efficient, nickel-catalyzed, Kumada
coupling reactions of aryl halides.
Scheme 1. (1) Classical Catalytic Cycle between Ni(0) and
Ni(II), (2) Proposed Cycle between Ni(I) and Ni(III), (3)
Oxidation States in the Oxidative Addition to a Monomeric
Ni(I) Center, and (4) Oxidation States of the Dimeric Ni
Centers
INTRODUCTION
■
Considerable progress has been made during the course of the
last three decades toward the development of organotransition-
metal-catalyzed cross-coupling reactions of aryl halides, using
various metal catalysts or catalyst precursors with ligands and/
or cocatalysts. These methods have been successfully applied to
the synthesis of a large number of organic products for
electronic devices, liquid crystals, and medicinal chemistry.1
The Kumada−Tamao−Corriu cross-coupling reaction of an
aryl or alkyl halide with a Grignard reagent in the presence of a
Ni, Pd, Co, or Fe catalyst is one of the most useful reactions for
the formation of aryl−aryl or alkyl−aryl bonds under mild
conditions.1
A general catalytic system using a mixture of Ni(cod)2 and a
bulky N-heterocyclic carbene (NHC) ligand, such as IPr or
IMes, is widely employed in various cross-coupling reactions
including the Kumada−Tamao−Corriu cross-coupling reac-
tions.2 Despite the wide variety of substrates and applications
that have been developed for coupling reactions involving
Grignard reagents, very little is currently known about the
mechanisms of these reactions. The classical catalytic cycle
proposed for these transformations is shown in Scheme 1(1).3
In contrast, the results of several studies involving Ni and Pd
catalysts have provided evidence that radical pathways,
including single-electron-transfer processes, are involved in
the cross-coupling reactions.4 We have developed a well-
defined nickel(I) compound bearing a bulky NHC, NiCl(IPr)2
(where IPr is 1,3-bis(2,6-diisopropylphenyl)imidazol-2-yli-
dene), which was active in the Kumada coupling reaction of
aryl halides.5 The active nickel(I) species could be generated in
situ in the cross-coupling reactions reported above, because this
nickel(I) complex, NiCl(IPr)2, can be formed from a mixture of
Ni(cod)2, IPr, and an aryl halide.5 Several other well-defined
monovalent nickel complexes have also been reported to act in
a range of other catalytic cycles.6 Scheme 1(2) shows a
proposed catalytic cycle between Ni(I) and Ni(III) species. It is
noteworthy, however, that there have been no studies in the
literature providing clear evidence of how these reactions
proceed, including, for example, reports pertaining to the
structures of possible intermediates. This lack of information
prompted us to investigate the mechanism of the Kumada
coupling reaction using monovalent nickel species to develop a
deeper mechanistic understanding of the reaction. The work in
Received: June 4, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
this study is based on the hypothesis that a dinuclear nickel(I)
species would be more plausible as an intermediate in these
reactions than a mononuclear nickel(I) species (Scheme 1(3),
(4)). This hypothesis was formed on the basis that the unpaired
electrons on the nickel centers could readily form metal−metal
bonds in the presence of an appropriately bulky ligand, which
could stabilize the formation of coordinatively unsaturated
dinuclear compounds that would be capable of performing as
highly active and efficient cooperative catalysts.
Dinuclear complexes containing two linked metal atoms have
attracted considerable attention as potential catalysts for a
variety of organic transformations, because synergistic and
cooperative effects by the multinuclei centers in catalysis can
activate inert and/or specific bonds much more efficiently than
the corresponding mononuclear catalysts.7−9 However, there
are very few examples in the literature of dinuclear catalysts that
have been designed and applied to cross-coupling reactions. A
dinuclear nickel catalyst using a multidentate ligand was used to
catalyze the Negishi cross-coupling reaction of an aryl chloride,
where it performed much more effectively than a mononickel
catalyst system.10 In palladium chemistry, dinuclear Pd(I)
complexes have been reported to catalyze the Buchwald−
Hartwig amination of aryl halides, even at room temperature.11a
Schoenebeck et al. also recently studied catalytic cross-coupling
reactions of aryl halides and the mechanism using dinuclear
Pd(I) complexes.11b−d
With regard to the dinuclear nickel(I) complexes discussed in
this report, the oxidation states of the nickel atoms could be
regarded as the most important feature of their catalytic cycle.
Oxidative addition to a mononuclear nickel(I) complex would
lead to a formal change in oxidation state of the nickel atom
from +1 to +3 (Scheme 1(3)).3 In contrast, oxidative addition
to a dinuclear nickel(I) complex would lead to a change in the
formal oxidation states of both nickel atoms from +1 to +2.
Therefore, we hypothesized that the two stable nickel(II)
centers are generated favorably in this dinickel(I) catalyst
system (Scheme 1(4)).
Herrmann et al. reported that the catalytic activity of an in
situ generated active nickel species in the Kumada−Tamao−
Corriu cross-coupling of aryl fluoride was strongly dependent
upon the amount of IPr added to the system.15 The addition of
1 equiv of NHC to a nickel precursor provided a nickel species
bearing one NHC ligand that was much more active than the
corresponding nickel species bearing two NHC ligands
generated by the addition of 2 equiv of NHC. As noted
above, NiCl(IPr)2 can act as an active catalyst but has two
molecules of IPr.4 Therefore, it remains unclear how a nickel
catalyst bearing only one IPr ligand was more active in the
cross-coupling reactions. The results of our most recent study
showed that the complex formed from mixed IPr/PPh3 and
nickel(I) exhibited higher activity in the Kumada coupling
reaction of aryl halides than NiCl(IPr)2.6b Here, we found that
the oxidative addition of aryl chloride to an in situ generated
nickel(0) complex bearing an IPr ligand gave a coordinatively
unsaturated dinuclear nickel(I) complex, which exhibited
catalytic performance higher than that of the monomeric
nickel(I) catalyst NiCl(IPr)2. We wish to report the isolation
and structural elucidation of these dinickel complexes, as well as
an evaluation of their catalytic performance. The results of
mechanistic studies, including stoichiometric reactions, kinetic
experiments, and DFT calculations, will also be discussed.
RESULTS AND DISCUSSION
■
Synthesis and Structures of the Dinickel(I) Aryl
Complexes. The dinuclear nickel(I) oxidative addition
products [Ni(IPr)]2(μ-Cl)(μ-η2-C6H4R) (1a, R = 4-CH3; 1b,
R = 4-OCH3) were successfully prepared by addition of p-
chlorotoluene and p-chloroanisole to a solution of Ni(cod)2
and IPr (1 equiv with respect to nickel) in THF at room
temperature (Scheme 2). Instead of forming the expected
Scheme 2. Synthesis of Dinickel(I) Oxidative Addition
Products 1a,b
NHC ligands are some of the most versatile and useful
candidates for the construction of active catalysts, because they
are strongly electron donating (i.e., two electrons) and can also
generate strong trans effects.12 Bulky NHC ligands, such as IPr
and IMes (where IMes is 1,3-bis(2,4,6-trimethylphenyl)-
imidazol-2-ylidene), have been used to good effect in various
catalytic cross-coupling reactions, providing high levels of
efficiency and chemoselectivity. Among the many Ni(I)
complexes reported to date bearing a range of phosphorus
and other ligands,13 several recent reports have revealed that
bulky NHC ligands can also kinetically stabilize monovalent
nickel species and that the resulting systems can be used to
catalyze various reactions, including Kumada−Tamao−Corriu
coupling,4,6b,c Suzuki coupling,6a Buchwald−Hartwig amina-
tion,6b and hydrodehalogenation reactions.6d It is noteworthy
that mononuclear Ni(I) NHC complexes were used in all of
these examples. On the other hand, the synthesis and
stoichiometric reactions of a variety of dinuclear nickel(I) IPr
complexes having a Ni−Ni single bond have been reported.14
Taken together, these results provide a strong indication that
the efficiency of catalytic reactions using these ligands would
differ considerably from those using smaller ligands, most
probably because of differences in the mechanisms of these
reactions resulting from differences in the oxidation states of
the metal centers.
mononickel(II) adducts via the oxidative addition of the aryl
halides to a Ni(0) center, these two oxidative addition reactions
rather unexpectedly afforded 1a,b in a selective manner. The
isolated yields of 1a,b following their purification upon
recrystallization were 44 and 29%, respectively. Trace amounts
of the dinuclear μ-chloride [Ni(IPr)]2(μ-Cl)2 (2)16 and 4,4′-
dimethylbiphenyl or 4,4′-dimethoxybiphenyl were detected in
1
the crude H NMR spectra. Addition of only IPr to a solution
of Ni(cod)2 in THF gave a dark green solution, which most
likely contained an (IPr)Ni0 THF complex rather than a
(NHC)Ni0 dimer (reddish brown).17 The solution became
dark reddish brown following the addition of p-chlorotoluene,
indicating formation of the compound 1a. In the absence of IPr,
Ni(cod)2 did not react with p-chlorotoluene under these
conditions. Furthermore, the use of other nonpolar solvents,
including n-hexane and benzene-d6, resulted in the formation of
numerous byproducts, and the addition of 2 equiv of IPr to
B
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Ni(cod)2 followed by p-chlorotoluene resulted in the formation
of the mononuclear nickel(I) product NiCl(IPr)2 (4).4
Addition of 1 equiv of a Grignard reagent and aryl halide
independently to compound 1a in THF at room temperature
afforded dinickel(I) complexes, representing interesting trans-
formations. The reaction of 1a with p-tolylmagnesium chloride
for 18 h led to a transmetalation reaction at the bridging
chloride ligand, which yielded the corresponding biaryl complex
[Ni(IPr)]2(σ:η2-C6H4CH3)2 (3a) (Scheme 3(1)). The yield of
3a as dark green crystals following its purification upon
recrystallization from THF at −30 °C was 43%.
unsaturated 30e dinickel adducts. The oxidation state of each
nickel atom was found to be +1, and therefore in contrast to
that of the divalent mononickel analogues generated as a result
of the oxidative addition of aryl halides to Ni(0) complexes.20
The distance between the two nickel centers in 1a was found to
be 2.3954(5) Å (Figure 1), which suggested the existence of a
Scheme 3. Reactions of 1a (1) with Tolylmagnesium
Chloride and (2) with 4-Chlorotoluene and 4-Chloroanisole
and the Reverse Reaction
On the other hand, the reaction of 1a with 4-chlorotoluene
afforded the dinickel μ-chloride 2 (Scheme 3(2)). A similar
outcome was also observed for the reaction of 1a with several
other aryl halides, including p-chloroanisole and chlorobenzene.
The formation of 2 from the reactions of 1a with a range of aryl
halides suggested that the oxidative addition of the aryl
chlorides to 1a and the subsequent reductive elimination of the
biaryl products were occurring at the unsaturated dinickel
centers. Detailed experimental results and discussion are given
below. Interestingly, the reverse reaction from 2 to 1a was
demonstrated by the reaction of 2 with 1.0 equiv of p-
Figure 1. ORTEP drawings of (a) 1a and (b) 3a (thermal ellipsoids at
the 50% probability level). All of the hydrogen atoms have been
omitted for clarity. Half of the 3a molecule has been shown by a
symmetric operation.
1
tolylmagnesium chloride (Scheme 3(2)). A crude H NMR
spectrum of the reaction showed that 1a was being formed as
the major species, together with only small amounts of 3a and
2. It is noteworthy that the dinuclear framework in 1a remained
intact even after the stoichiometric reactions.
single bond.14 One of the CC bonds in the σ-aryl moiety was
coordinated to one of the two metal centers in the η2 mode,
and the ring structure was close to that of a cyclohexatriene, in
that the C3−C4 and C5−C6 bond lengths were 1.382(5) and
1.375(4) Å, shorter than the C2−C3, C4−C5, and C6−C1
bond lengths, which were 1.417(4), 1.406(4), and 1.431(4) Å,
respectively. The C1−C2 bond length was elongated to
1.423(4) Å because it was π-coordinated to the nickel atom.
However, solution-phase NMR analyses of compound 1a
suggested that the protons and carbons belonging to this aryl
moiety, as well as those of the IPr ligand, were symmetrical.21
The links between the two nickel atoms and the carbene carbon
atoms were slightly bent, as shown by the Ni2−Ni1−
C8(carbene) and Ni1−Ni2−C35(carbene) bond angles,
which were 158.50(8) and 167.98(8)°, respectively. The
generation of a space-filling model of 1a from the crystal
structure data revealed that the phenyl rings and the methyl
groups of the two IPr ligands formed a pocket over the two
nickel atoms and tightly surrounded the σ-tolyl moiety (Figure
2). The bending of the Ni−Ni−C(carbene) bond angles could
therefore be explained in terms of the steric hindrance between
the IPr and the tolyl groups. Depending on the leaning
The 1H NMR spectra of 1a,b showed characteristic
diamagnetic high-field signals at δ 6.07 (4H in 1a) or δ 6.17
and 6.72 (2H + 2H in 1b), which were attributed to the aryl
protons of the bridging tolyl or anisyl group. These signals
indicated the occurrence of a shielding effect derived from the
aryl rings of the IPr ligands, where the bridging aryl protons
were located in close proximity to the aryl rings of IPr. The
integrated ratio of the signals assigned as the methyl protons of
the σ-tolyl or anisyl moiety and the isopropyl methine protons
of the carbene ligand was 3:8, which indicated that the ratio of
the σ-aryl moiety and the carbene ligand was 2:1. This
observation suggested formation of an oxidative addition
product bearing two NHC ligands in a manner similar to
that reported by Radius et al.18,19 for a series of nickel
biscarbene complexes. However, the actual structures of 1a,b,
which were determined by X-ray crystallography using single
crystals derived from hexane solutions, revealed, rather
unexpectedly, that compounds 1a,b were coordinatively
C
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 2. Kumada Cross-Coupling of Aryl Halides Mediated
by Ni(I) Complexes
a
entry
cat.
R
X
additive
yield (%)
1
2
3
4
5
1a
1a
4
OMe
OMe
OMe
OMe
Ph
Cl
F
85
42
33
93
89
Figure 2. Space-filling models of 1a generated from the crystallo-
graphic data (yellow mesh, σ-tolyl group; gray, carbon; white,
hydrogen; green, nickel): (a) top view; (b) side view.
Cl
Br
Br
IPr (5 mol %)
IPr (5 mol %)
IPr (5 mol %)
4
4
direction of the σ-aryl moiety in the fluxional motion in
solution, the geometry of each nickel center could change
slightly, which would result in a change in the Ni−Ni−
C(carbene) bond angle. The crystal structure of 1b was also
determined and was found to be quite similar to that of 1a (see
the Supporting Information).
A single-crystal X-ray diffraction study of 3a was also
conducted and revealed successful incorporation of the bridging
4-tolyl group instead of chlorine. The structural features of 3a
were also found to be similar to those of 1a, as shown in Figure
1b. The Ni−Ni distance was determined to be 2.4067(8) Å
(Table 1), which indicated the presence of a bonding
a
Yields were determined after silica gel column chromatography as an
average of at least two runs.
exhibited moderate activity toward the inactive substrate 4-
fluoroanisole and gave a 42% yield of the corresponding biaryl
product following purification by silica gel column chromatog-
raphy (Table 2, entry 2). It is noteworthy that the Kumada
coupling of anisole derivatives can also yield unwanted
byproducts as a result of the elimination of the methoxy
group.24 Free IPr (5 equiv relative to 4) was added to the
reaction mixture to avoid the formation of the dinickel(I)
complex 2 in solution as a result of the slow equilibrium
reaction between 4 and 2. Furthermore, the rapid Kumada
coupling of aryl chlorides in the presence of 2 was briefly
discussed in our previous report.6b
The above profiles of the Kumada coupling reactions are in
good agreement with those in the literature using nickel
precatalysts and the imidazolium salt IPr/HCl.15 In our
preliminary study, stoichiometric reaction of p-tolylmagnesium
chloride in THF with the monomeric Ni(I) complex 4 gave not
the transmetalated Ni(I) complex [(IPr)2NiPh] or the
dinickel(I) complex 1a but the reduced Ni(0) compound
[(IPr)2Ni] (Scheme 4). It can be noted that the added amount
of IPr strongly affects the catalytic processes in the Kumada
cross-coupling reactions.
Table 1. Representative Bond Distances (Å) and Angles
(deg) of 1a and 3a
1a
3a
Bond Distances (Å)
2.3954(5)
Ni1−Ni2
Ni1−C1
Ni2−C1
Ni1−C2
Ni1−C8
Ni2−C35
C1−C2
C2−C3
C3−C4
C4−C5
C5−C6
C6−C1
Ni1−Ni1′
Ni1−C1
Ni1−C6
Ni1−C8
C1−C2
C2−C3
C3−C4
C4−C5
C5−C6
C6−C1
2.4067(8)
1.936(3)
2.376(3)
1.894(3)
1.432(4)
1.390(4)
1.395(5)
1.381(5)
1.415(5)
1.404(4)
1.983(3)
1.914(3)
2.210(3)
1.910(3)
1.855(3)
1.423(4)
1.417(4)
1.382(5)
1.406(4)
1.375(4)
1.431(4)
Scheme 4. Preliminary Stoichiometric Reactions of 2 and 4
with Phenylmagnesium Chloride
Bond Angles (deg)
Ni2−Ni1−C8
158.50(8)
167.98(8)
65.29(2)
75.8(1)
Ni1′−Ni1−C8
Ni1−C1−Ni1′
160.12(9)
76.3(1)
Ni1−Ni2−C35
Ni1−Cl1−Ni2
Ni1−C1−Ni2
interaction between the two nickel atoms, and one of the
CC bonds of each σ-aryl group was coordinated to a nickel
center, being similar to that observed in 1a. Other bimetallic
coordination effects for μ-σ-aryl groups have been reported in
gold complexes.22 Johnson et al. reported coordination in
dinickel(I) complexes similar to that in complexes 1 and 3.23
Catalytic Application of 1a for the Kumada Coupling.
Compound 1 was found to catalyze the Kumada cross-coupling
reaction of weakly active aryl halides. Compounds 1a and 4
exhibited clear differences in their catalytic activity, as shown in
Table 2. For example, the coupling of 4-chloroanisole with
phenylmagnesium chloride in the presence of 1a for 18 h at
ambient temperature gave 4-methoxybiphenyl in 85% yield
(Table 2, entry 1). In contrast, the use of compound 4 as a
catalyst resulted in a much lower yield of the cross-coupling
product of 33% (Table 2, entry 3). Compound 1a also
Possible Dinickel Cycles in the Kumada Coupling
Reaction. The results presented above on the reactivity of the
dinickel(I) species suggested the possibility of two catalytic
cycles in the Kumada cross-coupling of aryl halides involving
the dinickel species 1−3, as shown in Scheme 5. One of these
cycles would involve the reaction of compounds 1 and 2.
Briefly, the oxidative addition of an aryl halide to 1 would result
in the formation of a metastable dinickel(II) species, which
would smoothly eliminate the corresponding biaryl product to
give 2. Compound 1 would then be regenerated via a
transmetalation reaction with arylmagnesium chloride. A similar
D
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
thoxybiphenyl, were unexpectedly detected by GC-MS analysis
in this reaction (Scheme 7). The product ratio of these three
Scheme 5. Two Possible Reaction Mechanisms in the
Kumada Coupling of Aryl Halides
Scheme 7. Stoichiometric Reaction of 1a with p-
Chloroanisole
process involving 3 as an intermediate could also be possible. In
this case, the reaction of a Grignard reagent with 1 instead of an
aryl halide would result in the formation of complex 3 via a
transmetalation process. The reductive elimination of biaryls in
the presence of an aryl halide would then form 1 as the
oxidative addition product (Scheme 5).
Several experiments were conducted to verify the possibility
of the proposed cycles. The first of these experiments involved
the reaction of 3a with 2 equiv of chlorobenzene in THF at
room temperature for 21 h. Notably, the major product of this
reaction was not compound 1 but 2 (Scheme 6). It is possible
compounds was 13:37:50. In the catalytic process as given in
Table 2, no homocoupling products from aryl halides were
detected, showing that the stoichiometric reaction products
using 1a are not consistent with those in the catalytic reactions.
Phenylmagnesium chloride was also added to a reaction
mixture consisting of 1a and a stoichiometric amount of 4-
chloro- or 4-bromoanisole. After 18 h at room temperature, all
of the starting aryl halides had been consumed, and the yields
of the extracted biaryl products were determined by GC-MS
analysis. All six of the possible biaryl products were detected in
this reaction (Scheme 8). Surprisingly, the 4-methoxybiphenyl
Scheme 6. Stoichiometric Reaction of 3a with
Chlorobenzene
Scheme 8. Stoichiometric Reaction of 1a with p-Haloanisole
and Phenylmagnesium Chloride
that compound 1 is generated from 3 and the subsequent
reaction of 1 with chlorobenzene readily occurs to yield 2 in
this reaction. In addition, only 4,4′-dimethylbiphenyl and
biphenyl were observed as the organic products, and no 4-
methylbiphenyl was detected by GC-MS (Scheme 6). We could
not detect 1c in this reaction even when a smaller amount of
chlorobenzene was added to 3. Given that compound 3 was
found to be stable in the absence of chlorobenzene, the
reductive elimination of 4,4′-dimethylbiphenyl would most
likely occur in a concerted manner involving the coordination
and oxidative addition of chlorobenzene.25 Given that no 4-
methylbiphenyl was formed as a cross-coupling product from
the p-tolyl group in 3 and chlorobenzene, it is likely that any
other routes responsible for the separation to monomeric p-
tolylnickel(I) species from 3 would be negligible during the
oxidative addition of chlorobenzene. The reductive elimination
of two σ-aryl groups from dinickel centers has also been
reported in the literature.23
Product Distribution of Biaryls in the Stoichiometric
Reactions. Complex 1a was treated with 1 equiv of 4-
chloroanisole in THF for 6 h at room temperature to determine
where the bridging σ-aryl group from 1a ended up in the cross-
coupling and/or homocoupling products. It is noteworthy that
all of the possible coupling products, including 4,4′-
dimethylbiphenyl, 4-methyl-4′-methoxybiphenyl, and 4′4-dime-
cross-coupling product was formed in the highest yield (71%)
from 4-haloanisole, despite only 1 equiv of 4-haloanisole being
added to 1a. Most notably, the yield of 4-methyl-4′-
methoxybiphenyl was much higher than that of 4-methyl-
biphenyl (e.g., 27 vs 2% from 4-bromoanisole and 9 vs 2% from
4-chloroanisole, respectively) (Figure 3). The former of these
two products could also be obtained by the oxidative addition
of haloanisole to 1a (Scheme 7), whereas the latter of the two
products could be derived from the reaction of 1a with
phenylmagnesium chloride via the formation of bis(σ-aryl)-
dinickel complex 3 (Scheme 6), according to the above
stoichiometric reactions. This result therefore confirmed that
the oxidative addition process from 1a to 2 was preferred over
the transmetalation process from 1a to 3. This result was
especially interesting because 4-chloroanisole is usually less
active toward the oxidative addition of carbon−chlorine bonds.
Reports pertaining to the occurrence of competition between
oxidative addition and transmetalation reactions in catalytic
cycles are scarce, and the findings presented in the current
E
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
nickel species was being formed during the course of both of
these reactions and that this unknown active species was
responsible for the formation of the catalytic product 4-
methoxybiphenyl.
Kinetics and Theoretical Studies in the Oxidative
Addition of 4-Chlorotoluene to Dinickel(I) Species 1a.
The results of a detailed investigation of the processes involved
in the oxidative addition of aryl halides to the dinickel(I)
systems, such as 1a and 2, could provide significant
understanding of how organometallic reactions can proceed
on such unsaturated dinuclear systems. It is noteworthy that
oxidative addition reactions to monovalent dinickel species are
uncommon in comparison with normal zerovalent nickel
species. Several different pathways could be proposed to
account for formation of 2 and the biaryl products formed
during this multistep reaction. For example, a mononickel(III)
oxidative adduct could be formed, according to a pathway
similar to that proposed in the monomeric nickel(I) system.6
Alternatively, the addition of an aryl halide species to Ni(I)−
Ni(I) could lead to the formation of the dinickel Ni(II)−Ni(II)
system (Scheme 1(4)). It was envisaged that the kinetic
parameters of the transformation could be estimated by the
monitoring of a low-temperature reaction by NMR spectros-
copy. Because the reaction proceeds slowly, even at low
temperature, it was quite difficult to quench the reaction. After
several attempts, the pseudo-first-order reaction rates of 1a
were determined experimentally using a large excess (30 equiv)
of 4-chlorotoluene, which could be readily removed under
reduced pressure to stop the reaction, at temperatures of −81,
Figure 3. Product distribution of the biaryl derivatives in the
stoichiometric reactions of 1a with MeO(C6H4)X (X = Cl, Br) and
PhMgCl.
report therefore represent an important development in our
understanding of the reactivity of monovalent nickel complexes.
Although no supporting results were generated during the
course of this study to explain why the rate of the
transmetalation reaction was much slower than that of the
oxidative addition, it is envisaged that the electron-rich
nickel(I) center would disfavor interaction with the electron-
rich aromatic carbon of the Grignard reagent. It is also possible
that a facile electron transfer process could have occurred from
the electron-rich nickel to the aryl halide acceptor.26
The product ratios of 4-methyl-4′-methoxybiphenyl and 4-
methylbiphenyl were low at 9 and 2%, respectively, in Figure 3,
when 4-chloroanisole was used as the substrate. In contrast, the
product ratio of 4-methoxybiphenyl generated from 4-
chloroanisole was high at 71%. Furthermore, 1H NMR analysis
of the crude reaction mixtures showed that the majority of the
compound 1a remained intact after the reactions with 4-
chloroanisole, as shown in Figure 4a, whereas compound 1a
1
−70, −55, and −43 °C. The H NMR spectra of the reaction
mixture contained an independent signal at δH 6.06, which was
assigned as the bridging aryl protons in 1a. This signal rapidly
diminished as the reaction proceeded, which indicated that the
reaction could be regarded as an irreversible process involving
the oxidative addition of 4-chlorotoluene to 1a, followed by the
reductive elimination of 4,4′-dimethylbiphenyl to form 2. The
observed rate constant was defined as kobs (Scheme 9). The
concentration of 1a (C1a) and the time could therefore be
plotted according to eq 1.
Scheme 9. Kinetic Parameter kobs in the Stoichiometric
Reaction of 1a
Figure 4. 1H NMR spectra (400 MHz, in C6D6, room temperature) of
the stoichiometric reaction mixtures using (a) 4-chloroanisole and (b)
4-bromoanisole.
Four different kobs values were determined depending on the
reaction temperatures to give the corresponding Arrhenius plot
(Figure 5). The resulting ΔH⧧, ΔS⧧, and ΔG298 values were
⧧
was converted almost exclusively to the bromide analogue of
compound 2, [Ni(IPr)]2(μ-Br)2 (2′),13i when 4-bromoanisole
was added to the reaction mixture (Figure 4b). The remarkable
contrast in these results indicated that 1a was in no way
involved in the formation of the catalytic product 4-
methoxybiphenyl from 4-chloroanisole. It was therefore
envisaged that a significant amount of some highly active
+2.40 0.12 kcal mol−1 (+10.1 0.5 kJ mol−1), −56.52 0.52
cal K−1 mol−1 (−237.4 2.2 J K−1 mol−1), and +19.3 0.3
kcal mol−1 (+80.9
1.1 kJ mol−1), respectively. The most
striking feature of this reaction was its large negative activation
entropy, which was over −50 cal K−1 mol−1. The Gibbs free
activation energy ΔG⧧ was therefore largely dependent on the
activation entropy, which suggested that the rate-determining
F
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
reaction (Figures 6 and 7). When a more bulky 1,3-
bis(phenyl)imidazol-2-ylidene (IPh) was employed in the
DFT calculations, many processes involving subtle rotation
movement of the phenyl rings in IPh should be taken into
account. However, the overall sequence of the reaction steps
using IMe was similar to that with IPh. Therefore, the steric
effect of these NHC ligands may not affect the overall sequence
so much even when the bulkier NHC ligand is employed for
the nickel(I) complexes. Similar dinuclear processes have also
been calculated by Schoenebeck et al.11b−d
In the initial step, chlorobenzene was added to complex A to
form intermediate B, where the aromatic ring was coordinated
to the complex in a η2 fashion via the carbon atoms at C2 and
C3. The location of the transition state TSAB revealed a
moderate activation barrier of 15.5 kcal mol−1 (64.8 kJ mol−1).
The chlorobenzene group subsequently underwent a ring
slippage rearrangement to form C, where the aromatic ring was
still η2 coordinated to the complex, but now via the carbon
atoms at C1 and C2. During the course of this process, the Ni−
Ni bond distance increased from 2.35 Å in A to 2.62 Å in C,
with the concomitant activation of the C−Cl bond.
Furthermore, the C−Cl bond distance increased from 1.75 Å
in both the free chlorobenzene and B to 1.82 Å in C. This
process was found to be slightly endergonic by 4.7 kcal mol−1
(19.7 kJ mol−1) with respect to the initial reactants. During the
following oxidative addition step, the C−Cl bond was cleaved
and a new Ni−Cl bond was formed to give intermediate D,
which no longer possessed a Ni−Ni bond (3.30 Å).
Furthermore, both of the arene rings in D were coordinated
to nickel in a σ fashion through a single C atom. The formation
of D from C was found to strongly exergonic, releasing 44.9
kcal mol−1 (187.9 kJ mol−1). The Ni−Cl bond started to form
Figure 5. Arrhenius plot of the low-temperature reaction of 1a with 4-
chlorotoluene.
step was more likely to be the coordination of the aryl halide to
the nickel species rather than the oxidative addition, reductive
elimination, or the reaction responsible for the separation to a
monomeric species from the dimeric species (if possible).
These results were therefore consistent with the presence of the
two bulky NHC ligands around the dinickel centers, which
could hinder the approach of an aryl halide molecule, as
indicated in Figure 2. However, this suggestion is dependent on
the dinickel framework being maintained during the reaction.
0
ln C1a = −kobst + ln C1a
kobs = kC4‐chlorotoluene
(1)
The large negative activation entropy observed for this
reaction prompted us to conduct a theoretical investigation of
the reaction mechanism using density functional theory (DFT).
DFT/PBE0 calculations (free energies in kcal mol−1) based on
the use of the complex [Ni(IMe)]2(μ-Cl)(μ:η1,η2-C6H5) (A)
(with the simpler model ligand 1,3-bis(methyl)imidazol-2-
ylidene (IMe), instead of IPr) and chlorobenzene (ClBz) as a
model substrate provided a conceivable pathway for the
Figure 6. Reaction profile of the computed relative Gibbs free energies (kcal mol−1) for the reaction of [Ni(IMe)]2(μ-Cl)(μ:η1,η2-C6H5) (A) with
chlorobenzene (ClBz) to give intermediate E (Ni−Ni bond distances in Å).
G
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 7. Reaction profile of the computed relative Gibbs free energies (kcal mol−1) for the reaction of intermediate E to give the final products
[Ni(IMe)]2(μ-Cl)2 (I) and biphenyl (Ni−Ni bond distances in Å).
during this step, going from 3.12 Å in C to 2.76 Å in TSCD and
ultimately reaching 2.25 Å in D. This step was accompanied by
the concomitant cleavage of the C−Cl bond, with the C···Cl
distance increasing from 1.82 Å in C to 2.35 Å in TSCD and
finally becoming 3.00 Å in D.
(−91.6 kJ mol−1)). The overall reaction from A to I was found
to be exergonic by −52.3 kcal mol−1 (−218.8 kJ mol−1).
Remarkably, the relative Gibbs free energy of the most stable
intermediate D, −40.2 kcal mol−1 (−168.2 kJ mol−1), was still
12 kcal mol−1 (50.2 kJ mol−1) higher than that of the final
products, [Ni(IMe)]2(μ-Cl)2 (I) and the biaryl compound. The
result suggested that the compound D is the metastable
intermediate over the oxidative addition and reductive
elimination processes. The structure of D revealed that the
terminal σ-aryl moiety located at the position adjacent to that of
the carbene ligand could cause steric repulsion between the
carbene and the σ-aryl moiety. Given that there was no bonding
interaction between the nickel atoms in D, the equilibrium for
the cleavage of the Ni−Cl bond by disproportionation could
ultimately lead to the formation of the mixed biaryl products
(Scheme 10).
In the following step, one of the two η1-bound aryl ligands
was brought into a bridging position (μ-η1,η2-C6H5) to give
intermediate E via TSDE. The bridging aryl moiety in
intermediate E was η2-coordinated to a Ni center containing
a σ-coordinated aryl ring. This process required free activation
energy of 26.2 kcal mol−1 (109.6 kJ mol−1) from D and was
endergonic by 23.9 kcal mol−1 (100.0 kJ mol−1), but the free
energy of the transition state TSDE was still 14.0 kcal mol−1
(58.8 kJ mol−1) lower than that of A + ClBz. The Ni−Ni
distance was reduced to 2.84 Å following this step. During the
following two steps (i.e., E → F and then F → G), there was a
reorientation of the μ-η1,η2-C6H5-bound aryl ligand, which
brought the aryl moieties together to enable the formation of
the C−C bond of the biphenyl product (note that the C−C
bond distance in E was 3.13 Å). In F, the C−C bond distance
was reduced to 2.73 Å, and the aryl ligand was coordinated in a
μ-η1 fashion. However, in G, the aryl unit was once again bound
to the Ni in a μ-η1,η2 mode but was now also η2 coordinated to
the other Ni center bearing no σ-aryl ring. Finally, the
formation of the C−C bond of the biphenyl product occurred
via TSGH (dC−C = 1.96 Å) to give complex H (dC−C = 1.48 Å),
where the newly formed biphenyl (Ph-Ph) molecule was
coordinated in a μ-η2,η2 fashion. This process was determined
to be facile, with a free energy barrier of only 14.1 kcal/mol
using IMe as the model spectator ligand. In the final step the
Ph-Ph ligand was released to give the complex [Ni(IMe)]2(μ-
Cl)2 (I) in an exergonic reaction (ΔG = −21.9 kcal mol−1
Scheme 10. Possible Disproportionation Equilibrium
The initial coordination of the chlorobenzene substrate to
the nickel atom had the largest activation barrier of all of the
steps in the process, and its activation entropy was estimated to
be −70 J K−1 mol−1. The theoretical pathway in Figure 6 could
be consistent with the real pathway, because the rate-limiting
process would be the coordination of the aryl halide also in
calculations. As indicated by the kinetics, the more sterically
hindered IPr ligand would most likely require considerable
H
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
negative activation entropy. Furthermore, the coordination of
the aryl halide would most probably require activation energy
higher than that calculated using IMe. In any case, the most
important feature of this calculation is that the whole process is
conducted on dinickel centers, which supports our hypothesis
that this catalytic cycle involves both [Ni(I)−Ni(I)] and
[Ni(II)−Ni(II)] species.
Alternative Catalytic Cycle Involving 2. In the
stoichiometric reaction of 1a with both 4-chlorotoluene and
phenylmagnesium chloride, the consumption of 1a was much
slower than that of 4-chlorotoluene, which was consumed to
form the product in catalysis (Figure 4). Furthermore, the
distribution of the mixed biaryl products in the stoichiometric
reaction of 1a or 3 was completely different from that observed
after the catalytic Kumada cross-coupling reaction, yielding only
one cross-coupling product, where the stoichiometric reactions
afforded all of the possible biaryl products as a result of a
disproportionation reaction (Scheme 10). These experiments
revealed that the catalytic cycles proposed above between 1a
and 2 and between 1a and 3 are not involved in the real
catalytic cycle. On the other hand, these results strongly
support the following alternative pathway as shown in Scheme
11. Oxidative addition of aryl halide to complex 2 would lead to
activation pathway, including the coordination and oxidative
addition of chlorobenzene to dinickel(I) dichloride (see the
Supporting Information). The initiation process from the
mixture of Ni(cod)2, IPr, and aryl halide easily affords 1 and
then 2 (Scheme 11), as the result of the reaction of 1 with aryl
halide, indicating that this alternative cycle is also readily
accessible from the general in situ catalyst system using nickel
precursor and IPr.
Although we cannot determine the intermediary compound
X from the reaction of 2 with aryl halides, we obtained a
oxidative addition product from the stoichiometric reaction of 2
with 3,5-dichloropyridine at room temperature (Scheme 12). A
Scheme 12. Oxidative Addition Product 5 from 2 and 3,5-
Dichloropyridine and Proposed Intermediate X′ in the
Reaction
Scheme 11. Alternative Possible Catalytic Cycle Mediated by
2 and the Initiation Process Involving the Catalysis Reaction
of 1 with Aryl Halide
single crystal suitable for X-ray crystallography was obtained
following the concentration of a reaction mixture of 2 with 3,5-
dichloropyridine. The divalent tetranickel complex 5 was
successfully determined by X-ray diffraction analysis (see the
Supporting Information). That is, oxidative addition of aryl
chloride on 2 can occur, suggesting the possibility of formation
of the intermediate X in the catalytic cycle from 2 and aryl
halide.
The most important point found in these mechanistic studies
is that the oxidation and reduction processes in these
dinickel(I) complexes are most likely distinct from the
corresponding mononuclear Ni(I) and Ni(III) redox processes.
Ni(III) complexes are generally synthesized by electrochemical
oxidation or the oxidation of Ni(II) complexes with dioxygen.27
In contrast, the oxidation states of the two nickel(I) centers
increased from +1 to +2 in the current study during the
oxidative addition of the aryl halide (Scheme 1). The bridging
halogen ligand could play an important role in maintaining the
proximal positions of both the nickel centers. Furthermore, the
bulky NHC ligand IPr kinetically stabilized the dinuclear
nickel(I) framework in a way that cannot be obtained for the
smaller IMes ligand. Dinuclear systems of this type are
consistent with the large negative entropy (ΔS⧧ = −56.52
0.52 cal K−1 mol−1) observed in the reaction of the similar
dinickel(I) system 1a with aryl halide followed by the reductive
elimination of the corresponding biaryl product.
the formation of a dinickel(II) adduct (X), and subsequent
transmetalation with a Grignard reagent would give a nickel(II)
diaryl species (D), which would undergo a facile reductive
elimination reaction to give the biaryl product with the
concomitant regeneration of complex 2. This alternative cycle
was believed to be much more likely than any of the other
cycles for the following reasons. In comparison with complex
1a, complex 2 seems to be sterically less hindered, and the
reactions with aryl halide on 2 to give the intermediate X may
not require such a large negative entropy, which was observed
in the kinetic study of 1a with 4-chlorotoluene. Therefore, this
process would occur with greater ease than the corresponding
reaction onto the more hindered compound 1a. These
reactions including oxidative addition and reductive elimination
with the complex 2 in Scheme 11 are regarded as the processes
similar to the calculated ones, which are shown in Figure 6.
Additionally, a theoretical study of the process from 2 to X was
also conducted using DFT calculations with the 3-21G basis set
for C, H, and N, 6-31G* for Cl, and CEP/LanL2DZ for Ni.
The results of this study led to the proposal of a similar dinickel
CONCLUSIONS
■
Intermediary monovalent dinickel complexes have been found
using a nickel(0) catalyst precursor and a bulky N-heterocyclic
carbene ligand in the Kumada−Tamao−Corriu cross-coupling
reaction of aryl halides. The compounds were obtained by
oxidative addition of aryl halides to a Ni(0) precursor and were
coordinatively unsaturated (30e), dinuclear μ-σ-aryl-μ-chloro
I
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
nickel(I) species, which could react with both aryl halides and
Grignard reagents. Further reaction with aryl halides yielded the
C(Ar)−C(Ar) bond coupled biaryl products together with a
(μ-chloro)nickel(I) dimer, whereas transmetalation with a
Grignard reagent afforded bis(μ-σ-aryl)dinickel(I) complexes.
Both of these dinuclear complexes had the same dinickel(I)
framework as the complex 2 and can be regarded as at least the
analogues of the real active species in the catalytic cycle.
Therefore, stoichiometric reactions of these complexes
concerning the catalytic reaction were investigated. Experi-
ments showing faster oxidative addition in comparison to
transmetalation on the dinickel(I) complexes indicated the
possible reaction order of these complexes in the presence of
both aryl halide and Grignard reagent. Kinetics for the oxidative
addition of an aryl halide and the subsequent reductive
elimination of a biaryl product showed large negative entropy,
which suggested that the coordination of the aryl halide could
be the rate-determining step as a consequence of the steric
hindrance derived from the two bulky NHC ligands.
Experimental details, DFT calculation details, crystal
structures of 1b and 4 (PDF)
Full crystallographic descriptions (CIF)
AUTHOR INFORMATION
Corresponding Author
*E-mail for K.M.: kmatsuba@fukuoka-u.ac.jp.
ORCID
Kouki Matsubara: 0000-0003-3909-7738
Karl Kirchner: 0000-0003-0872-6159
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Japan Society for the
Promotion of Science (Grant-in-Aid for Scientific Research
22550104 and 25410123) and a fund from the Central
Research Institute of Fukuoka University (No. 117106).
Although the semistable dinickel(II) intermediate, which
would be formed by the reaction of the dinuclear μ-σ-aryl μ-
chloro nickel(I) species with aryl halide, cannot be uncovered
in this study, theoretical studies strongly suggested that
coordination and oxidative addition of aryl halide onto the
unsaturated nickel(I) species led to formation of the nickel(II)
intermediate, retaining its dinuclear framework. It was of
interest that the Gibbs energy of the divalent intermediate was
still higher than that of the final (μ-chloro)nickel(I) dimer,
making it possible to smoothly restart the catalysis. However,
this cycle was negligible in the real catalytic system, as
concluded from the experimental results.
REFERENCES
■
(1) Cross-coupling reactions, recent reviews: (a) Metal-Catalyzed
Cross-Coupling Reactions, 2nd ed.; De Meijere, A., Francois, D., Eds.;
Wiley-VCH: Weinheim, Germany, 2004. (b) Han, F.-S. Chem. Soc.
Rev. 2013, 42, 5270−5298. (c) Suzuki, A. Angew. Chem., Int. Ed. 2011,
́
50, 6723−6737. (d) Phapale, V. B.; Cardenas, D. J. Chem. Soc. Rev.
2009, 38, 1598−1607. (e) Jana, R.; Pathak, T. P.; Sigman, M. S. Chem.
Rev. 2011, 111, 1417−1492. (f) Frisch, A. C.; Beller, M. Angew. Chem.,
Int. Ed. 2005, 44, 674−688. (g) Corbet, J.-P.; Mignani, G. Chem. Rev.
2006, 106, 2651−2710. (h) Nicolaou, K. C.; Bulger, P. G.; Sarlah, D.
Angew. Chem., Int. Ed. 2005, 44, 4442−4489.
(2) For example, see: (a) Bohm, V. P. W.; Gstottmayr, C. W. K.;
̈
̈
On the other hand, the above results were very helpful in
understanding the most possible catalytic cycle via the
formation of the dinickel(I) μ-chloride complex 2 as the key
compound in the catalytic process. It would allow for the faster
oxidative addition of an aryl halide to form the divalent dinickel
species, in comparison with the reaction with the starting
dinuclear μ-σ-aryl μ-chloro nickel(I) complex 1, because of
possessing the less-hindered nickel centers in 2, which would be
immediately regenerated by transmetalation of the divalent
dinickel species with a Grignard reagent and subsequent
reductive elimination to give the cross-coupling products,
enabling a fast catalytic process. It was quite interesting that the
dinuclear system in oxidative addition of aryl halide and
reductive elimination of biaryl works efficiently in the cross-
coupling reaction. The DFT calculations for the mechanism
revealed that oxidative addition occurs on one of the
unsaturated nickel centers and then one of the terminal aryl
groups migrates to the bridging position to couple with the aryl
group on the other nickel. Moreover, the biaryl just after the
C−C bond formation coordinates to both nickel centers. Such
cooperative processes make the catalytic cross-coupling occur
smoothly without a large energy barrier, led by the dinuclear
system. We are conducting studies to apply the catalyst system
to other cross-coupling reactions, these being generally
challenging processes for nickel catalysts.
Weskamp, T.; Herrmann, W. A. Angew. Chem., Int. Ed. 2001, 40,
3387−3389. (b) Tekavec, T. N.; Zuo, G.; Simon, K.; Louie, J. J. Org.
Chem. 2006, 71, 5834−5836. (c) Kuhl, S.; Schneider, R.; Fort, Y. Adv.
Synth. Catal. 2003, 345, 341−344. (d) Kim, C.-B.; Jo, H.; Ahn, B.-K.;
Kim, C. K.; Park, K. J. Org. Chem. 2009, 74, 9566−9569. (e) Lohre, C.;
Droge, T.; Wang, C.; Glorius, F. Chem. - Eur. J. 2011, 17, 6052−6055.
̈
(f) Ritleng, V.; Henrion, M.; Chetcuti, M. J. ACS Catal. 2016, 6, 890−
906.
(3) Kumada coupling catalytic cycle: Herrmann, W. A. In Applied
Homogeneous Catalysis with Organometallic Compounds; Cornils, B.,
Herrmann, W. A., Eds.; Wiley-VCH: Weinheim, Germany, 2002;
Chapter 3, pp 824−825.
(4) Radical catalysis: (a) Manolikakes, G.; Knochel, P. Angew. Chem.,
Int. Ed. 2009, 48, 205−209. (b) Ford, L.; Jahn, U. Angew. Chem., Int.
Ed. 2009, 48, 6386−6389. (c) Gao, C.-Y.; Cao, X.; Yang, L.-M. Org.
Biomol. Chem. 2009, 7, 3922−3925. (d) Jones, G. D.; McFarland, C.;
Anderson, T. J.; Vicic, D. A. Chem. Commun. 2005, 4211−4213.
(e) Jones, G. D.; Martin, J. L.; McFarland, C.; Allen, O. R.; Hall, R. E.;
Haley, A. D.; Brandon, R. J.; Konovalova, T.; Desrochers, P. J.; Pulay,
P.; Vicic, D. A. J. Am. Chem. Soc. 2006, 128, 13175−13183. (f) Ren, P.;
Vechorkin, O.; von Allmen, K.; Scopelliti, R.; Hu, X. J. Am. Chem. Soc.
2011, 133, 7084−7095. (g) Breitenfeld, J.; Ruiz, J.; Wodrich, M. D.;
Hu, X. J. Am. Chem. Soc. 2013, 135, 12004−12012.
(5) Miyazaki, S.; Koga, Y.; Matsumoto, T.; Matsubara, K. Chem.
Commun. 2010, 46, 1932−1934.
(6) (a) Zhang, K.; Conda-Sheridan, M.; Cooke, S. R.; Louie, J.
Organometallics 2011, 30, 2546−2552. (b) Nagao, S.; Matsumoto, T.;
Koga, Y.; Matsubara, K. Chem. Lett. 2011, 40, 1036−1038. (c) Page,
M. J.; Lu, W. Y.; Poulten, R. C.; Carter, E.; Algarra, A. G.; Kariuki, B.
M.; Macgregor, S. A.; Mahon, M. F.; Cavell, K. J.; Murphy, D. M.;
Whittlesey, M. K. Chem. - Eur. J. 2013, 19, 2158−2167. (d) Davies, C.
J. E.; Page, M. J.; Ellul, C. E.; Mahon, M. F.; Whittlesey, M. K. Chem.
Commun. 2010, 46, 5151−5153.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00451.
(7) (a) Cariou, R.; Graham, T. W.; Dahcheh, F.; Stephan, D. W.
Dalton Trans. 2011, 40, 5419−5422. (b) Fischer, R.; Langer, J.;
J
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Malassa, A.; Walther, D.; Gorls, H.; Vaughan, G. Chem. Commun.
(21) These results suggested that the aryl moiety is undergoing a fast
flip-flop motion between the two metal atoms in solution at room
temperature. However, we do not have any evidence supporting the
solution structure containing the η2-CC interaction to the nickel
center.
(22) (a) Heckler, J. E.; Zeller, M.; Hunter, A. D.; Gray, T. G. Angew.
Chem., Int. Ed. 2012, 51, 5924−5928. (b) Osawa, M.; Hoshino, M.;
Hashizume, D. Dalton Trans. 2008, 2248−2252.
(23) (a) Beck, R.; Johnson, S. A. Chem. Commun. 2011, 47, 9233−
9235. (b) Keen, A. L.; Doster, M.; Johnson, S. A. J. Am. Chem. Soc.
2007, 129, 810−819.
(24) Yu, D.-G.; Li, B.-J.; Shi, Z.-J. Acc. Chem. Res. 2010, 43, 1486−
1495.
(25) Morrell, D. G.; Kochi, J. K. J. Am. Chem. Soc. 1975, 97, 7262−
7270.
(26) Tsou, T. T.; Kochi, J. K. J. Am. Chem. Soc. 1979, 101, 6319−
6332.
(27) (a) Iluc, V. M.; Miller, A. J. M.; Anderson, J. S.; Monreal, M. J.;
Mehn, M. P.; Hillhouse, G. L. J. Am. Chem. Soc. 2011, 133, 13055−
́
13063. (b) Gennari, M.; Orio, M.; Pecaut, J.; Bothe, E.; Neese, F.;
̈
2006, 2510−2512. (c) Ohno, K.; Arima, K.; Tanaka, S.; Yamagata, T.;
Tsurugi, H.; Mashima, K. Organometallics 2009, 28, 3256−3263.
(d) Rodriguez, B. A.; Delferro, M.; Marks, T. J. Organometallics 2008,
27, 2166−2168. (e) Velian, A.; Lin, S.; Miller, A. J. M.; Day, M. W.;
Agapie, T. J. Am. Chem. Soc. 2010, 132, 6296−6297. (f) Hruszkewycz,
D. P.; Wu, J.; Hazari, N.; Incarvito, C. D. J. Am. Chem. Soc. 2011, 133,
3280−3283. (g) Xue, F.; Zhao, J.; Hor, T. S. A. Dalton Trans. 2013,
42, 5150−5158. (h) Kalvet, I.; Bonney, K. J.; Schoenebeck, F. J. Org.
Chem. 2014, 79, 12041−12046.
(8) Reviews, for example: (a) van der Vlugt, J. I. Eur. J. Inorg. Chem.
2012, 2012, 363−375. (b) Dyson, P. J. Coord. Chem. Rev. 2004, 248,
2443−2458. (c) Nagashima, H. Monatsh. Chem. 2000, 131, 1225−
1239. (d) Suss-Fink, G.; Therrien, B.; Vieille-Petit, L.; Tschan, M.;
̈
Romakh, V. B.; Ward, T. R.; Dadras, M.; Laurenczy, G. J. Organomet.
Chem. 2004, 689, 1362−1369. (e) Metal Clusters in Chemistry;
́
Braunstein, P., Rose, J., Eds.; Wiley-VCH: Weinheim, Germany, 1999.
(9) Takao, T.; Suzuki, H. Coord. Chem. Rev. 2012, 256, 695−708 and
references cited therein..
(10) Xi, Z.; Zhou, Y.; Chen, W. J. Org. Chem. 2008, 73, 8497−8501.
(11) (a) Stambuli, J. P.; Kuwano, R.; Hartwig, J. F. Angew. Chem., Int.
Ed. 2002, 41, 4746−4748. (b) Bonney, K. J.; Schoenebeck, F. Chem.
Soc. Rev. 2014, 43, 6609−6638. (c) Proutiere, F.; Lyngvi, E.; Aufiero,
M.; Sanhueza, I. A.; Schoenebeck, F. Organometallics 2014, 33, 6879−
6884. (d) Yin, G.; Kalvet, I.; Schoenebeck, F. Angew. Chem., Int. Ed.
2015, 54, 6809−6813.
Collomb, M.-N.; Duboc, C. Inorg. Chem. 2011, 50, 3707−3716.
(c) Green, B. J.; Tesfai, T. M.; Xie, Y.; Margerum, D. W. Inorg. Chem.
2004, 43, 1463−1471. (d) Lee, C.-M.; Chuang, Y.-L.; Chiang, C.-Y.;
Lee, G.-H.; Liaw, W.-F. Inorg. Chem. 2006, 45, 10895−10904. (e) Lee,
C.-M.; Chiou, T.-W.; Chen, H.-H.; Chiang, C.-Y.; Kuo, T.-S.; Liaw,
W.-F. Inorg. Chem. 2007, 46, 8913−8923.
(12) Recent reviews: (a) Valente, C.; C
D.; Sayah, M.; Organ, M. G. Angew. Chem., Int. Ed. 2012, 51, 3314−
3332. (b) Díez-Gonzalez, S.; Marion, N.; Nolan, S. P. Chem. Rev. 2009,
109, 3612−3676. (c) Clavier, H.; Nolan, S. P. Chem. Commun. 2010,
46, 841−861. (d) Corberan, R.; Mas-Marza, E.; Peris, E. Eur. J. Inorg.
̧alimsiz, S.; Hoi, K. H.; Mallik,
́
́
́
Chem. 2009, 2009, 1700−1716. (e) Schuster, O.; Yang, L.;
Raubenheimer, H. G.; Albrecht, M. Chem. Rev. 2009, 109, 3445−
3478. (f) Hahn, F. E.; Jahnke, M. C. Angew. Chem., Int. Ed. 2008, 47,
3122−3172.
(13) Ni(I) complexes: (a) Kitiachvili, K. D.; Mindiola, D. J.;
Hillhouse, G. L. J. Am. Chem. Soc. 2004, 126, 10554−10555.
(b) Melenkivitz, R.; Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem.
Soc. 2002, 124, 3846−3847. (c) Holland, P. L.; Cundari, T. R.; Perez,
L. L.; Eckert, N. A.; Lachicotte, R. J. J. Am. Chem. Soc. 2002, 124,
14416−14424. (d) Eckert, N. A.; Dinescu, A.; Cundari, T. R.; Holland,
P. L. Inorg. Chem. 2005, 44, 7702−7704. (e) Fujita, K.; Rheingold, A.
L.; Riordan, C. G. Dalton Trans. 2003, 2004−2008. (f) Eaborn, C.;
Hill, M. S.; Hitchcock, P. B.; Smith, J. D. Chem. Commun. 2000, 691−
692. (g) Ito, M.; Matsumoto, T.; Tatsumi, K. Inorg. Chem. 2009, 48,
2215−2223. (h) Marlier, E. E.; Tereniak, S. J.; Ding, K.; Mulliken, J.
E.; Lu, C. C. Inorg. Chem. 2011, 50, 9290−9299. (i) Laskowski, C. A.;
Bungum, D. J.; Baldwin, S. M.; Ciello, S. A. D.; Iluc, V. M.; Hillhouse,
G. L. J. Am. Chem. Soc. 2013, 135, 18272−18275.
(14) Laskowski, C. A.; Hillhouse, G. L. Organometallics 2009, 28,
6114−6120.
(15) Bohm, V. P. W.; Weskamp, T.; Gstottmayr, C. W. K.;
̈
̈
Herrmann, W. A. Angew. Chem., Int. Ed. 2000, 39, 1602−1604.
(16) Dible, B. R.; Sigman, M. S.; Arif, A. M. Inorg. Chem. 2005, 44,
3774−3776.
(17) (a) Lee, C. H.; Laitar, D. S.; Muller, P.; Sadighi, J. P. J. Am.
̈
Chem. Soc. 2007, 129, 13802−13803. (b) Hoshimoto, Y.; Hayashi, Y.;
Suzuki, H.; Ohashi, M.; Ogoshi, S. Organometallics 2014, 33, 1276−
1282.
(18) (a) Schaub, T.; Backes, M.; Radius, U. J. Am. Chem. Soc. 2006,
128, 15964−15965. (b) Zell, T.; Fischer, P.; Schmidt, D.; Radius, U.
Organometallics 2012, 31, 5065−5073. (c) Radius, U.; Bickelhaupt, F.
M. Coord. Chem. Rev. 2009, 253, 678−686.
(19) Because complex 3a was detected as the minor product
generated in the formation of 3b in solution, cogenerated 3c cannot be
identified.
(20) (a) McGuinness, D. S.; Cavell, K. J. Organometallics 1999, 18,
1596−1605. (b) Schaub, T.; Fischer, P.; Meins, T.; Radius, U. Eur. J.
Inorg. Chem. 2011, 2011, 3122−3126.
K
DOI: 10.1021/acs.organomet.6b00451
Organometallics XXXX, XXX, XXX−XXX