Highly efficient and stable palladium/imidazolium salt-phosphine catalysts for Suzuki-Miyaura cross-coupling of aryl bromides

Article abstract of DOI:10.1016/j.molcata.2006.04.025

A robust palladium catalyst system supported by the hybrid ligand N-heterocyclic carbene along with a phosphine derived from ferrocene has been discovered for the Suzuki cross-coupling reaction. This novel system effectively catalyzes the cross-coupling of aryl bromides and phenylboronic acid with up to 20,000 TONs and 10,000 h^-1 TOFs to produce biaryl products in excellent yields.

Full text of DOI:10.1016/j.molcata.2006.04.025

Journal of Molecular Catalysis A: Chemical 259 (2006) 7–10
Short communication
Highly efficient and stable palladium/imidazolium salt-phosphine catalysts
for Suzuki–Miyaura cross-coupling of aryl bromides
Ji-Cheng Shia,b,∗, Peng-Yu Yanga, Qingsong Tongb, Yang Wub, Yiru Pengb
a State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, China
b College of Chemistry and Materials Science, Fujian Normal University, Fuzhou 350007, China
Received 21 February 2006; received in revised form 5 April 2006; accepted 6 April 2006
Available online 13 July 2006
Abstract
A robust palladium catalyst system supported by the hybrid ligand N-heterocyclic carbene along with a phosphine derived from ferrocene has been
discovered for the Suzuki cross-coupling reaction. This novel system effectively catalyzes the cross-coupling of aryl bromides and phenylboronic
acid with up to 20,000 TONs and 10,000 h−1 TOFs to produce biaryl products in excellent yields.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Cross-coupling; Palladium; N-Heterocyclic carbene; Phosphine; Biaryl
The Suzuki–Miyaura cross-coupling reaction is an efficient
method for the preparation of biaryl compounds [1]. It has been
applied to the syntheses of polymers, materials, liquid crys-
tals [2], pharmaceutical compounds and natural products [3].
Initially, arylphosphines were used as ligands for palladium-
catalyzed reactions [4]. Recently, two classes of compounds
have been reported as more efficient ligands: the monoden-
tate, bulky and electron-donating tertiary phosphines [5], and
the nucleophic N-heterocyclic carbene (NHC) derivatives [6].
The application of the NHC ligand to the palladium-catalyzed
Suzuki–Miyaura cross-coupling was first reported by Her-
rmann in 1998 [7]. Subsequently, palladium complexes of some
NHCs, such as 1,3-bisadamantylimidazol-2-ylidene (1) [8a]
and 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene (2) [8b],
were found to be highly active catalyst systems for coupling
reactions using aryl chlorides at room temperature [9]. However,
these NHCs ligands have two major drawbacks: (1) the loading
of palladium catalyst is relatively high [8,9]; (2) some of the cat-
alysts were unstable, resulting in the precipitation of palladium
black even under mild reaction conditions [8a]. Therefore, it is
desirable to develop a new NHC-palladium catalyst system that
is both highly active and stable under the reaction conditions.
It is believed that the high catalytic activities of palladium
catalysts using the N-heterocyclic carbenes 1 and 2 as ligand
arise from their strong ␴-donating character and their bulkiness.
The severe steric demanding of the ligands 1 and 2 makes a
12-electon Pd(0) species possible. The electron richness and the
readily available coordination sites of the 12e species greatly
facilitate the oxidative addition of palladium to aromatic chlo-
rides. The high reactivity of 12e Pd(0) species, however, may be
also responsible for the instability of catalysts and the side reac-
tions via ␤-H elimination. Theoretical calculations indicated that
certain bidentate chelating ligands could improve the stability of
catalyst, and favor the oxidative-addition process [10]. Hybrid
N-heterocyclic carbene-phosphines are one type of the bidentate
ligands [11]. A few examples using such ligands were applied
to the Heck reaction and the Suzuki–Miyaura reaction by Nolan
[11e] and Zhou [11f]. Herein we present our preliminary results
in developing such a highly active and stable palladium catalysts
supported by N-heterocyclic carbenes for the Suzuki–Miyaura
reaction.
In the palladium-catalyzed cross-coupling, the proper bite
angle (around 102◦) induced by chelating ligands plays a cru-
cial role on the activity and selectivity [12]. Josiphos derived
from the ferrocenyl derivatives [13] 3 (X = PR2) ligating to pal-
ladium with a proper bite angle exhibits excellent performance
on the palladium-catalyzed cross-coupling for C–N formation
[14]. We envisioned that compound 4 could be a good precur-
sor for the NHC carbene bidentate ligand. The readily available
∗
Corresponding author. Tel.: +86 591 8346 5225.
E-mail address: jchshi@fjnu.edu.cn (J.-C. Shi).
1381-1169/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2006.04.025

8
J.-C. Shi et al. / Journal of Molecular Catalysis A: Chemical 259 (2006) 7–10
Table 1
Ligand effect in Suzuki–Miyaura cross-coupling of 4-bromotoluene with
phenylboronic acida
Entry
Ligand
Time (h)
Conversionb (%)
Yieldc (%)
1
2
3
4
5
6
7
8
4a
4b
4c
4d
4e
4f
6
6
1
1.5
1.5
3
6
1
62
72
>99
>99
>99
90
/
/
96
94
97
85
/
Scheme 1. Phosphine-functionalized imidazolium salts.
4g
5
56
>99
racemic ferrocenyl derivative 3 can be easily converted to imi-
dazolium salt 4 in high yield [11d]. Compound 4, a precursor of
carbene, provided us a good module for the development of a
stable and highly active catalyst system (Scheme 1). The scaf-
fold of imidazolium 3 should result in a type of hybrid ligands
4c–h with a proper bite angle of the chelating carbene-phosphine
to palladium [15]. In addition, the degree of steric hindrance of
the synthesized imidazolium salts can be finely tuned through
the size of the substituents on the imidazoles used. For confirm-
ing the chelating effect, the imidazolium salts without a side
arm (4a and 4b) and with non-coordinated side arm were also
synthesized (4g).
First, we tested the newly synthesized imidazolium salts as a
ligand with Pd(OAc)2 for the cross-coupling reaction between
4-bromotoluene and phenylboronic acid using dioxane as sol-
vent, Cs2CO3 as base, and at 80 ◦C [16]. Using imidazolium
salts 4a and 4b without a phosphine side arm, the coupling
reaction produced, after 3 h, the desired product in 62–72% of
yields as determined by GC (Table 1, entries 1 and 2). No further
conversion of 4-bromotoluene with extension of reaction time
and appearance of palladium black indicated the decomposi-
tion of the catalyst supported by unidentate imidazol-2-ylidene
derived from 4a and 4b [17]. Employing a reaction protocol
based the reported conditions [16], in which the Pd/L ratio is
1:2, did not increase the conversion significantly. To our delight,
with phosphine side arm, the imidazolium salts 4c, 4d, and 4e
not only significantly increased the stability of the catalyst but
also enhanced the rate of the reaction, resulting in complete
conversions of 4-bromotoluene with 94–97% of isolated yields
(Table 1, entries 3–5). It should be mentioned that the size of the
substituent introduced from the imidazole also plays a steric
role in the catalytic performance to some extent. For exam-
ple, with 2,6-diisopropylphenyl substitute (4f), the conversion
95
a
Reaction conditions: 1.0 mmol of 4-bromotoluene, 1.5 mmol of phenyl-
boronic acid, 2.0 mmol of Cs2CO3, 1.0 mol% Pd(OAc)2, 1.0 mol% ligand, 2 mL
of dioxane, 80 ◦C.
b
GC yield.
Isolated yield.
c
of 4-bromotoluene was not completed even after 3 h. As the
phosphine was oxidized to phosphine oxide, the performance of
the imidazolium salt 4g was similar to that of the imidazolium
salts 4a and 4b without a side arm, implying the contribution
of chelating effect to the stability of the catalyst system. For
comparison, the palladium complex 5 (Scheme 2) was also syn-
thesized and tested for the reaction; its catalytic activity (Table 1,
entry 8) is comparable to that formed in situ. Since two steps
were needed to prepare the complex 5, most of catalytic studies
were carried out using the catalysts formed in situ. It should be
noted that, with the prepared catalyst or once the catalyst formed
in situ, the coupling reaction can be carried out under air and
no visible changes of the performance of the palladium catalyst
were observed.
A brief study of the influence of the base indicated that a
range of bases are suitable for the catalyst system. In dioxane,
potassium carbonate, sodium hydroxide, sodium tert-butoxide
are as effective as cesium carbonate (Table 2, entries 1, 4, 5
and 6), although 1–3 h of longer reaction time was needed to
completely consume 4-bromotoluene. In turn, KF, a very effec-
tive additive for N-heterocyclic carbenes, proved to be less
effective in our case, supporting 70% of conversion (Table 2,
entry 9). Other bases such as lithium carbonate, sodium car-
bonate, sodium acetate, potassium acetate, and potassium phos-
phate proved to be inefficient for coupling 4-bromotoluene with
phenylboronic acid. Various solvents, such as THF, DME, and
Scheme 2. Synthesis of the palladium complexes 5.

J.-C. Shi et al. / Journal of Molecular Catalysis A: Chemical 259 (2006) 7–10
9
Table 2
Table 3
Effect of base on Pa(OAc)2/4c-catalyzed cross-coupling of 4-bromotoluene with
Pa(OAc)2/4c-catalyzed cross-coupling of aryl halides with phenylboronic acida
phenylboronic acida
Entry
Aryl halide
Time
(h)
Conversion
(%)b
Yield
(%)c
Entry
Base
Time (h)
Conversionb (%)
Yieldc (%)
1
2
3
4
5
6
7
8
9
10
11d
12d
13d
Bromobenzene (6a)
2
2
2
6
2
2
2
2
2
12
24
24
24
>99
>99
>99
>99
>99
>99
90
>99
>99
92
94
96
93
95
93
97
85
95
94
85
81
76
77
1
2
3
4
5
6
7
8
9
Cs2CO3
Li2CO3
Na2CO3
K2CO3
NaOH
1
20
24
2
3
2
20
20
4
20
24
24
>99
<2
28
>99
98
97
/
/
96
93
95
/
/
/
/
/
/
4-Bromotoluene (6b)
3-Bromotoluene (6c)
2-Bromotoluene (6d)
4-Bromobenzotrifuoride (6e)
4-Bromoacetophenone (6f)
4-Bromo-m-xylene (6g)
2-Bromobenzonitrile (6h)
4-Bromoanisole (6i)
2-Bromopyridine (6j)
4-Chlorobenzotrifuoride
4-Chloroacetophenone
2-Chlorobenzonitrile
NaOBut
NaOAc
KOAc
98
<2
<5
70
11
<2
<2
KF
K3PO4
Et3N
10
11
12
90
83
85
Hunig’s base
a
Reaction conditions: 1.0 mmol of 4-bromotoluene, 1.5 mmol of phenyl-
boronic acid, 2.0 mmol of base, 1.0 mol% Pd(OAc)2, 1.0 mol% ligand, 2 mL
of dioxane, 80 ◦C.
a
Reaction conditions: 1.0 mmol of aryl halide, 1.5 mmol of phenylboronic
acid, 2.0 mmol of K2CO3, 0.5 mol% Pd(OAc)2, 0.5 mol% ligand 4c, 2 mL of
dioxane, 80 ◦C.
b
GC yield.
Isolated yield.
b
Determined by GC.
c
c
Yields for isolated products of >95% purity judged by 1H NMR.
d
2.0 mol% of catalyst, 2 equiv. of Cs2CO3, 110 ◦C.
to some extent toluene, are also suitable for the catalyst system.
Pd(CH3CN)2Cl2 [(C3H5)PdCl]2 (C3H5 = allyl) were all good
alternative Pd sources for Pd(OAc)2. However, no reaction was
observed with Pd2(dba)3.
As shown in Table 3, the procedure employing the 0.5 mol%
of Pd(OAc)2/4c system is extremely effective in coupling of a
wide range of aryl bromides with phenylboronic acid. Sterically
congested substrates also led to excellent yields (entries 4, 7, 8
and 10). As expected, nitrile and keto functional groups were
found to be compatible under these reaction conditions (entries
5, 6 and 8). In fact, good yields could be obtained from the deac-
tivated 4-bromoanisole as substrate (entry 9) [18]. Furthermore,
activated aryl chlorides can also couple with phenylboronic
acid in high yields although 2 mol% of the catalyst system was
required. In all cases examined, only traces of homocoupling
and dehalogenation product were found (<2%).
The efficiency of the Pd(OAc)2/4c catalyst system in the
Suzuki–Miyaura reaction was further evaluated and the results
are summarized in Table 4. For the coupling of bromoben-
zene (6a), 4-bromotoluene (6b), and 3-bromotoluene (6c) with
phenylboronic acid in dioxane, the catalyst could be decreased
Table 4
Screening the catalytic efficiency of the Pd(OAc)2/4c catalyst system for the Suzuki–Miyaura reactiona
Entry
ArBr
Pd (mol%)
Time (h)
Conversion (%)
Yield (%)b
Tonnes
1
2c
3
4
5
6
7
8
6a
6a
6b
6c
6d
6g
6i
0.005
0.005
0.005
0.005
0.025
0.025
0.05
24
2
>99
>99
>99
>99
>99
>99
92
99
99
99
99
99
99
90
98
20000
20000
20000
20000
4000
4000
1840
3960
24
24
16
16
24
24
6j
0.025
99
a
Reaction conditions: aryl bromide (1.0 equiv.), phenylboronic acid (1.5 equiv.), K2CO3 (2.0 equiv.), Pd(OAc)2 (0.005 mmol), 4c (0.005 mmol), dioxane/aryl
bromides (2 mL/mmol), 110 ◦C.
b
Yields for isolated products of >95% purity judged by 1H NMR.
In 95% aqueous ethanol, at refluxing.
c

10
J.-C. Shi et al. / Journal of Molecular Catalysis A: Chemical 259 (2006) 7–10
to 0.005 mol% and almost quantitative yields could be obtained
at 110 ◦C (Table 4, entries 1, 2 and 3), corresponding to
20,000 TONs. However, the efficiencies of the Pd(OAc)2/4c cat-
alyst system decreased to some extent for the substrates with
steric hindrance. For example, 0.025 mol% of the catalyst was
required to reach 100% of conversion for the substrates with one
ortho-methyl substituent (Table 4, entries 5 and 6). High TONs
could also be achieved for the deactivated 4-bromoanisole with
92% of conversion (Table 4, entry 7) and for the heterocyclic
substrate (Table 4, entry 8).
Gratifyied by the low loading of the Pd(OAc)2/4c catalyst,
we tested the coupling of 4-bromotoluene with phenylboronic
acid in 95% aqueous ethanol, which is a cheaper and envi-
ronmentally benign solvent for industry, particularly for the
pharmaceutical industry. Thus, with a 0.005 mol% loading of
Pd(OAc)2/4c, a complete conversion of 4-bromotoluene with
99% yield was realized within 2 h in refluxing in 95% aqueous
ethanol (Table 4, entry 2). The data indicated that the TOFs is
as high as ∼10,000 h−1 (average) in the condition of complete
conversion.
In summary, we have developed a novel and robust palladium
catalyst system supported by the hybrid ligand N-heterocyclic
carbene along with a phosphine derived from ferrocene. In
the course of the study, we have applied successfully the
chelating effect, the bite angle, and the steric hindrance to
the design of the novel bitendate ligand. This type of biden-
tate ligand is not only very active but also very stable under
the reaction conditions with TONs up to 20,000 and TOFs to
10,000 h−1. A wide range of bases and solvents are suitable
for the catalyst system. We have found that the chelating effect
is a promising tool for developing highly stable palladium/N-
heterocyclic carbene catalysts for the Suzuki–Miyaura cross-
coupling, although the activity of the present catalyst system
toward arylchlorides is still low. Investigation of labile coordi-
nating atoms cis to N-heterocyclic carbene, aiming at increasing
the activity toward arylchlorides and retaining the stability, is
underway.
Acknowledgments
This work was financially supported by the Knowledge Inno-
vation Program of the Chinese Academy of Sciences (DICP
R200309), the National Basic Research Program of China
(2003CB61615803) and the National Natural Science Founda-
tion of China (20473089).
References
[1] (a) N. Miyaura, Top. Curr. Chem. 219 (2002) 11;
(b) A. Suzuki, J. Organomet. Chem. 576 (1999) 147.
[2] (a) J.J. Roncali, Chem. Rev. 92 (1992) 711;
(b) K. Yamamura, S. Ono, H. Ogoshi, H. Masuda, Y. Kuroda, Synlett.
(1989) 18.
[3] K.G.B. Torssell, Natural Product Chemistry, Wiley, Chichester, 1983.
[4] (a) A.F. Littke, G.C. Fu, Angew. Chem. Int. Ed. 41 (2002) 4176;
(b) R.B. Bedford, G.S.J. Cazin, D. Holder, Coord. Chem. Rev. 248 (2004)
2282.
[5] (a)S.D. Walker, T.E. Barder, J.R. Martinelli, S.L. Buchwald, Angew. Chem.
Int. Ed. 43 (1871);
(b) A.F. Littke, G.C. Fu, Angew. Chem. Int. Ed. 37 (1998) 1433;
(c) J.F. Jensen, M. Johannsen, Org. Lett. 5 (2003) 3025;
(d) C. Baillie, L. Zhang, J. Xiao, J. Org. Chem. 69 (2004) 7779.
[6] (a) W.A. Herrmann, C. Ko¨cher, Angew. Chem. Int. Ed. 36 (1997) 2162;
(b) D. Bourissou, O. Guerret, F.P. Gabbai, G. Bertrand, Chem. Rev. 100
(2000) 39;
(c) L. Jafarpour, S.P. Nolan, Adv. Organomet. Chem. 557 (1998) 93;
(d) S.B. Park, H. Alper, Org. Lett. 5 (2003) 3209.
[7] (a) W.A. Herrmann, C.-P. Reisinger, M. Spiegler, J. Organomet. Chem. 557
(1998) 93;
(b) C. Zhang, J. Huang, M.L. Trudell, S.P. Nolan, J. Org. Chem. 64 (1999)
3804.
[8] (a) C.W.K. Gsto¨ttmayr, V.P.W. Bo¨hm, E. Herdtweck, M. Grosche, W.A.
Herrmann, Angew. Chem. Int. Ed. 41 (2002) 1363;
(b) O. Navarro, R.A. Kelly III, S.P. Nolan, J. Am. Chem. Soc. 125 (2003)
16194.
[9] G. Altenhoff, R. Goddard, C.W. Lehmann, F. Glorius, J. Am. Chem. Soc.
126 (2004) 15195.
[10] M.-D. Su, S.-Y. Chu, Inorg. Chem. 37 (1988) 3400.
[11] (a) W.A. Herrmann, C. Ko¨cher, L.J. Goo␤en, G.R.J. Artus, Chem. Eur. J.
2 (1996) 1627;
(b) A.A. Danopoulos, S. Winston, T. Gelbrich, M.B. Hursthouse, R.P.
Tooze, Chem. Commun. (2002) 482;
(c) S. Gischig, A. Togni, Organometallics 23 (2004) 2479;
(d) H. Seo, B.Y. Kim, J.H. Lee, H.-J. Park, S.U. Son, Y.K. Chung,
Organometallics 22 (2003) 4783;
1. Experimental
(e) C. Yang, H.M. Lee, S.P. Nolan, Org. Lett. 3 (2001) 1511;
(f) A.-E. Wang, J. Zhong, J.-H. Xie, K. Li, Q.-L. Zhou, Adv. Synth. Catal.
346 (2004) 595.
1.1. Suzuki–Miyaura cross-coupling of aryl halides with
phenylboronic acid: general procedure
[12] M. Kranenburg, P.C.J. Kamer, P.W.N.M. von Leeuwen, Eur. J. Inorg. Chem.
(1998) 155.
In air, imidazolium salt (0.005 mmol), Pd(OAc)2 (0.005
mmol) and K2CO3 (2.0 mmol) were added to an oven-dried
Schlenk tube equipped with a stir bar. The Schlenk tube was
fitted with a rubber septum, evacuated, and then refilled with
nitrogen. 1,4-Dioxane (1.0 mL) was added via syringe. The
mixture was stirred for 30 min at room temperature. Then aryl
halide (1.0 mmol), phenylboronic acid (1.5 mmol), undecane
(0.5 mmol) and again 1,4-dioxane (1.0 mL) were added. The
tube was sealed and immersed in oil bath at 80 ◦C and the course
of the reaction was monitored by GC. After being quenched, and
thereaction mixture was dilutedwithether, filteredthroughapad
of silica gel with copious washings, concentrated, and purified
by column chromatography.
[13] (a) T. Hayashi, T. Mise, M. Fukushima, M. Kagotani, N. Nagashima, Y.
Hamada, A. Matsumoto, S. Kawakami, M. Konishi, K. Yamamoto, M.
Kumada, Bull. Chem. Soc. Jpn. 53 (1980) 1138;
(b) H. Seo, H.-J. Park, B.Y. Kim, J.H. Lee, S.U. Son, Y.K. Chung,
Organometallics 22 (2003) 618.
[14] B.C. Hamann, J.F. Hartwig, J. Am. Chem. Soc. 120 (1998) 7369.
[15] The bite angle of the chelating carbene-phosphine to rhodium is 97.2◦ (see
Ref. [9d]).
[16] G.A. Grasa, M.S. Viciu, J. Huang, C. Zhang, M.L. Trudell, S.P. Nolan,
Organometallics 21 (2002) 2866.
[17] H. Yu, J. Shi, H. Zhang, P. Yang, X. Wang, Z. Jin, J. Mol. Catal. A: Chem.
250 (2006) 15.
[18] (a) H.M. Lee, J.Y. Zeng, C.-H. Hu, M.-T. Lee, Inorg. Chem. 43 (2004)
6822;
(b) H.M. Lee, P.L. Chiu, J.Y. Zao, Inorg. Chim. Acta 357 (2004) 4313.