Pd(quinoline-8-carboxylate)2 as a low-priced, phosphine-free catalyst for heck and suzuki reactions

Article abstract of DOI:10.1021/jo701783k

(Chemical Equation Presented) N,O-Bidentate compounds were systematically evaluated as phosphine-free ligands for Pd-catalyzed C-C bond-formation reactions through kinetic measurements. Pd(quinoline-8-carboxylate)₂ was identified as one of the most efficient, yet still low-priced, phosphine-free catalysts for Heck as well as Suzuki reactions of unactivated aryl bromides with high turnover numbers up to ca. 10,000.

Full text of DOI:10.1021/jo701783k

ligands (e.g., NHC) are less expensive than phosphines. On the
other hand, Yao et al. reported in 200317 that Pd(OAc)2, albeit
being very simple, could also efficiently catalyze the Heck
reaction where the acetate anion was proposed to function as
ligand. This interesting finding was consistent with the theory
of Amatore and Jutand18 who claimed that coordination by
anionic carboxylates could enhance the activity of Pd toward
oxidative addition.
Prompted by Yao’s pioneering discovery, we hypothesized
that it was possible to design phosphine-free Pd catalysts more
efficient, yet still low-priced, than Pd(OAc)2 by strategically
incorporating an additional coordinating site into the acetate
anion (Scheme 1). This hypothesis was partly validated by our
Pd(quinoline-8-carboxylate)2 as a Low-Priced,
Phosphine-Free Catalyst for Heck and Suzuki
Reactions
Xin Cui,†,‡ Juan Li,†,‡ Zhi-Ping Zhang,‡ Yao Fu,‡
Lei Liu,*,†,‡ and Qing-Xiang Guo*,‡
Department of Chemistry, Key Laboratory of Bioorganic
Phosphorus Chemistry and Chemical Biology, Tsinghua
UniVersity, Beijing 100084, China, Department of Chemistry,
UniVersity of Science and Technology of China,
Hefei 230026, China
lliu@mail.tsinghua.edu.cn; qxguo@ustc.edu.cn
SCHEME 1. Improved Design of Phosphine-Free Ligands
for Pd Catalysis by Incorporating an Additional
Coordinating Atom
ReceiVed August 15, 2007
N,O-Bidentate compounds were systematically evaluated as
phosphine-free ligands for Pd-catalyzed C-C bond-formation
reactions through kinetic measurements. Pd(quinoline-8-
carboxylate)2 was identified as one of the most efficient, yet
still low-priced, phosphine-free catalysts for Heck as well
as Suzuki reactions of unactivated aryl bromides with high
turnover numbers up to ca. 10,000.
recent study on the use of amino acids as ligands for Pd-
catalyzed Heck reactions19 and an earlier related study by Reetz
(5) (a) Tao, B.; Boykin, D. W. J. Org. Chem. 2004, 69, 4330. (b) Li,
J.-H.; Liu, W.-J. Org. Lett. 2004, 6, 2809. (c) Li, J.-H.; Liu, W.-J.; Xie,
Y.-X. J. Org. Chem. 2005, 70, 5409. (d) Li, J.-H.; Liang, Y.; Wang, D.-P.;
Liu, W.-J.; Xie, Y.-X.; Yin, D.-L. J. Org. Chem. 2005, 70, 2832. (e) Xie,
Y.-X.; Li, J.-H.; Yin, D.-L. Chin. J. Org. Chem. 2006, 26, 1155.
(6) (a) Grasa, G. A.; Hillier, A. C.; Nolan, S. P. Org. Lett. 2001, 3, 1077.
(b) Wu, K.-M.; Huang, C.-A.; Peng, K.-F.; Chen, C.-T. Tetrahedron 2005,
61, 9679.
(7) (a) Buchmeiser, M. R.; Wurst, K. J. Am. Chem. Soc. 1999, 121,
11101. (b) Kawano, T.; Shinomaru, T.; Ueda, I. Org. Lett. 2002, 4, 2545.
(c) Najera, C.; Gil-Moito, J.; Karlstrum, S.; Falvello, L. R. Org. Lett. 2003,
5, 1451.
(8) (a) Mino, T.; Shirae, Y.; Sakamoto, M.; Fujita, T. Synlett 2003, 882.
(b) Mino, T.; Shirae, Y.; Sakamoto, M.; Fujita, T. J. Org. Chem. 2005, 70,
2191. (c) Mino, T.; Shirae, Y.; Sasai, Y.; Sakamoto, M.; Fujita, T. J. Org.
Chem. 2006, 71, 6834.
(9) Li, S. H.; Xie, H. B.; Zhang, S. B.; Lin, Y. J.; Xu, J. N.; Cao, J. G.
Synlett 2005, 1885.
Pd-catalyzed C-C bond-formation reactions such as the Heck
and Suzuki cross couplings are playing important roles in
modern organic synthesis.1 These cross-coupling reactions are
usually performed with 1-5 mol % of Pd catalyst along with
phosphine ligands, which sometimes creates practical problems
because organophosphines tend to be expensive, poisonous, and
air sensitive.2 Accordingly, an interesting current challenge is
to develop Pd catalysts that can utilize inexpensive phosphine-
free ligands. In this regard a number of ligands including
N-heterocyclic carbenes (NHC),3 oxazolines,4 amines,5 Schiff
bases,6 pyridines,7 hydrazones,8 guanidines,9 pyrazoles,10 tet-
razoles,11 quinolines,12 carbazones,13 imidazoles,14 thioureas,15
and 1,3-dicarbonyl compounds16 have been examined for Pd
catalysis recently. Note that not all of these phosphine-free
(10) Mukherjee, A.; Sarkar, A. Tetrahedron Lett. 2005, 46, 15.
(11) Gupta, A. K.; Song, C. H.; Oh, C. H. Tetrahedron Lett. 2004, 45,
4113.
(12) Iyer, S.; Kulkarni, G. M.; Ramesh, C. Tetrahedron 2004, 60, 2163.
(13) Kovala-Demertzi, D.; Yadav, P. N.; Demertzis, M. A.; Jasiski, J.
P.; Andreadaki, F. J.; Kostas, I. D. Tetrahedron Lett. 2004, 45, 2923.
(14) (a) Park, S. B.; Alper, H. Org. Lett. 2003, 5, 3209. (b) Xiao, J. C.;
Twamley, B.; Shreeve, J. M. Org. Lett. 2004, 6, 3845.
(15) (a) Dai, M. J.; Liang, B.; Wang, C. H.; You, Z. J.; Xiang, J.; Dong,
G. B.; Chen, J. H.; Yang, Z. AdV. Synth. Catal. 2004, 346, 1669. (b) Yang,
D.; Chen, Y. C.; Zhu, N. Y. Org. Lett. 2004, 6, 1577. (c) Chen, W.; Li, R.;
Han, B.; Li, B. J.; Chen, Y. C.; Wu, Y.; Ding L. S.; Yang, D. Eur. J. Org.
Chem. 2006, 1177.
† Tsinghua University.
‡ University of Science and Technology of China.
(1) Cacchi, S.; Fabrizi, G. Chem. ReV. 2005, 105, 2873.
(2) (a) Farina, V. AdV. Synth. Catal. 2004, 346, 1553. (b) Phan, N. T.
S.; Van Der Sluys, M.; Jones, C. W. AdV. Synth. Catal. 2006, 348, 609.
(3) (a) Gstottmayr, C. W. K.; Bohm, V. P. W.; Herdtweck, E.; Grosche,
M.; Herrmann, W. Angew. Chem., Int. Ed. 2002, 41, 1363. (b) Navarro,
O.; Kelly, R. A., III; Nolan, S. P. J. Am. Chem. Soc. 2003, 125, 16194.
(4) (a) Tao, B.; Boykin, D. W. Tetrahedron Lett. 2002, 43, 4955. (b)
Gossage, P. A.; Jenkins, H. A.; Yadav, P. N. Tetrahedron Lett. 2004, 45,
7689. (c) Lee, S. J. Organometal. Chem. 2006, 691, 1347.
(16) Cui, X.; Li, J.; Liu, L.; Guo, Q. X. Chin. Chem. Lett. 2007, 18,
625.
10.1021/jo701783k CCC: $37.00 © 2007 American Chemical Society
Published on Web 10/31/2007
9342
J. Org. Chem. 2007, 72, 9342-9345

FIGURE 1. Initial velocity of the Heck reaction between PhBr and styrene. Conditions: PhBr (5 mmol), styrene (7.5 mmol), K2CO3 (10 mmol),
DMF (5 mL), Pd(OAc)2 ) 0.005 mol, Pd(OAc)2:ligand ) 1:1 or 1:2, temperature ) 130 °C, under Ar. The values in parentheses are the GC yields
measured after each reaction is stirred for 12 h.
et al.20 Here we further extended our search for more efficient
phosphine-free ligands for Pd catalysis by systematically
evaluating various N,O-bidentate compounds. Through detailed
kinetic measurements we demonstrated the significant effect of
incorporating an additional coordinating site into the acetate.
We also found that Pd(quinoline-8-carboxylate)2 constituted one
of the most efficient, yet low-priced, phosphine-free catalysts
for the Heck and Suzuki reactions with fairly high turnover
numbers up to ca. 10,000 for unactivated aryl bromides.
In designing our experiments we fixed the solvent as DMF.21
By using Pd(OAc)2 in combination with 1 or 2 equiv of an N,O-
bidentate ligand, we monitored the progress of the Heck reaction
between phenyl bromide and styrene by measuring the GC
yields at different intervals after the reaction was initiated
(Figure 1). It was found that in the absence of any other ligand,
Pd(OAc)2 indeed could catalyze the Heck reaction, but the
maximum yield was 61% (compared to 74% as reported by Yao
et al.17 under similar conditions). Adding ligand 1 increased
the yield to 76%, whereas adding ligand 2 further increased
the yield to 92%. Evidently the attachment of an amino group
to acetate could produce a significant effect on the catalysis.
As aforementioned,18 coordination to Pd by anionic carboxy-
lates could facilitate oxidative addition. In agreement with this
theory, our experiments showed that adding 3 or 4 dramatically
decreased the yields where the carboxylate oxygen was replaced
by an amide or alcohol oxygen. Furthermore, it was found that
the nature of the nitrogen atom also had a strong effect on the
catalysis. Thus, ligand 5 gave a yield of 78% that was
comparable to that of ligand 1. Ligands 6-9 gave yields that
were much lower than the yield provided by Pd(OAc)2. Ligand
10, on the other hand, gave a yield (63%) that was slightly better
than that of Pd(OAc)2. These data indicated that alkylamine and
pyridine were good neighboring groups to enhance the effect
of a carboxylate group, whereas arylamine, methoxyamine, and
amidine were not.
Given the fact that 2 was better than 1, we next added a
methylene group into 10, ending up with ligand 13. It was found
that 13 gave a GC yield of 95% that was higher than the yield
provided by ligand 2. Encouraged by this finding, we then
incorporated a phenyl ring into 13 and obtained ligand 15. To
our delight, this particular ligand was found to give the highest
GC yield (i.e., 96%) among all the N,O-bidentate ligands that
we had examined. Noteworthy control experiments with ligands
(17) Yao, Q.; Kinney, E. P.; Yang, Z. J. Org. Chem. 2003, 68, 7528.
(18) Amatore, C.; Jutand, A. Acc. Chem. Res. 2000, 33, 314.
(19) Cui, X.; Li, Z.; Tao, C.-Z.; Xu, Y.; Li, J.; Liu, L.; Guo, Q.-X. Org.
Lett. 2006, 8, 2467.
(20) Reetz, M. T.; Westermann, E.; Lohmer, R.; Lohmer, G. Tetrahedron
Lett. 1998, 39, 8449.
(21) DMA was used in Yao’s study (ref 17), but DMF is more user-
friendly than DMA in terms of solvent removal.
J. Org. Chem, Vol. 72, No. 24, 2007 9343

a
TABLE 1. Heck-Type Reactions Catalyzed by Pd(quinoline-8-carboxylate)2
yieldb
(%)
yieldb
(%)
entry
ArX
R
entry
ArX
R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
PhBr
PhBr
Ph
94
91
97
35
91
88
78
73
93
82
92
39
90
92
15
16
17
18
19
20
21
22
23
24
25
26
27
4-NC-C6H4Br
4-NC-C6H4Br
4-Ph-C6H4Br
4-Ph-C6H4Br
3-F3C-C6H4Br
3-F3C-C6H4Br
3-bromopyridine
3-bromopyridine
PhI
Ph
87
71
71
74
98
92
96
98
92
87
73
0
CO2Bun
Ph
CO2Bun
Ph
4-O2N-C6H4Br
4-O2N-C6H4Br
4-Me-C6H4Br
4-Me-C6H4Br
4-MeO-C6H4Br
4-MeO-C6H4Br
4-OHC-C6H4Br
4-OHC-C6H4Br
4-MeOC-C6H4Br
4-MeOC-C6H4Br
3-NC-C6H4Br
3-NC-C6H4Br
CO2Bun
Ph
CO2Bun
Ph
CO2Bun
Ph
CO2Bun
Ph
CO2Bun
Ph
CO2Bun
Ph
CO2Bun
Ph
PhI
CO2Bun
Ph
4-MeO-C6H4I
PhCl
4-NO2-C6H4Cl
CO2Bun
Ph
Ph
Ph
35
CO2Bun
a Reaction conditions: Aryl halide (5 mmol), olefin (7.5 mmol), K2CO3 (10 mmol), DMF (10 mL), catalyst ) 18, under Ar. b Isolated yield.
11, 12, and 14 demonstrated that the presence of a carboxylate
group was crucial for the catalysis efficiency.
that 18 constituted a simple, yet efficient, phosphine-free catalyst
for many Heck-type reactions of aryl bromides. Both the
electron-rich (deactivated) and electron-poor (activated) aryl
bromides could be efficiently converted to the desirable products
in high yields. The turnover numbers in most of the cases were
about 104. Besides aryl bromides, aryl iodides could also be
successfully used under the same reaction conditions (entries
23-25). Nonetheless, 18 was not active enough to handle an
aryl chloride unless it was activated by electron-withdrawing
substituents (such as 4-NO2).
The optimal ligand:Pd ratio was found to be 2:1 for ligands
2, 13, and 15. As a result we synthesized the Pd(II) salts of 2,
13, and 15 by mixing 2 equiv of sodium N,N-dimethyl â-alanate,
2-pyridylacetate, or quinoline-8-carboxylate with 1 equiv of K2-
PdCl4 in water, which provided the target Pd complexes 16,
17, and 18 as solid precipitates. Using the pure Pd salts as
catalyst we next lowered the Pd loading to 0.01 mol % in the
cross coupling between phenyl bromide and styrene. It was
found that the maximum turnover numbers for catalysts 16, 17,
and 18 were 1600, 3800, and 9400, respectively (Figure 2). The
turnover number of catalyst 18 is about one magnitude lower
than the best phosphine-free Pd catalysts currently available
(e.g., Pd pincers and Pd NHC complexes).22 Nonetheless, the
turnover number of 18 (ca. 104) is sufficiently high from
practical considerations.
TABLE 2. Suzuki Reactions Catalyzed by
a
Pd(quinoline-8-carboxylate)2
catalyst (0.01 mol%)
ArX + R-B(OH)2 K PO , EtOH/H O, 50 °C, 5 h8 Ar-R
3
4
2
yieldb
(%)
entry
ArX
R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
PhBr
4-Me-C6H4-
96
98
93
95
95
93
89
91
96
99
97
88
94
4-Me-C6H4Br
4-MeO-C6H4Br
PhBr
4-MeCO-C6H4Br
4-MeCO-C6H4Br
4-O2N-C6H4Br
4-O2N-C6H4Br
4-NC-C6H4Br
4-MeCO-C6H4I
PhI
4-MeO-C6H4I
PhI
4-Me-PhCl
4-O2N-C6H4Cl
Ph-
Ph-
4-MeO-C6H4-
Ph-
4-MeO-C6H4-
4-MeO-C6H4-
Ph-
Ph-
Ph-
4-Me-C6H4-
Ph-
4-MeO-C6H4-
Ph-
Ph-
0
Trace
a Conditions: aryl halide (5 mmol), boronic acid (7.5 mmol), K3PO4
(10 mmol), EtOH/H2O (7.5/7.5 mL), catalyst ) 18, in air. b Isolated yield.
Having found that 18 was an excellent catalyst for the Heck
reactions, we next examined whether 18 could also facilitate
the Suzuki reactions. To our satisfaction, it was found that 18
was active enough to promote the Suzuki reactions under
relatively mild conditions in air and in the presence of water
FIGURE 2. Maximum turnover numbers given by the best three N,O-
bidentate ligands.
To explore the scope of the reactions catalyzed by 18, we
next examined the application of 18 to the cross coupling of a
variety of aryl halides with styrene and n-butyl acrylate under
0.01 mol % catalyst loading (See Table 1). The results indicated
(22) (a) Weck, M.; Jones, C. W. Inorg. Chem. 2007, 46, 1865. (b) Yao,
Q.; Zabawa, M.; Woo, J.; Zheng, C. J. Am. Chem. Soc. 2007, 129, 3088.
(c) Dupont, J.; Consorti, C. S.; Spencer, J. Chem. ReV. 2005, 105, 2527.
9344 J. Org. Chem., Vol. 72, No. 24, 2007

precipitate was filtered and dried under vacuum to afford a yellow
solid which could not dissolve in the common organic solvents
(69%). Elemental analysis calcd for C20H12N2O4Pd: C 53.33; H
2.67; N 6.22 (O 14.22, Pd 23.56). Found: C 52.93; H 2.67; N
6.16.
(Table 2). Both electron-rich and electron-poor aryl bromides
could be successfully converted to the desirable products very
rapidly (ca. 5 h). The turnover numbers were as high as ca.
104. Furthermore, aryl iodides were excellent substrates for the
reaction (entries 10-13), although aryl chlorides could not be
converted effectively (entries 14-15). It is worth noting that
phosphine-free Suzuki reactions were achieved previously either
by using expensive Pd pincers and Pd NHC complexes,23 or by
using low-priced ligands but with a relatively high load of Pd
(i.e., >1 mol %).24 In comparison, the new phosphine-free
catalyst (i.e., 18) described here is advantageous from the
economic point of view.
To summarize, in the present study we systematically tested
the hypothesis that incorporating an additional coordination site
to the acetate anion would provide more efficient yet low-priced
ligands for Pd catalyzed C-C bond-formation reactions than
Pd(OAc)2. It was found that alkylamine and pyridine were good
neighboring groups to enhance the effect of a carboxylate group
in Pd catalysis, whereas arylamine, methoxyamine, and amidine
were not. The best N,O-bidentate ligand turned out to be
quinoline-8-carboxylate, and Pd(quinoline-8-carboxylate)2 con-
stituted an ideal Pd catalyst25 that was structurally simple,
synthetically inexpensive, and chemically stable. Synthetic
examinations demonstrated that Pd(quinoline-8-carboxylate)2
could smoothly catalyze the Heck and Suzuki reactions of
unactivated aryl bromides and iodides with fairly high turnover
numbers up to ca. 104.
Typical Experimental Procedure for the Heck Reaction. A
mixture of phenyl bromide (5 mmol), styrene (7.5 mmol), 18 (0.5
× 10-3 mmol), and K2CO3 (10 mmol) in 10 mL of dry DMF was
stirred under Ar at 130 °C for the desired time. The mixture was
washed by water and extracted by ether, and then the solvent was
evaporated. The residue was purified by flash column chromatog-
1
raphy to afford the desired product. (E)-1,2-Diphenylethene: H
NMR (400 MHz, CDCl3): δ 7.11 (s, 2H), 7.24-7.28 (m, 2H), 7.35
(t, 4H, J ) 7.5 Hz), 7.52 (d, 4H, J ) 7.7 Hz).26
Typical Experimental Procedure for the Suzuki Reaction. A
mixture of p-methylphenyl bromide (5 mmol), phenyl boronic acid
(7.5mmol), 18 (0.5 × 10-3 mmol), and K3PO4 (10 mmol) in
methanol/H2O (7.5 mL/7.5 mL) was stirred at 50 °C for the desired
time. The mixture was washed by water and extracted by ether,
and then the solvent was evaporated. The residue was purified by
flash column chromatography to afford the desired product.
4-Methylbiphenyl: 1H NMR (400 MHz, CDCl3): δ 2.39 (s, 3H),
7.25 (d, 2H, J ) 8.0 Hz), 7.33 (m, 1H), 7.42 (t, 2H, J ) 7.6 Hz);
7.49 (d, 2H, J ) 8.2 Hz), 7.57 (m, 2H).27
Acknowledgment. We thank NSFC (No. 20472079) for the
financial support.
Supporting Information Available: Experimental procedures,
analytical data, and scanned spectra. This material is available free
of charge via the Internet at http://pubs.acs.org.
Experimental
JO701783K
Synthesis of Pd(quinoline-8-carboxylate)2. A mixture of
2 equiv of sodium quinoline-8-carboxylate and 1 equiv of
K2PdCl4 was stirred in water at room temperature for 10 min. The
(25) Strictly speaking, Pd(quinoline-8-carboxylate)2 was a precatalyst
because Pd(II) must be reduced to Pd(0) to catalyze the Heck and Suzuki
reactions. For a mechanistic study, see: Cardenas, D. J.; Martin-Matute,
B.; Echavarren, A. M. J. Am. Chem. Soc. 2006, 128, 5033.
(26) Yang, Y.-F.; Zhuang, M.; Zeng, C.-X.; Huang, C.-B.; Luo, M.-F.
Chin. J. Chem. 2006, 24, 1309.
(23) A very recent example: Navarro, O.; Marion, N.; Mei, J.; Nolan,
S. P. Chem. Eur. J. 2006, 12, 5142.
(24) Some examples: (a) Zhang, G. Synthesis 2005, 537. (b) Najera,
C.; Gil-Molto, J.; Karlstrom, S. AdV. Synth. Catal. 2004, 346, 1798.
(27) Wang, L.; Li, P.-H. Chin. J. Chem. 2006, 24, 770.
J. Org. Chem, Vol. 72, No. 24, 2007 9345