Ni ion-containing ionic liquid salt and Ni ion-containing immobilized ionic liquid on silica: Application to Suzuki cross-coupling reactions between chloroarenes and arylboronic acids

Article abstract of DOI:10.1016/j.jcat.2006.06.020

Bis(1-n-butyl-3-methyl-imidazolium)tetrachloronickelate ([Bmim]₂[NiCl₄]), which is a nickel ion-containing ionic liquid, and immobilized nickel ion-containing ionic liquid (ImmNi²⁺_IL) on silica surface were prepared and applied as new catalysts for Suzuki cross-coupling reactions between aryl chlorides and arylboronic acids. It was found that pretreatment of the catalysts and the addition of triphenylphosphine to the reaction system greatly promoted the reactions. Before the addition of substrates, [Bmim]₂[NiCl₄] should be treated with K₃PO₄ in dioxane at a refluxing temperature for 1 h, and ImmNi²⁺_IL should be treated with NaO^tBu in dioxane at room temperature for 30 min. An active precursor for the catalytic reactions was found to be a Ni carbene species formed by pretreatment with the bases as characterized by NMR and EXAFS. Reusability was confirmed for ImmNi²⁺_IL.

Full text of DOI:10.1016/j.jcat.2006.06.020

Journal of Catalysis 242 (2006) 357–364
www.elsevier.com/locate/jcat
Ni ion-containing ionic liquid salt and Ni ion-containing immobilized ionic
liquid on silica: Application to Suzuki cross-coupling reactions
between chloroarenes and arylboronic acids
Chongmin Zhong a, Takehiko Sasaki b,∗, Mizuki Tada a, Yasuhiro Iwasawa a
a
Department of Chemistry, Graduate School of Science, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
b
Department of Complexity Science and Engineering, Graduate School of Frontier Sciences, The University of Tokyo, 5-1-5, Kashiwanoha,
Kashiwa, Chiba 277-8561, Japan
Received 28 March 2006; revised 17 June 2006; accepted 20 June 2006
Abstract
Bis(1-n-butyl-3-methyl-imidazolium)tetrachloronickelate ([Bmim] [NiCl ]), which is a nickel ion-containing ionic liquid, and immobilized
2
4
2+
nickel ion-containing ionic liquid (ImmNi _IL) on silica surface were prepared and applied as new catalysts for Suzuki cross-coupling reactions
between aryl chlorides and arylboronic acids. It was found that pretreatment of the catalysts and the addition of triphenylphosphine to the reaction
system greatly promoted the reactions. Before the addition of substrates, [Bmim] [NiCl ] should be treated with K PO in dioxane at a refluxing
2
4
3
4
2+
t
temperature for 1 h, and ImmNi _IL should be treated with NaO Bu in dioxane at room temperature for 30 min. An active precursor for the
catalytic reactions was found to be a Ni carbene species formed by pretreatment with the bases as characterized by NMR and EXAFS. Reusability
2+
was confirmed for ImmNi _IL.
© 2006 Elsevier Inc. All rights reserved.
Keywords: Ionic liquid; Suzuki cross-coupling reaction; Chloroarene; Immobilization; Ni; Phenylboronic acid; EXAFS
1. Introduction
nary ammonium-based salts are typical ionic liquids studied
widely because of their ease of synthesis.
Metal ion-containing ionic liquids find various applications
to catalysts, but there have been few syntheses and structure
analyses of ionic liquids involving metal ions. In many cases,
the properties of metal ion-containing ionic liquids were studied
with samples prepared by mixing chlorine- or bromine-based
ionic liquids with metal chlorides or bromides in various mo-
lar ratios. However, ionic liquids prepared by this method are
usually mixtures containing several types of anionic species. It
is well known that ionic liquids have the tendency to form rel-
atively stable supercooled states, especially in the presence of
impurities, indicating the need for careful handling and treat-
ment during syntheses and measurements.
The nonaqueous polar property of ionic liquids makes them
promising alternative solvents to traditional organic solvents
and water as an immobilizing phase for transition metal cata-
lysts in homogeneous or biphasic systems, in addition to other
versatile functionalities such as electrolytes in electrochem-
istry and eco-friendly nonvolatile solvent in organic synthesis
[1–7]. The completely ionic nature of ionic liquids makes them
useful as solvents for highly charged complexes, and the low
vapor pressure of ionic liquids makes it possible to use them
in vacuum and as “green” solvents in industrial processes [8].
They are also considered new precursor materials for catalysts
[9–12]. Dialkylimidazolium and N-alkyl pyridinium or quater-
To obtain definite information on the composition, struc-
ture, and physicochemical properties of ionic liquids, it is very
important to prepare pure ionic liquids with single and clear
compositions. Based on these research interests, we prepared
*
Corresponding author. Fax: +81 4 7136 3910.
E-mail address: takehiko@k.u-tokyo.ac.jp (T. Sasaki).
0021-9517/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2006.06.020

358
C. Zhong et al. / Journal of Catalysis 242 (2006) 357–364
a series of metal ion-containing ionic liquids and characterized
their structures by X-ray single crystal structure analysis and
thermal and spectroscopic methods [11]. A series of metal ion-
containing ionic liquids [Bmim]2[MCl4] (M = Sn, Cu, Ni, Mn,
Fe, Co, Zn, and Pt) and [Bmim]2[ZrCl6] were synthesized and
single crystal structures were determined. The melting point
of the synthesized salts and the degree of the local symmetry
of metal chloride anions were found to be closely correlated
with each other. Ionic conductivity was measured for super-
cooled states of the synthesized salts and it was found that these
salts behaved as ionic liquids unambiguously. Moreover, the
metal ion-containing ionic liquids were covalently immobilized
on silica surface (Imm_IL) and characterized by EXAFS and
DR UV/vis [12]. It was found that Cu2+-containing ionic liq-
uid (ImmCu2+_IL) gave high activities for Kharasch addition
reactions [12]. There are previous reports on immobilization
of ionic liquids involving FeCl3, SnCl2, and AlCl3 [13–17].
Valkenberg et al. [13,16] introduced the term “novel Lewis acid
catalysts” (NLACs) in the immobilization of ionic liquids; our
method [12] for immobilization can be classified as NLAC II
(immobilization via cation).
and immobilized ionic liquid (ImmNi2+_IL) to the Suzuki
cross-coupling reactions between chloroarenes and arylboronic
acids.
2. Experimental
2.1. Materials
Anhydrous 1-methylimidazole (>99%) and 3-trimethoxy-
silylpropyl chloride (>97%), o-tolylboronic acid, 2,6-dimeth-
ylphenylboronic acid, 3-thiopheneboronic acid, and anhydrous
potassium phosphate (97%) were purchased from Aldrich. An-
hydrous NiCl2 (99.9%), 1-chlorobutane (98.0%), and all other
chemicals were purchased from WAKO. K3PO4 was crushed
into a fine powder and dried under vacuum at 413 K for 2 h
before use. All dehydrated solvents stored under nitrogen at-
mosphere were purchased from WAKO and used without fur-
ther purification except 1,4-dioxane, which was dehydrated
again before use with metal sodium and stored under nitrogen
atmosphere. Aerosil 300 (surface area 300 m2/g), supplied by
Japan Aerosil Co., use as support. The calcined support was
kept in dry nitrogen.
In this study we focus on Ni-catalyzed Suzuki cross-
coupling reactions using Ni ion-containing ionic liquid and Ni
ion-containing immobilized ionic liquid on silica surface as an
extension of our study [11,12]. The Suzuki cross-coupling re-
action [18,19] between arylboronic acids and aryl halides or
triflates is one of the most important methods for synthesizing
biaryls, which are useful for natural products, polymers, ad-
vanced materials, and target molecules of medicinal chemistry.
In ionic liquid media, a Pd-catalyzed Suzuki cross-coupling
reaction has been reported for bromoarenes [20] in a homoge-
neous system and for iodoarenes immobilized on a resin [21].
Although palladium-catalyzed Suzuki coupling reactions can
achieve high yields when aryl bromides, iodides, and triflates
are used, chloroarenes, which are economical and easily avail-
able substrates, have been used only rarely, because the oxi-
dation addition of chloroarenes to palladium(0) is too slow to
allow development of the catalytic cycle. Recently several re-
search groups solved this problem with Pd. The application
of stable carbene, imidazol-2-ylidene, to palladium-catalyzed
Suzuki coupling reactions [22–24] makes the Pd catalysts
active to chloroarenes. Layered double hydroxide-supported
nanopalladium catalyst [25] has been reported to be active to
chloroarenes. Recently, hydroxyapatite-bound palladium com-
plex was reported to catalyze the Suzuki cross-coupling reac-
tion with chloroarene in the presence of TBAB [26]. Nickel
catalysts, which are much cheaper than precious metals like pal-
ladium, have been reported to be active for the Suzuki reactions
of chloroarenes in the presence of four equivalents of phosphine
ligands [27–32]. However, these systems suffer from problems
involved with using strong reducing agents, a large amount of
catalyst (up to 10 mol%), expensive phosphine ligands, and pro-
longed reaction times, which do not favor practical large-scale
use. These issues are awaiting resolution by newly designed
economically feasible catalysts.
2.2. X-ray fluorescence
X-ray fluorescence analysis (XRF) was done using a SEA-
2010 spectrometer (Seiko Electronic Industrial Co.). Physical
mixtures of metal chloride salts and support Aerosil 300 in var-
ious ratios were used to make calibration lines to estimate the
metal loadings in immobilized ionic liquids.
2.3. EXAFS measurements
Measurements of extended X-ray absorption fine structure
(EXAFS) at the Ni K-edge were carried out at the Photon
Factory in the Institute of Materials Structure Science, High-
Energy Accelerator Research Organization (KEK-MSS-PF).
The measurements were made in a transmission mode at BL-
12C at room temperature. Synchrotron radiation from the stor-
age ring was monochromatized by a Si(111) channel-cut crys-
tal. Ionization chambers used as detectors for incident X-ray
(I0) and transmitted ray (I) were filled with Ar/N2 mixture
(1:1) gas and Ar gas, respectively. The samples were handled
in dry nitrogen. The EXAFS raw data were analyzed with the
UWXAFS analysis package [33], including the background
subtraction program AUTOBK [34] and the curve-fitting pro-
gram FEFFIT [35]. Backscattering amplitudes and phase shifts
were calculated using FEFF8.0 [36].
2.4. Catalyst preparation
The catalyst preparation procedures are summarized in
Scheme 1 [11,12]. All processes were conducted under ni-
trogen atmosphere. 1-n-Butyl-3-methylimidazolium chloride
([Bmim]Cl, 1) was obtained by refluxing a mixture of N-meth-
ylimidazole and 1-chlorobutane (R1). Purification of the ob-
tained product was achieved by recrystallization. Bis(1-n-
butyl-3-methylimidazolium) tetrachloronickelate(II) ([Bmim]2
In this work we present for the first time the catalytic ap-
plications of Ni2+-containing ionic liquid ([Bmim]2[NiCl4])

C. Zhong et al. / Journal of Catalysis 242 (2006) 357–364
359
2.5. Catalytic reactions
All experiments were carried out under nitrogen atmosphere
using standard Schlenk techniques. The yields were referred to
gas chromatography (GC) yields based on chloroarenes. The
products were analyzed with a Shimadzu gas chromatograph
(GC-14B) with a flame ionization detector using a 30 m ×
0.25 mm NEUTRA Bond-1 column. GC-mass spectroscopy
(MS) measurements were performed on Shimadzu GCMS-
QP5000 for assigning GC peaks.
[Bmim]2[NiCl4]-catalyzed Suzuki reactions without pre-
treatment were examined by mixing [Bmim]2[NiCl4]
(0.02 mmol), K3PO4 (0.85 g, 4 mmol), phosphine, chloroarene
(2 mmol), arylboronic acid (2.4 mmol), and solvent in a 50-ml
Schlenk tube and stirring at 353 K for a prespecified time. It
was found that the reaction without any pretreatment resulted
in very low activity; consequently, a way of pretreating the cat-
alysts to promote the reaction was sought. [Bmim]2[NiCl4]
(0.02 mmol) and K3PO4 (0.85 g, 4 mmol) were mixed in a
50-ml Schlenk tube, to which solvent was added, and the mix-
ture was refluxed for 1 h. After the pretreatment, phosphine,
chloroarene (2 mmol), and arylboronic acid (2.4 mmol) were
added, and the vessel was sealed with a septum wrapped with
wire. The mixture was stirred at 353 K for a prespecified time.
Immobilized catalysts were also used for the Suzuki reac-
tions with various pretreatment methods. ImmNi2+_IL
(0.04 mmol) and K3PO4 (0.85 g, 4 mmol) or NaOtBu
(0.16 mmol, 16 mg) were mixed in a 50-ml Schlenk tube, af-
ter which the solvent was added and the mixture was stirred
at refluxing temperature (in the case of K3PO4) or room tem-
perature (in the case of NaOtBu) for a prespecified time. After
the pretreatment, phosphine, chloroarene (2 mmol), and aryl-
boronic acid (2.4 mmol) were added, and the vessel was sealed
with a septum wrapped with a wire.
2+
Scheme 1. Preparation steps for [Bmim] [NiCl ] and ImmNi _IL.
2
4
[NiCl4], 2) was obtained by refluxing the compound 1 and an-
hydrous nickel(II) chloride dissolved in dry acetonitrile (R2),
then recrystallizing from acetonitrile solution.
For the preparation of immobilized ionic liquid catalysts,
1-methyl-3-(3-trimethoxysilylpropyl) imidazolium chloride (3)
was synthesized by refluxing the mixture of N-methylimidazole
and 3-trimethoxysilylpropyl chloride (R3).
3. Results and discussion
In a glovebox, pretreated Aerosil 300 and the compound (3)
(weight ratio 1:1) were dispersed in dehydrated toluene in a
Schlenk tube, and the mixture was refluxed for 48 h in nitro-
gen (R4). Toluene was removed under reduced pressure, and
the resulting material was transferred to a Soxhlet tube in nitro-
gen. The excess ionic liquid was removed by Soxhlet extrac-
tion with dichloromethane. The resulting solid is denoted as
Imm_IL (4). Elemental analysis was performed for this sam-
ple. Anal. found: C, 7.36; H, 1.92; N, 1.54. The density of
immobilized ionic liquid functional groups was estimated as
ca. 1.2 nm2 based on the nitrogen percentage in the elemen-
tal analysis. Note that no N-containing solvent was used for the
preparation of Imm_IL.
3.1. EXAFS analysis for ImmNi2+_IL
The crystal structures of [Bmim]2[NiCl4] has been deter-
mined by X-ray single crystal structure analysis, which will be
reported separately [11]. [Bmim]2[NiCl4] crystallizes in space
group Cc of a monoclinic structure with a = 9.450(1) Å, b =
16.410(1) Å, c = 15.439(2) Å, β = 104.946(2)◦, and Z = 4.
Its asymmetric unit (Fig. 1) contains two [Bmim]+ cations and
one [NiCl4]2− anion. The two butyl chains on the two imida-
zolium rings cross each other at nearly right angles. The anion
[NiCl4]2− has a distorted tetrahedron, and the mean angle of
Cl–Ni–Cl is 115.8◦. The mean Ni–Cl bond length of [NiCl4]2−
is 2.249 Å.
In the next step, Imm_IL was added to an acetonitrile so-
lution of NiCl2 in a 50-ml Schlenk tube and refluxed for 24 h
(R5). Acetonitrile was removed at a reduced pressure, and ex-
cess NiCl2 was removed by Soxhlet extraction with acetoni-
trile for 48 h. The prepared immobilized ionic liquid is de-
noted as ImmNi2+_IL (5). The Ni loading was determined by
XRF to be 3.3 wt%. The amount of NiCl2 was calculated as
The ImmNi2+_IL was characterized by Ni K-edge EXAFS.
To determine the coordination number (CN) around Ni, the
amplitude reduction factor S02 should be determined using a
reference compound with a known structure. The structure
of ionic liquid [Bmim]2[NiCl4] has already been determined
as mentioned above, which has a coordination number of 4;
thus [Bmim]2[NiCl4] was used as a reference compound
0.6 Ni ion nm−2
.

360
C. Zhong et al. / Journal of Catalysis 242 (2006) 357–364
ImmNi2+_IL is shown in Fig. 2b. Within the fitting uncer-
tainty, the distance and coordination number of Ni–Cl bond
in ImmNi2+_IL are the same as those in [Bmim]2[NiCl4], indi-
cating that the local structure around the Ni ion in ImmNi2+_IL
is almost the same as that in [Bmim]2[NiCl4]. The schematic
view of ImmNi2+_IL is shown at the bottom of Scheme 1, with
the nickel ion held between two imidazolium ions.
3.2. Suzuki cross-coupling reactions with the ionic liquid
[Bmim]2[NiCl4]
When [Bmim]2[NiCl4] was used as a catalyst for the Suzuki
coupling reaction without any treatment, the yield was quite
low. Grasa et al. [22] and McLachlan et al. [37] have shown that
Pd/imidazolium catalysts should be pretreated before substrates
are added, to obtain a catalytically active solution. In our initial
study, the pretreatment of [Bmim]2[NiCl4] with K3PO4 in diox-
ane at a refluxing temperature for 1 h greatly improved catalyst
activity; the catalyst seemed to change into a different chemical
species. Various bases were used for the pretreatment, and an-
hydrous K3PO4 was found to be the best base for performance.
Indeed, anhydrous K3PO4 has been most widely used in nickel-
catalyzed Suzuki reactions in dioxane [27–31,38]. Based on the
above information, the reaction conditions were optimized us-
ing anhydrous K3PO4 as base and dioxane as solvent, whereas
the pretreatment of [Bmim]2[NiCl4] with K3PO4 at refluxing
temperature for 1 h before the addition of arylchloride and aryl-
boronic acid was carried out in dioxane (Table 2).
Fig. 1. The ORTEP drawing of the asymmetric unit of [Bmim] [NiCl ] at the
50% probability level. Hydrogen atoms are omitted for clarity.
2
4
(a)
It is known that nickel phosphine complexes are active for
the Suzuki cross-coupling of chloroarenes in the existence of
additional phosphine ligands. Although in the absence of any
phosphines, [Bmim]2[NiCl4] was active for the substrates of
iodoarenes and bromoarenes, it gave a low yield of 23% for
4-chlorotoluene (Table 2, entry 1). To improve the activity,
the effect of phosphines on the yield was investigated (entries
2–10). The addition of PPh3 in various equivalents with respect
to nickel produced a great increase in yield upon the increment
of the equivalent of PPh3, with yield reaching 96% at the equiv-
alent of 2 (entry 6). Increasing the amount of PPh3 over 3 equiv-
alents had no further effect on yield (entry 7; see also Fig. 3a).
The effects of different phosphines in two equivalents were also
investigated (entries 8–10; Fig. 3b). Dppf, the most widely used
ligand in Ni catalytic systems, gave a yield of only 15%. Dppf’s
low activity might be due to its bulky steric effect, which is
incompatible with ligands from ionic liquid. For this reason,
PPh3, the most common, inexpensive, and stable phosphine,
was the best choice. Under the same conditions as those for
entry 6, however, a Ni-phosphine complex NiCl2(PPh3)2 gave
a yield of 10% (entry 18). Although [Bmim]2[NiCl4] without
PPh3 was a little more active (yield 23%) than NiCl2(PPh3)2,
(b)
3
Fig. 2. k -Weighted Fourier transform (amplitude (thick curve) and imag-
inary part (thin curve)) of Ni K-edge EXAFS. (a) [Bmim] [NiCl ] and
(b) ImmNi _IL (3.3 wt% Ni). Observed data: solid curves; fitting data: dotted
curves.
2
4
2+
for determining the local structure of ImmNi2+_IL. The k3-
weighted EXAFS Fourier transform for [Bmim]2[NiCl4] is
shown in Fig. 2a. The parameter S02 was determined to be
0.83 by curve-fitting the experimental EXAFS function, in
which the coordination number was fixed at 4. Then the ob-
tained S02 (0.83) was used in curve-fitting the EXAFS data
for ImmNi2+_IL. The structural parameters for the reference
compound [Bmim]2[NiCl4] and ImmNi2+_IL are summarized
in Table 1. The k3-weighted EXAFS Fourier transform for
Table 1
2+
EXAFS fitting results for [Bmim] [NiCl ] and ImmNi _IL
2
4
−1
−5
2
−1
−1
2
0
Bond
CN
4.00
R (10 nm)
DW (10 nm )
ꢀk (10 nm
)
ꢀR (10 nm)
S
ꢀE (eV)
R (%)
0
f
[Bmim] [NiCl ]
Ni–Cl
Ni–Cl
2.267 0.003
2.268 0.003
4.6 0.3
6.1 0.4
3.0–13.0
3.0–13.0
1.0–3.0
1.0–3.0
0.83 0.04
0.83
1.6 0.6
2.1 0.6
0.53
0.65
2
4
2+
ImmNi _IL
3.6 0.2

C. Zhong et al. / Journal of Catalysis 242 (2006) 357–364
361
a
Table 2
Optimization of reaction conditions for the Suzuki cross coupling reaction between 4-chlorotoluene and phenylboronic acid on the [Bmim] [NiCl ] catalyst
2
4
a
b
b
c
Entry
Phosphine
None
Solvent
Yield (%)
Entry
Phosphine
Solvent
Yield (%)
1
2
3
4
5
6
7
Dioxane (0.63)
Dioxane (0.63)
Dioxane (0.63)
Dioxane (0.63)
Dioxane (0.63)
Dioxane (0.63)
Dioxane (0.63)
23
25
40
44
61
96
94
11
12
13
14
15
16
17
PPh (2.0)
PPh (2.0)
PPh (2.0)
3
PPh (2.0)
PPh (2.0)
3
PPh (2.0)
PPh (2.0)
Dioxane (1.0)
Dioxane (1.7)
Dioxane (2.8)
Toluene (0.63)
Acetonitrile (0.63)
DMF (0.63)
71
64
32
74
57
51
20
3
PPh (0.5)
PPh (1.0)
PPh (1.5)
PPh (1.8)
PPh (2.0)
PPh (3.0)
3
3
3
3
3
3
3
3
Dioxane/H O
3
3
2
(0.5/0.13)
Dioxane (0.63)
Dioxane (0.63)
d
8
9
10
PCy (2.0)
Dioxane (0.63)
Dioxane (0.63)
Dioxane (0.63)
2
21
15
18
19
NiCl (PPh )
3 2
10
64
3
2
e
P(NMe ) (2.0)
PPh (2.0)
2 3
3
Dppf (2.0)
a
Reaction conditions: [Bmim] [NiCl ], 0.04 mmol (1.0 mol%); 4-chlorotoluene, 4 mmol; phenylboronic acid, 4.8 mmol; K PO , 8 mmol; 353 K; 20 h.
2
4
3
4
b
c
d
e
The amount of phosphine was expressed as the molar ratio with respect to Ni.
The amount of solvent was expressed as the volume (ml) with respect to 1 mmol of chlorotoluene.
1 mol% was used.
No pretreatment was done.
(a)
(b)
(c)
(d)
Fig. 3. Summary of reaction conditions for the Suzuki coupling reaction catalyzed by [Bmim] [NiCl ]. (a) Effect of molar ratio of PPh to Ni, (b) effect of
2
4
3
phosphines, (c) effect of volume of dioxane, (d) effect of solvent.
[Bmim]2[NiCl4] in the presence of 2 equivalents of PPh3 lig-
ands showed the highest activity (yield 96%, entry 6).
ane also resulted in a low yield of 20% (entry 17). The ratio
of 0.6 ml per mmol of chlorotoluene is necessary to keep the
reaction system in good stirring state.
No studies on the effect of solvent volume have been re-
ported to date. When the ratio of dioxane increased from ca. 0.6
to 1.0 ml per mmol of chlorotoluene, the yield decreased to 71%
(entry 11); at ca. 2.8 ml per mmol of chlorotoluene, the yield de-
creased dramatically, to 32% (entry 13; Fig. 3c). The effects of
other solvents at the ideal ratio of 0.63 ml per mmol of chloro-
toluene are reported in Table 3 (entries 14–16) and Fig. 3d. The
lower yield in toluene (74%) might be due to the low solubility
of K3PO4 in the less polar solvent. Although K3PO4 has greater
solubility in acetonitrile and DMF than in toluene and dioxane,
the coordinating properties of DMF and acetonitrile may cause
a poisoning effect on the catalyst. Adding water to the diox-
The time dependence of the yield under the optimized condi-
tion is shown in Fig. 4. There was an induction period of about
7 min during which the catalyst was reduced to Ni(0), the active
species in the Suzuki reaction. The turnover frequency (TOF)
for the Suzuki cross-coupling reaction between chlorotoluene
and phenylboronic acid was calculated as 255 h−1—to the best
of our knowledge, the highest value yet reported for Ni cata-
lysts.
The cross-coupling reactions between several substituted
arylboronic acids and aryl chlorides were performed in a
standard set of reaction conditions. As shown in Table 3,

362
C. Zhong et al. / Journal of Catalysis 242 (2006) 357–364
Table 3
a
[Bmim] [NiCl ]/PPh catalyzed Suzuki cross coupling reactions
2
4
3
b
Entry R
R
Catalyst
(mol% Ni)
Time Yield
c
e
d
(%)
(h)
1
2
3
4
5
6
7
8
9
10
11
12
13
4-CH CO–
3
4-CN–
4-CHO–
0.5
0.5
1.0
0.5
0.5
0.5
1.0
1.0
1.0
1.0
1.0
1.0
2
2
2
2
2
20
20
18
18
24
18
18
18
100
100
87
100
100
96
94
83
90
37
4-CF –
3
4-(PhCO)–
4-CH –
3
4-CH O–
3
4-CH OOC–
3
2-CH –
3
2-CH – 6-CH –
3
3
3
4-CH –
2-CH –
83
0
19
Fig. 5. Ni K-edge EXAFS data for the brown complex ((A) k -weighted
3
3
3
4-CH –
2-CH – 6-CH –
EXAFS oscillation, (B) k -weighted Fourier transform) and the yellow com-
3
3
3
3
3
4-CH –
3-thiopheneboronic acid 1.0
plex ((C) k -weighted EXAFS oscillation, (D) k -weighted Fourier transform).
Thick solid curve and thick dotted curve represent observed and fitted absolute
values, respectively. Thin solid curve and thin dotted curve represent observed
and fitted imaginary parts, respectively.
3
a
Pretreatment was conducted by refluxing the mixture of [Bmim] [NiCl ]
2
4
and K PO in dioxane for 1 h, and then PPh , substrate and arylboronic acid
3
4
3
were added to start the reaction. Reaction conditions: PPh /Ni = 2; 3.0 mmol
3
aryl chlorides; 3.6 mmol arylboronic acid; 6.0 mmol K PO . Reactions were
3
4
carried out at 353 K under nitrogen atmosphere.
the instability of the substituents under basic reaction condi-
tions. Chloroarenes with an electron-donating group, such as
–Me and –OMe, also gave excellent yields (>90%). Ortho-
monosubstituted chloroarene readily coupled with phenyl-
boronic acid, and vice versa (entries 9 and 11). However,
ortho-dimethyl groups were not tolerated (entries 10 and 12).
Arylboronic acid with heteroatoms could not readily couple
with chloroarene (entry 13).
b
Vacancy represents no specific substitutional group.
Catalyst mol%Ni was defined with respect to arylchloride.
Reaction time was not optimized.
c
d
e
Yield was determined by GC.
3.3. Nature of the pretreated species
The pretreated species was characterized by NMR and
EXAFS. When the blue [Bmim]2[NiCl4] was pretreated with
K3PO4 in dioxane, a brown color complex was obtained in the
first 30 min, with a yellow complex formed in the subsequent
30 min. The 13C NMR (125 MHz, CDCl3) spectrum of the yel-
low complex showed a resonance from the carbene carbons at
δ = 166.8 ppm, which is in the range for Ni(II) carbene com-
plex (160–170 ppm) [39–44]. The reported chemical shift of
carbene carbon in Ni(0) carbene complexes ranges from 190
to 225 ppm [45,46]. The carbene complexes were also char-
acterized with Ni K-edge EXAFS (brown complex, Figs. 5A
and 5B; yellow complex, Figs. 5C and 5D). The fitting results
are presented in Table 4. The Ni atom of the brown com-
Fig. 4. Variation of yield with reaction time in the [Bmim] [NiCl ]/2PPh cat-
alytic system at 353 K.
2
4
3
[Bmim]2[NiCl4]/PPh3 was able to catalyze the Suzuki reactions
for substrates with various functional groups. Chloroarenes
with electron-withdrawing substituents (entries 1, 2, 4, and 5)
gave 100% yields of cross-coupling products after 2 h using
0.5 mol% catalyst. The slightly lower yield for substrates with
having –CHO and –COOMe (entries 3 and 8) may be due to
Table 4
EXAFS curve-fitting results for the brown and yellow color complexes
a
b
Bond
Brown color complex
CN
Yellow color complex
−1
−5
2
−1
−5
2
R (10 nm)
DW (10 nm )
CN
R (10 nm)
DW (10 nm )
Ni–Cl
Ni–C
2
1
2.22 0.02
1.88 0.06
3
0
4
7
1.4 1.3
4
1
1.92 0.02
3
4
a
−1
−1
2
ꢀE = 8 2 eV; ꢀk = 30–117 nm ; ꢀR = 0.10–0.30 nm; S = 0.73; R = 4.70%.
0
f
0
2
b
ꢀE = 8 3 eV; ꢀk = 30–120 nm ; ꢀR = 0.10–0.20 nm; S = 0.73; R = 0.73%.
0
f
0

C. Zhong et al. / Journal of Catalysis 242 (2006) 357–364
363
a
plex was coordinated by two chlorine atoms and two carbon
Table 5
2+
ImmNi _IL/PPh catalyzed Suzuki cross coupling reactions (2 mol% Ni)
3
atoms at distances of 0.222 and 0.188 nm, respectively. The Ni
atom of the yellow complex has four nearest carbons located at
0.192 nm. These results show that the 1-h pretreatment changed
[Bmim]2[NiCl4] to a Ni(II) complex with carbene ligands cor-
responding to dehydrogenated imidazolium cations. Herrmann
et al. also reported [44] NiCl2(carbene)2 species with a Ni–
Cl distance of 0.218 nm and a Ni–C distance of 0.191 nm,
corresponding to the brown complex. Douthwaite et al. [43]
reported Ni(II)(carbene)4 species with an average Ni–C dis-
tance of 0.193 nm, corresponding to the yellow complex in the
present study.
b
ꢀb
c
d
Entry
R
R
Time (h)
Yield (%)
1
2
3
4
5
4-CH CO–
3
4-CN–
4-CHO–
4-(PhCO)–
4-CF –
1
1
2
3
5
100
100
100
90
94
71
94
70
96
87
95
6
3
6
4-CH OOC–
5
5
3
7
4-CH –
3
3.4. Suzuki cross-coupling reactions on the immobilized ionic
liquid ImmNi2+_IL
8
9
4-CH –
2-CH –
3
0.33
10
7
10
20
3
10
11
12
13
14
4-CH O–
3
2-CH –
2-CH – 6-CH –
3-thiopheneboronic acid 15
20
3
Although [Bmim]2[NiCl4]/PPh3 showed high activity and
selectivity in the Suzuki cross-coupling reactions, separating
the catalyst from the reaction mixture proved difficult. The im-
mobilized catalyst ImmNi2+_IL, which can address this prob-
lem, was applied to the Suzuki cross-coupling reactions.
The effects of pretreatment and PPh3 were examined as
follows. In the absence of PPh3, 10 h of pretreatment of
ImmNi2+_IL with K3PO4 in refluxing dioxane gave a very
low yield of 2%, and 24 h of pretreatment gave a low yield
of 46%. The low solubility of K3PO4 in the solvent and the
presence of the support surface mandated a greatly prolonged
time for pretreating the immobilized catalyst. However, the ad-
dition of PPh3 had no effect on the yield, unlike the case of
[Bmim]2[NiCl4]. The prolonged pretreatment may change the
ionic liquid into a species that cannot interact with PPh3. To de-
crease the pretreatment time, a strong base NaOtBu was applied
in this system. Pretreatment with NaOtBu for 30 min at room
temperature affected the reaction a little, resulting in a yield of
52%; the addition of PPh3 further enhanced the yield to 84%.
If 2 mol% of ImmNi2+_IL was used, the highest yield of 94%
at 373 K was obtained; however, the yield decreased to 13% at
353 K.
3
3
31
9
2-CH – 6-CH –
3
3
a
2+
Pretreatment was conducted by stirring the mixture of ImmNi _IL and
t
NaO Bu (4 equiv(Ni)) in dioxane at RT for 30 min. Reaction conditions:
PPh /Ni = 2; 1.5 mmol aryl chlorides; 1.8 mmol arylboronic acids; 3.0 mmol
3
K PO ; 373 K; reactions were carried out under N .
3
b
4
2
Vacancy represents no specific substitutional group.
Reaction time was not optimized.
Yields determined by GC.
c
d
with dichloromethane, then performing the next reaction by
adding 1 equivalent of base, phosphine, substrates, and solvent.
In the first reuse, the yield was maintained at 94%. The re-
action of the filtrate was examined after adding reactants and
base, and negligible product was found, indicating that leach-
ing of the immobilized catalyst hardly occurs. The yield was
reduced to 69% on the second reuse and 64% on the third reuse,
probably because the relative content of catalysts decreased.
These results indicate reusability of the immobilized catalyst
ImmNi2+_IL.
4. Conclusion
Under the optimized reaction conditions, the TOF of the
ImmNi2+_IL/PPh3 catalytic system was determined to be
162 h−1 based on the yield in the first 20 min. The lower TOF
value for the immobilized catalyst is ascribed to immobilization
of the active sites on the support and the limited accessibility of
the reactants compared with the ionic liquid catalysts. Cross-
coupling reactions between several substituted arylboronic
acids and aryl chlorides were conducted; the performances are
given in Table 5. ImmNi2+_IL/PPh3-catalyzed Suzuki reac-
tions were exceptionally tolerant to various functional groups
on both aryl chlorides and arylboronic acids. Comparing Ta-
ble 3 with Table 5 shows very similar activity patterns in cat-
alytic systems ImmNi2+_IL/PPh3 and [Bmim]2[NiCl4]/PPh3
for the same substrates, except for p-chlorobenzaldehyde,
which gave a higher yield with ImmNi2+_IL/PPh3 (100%) than
with [Bmim]2[NiCl4]/PPh3 (87%) (entry 3 in both tables).
Reuse of ImmNi2+_IL was examined up to the third cy-
cle. The reuse procedure involved retrieving solid-phase in-
cluding catalyst and base by filtration, followed by washing
New [Bmim]2[NiCl4] and ImmNi2+_IL were synthesized
and applied as the first examples of Ni ion-containing ionic liq-
uids and immobilized ionic liquids to Suzuki cross-coupling
reactions between substrates with various functional groups.
Both of the catalysts showed very high activity and selectiv-
ity for the cross-coupling reactions between aryl chlorides and
arylboronic acids in the presence of 2 equivalents of PPh3.
A Ni carbene species produced by pretreating the catalysts
with K3PO4 or NaOtBu in dioxane was suggested to be a cat-
alytically active species for the Suzuki cross-coupling reaction
as characterized by NMR and EXAFS. The reaction condi-
tions were optimized to find the best yield of the reactions.
The activity of the catalyst [Bmim]2[NiCl4] was the highest
yet reported for Ni catalysts. The ImmNi2+_IL catalyst was
reusable, albeit with a slightly decreased yield, whereas sepa-
rating [Bmim]2[NiCl4] from the reaction products proved very
difficult.

364
C. Zhong et al. / Journal of Catalysis 242 (2006) 357–364
Acknowledgments
[24] W.A. Herrmann, C.-P. Reisinger, M. Spiegler, J. Organomet. Chem. 557
(1998) 93.
[25] B.M. Choudary, S. Madhi, N.S. Chowdari, M.L. Kantam, B. Sreedhar,
J. Am. Chem. Soc. 124 (2002) 14,127 .
[26] K. Mori, T. Hara, M. Oshiba, T. Mizugaki, K. Ebitani, K. Kaneda, New J.
Chem. 29 (2005) 1174.
[27] S. Saito, S. Oh-tani, N. Miyaura, J. Org. Chem. 62 (1997) 8024.
[28] A.F. Indolese, Tetrahedron Lett. 38 (1997) 3513.
[29] J.-C. Galland, M. Saviganac, J.-P. Genet, Tetrahedron Lett. 40 (1999)
2323.
This work was supported by the 21st Century COE Program
of MEXT, Japan. T.S. was supported by a Grand-in-Aid for Sci-
entific Research (14703004) from MEXT. The XAFS measure-
ments were performed at KEK-IMSS-PF with the approval of
the Photon Factory Advisory Committee (project 2004G081).
References
[30] S. Saito, M. Sakai, N. Miyaura, Tetrahedron Lett. 37 (1996) 2993.
[31] B.H. Lipshutz, J.A. Sclafani, P.A. Blomgren, Tetrahedron 56 (2000) 2139.
[32] F. Firooznia, C. Gude, K. Chan, Y. Satoh, Tetrahedron Lett. 39 (1998)
3985.
[33] E.A. Stern, M. Newville, B. Ravel, Y. Yacoby, D. Haskel, Physica B 208–
209 (1995) 117.
[34] M. Newville, P. Livins, Y. Yacoby, E.A. Stern, J.J. Rehr, Phys. Rev. B 47
(1995) 14,126.
[35] M. Newville, B. Ravel, D. Haskel, J.J. Rehr, E.A. Stern, Y. Yacoby, Phys-
ica B 208–209 (1995) 154.
[1] C.L. Hussey, Pure Appl. Chem. 60 (1988) 1763.
[2] T. Welton, Chem. Rev. 99 (1999) 2071.
[3] P. Wasserscheid, W. Keim, Angew. Chem. Int. Ed. 39 (2000) 3772.
[4] R. Hagiwara, Y. Ito, J. Fluorine Chem. 105 (2000) 221.
[5] J. Dupont, R.F. de Souza, P.A.Z. Suarez, Chem. Rev. 102 (2002) 3667.
[6] C.M. Gordon, Appl. Catal. A Gen. 222 (2001) 101.
[7] H. Olivier-Bourbigou, L. Magna, J. Mol. Catal. A 182–183 (2002) 419.
[8] K.R. Seddon, J. Chem. Technol. Biotechnol. 68 (1997) 351.
[9] J.E.L. Dullius, P.A.Z. Suarez, S. Einloft, R.F. de Souza, J. Dupont, Or-
ganometallics 17 (1998) 815.
[36] A.L. Ankudinov, B. Ravel, J.J. Rehr, S.D. Conradson, Phys. Rev. B 58
(1998) 7565.
[37] F. McLachlan, C.J. Mathews, P.J. Smith, T. Welton, Organometallics 22
(2003) 5350.
[10] D. Zhao, Z. Fei, T.J. Geldbach, R. Scopelliti, P.J. Dyson, J. Am. Chem.
Soc. 126 (2004) 15,876.
[11] C. Zhong, T. Sasaki, A. Jimbo-Kobayashi, E. Fujiwara, A. Kobayashi,
M. Tada, Y. Iwasawa, J. Phys. Chem. B, submitted for publication.
[12] T. Sasaki, C. Zhong, M. Tada, Y. Iwasawa, Chem. Commun. (2005) 2506.
[13] M.H. Valkenberg, C. deCastro, W.F. Hölderich, Green Chem. 4 (2002) 88.
[14] M.H. Valkenberg, C. deCastro, W.F. Hölderich, Appl. Catal. A Gen. 215
(2001) 185.
[15] C. deCastro, E. Sauvage, M.H. Valkenberg, W.F. Hölderich, J. Catal. 196
(2000) 86.
[16] M.H. Valkenberg, C. deCastro, W.F. Hölderich, Top. Catal. 14 (2001) 139.
[17] C.P. Mehnert, R.A. Cook, N.C. Dispenziere, M. Afework, J. Am. Chem.
Soc. 124 (2002) 12,932 .
[38] D. Zim, V.R. Lando, J. Dupont, A.L. Monteiro, Org. Lett. 3 (2001) 3049.
[39] D.S. McGuinness, W. Mueller, P. Wasserscheid, K.J. Cavell, B.W. Skelton,
A.H. White, U. Englert, Organometallics 21 (2002) 175.
[40] M.V. Baker, B.W. Skelton, A.H. White, C.C. Williams, Organometal-
lics 21 (2002) 2674.
[41] D.S. Clyne, J. Jin, E. Genest, J.C. Galucci, T.V. RajanBabu, Org. Lett. 2
(2000) 1125.
[42] W.A. Herrmann, J. Schwarz, M.G. Gardiner, M. Spiegler, J. Organomet.
Chem. 575 (1999) 80.
[18] N. Miyaura, A. Suzuki, Chem. Rev. 95 (1995) 2457.
[19] A. Suzuki, J. Organomet. Chem. 576 (1999) 147.
[20] C.J. Mathews, P.J. Smith, T. Welton, Chem. Commun. (2000) 1249.
[21] J.D. Revell, A. Ganesan, Org. Lett. 4 (2002) 3071.
[22] G.A. Grasa, M.S. Viciu, J. Huang, C. Zhang, M.L. Trudell, S.P. Nolan,
Organometallics 21 (2002) 2866.
[43] R.E. Douthwaite, D. Haussinger, M.L.H. Green, P.J. Silcock, P.T. Gomes,
A.M. Martins, A.A. Danopoulos, Organometallics 18 (1999) 4584.
[44] W.A. Herrmann, G. Gerstberger, M. Spiegler, Organometallics 16 (1997)
2209.
[45] A.J. Arduengo, S.F. Gamper, J.C. Calabrese, F. Davidson, J. Am. Chem.
Soc. 116 (1994) 4391.
[23] G. Altenhoff, R. Goddard, C.W. Lehmann, F. Glorius, Angew. Chem. Int.
Ed. 42 (2003) 3690.
[46] Y. Sato, R. Sawaki, M. Mori, Organometallics 20 (2001) 5510.