Synthesis and application of novel ionic phosphine ligands with a cobaltocenium backbone

Article abstract of DOI:10.1016/j.jorganchem.2007.05.046

Six novel ionic phosphine ligands with a cobaltocenium backbone, 1,1′-bis(dicyclohexylphosphino) cobaltocenium hexafluorophosphate (di- cypc⁺ PF₆^-) (1a), 1,1′-bis(di-iso-propylphosphino) cobaltocenium hexafluorophosphate (

Full text of DOI:10.1016/j.jorganchem.2007.05.046

Journal of Organometallic Chemistry 692 (2007) 3914–3921
www.elsevier.com/locate/jorganchem
Synthesis and application of novel ionic phosphine ligands
with a cobaltocenium backbone
*
Guang-Ao Yu, Yong Ren, Jin-Tao Guan, Yan Lin, Sheng Hua Liu
Key Laboratory of Pesticide and Chemical Biology of Ministry of Education, College of Chemistry, Central China Normal University, Wuhan 430074, China
Received 18 March 2007; received in revised form 24 May 2007; accepted 25 May 2007
Available online 7 June 2007
Abstract
Six novel ionic phosphine ligands with a cobaltocenium backbone, 1,10-bis(dicyclohexylphosphino) cobaltocenium hexafluorophos-
phate (di-cypcþPF6ꢀ) (1a), 1,10-bis(di-iso-propylphosphino) cobaltocenium hexafluorophosphate (di-isoppcþPFꢀ6 ) (2a), 1,10-bis(di-tert-
butylphosphino) cobaltocenium hexafluorophosphate (di-tbpcþPFꢀ) (3a), and the monophosphine ligand CcþPR2PFꢀ (Cc+
=
6
6
cobaltocenium; R = Cy, 1b; R = i-Pr, 2b; R = t-Bu, 3b) were synthesized and characterized by elemental analysis, spectroscopy, and
X-ray diffraction techniques. These ligands are air-stable and useful for Suzuki coupling reactions in the ionic liquid 1-butyl-3-methyl-
imidazolium hexafluorophosphate (BMIMþPF6ꢀ), enabling high catalytic activity.
ꢀ 2007 Elsevier B.V. All rights reserved.
Keywords: Ionic phosphine ligands; Cobaltocenium; Synthesis; Suzuki coupling reactions; Ionic liquids
1. Introduction
bonds and has found widespread use in organic synthesis
[3]. However, the separation of the products from the reac-
tion mixture, the recovery of the catalysts, and the need for
organic solvents are the major disadvantages in the Suzuki
cross-coupling reactions. For these reasons, many Suzuki
coupling reactions are not used on an industrial scale
despite their benefits. To overcome these problems, some
heterogeneous Pd-catalyzed Suzuki coupling systems have
been introduced, such as the use of aqueous media [4]
and ionic liquids [3b,5]. The use of ionic liquids as solvents
for the Suzuki coupling reactions is a very attractive
approach [5,6]. The products can be separated from ionic
liquids by extraction with organic solvents and the catalytic
solution can be reused. Generally, neutral phosphine
ligands are used to complex the palladium species, resulting
in excellent results for the palladium-catalyzed Suzuki cou-
pling reactions [5a]. However, the separation of the prod-
ucts with neutral phosphine ligands and Pd-catalyst is
difficult because neutral phosphine ligands and catalyst
easily dissolve in organic solvents. To overcome these
problems, a new catalyst system consisted of PdCl2/ionic
phosphine ligand/ionic liquid was suggested.
Metallocene derivatives have found widespread use as
ligands. In particular, phosphine ligands based on substi-
tuted ferrocenes have been widely used [1]. Phosphine
ligands based on substituted cobaltocenium are difficult
to be synthesized in large scale, and only 1,10-bis(diph-
enylphosphino)
cobaltocenium
hexafluorophosphate
(dppcþPFꢀ6 ) has been used in catalytic hydroformylation
[2]. Compared with the isoelectronic ferrocenes, cobaltoce-
nium cations show different reactivity on the cyclopenta-
dienyl rings due to their electron-withdrawing properties.
Owing to their ionic character, they show good solubility
in polar solvents and ionic liquids. These properties stimu-
lated our interest in synthesizing novel cobaltoceniumyl-
phosphines and in testing them as ligands in Suzuki
coupling reactions in ionic liquids.
The palladium-catalyzed Suzuki coupling reaction is one
of the most efficient methods for the construction of C–C
*
Corresponding author.
E-mail address: chshliu@mail.ccnu.edu.cn (S.H. Liu).
0022-328X/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.05.046

G.-A. Yu et al. / Journal of Organometallic Chemistry 692 (2007) 3914–3921
3915
In the present article, we describe the synthesis and char-
acterization of six novel ionic phosphine ligands with a
cobaltocenium backbone and their successful applications
in Suzuki coupling reactions using ionic liquids as solvent.
2. Results and discussion
2.1. Synthesis and characterization of ionic phosphine ligand
We synthesized 1,10-bis(di-cyclohexylphosphino) cobalt-
ocenium hexafluorophosphate (di-cypcþPF6ꢀ, 1a) following
the method given by Wasserscheid [2]. The substituted
cyclopentadiene C5H5PCy2 was obtained from NaCp; sub-
sequent deprotonation with n-BuLi generated the anionic
ligand, and in situ reaction with CoCl2 afforded bidentate
cobaltocenylphosphine. Oxidation of this cobaltocenyl-
phosphine to the ionic phosphine is difficult because the
phosphine can also be easily oxidized to phosphine oxide
[7]. Using the smooth anaerobic oxidation method, we
were able to obtain 1a in high yield free of phosphine
oxides. The cobaltocenylphosphine reacted with C2Cl6 to
generate the green 1,10-bis(di-cyclohexylphosphino)cobalt-
ocenium chloride, and anion exchange with NH4PF6 led
to air-stable yellow solid 1a. Besides the bidentate
phosphine ligand, a small amounts of monophosphine
CcþPðCyÞ2PFꢀ6 1b (Cc+ = cobaltocenium) was also
obtained.
Fig. 1. ORTEP diagram of 1a. Thermal ellipsoids are shown at the 50%
level. Selected bond lengths (A) and angles (ꢁ): Co(1)–C(13) = 2.058(12),
C(13)–P(2) = 1.850(14), C(12)–P(2) = 1.908(14), C(23)–P(1) = 1.877(13),
C(30)–P(1) = 1.819(13); C(30)–P(1)–C(29) = 102.8(5), C(30)–P(1)–C(23) =
105.0(6).
˚
Using this procedure, we were able to obtain 2a
1,10-bis(di-iso-propylphosphino) cobaltocenium hexaflu-
orophosphate (di-isoppcþPFꢀ6 ) and 3a 1,10-bis(di-tert-buty-
angle of 8.1(7)ꢁ for 1a, 1.1(2)ꢁ for 2a, and 4.8(2)ꢁ for 3a, (II)
the average distance between the Co atom and the center of
˚
˚
˚
Cp ring is 1.6573 A for 1a, 1.6430 A for 2a, and 1.6445 A
for 3a, (III) the two R2P substituents in the cation {[g5-
R2PC5H4]2Co} of 1a, 2a or 3a are trans to each other
lylphosphino)
cobaltocenium
hexafluorophosphate
(di-tbpcþPFꢀ6 ), together with small amounts of 2b (CcþP-
ði-PrÞ2PFꢀ6 ) and 3b (CcþPðt-BuÞ2PFꢀ6 ), respectively (see
Scheme 1).
+
with respect to the Co metal center.
These ionic ligands were fully characterized by elemental
analysis, 1H NMR, 31P NMR and X-ray diffraction analy-
sis. The H NMR spectra of 1a, 2a and 3a show two sing-
2.2. Catalysis in ionic liquid
1
The catalytic systems were prepared in situ by mixing
PdCl2 with the ligand in BMIMþPF6ꢀ. The results obtained
in Suzuki coupling reaction of phenyl bromide with phenyl
boronic acid using different ligands are summarized in
Table 2. We found that the six ligands are stable in air
and active for the Suzuki coupling reaction. The trend in
the catalytic activity of the bidentate ligands for Suzuki
coupling reaction is 1a < 2a < 3a; The catalytic activity of
lets at 5.81 and 5.97 ppm, 5.49 and 5.76 ppm, 5.67 and
5.89 ppm, respectively, for their H3/H4 and H2/H5 protons
of the substituted Cp rings.
Crystals of these ligands were obtained by slow diffusion
of hexane into acetone solutions of the ligands. The molec-
ular structures of 1a, 2a, 3a, 1b, 2b are shown in Figs. 1–5,
respectively (for the information of 3b, see Ref. [8]). As can
be seen in Figs. 1–3, the structures of the three bidentate
ligands are very similar to each other. For example, (I)
the two substitute Cp rings are not parallel with a dihedral
monophosphine ligands has
a similar trend (1b <
2b < 3b). This is in good accordance with reported results
for Cp2Fe(PR2)2 [9].
1. n-BuLi
2.CoCl2
PR2
PR2
Co+ PF6
PR2
PR2
NH4PF6
NaCp
Co+ Cl-
+
ClPR2
-
Co+ Cl-
-
Co+ PF6
+
+
3.C2Cl6
PR2
PR2
1a R= Cy
2a R= i-Pr
3a R= t-Bu
1b R= Cy
2b R= i-Pr
3b R= t-Bu
Scheme 1.

3916
G.-A. Yu et al. / Journal of Organometallic Chemistry 692 (2007) 3914–3921
Fig. 2. ORTEP diagram of 2a. Thermal ellipsoids are shown at the 50%
˚
level. Selected bond lengths (A) and angles (ꢁ): Co(1)–C(1) = 2.063(3),
Fig. 4. ORTEP diagram of 1b. Thermal ellipsoid are shown at the 50%
level. Selected bond lengths (A) and angles (ꢁ): Co(1)–C(1) = 2.032(5),
C(6)–P(1) = 1.833(4), C(11)–P(1) = 1.842(4), C(17)–P(1) = 1.864(4), C(6)–
C(1)–P(1) = 1.832(3), C(6)–P(1) = 1.868(3), C(12)–P(2) = 1.834(3), C(17)–
P(2) = 1.872(3); C(1)–P(1)–C(9) = 102.67(13), C(1)–P(1)–C(6) = 97.92(13).
˚
P(1)–C(11) = 103.70(17), C(6)–P(1)–C(17) = 101.88(17).
Fig. 5. ORTEP diagram of 2b. Thermal ellipsoid are shown at the 50%
˚
level. Selected bond lengths (A) and angles (ꢁ): Co(1)–C(1) = 2.060(4),
C(14)–P(1) = 1.851(6), C(11)–P(1) = 1.847(6), C(1)–P(1) = 1.816(5), C(1)–
P(1)–C(11) = 101.8(2), C(1)–P(1)–C(14) = 99.2(3).
Fig. 3. ORTEP diagram of 3a. Thermal ellipsoids are shown at the 50%
˚
level. Selected bond lengths (A) and angles (ꢁ): Co(2)–C(1) = 2.074(3),
bromides with varying electronic properties. The aryl bro-
mides with electron-donating groups, such as methyl
(entries 2–4) and methoxy (entries 5–7), gave excellent
yields, while aryl bromide with electron-withdrawing
groups, such as nitro, trifluoromethyl and acetyl, gave
the coupling products almost quantitatively. The trend
in the reactivity is expected for Suzuki coupling. It was
noteworthy that 2-bromothiophene also gave the desired
product in an excellent yield with this catalytic system
(entry 17). In contrast Colacot found that bromothioph-
ene is a difficult substrate to couple when neutral phos-
phine-based catalysts were employed [9]. Additionally, a
C(1)–P(1) = 1.837(3), C(6)–P(1) = 1.882(4), C(10)–P(1) = 1.900(5), C(1a)–
P(1a) = 1.890(5), C(6a)–P(1a) = 1.829(3), C(1)–P(1)–C(6) = 105.79(17),
C(6)–P(1)–C(10) = 110.2(2).
Coupling of phenyl boronic acid with a variety of aryl
bromides was conducted using three bidentate ligands as
illustrated in Table 3. Our first model, bromobenzene,
gave an excellent result of a 98% conversion of starting
aryl bromide and an isolated purified product yield of
92%, utilizing 1.0 mol% catalyst (entry 1). Then we tested
the scope of the reaction with a range of substituted aryl

G.-A. Yu et al. / Journal of Organometallic Chemistry 692 (2007) 3914–3921
3917
Table 1
Selected crystal data, data collection, and structure refinements for compounds 1a–3a, 1b, and 2b
1a 2a
3a
1b
2b
Formula
C34H44CoF6P3 C22H36CoF6P3 C26H44CoF6P3 C22H31CoF6P2 C16H23CoF6P2
Formula weight
Temperature (K)
Crystal color
Crystal system
Space group
718.53
293(2)
Yellow
Triclinic
P1
6.4358(6)
9.4851(8)
14.3753(13)
94.8580(10)
97.176(2)
99.3430(10)
854.15(13)
1
566.35
293(2)
Yellow
Monoclinic
P2(1)/c
8.0389(4)
22.6962(12)
14.7397(8)
90
97.9670(10)
90
2663.3(2)
4
1.412
622.45
293(2)
530.34
292(2)
Yellow
450.21
292(2)
Yellow
Monoclinic
P2(1)/c
19.9996(19)
14.9998(15)
14.4927(14)
90
110.038(2)
90
4084.5(7)
8
1.464
Yellow
Monoclinic
P2/c
11.7428(11)
10.0264(10)
26.965(3)
90
92.670(2)
90
3171.3(5)
2
Triclinic
ꢀ
P1
˚
a (A)
6.4097(8)
9.4244(12)
19.695(2)
81.078(2)
82.544(2)
81.751(2)
1156.1(2)
2
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
3
˚
V (A )
Z
Dc (mg mꢀ3
)
1.397
0.699
1.304
0.741
1.523
0.936
Absolute coefficient (mmꢀ1
)
0.875
1.045
F(000)
374
1176
1304
548
1840
h Range
2.87–28.24ꢁ
3157
1.66–28.26ꢁ
16,895
1.74–25.00ꢁ
15,577
2.11–26.00ꢁ
9275
1.74–25.99ꢁ
32,212
Number of reflections
2717
5271
5581
4465
7987
Number of independent reflections
R(int)
0.0325
0.988
0.0482
0.805
0.0286
1.094
R1 = 0.0565
wR2 = 0.1412
R1 = 0.0683
wR2 = 0.1487
0.0828
0.996
R1 = 0.0653
wR2 = 0.1585
R1 = 0.0863
wR2 = 0.1690
0.1027
0.959
Goodness-of-fit on F2
Final R indices
R1 = 0.0482
wR2 = 0.1188
R1 = 0.0588
wR2 = 0.1243
0.477, ꢀ0.196
R1 = 0.0442
wR2 = 0.0944
R1 = 0.0781
wR2 = 0.1013
0.265, ꢀ0.194
R1 = 0.0613
wR2 = 0.1706
R1 = 0.0937
wR2 = 0.1955
0.587, ꢀ0.640
[I > 2r(I)]
ꢀ3
˚
R indices (all data) largest difference in peak and hole (e A
)
0.434 , ꢀ0.275 1.234, ꢀ0.311
challenging problem in Suzuki coupling processes is the
ability to combine sterically hindered substrates. It is par-
ticularly difficult when coupling partners have large ortho
substituents. We found that it was efficient to couple the
sterically hindered 2-bromomesitylene with aryl boronic
acid (entry 18). The above procedure is not efficient for
the coupling of aryl chloride and aryl boronic acid, even
using 3a/PdCl2 as the catalyst, only 30% biphenyl product
was detected upon heating at 120 ꢁC for 3 h, while
Cp2Fe(PR2)2PdCl2 (R = i-Pr, t-Bu) was shown to be an
efficient catalyst for aryl chlorides [9,10]. Considering the
structural similarity of both kinds of ligands, their differ-
ent influence on the reaction activity has to be attributed
to electronic reasons. The electron density at the phospho-
rus atoms is significantly lower in the case of ionic ligand
due to the electron-withdrawing effect of the formal cobal-
t(III) central atom in the ligand. This interpretation is sup-
ported by former work [2].
The reusability of the present catalyst system was exam-
ined in the coupling reaction of bromobenzene with aryl
boronic acid. Unfortunately, the catalyst system can only
be reused 2–3 times (Table 4).
3. Conclusion
We describe the synthesis and characterization of six
novel ionic phosphine ligands with a cobaltocenium back-
bone by Wasserscheid’s method. These ionic phosphine
Table 2
Relative reactivities of ligand/PdCl2 for the coupling of bromobenzene and phenyl boronic acid in BMIM PF6
Br
+
B(OH)2
1.0 mol % PdCl2, 1.0 mol % ligand
BMIM PF6, 2.0 mol Na2CO3 ( aq )
Temperature (ꢁC)
Entry
Ligand
Time (h)
Yielda (%)
1
2
3
4
5
6
7
8
a
1a
1b
2a
2b
3a
3b
–
110
100
90
80
70
60
110
60
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
98
99
98
98
99
98
81
35
–
GC yield using hexadecane as internal standard.

3918
G.-A. Yu et al. / Journal of Organometallic Chemistry 692 (2007) 3914–3921
Table 3
Suzuki cross-coupling reaction of halobenzens with phenylbronic acid catalyzed by ligand/PdCl2 in BMIM PF6
Br
+
B(OH)2
1.0 mol % PdCl2, 1.0 mol % ligand
BMIM PF6, 2.0 mol Na2CO3 ( aq )
R
R
Entry
1
Substrate
Ligand
Temperature (ꢁC)
Time (h)
Yielda,b (%)
1a
2a
3a
110
80
60
0.5
0.5
0.5
98(92)
98(93)
98(92)
Br
2
3
1a
2a
3a
100
80
60
0.5
1
1.5
95(87)
99(94)
94(85)
Br
H3C
1a
2a
3a
100
80
60
0.5
1
1.5
98(94)
98(92)
98(90)
Br
CH3
4
5
1a
2a
3a
100
80
60
0.5
1
2
98(92)
98(92)
93(85)
Br
CH3
1a
2a
3a
100
80
60
2
2
1.5
98(92)
90(84)
94(85)
Br
H3CO
6
1a
2a
3a
100
80
60
1
1.5
1.5
98(90)
95(88)
95(86)
Br
Br
OCH3
7
8
9
1a
2a
3a
100
80
60
2
1.5
1.5
97(91)
92(86)
90(87)
OCH3
1a
2a
3a
100
80
60
0.5
1
1
99(92)
95(90)
92(86)
Br
O2N
1a
2a
3a
100
80
60
0.5
1
1
99(93)
96(90)
95(88)
Br
NO2
10
11
1a
2a
3a
100
80
60
0.25
0.5
0.5
99(92)
98(92)
94(85)
Br
NC
1a
2a
3a
100
80
60
0.25
0.5
0.5
99(92)
96(91)
96(89)
Br
CN
(continued on next page)

G.-A. Yu et al. / Journal of Organometallic Chemistry 692 (2007) 3914–3921
3919
Table 3 (continued)
Entry
12
Substrate
Ligand
Temperature (ꢁC)
Time (h)
Yielda,b (%)
1a
2a
3a
100
80
60
0.25
1.5
0.5
98(91)
98(92)
98(90)
Br
F3C
13
1a
2a
3a
100
80
60
0.25
1.5
0.5
95(90)
95(90)
95(91)
Br
CF3
14
15
1a
2a
3a
100
80
60
0.25
1
0.5
99(92)
95(92)
95(90)
Br
OHC
1a
2a
3a
100
80
60
0.25
1
0.5
99(92)
96(91)
97(91)
Br
CHO
16
17
1a
2a
3a
100
80
60
0.25
1
0.5
99(93)
98(92)
99(92)
Br
CHO
1a
2a
3a
100
80
60
0.25
0.5
0.5
99(94)
93(88)
95(90)
S
Br
18
1a
2a
3a
100
80
60
2
3
2
91(84)
95(86)
95(83)
Br
a
GC yield using hexadecane as internal standard.
Isolate yields are given in parentheses.
b
Table 4
Relative reactivities of ligand/PdCl2 for the coupling of bromobenzene and phenyl boronic acid in BMIM PF6
Br
+
B(OH)2
1.0 mol % PdCl2, 1.0 mol % ligand
BMIM PF6, 2.0 mol Na2CO3 ( aq )
Run
Ligand
1a (%)
2a (%)
3a (%)
1
2
3
a
98a
92a
60a
99a
95a
45a
98a
95a
20a
GC yield using hexadecane as internal standard.
ligands/PdCl2 are air-stable catalysts for aryl bromide
Suzuki coupling reactions. The trend in the reactivity of these
catalysts is 1a/PdCl2 < 2a/PdCl2 < 3a/PdCl2, 1b/PdCl2 <
2b/PdCl2 < 3b/PdCl2. The increase in activity is attributed
to the increase in electron density of the phosphine ligand.
Work is in progress in demonstrating the activities and reu-
sablities of these catalysts in Sonogashira, Heck and Negishi
coupling reactions.
4. Experimental
4.1. General
All reactions were carried out under an atmosphere of
highly purified nitrogen using standard Schlenk or vac-
uum-line techniques. THF was distilled under nitrogen
from sodium/benzophenone ketyl. All solvents were

3920
G.-A. Yu et al. / Journal of Organometallic Chemistry 692 (2007) 3914–3921
bubbled for at least 15 min before use. Elemental analyses
Compound 2b: Anal. Calc. for C16H23CoF6P2: C, 42.68;
H, 5.15. Found: C, 42.55; H, 5.26%. H NMR (CDCl3): d
1.10 (m, 12H, 4CH3), 2.06 (m, 2H, CH), 5.70 (s, 5H,
C5H5), 5.71 (s, 2H, C5H2 H2), 5.98 (s, 2H, C5H2H2)
ppm. 31P NMR (CDCl3): d ꢀ2.26 (s), ꢀ145.67 (quintet,
JP–F = 711 Hz) ppm.
1
1
were performed on an Elementar Vario El analyzer. H
NMR and 31P NMR spectra were obtained on a Varian
1
MERCURY Plus 400 MHz spectromer. H NMR chemi-
cal shifts are relative to TMS, and 31P NMR chemical
shifts are relative to 85% H3PO4.
4.2. Preparation of 1,10-bis(dicyclohexylphosphino)
cobaltocenium hexafluorophosphate (1a) and
(di-cyclohexylphosphino)cobaltocenium
hexafluorophosphate (1b)
4.4. Preparation of 1,10-bis(di-tert-butylphosphino)
cobaltocenium hexafluorophosphate (3a) and (di-tert-
butylphosphino)cobaltocenium hexafluorophosphate (3b)
Compound 3a and 3b were prepared similarly, with ClP
(t-Bu)2 instead of ClPCy2, in 12% (3a) and 8% yields (3b).
Compound 3a, Anal. Calc. for C26H44CoF6P3: C, 50.17;
To the freshly prepared NaCp (21 mmol) in THF,
ClPCy2 (21 mmol) dissolved in ether was added at ꢀ30 ꢁC.
The reaction mixture was stirred for 1.5 h at room tempera-
ture, then cooled to ꢀ78 ꢁC. n-BuLi (2.5 M in hexane,
21 mmol) was added dropwise. The mixture was stirred for
1.5 h at room temperature. CoCl2 (10.5 mmol) was added,
producing a dark brown solution. After stirring this solution
overnight under reflux, C2Cl6 (13 mmol) was added. The
resulting solution was stirred at room temperature for
another 10 min. After evaporation of all volatile substances
in vacuo, the residue was dissolved in 100 mL of CH2Cl2 and
filtered through Celite to remove LiCl. The filtrate was dried
in vacuo to obtain an oily, brown residue. The oil was
dissolved in 150 mL of CH2Cl2, NH4PF6 (11 mmol) in water
was added and stirred for 1 h, no precipitate was formed.
The solution was added to 150 mL of water, and extracted
with CH2Cl2 (3 · 30 mL). The combined organic layers were
dried over MgSO4, and filtered. After evaporation of solvent
in vacuo, the residue was subject to column chromatography
(silica gel, CH2Cl2) to give a yellow solid 1a (yield: 51%) and
a yellow solid 1b (yield: 14%).
1
H, 7.12. Found: C, 50.54; H, 7.11%. H NMR (CDCl3): d
1.24 (s, 18H, 6CH3), 1.27 (s, 18H, 6CH3), 5.67 (s, 4H,
C5H2H2), 5.89 (s, 4H, C5H2H2) ppm. 31P NMR (CDCl3):
d 23.25 (s), ꢀ145.90 (quintet, JP–F = 711 Hz) ppm.
Compound 3b: Anal. Calc. for C18H27CoF6P2: C,
45.20: H, 5.69. Found: C, 45.15; H, 5.32%. 1H NMR
(CDCl3): d1.24 (s, 9H, 3CH3), 1.26 (s, 9H, 3CH3), 5.71
(s, 5H, C5H5), 5.78 (s, 2H, C5H2H2), 6.04 (s, 2H,
C5H2H2) ppm. 31P NMR (CDCl3): d 24.32 (s), ꢀ145.88
(quintet, JP–F = 711 Hz) ppm.
4.5. X-ray crystal structure analysis
Crystals suitable for X-ray diffraction were grown by
slow diffusion of hexane into a solution of 1a in acetone
at room temperature. The crystals of 2a, 3a, 1b, 2b were
obtained by the similar method. A crystal was mounted
on a glass fiber, and the diffraction intensity data were col-
lected on a Bruker CCD 4K diffractometer with graphite-
˚
Compound 1a: Anal. Calc. for C34H52CoF6P3: C, 56.20;
monochromatized Mo Ka radiation (k = 0.71073 A). Lat-
1
H, 7.21. Found: C, 56.15; H, 7.18%. H NMR (CDCl3): d
tice determination and data collection were carried out
using SMART version 5.625 software. Data reduction and
absorption corrections were performed using SAINT version
6.45 and SADABS version 2.03. Structure solution and refine-
ment were performed using the SHELXTL version 6.14 soft-
ware package. All non-hydrogen atoms were refined
anisotropically. All hydrogen atoms were placed in ideal
positions and refined as riding atoms with relative isotropic
displacement parameters. Further crystallographic details
were summarized in Table 1.
1.13–1.96 (m, 44H, 4Cy): 5.81 (s, 4H, C5H2H2), 5.97 (s, 4H,
C5H2H2) ppm. 31P NMR (CDCl3): d ꢀ9.93 (s), ꢀ144.05
(quintet, JP–F = 711 Hz) ppm.
Compound 1b: Anal. Calc. for C22H31CoF6P2: C, 49.82;
1
H, 5.89. Found: C, 49.76; H, 5.83%. H NMR (CDCl3): d
1.03–1.88 (m, 22H, 2Cy): 5.65 (s, 5H, C5H5), 5.67 (s, 2H,
C5H2H2), 5.98 (s, 2H, C5H2H2) ppm. 31P NMR (CDCl3):
d ꢀ9.33(s), ꢀ144.50 (quintet, JP–F = 711 Hz) ppm.
4.3. Preparation of 1,10-bis(di-iso-propylphosphino)
cobaltocenium hexafluorophosphate (2a) and
(di-iso-propylphosphino)cobaltocenium
hexafluorophosphate (2b)
4.6. Catalytic experiments in ionic liquid
In a typical experiment (Table 2, entry 1), a 25 mL
round-bottom flask, equipped with a magnetic bar and a
condenser, was charged with PdCl2(0.0025 mmol), ligand
1a (0.0025 mmol) and BMIM+PF6ꢀ(5 mL). The mixture
was stirred at 110 ꢁC until a dark brown mixture was
obtained (at least 1 h); phenyl bromide (0.25 mmol), phenyl
boronic acid (0.375 mmol), and a solution of Na2CO3
(0.5 mmol) in water were added. The mixture was heated
with vigorous stirring until the peak of phenyl bromide
on GC disappeared, then cooled and extracted with diethyl
Compound 2a and 2b were prepared similarly, with
ClP(i-Pr)2 instead of ClPCy2, in 59% (2a) and 12% yields
(2b).
Compound 2a: Anal. Calc. for C22H36CoF6P3: C, 46.65;
1
H, 6.41. Found: C, 46.60; H, 6.38%. H NMR (CDCl3): d
1.07 (m, 24H, 8CH3), 2.05 (m, 4H, CH), 5.49 (s, 4H,
C5H2H2), 5.76 (s, 4H, C5 H2H2) ppm. 31P NMR (CDCl3):
d ꢀ3.03(s), ꢀ145.92 (quintet, JP–F = 711 Hz) ppm.

G.-A. Yu et al. / Journal of Organometallic Chemistry 692 (2007) 3914–3921
3921
[3] (a) R. Rajagopal, D.V. Jarikote, K.V. Srinivasan, Chem. Commun.
(2002) 616;
ether; the combined extracts were washed by brine, water
and dried. After removal of the solvent in vacuo, the mix-
ture was chromatographed on a silica pad to afford the dia-
ryl products in high purity as characterized by GC–MS, 1H
NMR spectroscopy.
(b) F. McLachlan, C.J. Mathews, P.J. Smith, T. Welton, Organo-
metallics 22 (2003) 5350;
(c) D. Zhao, Z. Fei, T.J. Geldbach, R. Scopelliti, P.J. Dyson, J. Am.
Chem. Soc. 126 (2004) 15876;
(d) J.-C. Xiao, J.M.J. Shreeve, Org. Chem. 70 (2005) 3072;
(e) W. Miao, T.H. Chan, Org. Lett. 5 (2003) 5003;
(f) J. McNulty, A. Capretta, J. Wilson, J. Dyck, G. Adjabeng, A.
Robertson, Chem. Commun. (2002) 1986;
4.7. Procedure for the catalyst recycling of Suzuki coupling
reaction
(g) A. Suzuki, Chem. Commun. (2005) 4759;
(h) L.-C. Liang, P.-S. Chien, M.-H. Huang, Organometallics 24
(2005) 353;
(i) S. Eichenseher, O. Delacroix, K. Kromm, F. Hampel, J.A.
Gladysz, Organometallics 24 (2005) 245.
After completion of the reaction and cooling to room
temperature, the products and the unreacted reagents were
extracted with diethyl ether, then washed with water and
dried in vacuum. The resulting viscous PdCl2/ligand/
BMIMþPFꢀ6 mixture was charged with fresh reagents (aryl
halide, phenyl boronic acid and the solution of Na2CO3 in
water) and heated at the required temperature with stirring
for the proper time.
[4] (a) L.R. Moore, K.H. Shaughnessy, Org. Lett. 6 (2004) 225;
(b) K.H. Shaughnessy, R.S. Booth, Org. Lett. 3 (2001) 2757;
(c) Y. Li, X.M. Hong, D.M. Collard, M.A. El-Sayed, Org. Lett. 2
(2000) 2385;
(d) G. Miao, P. Ye, L. Yu, C.M. Baldino, J. Org. Chem. 70 (2005)
2332;
(e) C. Najera, J. Gil-Molto, S. Karlstrom, L.R. Falvello, Org. Lett. 5
(2003) 1451;
5. Supplementary data
(f) R.B. Devasher, L.R. Moore, K.H. Shaughnessy, J. Org. Chem. 69
(2004) 7919;
(g) E.C. Western, K.H. Shaughnessy, J. Org. Chem. 70 (2005)
6378;
CCDC 640044, 640045, 640046, 640047 and 640048 con-
tain the supplementary crystallographic data for 1a, 2a, 3a,
1b and 2b. These data can be obtained free of charge via
http://www.ccdc.cam.ac.uk/conts/retrieving.html, or from
the Cambridge Crystallographic Data Centre, 12 Union
Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223 336
033; or e-mail: deposit@ccdc.cam.ac.uk.
(h) E.C. Western, J.R. Daft, E.M. Johnson II, P.M. Gannett,
K.H. Shaughnessy, J. Org. Chem. 68 (2003) 6767.
[5] (a) J.D. Revell, A. Ganesan, Org. Lett. 4 (2002) 3071;
(b) V. Calo, A. Nacci, A. Monopoli, F. Montingelli, J. Org. Chem.
70 (2005) 6040;
(c) R. Wang, B. Twamley, J.M. Shreeve, J. Org. Chem. 71 (2006)
426;
(d) P.D. Stevens, J. Fan, H.M.R. Gardimalla, M. Yen, Y. Gao,
Org. Lett. 7 (2005) 2085;
Acknowledgements
We gratefully acknowledge financial support from
National Natural Science Foundation of China
(No.20572029), New Century Excellent Talents in Univer-
sity (NCET-04-0743), the Cultivation Fund of the Key Sci-
entific and Technical Innovation Project, Ministry of
Education of China (No. 705039).
(e) L. Liu, Y. Zhang, B. Xin, J. Org. Chem. 71 (2006) 3994;
(f) J. Lemo, K. Heuze, D. Astruc, Org. Lett. 7 (2005) 2253;
(g) R.K. Arvela, N.E. Leadbeater, Org. Lett. 7 (2005) 2101;
(h) Y. Zheng, P.D. Stevens, Y. Gao, J. Org. Chem. 71 (2006)
537.
[6] (a) T. Welton, Chem. Rev. 99 (1999) 2071;
(b) J. Dupont, R.F. de Souza, P.A.Z. Suarez, Chem. Rev. 102 (2002)
3667.
References
[7] (a) A.W. Rudie, D.W. Lichtenberg, M.L. Katcher, A. Davison,
Inorg. Chem. 17 (1978) 2859;
(b) D.S. Shepard, B.F.G. Johnson, A. Harrison, S. Parsons, L.J.
Smidt Yellowo¨ess, S.J. Reed, J. Organomet. Chem. 563 (1998)
113.
[1] (a) G. Bandoli, A. Dolmella, Coord. Chem. Rev. 209 (2000) 161;
(b) L.-X. Dai, T. Tu, S.-L. You, W.-P. Deng, X.-L. Hou, Acc. Chem.
Res. 36 (2003) 659;
[8] Y. Ren, G.-A. Yu, S.-H. Liu, Acta Crystallogr., Sect. E 61 (2005)
m2714.
(c) R.C.J. Atkinson, V.C. Gibson, N.J. Long, Chem. Soc. Rev. 33
(2004) 313.
[9] T.J. Colacot, H.A. Shea, Org. Lett. 6 (2004) 3731.
[10] S. Teo, Z. Weng, T.S. Andy Hor, Organometallics 25 (2006) 1199.
[2] C.C. Brasse, U. Englert, A. Salzer, H. Waffenschmidt, P. Wasser-
scheid, Organometallics 19 (2000) 3818.