The palladium-catalyzed Suzuki and Heck coupling reactions of aryl halides are important and versatile methods for the generation of C-C bonds [1, 2, 3, 4, 5, 6, 7] and have been used extensively in the synthesis of natural products, pharmaceuticals, herbicides, and advanced materials [8, 9]. The Suzuki coupling reaction is useful because of its tractability, high stability, mild reaction conditions, and broad functional group tolerance, as well as the low toxicity of boron compounds [10, 11, 12, 13, 14]. Generally, palladium-catalyzed Heck and Suzuki reactions are performed in polar organic solvents under oxygen-free conditions because of the solubility of the substrate and instability of most catalysts. However, from economic and environmental standpoints, it is desirable to use water as a solvent instead of hazardous and expensive organic solvents [15]. It is therefore worthwhile developing new highly active catalysts that are not sensitive to air and moisture, as well as easy to recover and recycle [16, 17, 18, 19]. Some significant advances have been made in meeting these requirements, including the use of phase- transfer catalysts [20, 21, 22], water-soluble phosphine ligands [23, 24], water as a cosolvent [25, 26], and amphiphilic polymers for catalyst immobilization [27]. More recently, microwave heating [28] and ultrasonic irradiation [29] have been used as tools in synthetic chemistry to facilitate C-C coupling in water. However, many of these procedures use expensive and complex ligands, supports, or external assistance such as microwave heating or ultrasonic irradiation, and, with a few exceptions, most reactions involve hazardous organic cosolvents. To the best of our knowledge, only a few studies have shown that Suzuki and Heck coupling reactions could be carried out efficiently using ligand-free palladium catalysts under conventional heating in neat water. For instance, the protocol for a Suzuki coupling reaction using 2.5 mol% of Na2PdCl4, with the addition of sodium dodecyl sulfate as a surfactant, in pure water under aerobic conditions at 100 °C was reported by Ranu et al. [20]. Marziale et al. [30] developed an efficient method for Suzuki coupling reactions catalyzed by a palladacyclic catalyst precursor produced by ortho palladation of ([1,1′-biphenyl]-2-yloxy) diisopropylphosphine, enabling the reaction to be performed in neat water under air and at ambient temperature with Pd loadings of 0.04 mol%. Basu et al. [21] found that a combination of Pd(OAc)2 and tetra-n-butylammonium bromide in water at ambient temperature showed high catalytic activity in the reactions of aryl bromides or iodides with aryl trihydroxyborate salts of sodium at catalyst loading of 0.5 mol%. The key to the coupling reaction was the use of aryl trihydroxyborate salts of sodium to increase the substrate solubility in pure water. The protocol for Suzuki and Heck reactions catalyzed by 0.1 mol% palladium supported on a magnetic microgel in water at 90 °C was reported by Yu et al. [31]; however, the attractiveness of their method is limited because it requires an argon atmosphere. In addition, Roglans et al. [32] demonstrated that PdCl2(CH3CN)2 was an efficient catalyst for Suzuki and Heck coupling reactions in water at room temperature or 40 °C; however, arenediazonium salts were used as substrates, and the amount of palladium catalyst used was high, 5-10 mol%. From the academic and industrial viewpoints, the design of catalytic systems using simple ligand-free palladium catalysts without other external assistance in neat water is therefore still a promising route for establishing greener C-C coupling reactions [22].
In our previous studies, we synthesized a series of cyclopalladated ferrocenylimines, which showed satisfactory reactivities in Heck, Suzuki, Sonogashira, and Kumada reactions [33, 34, 35, 36, 37, 38, 39, 40, 41]. However, there has recently been considerable interest in the development of new and simple catalysts that can effectively catalyze C-C coupling reactions in water or aqueous media because such catalysts have the potential to be used in industrial systems. In this paper, the preparation and catalytic properties of cyclopalladated ferrocenylimines with ester groups are reported. The results show that these compounds effectively catalyze Heck and Suzuki coupling reactions with catalyst loadings of 0.1 mol% under ligand-free and ambient conditions in water or organic solvents.
All chemicals were purchased from commercial companies. N,N-Dimethylformamide (DMF), toluene, and methanol were dried with MgSO4, filtered, distilled, and stored over 4 Å molecular sieves. Melting points were measured using a WC-1 microscopic apparatus and were uncorrected. IR spectra were recorded with a Bruker VECTOR22 spectrometer using KBr pellets. Mass spectra were recorded using an LC-MSD-Trap- XCT instrument. 1H and 13C NMR spectra were recorded in CDCl3 using a Bruker DPX-400 spectrometer with tetramethylsilane (TMS) as the internal standard. The GC yields were determined using n-dodecane internal standards. GC analyses were carried out using a Shimadzu GC-2010 Plus instrument. All the Suzuki and Heck coupling reactions were performed without the protection of an inert gas.
Cyclopalladated ferrocenylimines were prepared as shown in Scheme 1.
A mixture of acetylferrocene (5.0 mmol), p- aminoethylbenzoate (5.0 mmol) or m-aminoethylbenzoate (5.0 mmol), molecular sieve (4 Å, 4.0 g), and toluene (80 mL) was refluxed for 24 h at 120 °C under a nitrogen atmosphere. The volume of the solution was reduced to ca. 10 mL. The products were separated and purified on a chromatography column with petroleum ether and ethyl acetate (2:1) as the eluent.
[(ƞ5-C5H5)Fe(ƞ5-C5H4CMe=NC6H4-p-COOEt)]. Yield, 80%; deep-red solid; m.p. 122-124 °C; IR (KBr pellet): 2922, 2846, 1704, 1627, 1594, 1100, 816, 714 cm−1; 1H NMR (400 MHz, CDCl3): δ 8.04-8.02 (2H, m, C6H4), 6.80 (2H, d, J = 8.4 Hz, C6H4), 4.79 (2H, t, J = 1.8 Hz, C5H4), 4.46 (2H, t, J = 1.8 Hz, C5H4), 4.40-4.35 (2H, m, CH2CH3), 4.23 (5H, s, C5H5), 2.08 (3H, s, CH3-C =N-), 1.40 (3H, t, J = 5.2 Hz, CH2CH3); 13C NMR (100 MHz, CDCl3): δ 168.1, 166.6, 156.3, 130.7, 125.1, 119.4, 83.0, 71.0, 69.9, 69.6, 68.4, 60.7, 58.5, 30.1, 18.3, 14.4; MS: Calcd for C21H21FeNO2 [M + 1]+ 376.1, found 376.7.
[(ƞ5-C5H5)Fe(ƞ5-C5H4CMe=NC6H4-m-COOEt)]. Yield, 85%; red oil; IR (KBr pellet): 2922, 2846, 1704, 1627, 1594, 1461, 1365, 880, 780, 700 cm−1; 1H NMR (400 MHz, CDCl3): δ 7.74 (1H, d, J = 7.1 Hz, C6H4), 7.40 (1H, s, C6H4), 7.30 (1H, t, J = 7.4 Hz, C6H4), 6.90 (1H, d, J = 7.7 Hz, C6H4), 4.79 (2H, s, C5H4), 4.45 (2H, m, C5H4), 4.30-4.25 (2H, m, CH2CH3), 4.20 (5H, s, C5H5), 2.22 (3H, s, CH3-C =N-), 1.29 (3H, t, J = 7.1 Hz, CH2CH3); 13CNMR (100 MHz, CDCl3): δ 166.0, 164.6, 160.4, 148.3, 131.0, 129.5, 120.6, 114.3, 80.8, 70.2, 68.3, 62.0, 60.9, 20.5, 14.1; MS: Calcd for C21H21FeNO2 [M + 1] + 376.1, found 376.7.
Cyclopalladated dimers (1a and 2a) were prepared as follows. A mixture of Li2PdCl4 (1.0 mmol), sodium acetate (1.0 mmol), and ferrocenylimine (1.1 mmol) in methanol (10 mL) was stirredfor 24 h at room temperature. The mixture was filtered and the obtained solid was washed with ethanol. The products were separated and purified by chromatography column, using dichloromethane and ethanol (2:1) as the eluent.
[PdCl{[(ƞ5-C5H5)]Fe[(ƞ5-C5H3)CMe=NC6H4-p-COOEt]}]2 1a. Yield, 90%; deep-red solid; m.p. > 218 °C (dec); IR (KBr pellet): 2925, 1274, 1168, 1104, 1017, 816 cm−1; 1H NMR (400 MHz, CDCl3):δ 8.09 (2H, d, J = 7.4 Hz, C6H4), 7.14 (2H, d, J = 7.2 Hz, C6H4), 4.66 (1H, s, C5H3), 4.41 (5H, s, C5H5), 4.38 (1H, s, C5H3), 4.33 (1H, s, C5H3) 4.27-4.21 (2H, m, CH2CH3), 1.99 (3H, s, CH3-C=N-), 1.43 (3H, t, J = 6.8 Hz, CH3CH2); 13C NMR (100 MHz, CDCl3): δ 179.8, 166.5, 165.1, 153.4, 131.4, 128.9, 122.4, 101.2, 91.5, 73.8, 71.3, 66.4, 65.9, 61.1, 53.2, 52.4, 21.6, 14.9; MS: Calcd for C42H40Cl2Fe2N2O4Pd2 [M−Cl]+ 994.9, found 995.1.
[PdCl{[(ƞ5-C5H5)]Fe[(ƞ5-C5H3)CMe=NC6H4-m-COOEt]}]2 2a. Yield, 80%; deep-red solid; m.p. > 260 °C (dec); IR (KBr pellet): 2925, 1274, 1168, 1104, 1017, 880, 780, 700 cm−1; 1H NMR (400 MHz, CDCl3): δ 7.76 (1H, d, J = 7.8 Hz, C6H4), 7.45 (1H, s, C6H4), 7.39 (1H, t, J = 7.8 Hz, C6H4), 6.95 (1H, d, J = 7.8 Hz, C6H4), 4.81-4.78 (2H, m, C5H5), 4.46-4.41 (2H, m, C5H3), 4.37 (2H, q, J = 7.1 Hz, CH2CH3), 4.23 (5H, s, C5H3), 2.08 (3H, s, CH3-C=N-), 1.38 (3H, t, J = 7.1 Hz, CH2CH3); 13C NMR (100 MHz, CDCl3): δ 179.8, 166.5, 165.1, 153.4, 131.4, 128.9, 122.4, 101.2, 91.5, 73.8, 71.3, 66.4, 61.1, 53.2, 52.4, 21.6, 14.9; MS: Calcd for C42H40Cl2Fe2N2O4Pd2 [M−Cl]+ 994.9, found 995.1.
Cyclopalladated ferrocenylimine monomers (1b and2b) were prepared as follows. A mixture of 1a or 2a (0.5mmol) and triphenylphsophine (1.5mmol) in dichloromethane (10 mL) was stirred for 1 h. The volume of the solution was reduced to ca. 3 mL. The products were purified on a chromatography column with dichloromethane as the eluent.
[PdCl{[(ƞ5-C5H5)]Fe[(ƞ5-C5H3)CMe=NC6H4-p-COOEt]}(PPh3)] 1b. Yield, 91%; deep-red solid; m.p. > 180 °C (dec); IR (KBr pellet): 2925, 1274, 1168, 1104, 1017, 810, 760, 684 cm−1; 1H NMR (400 MHz, CDCl3): δ 8.06 (2H, d, J = 8.4 Hz, C6H4), 7.81−7.71 (6H, m, PPh3), 7.43−7.35 (9H, m, PPh3), 7.08 (2H, d, J = 8.1 Hz, C6H4), 4.48 (1H, d, J = 2.2 Hz, C5H3), 4.35 (2H, m, CH2CH3), 4.16 (1H, t, J = 2.2 Hz, C5H3), 3.94 (5H, s, C5H5), 3.43 (1H, d, J = 2.2 Hz, C5H3), 2.08 (3H, s, CH3-C=N-), 1.36 (3H, t, J = 7.1 Hz, CH3CH2); 13CNMR (100 MHz, CDCl3): δ 179.8, 166.5, 165.1, 153.4, 135.0, 132.4, 131.4, 130.3, 128.9, 127.9, 122.4, 101.2, 91.5, 73.8, 71.3, 66.4, 61.1, 53.2, 52.4, 21.6, 14.9; MS: Calcd for C39H35ClFeNO2PPd [M−Cl]+ 742.1, found 742.1.
[PdCl{[(ƞ5-C5H5)]Fe[(ƞ5-C5H3)CMe=NC6H4-m-COOEt]}(PPh3)] 2b. Yield, 79%; deep-red solid; m.p. 138-140 °C; IR (KBr pellet): 2925, 1274, 1168, 1104, 1017, 880, 780, 700 cm−1; 1H NMR (400 MHz, CDCl3): δ 7.87 (1H, d, J = 7.8 Hz, C6H4), 7.80-7.71 (6H, m, PPh3), 7.65 (1H, s, C6H4), 7.48-7.37 (9H, m, PPh3), 7.37-7.34 (1H, m, C6H4), 7.26 (1H, s, C6H4), 5.29 (1H, s, C5H3), 4.48 (1H, d, J = 2.3 Hz, C5H3), 4.35 (2H, q, J = 7.1 Hz, CH2CH3), 4.16 (1H, t, J = 2.0 Hz, C5H3), 3.94 (5H, s, C5H5), 2.09 (3H, s, CH3-C=N-), 1.38 (3H, t, J = 7.1 Hz, CH2CH3); 13C NMR (100 MHz, CDCl3): δ 179.8, 166.0, 148.3, 132.4, 131.0, 130.3, 128.0, 127.9, 120.6, 119.9, 114.3, 101.2, 91.5, 73.8, 71.3, 66.4, 60.9, 53.2, 52.4, 14.1; MS: Calcd for C39H35ClFeNO2PPd [M−Cl]+ 742.1, found 742.1.
A mixture of aryl bromide (0.5 mmol), phenylboronic acid (0.6 mmol), base (1 mmol), catalyst 1a (0.1 mol%), and solvent (2 mL) was stirred at a specified temperature under air. The reaction mixture was stirred for 2 h and then diluted with water and ethyl acetate. The organic layer was separated, and the aqueous layer was extracted three times with ethyl acetate. The combined organic phase was dried with MgSO4 and filtered; the solvent was removed using a rotary evaporator. The crude product was transferred to a 10 mL volumetric flask to fix its quantity in ethyl acetate. Yields of the desired cross-coupling products were determined by GC using n-dodecane (10 mL, 0.5 mmol) as an internal standard. The products were isolated using thin-layer chromatography. The 1H NMR spectroscopy and melting point data of the products were identical to those of commercially available authentic samples.
An aryl halide (0.5 mmol) and styrene (2 mmol) were combined with a base (0.5 mmol), n-Bu4NBr (0.5 mmol), and catalyst 1a (0.1 mol%) in a small round-bottomed flask. The reaction mixture was stirred at a specified temperature under air for 6 h and then diluted with water and ethyl acetate. The organic layer was separated, and the aqueous layer was extracted three times with ethyl acetate. The combined organic phase was dried with MgSO4 and filtered, and the solvent was removed using a rotary evaporator. The crude product was transferred to a 10 mL volumetric flask to fix its quantity in ethyl acetate. Yields of the desired cross-coupling products were determined by GC using n-dodecane (10 mL, 0.5 mmol) as an internal standard. The products were isolated using thin-layer chromatography. The 1H NMR spectroscopy and melting point data of the products were identical to those of commercially available authentic samples.
The coupling of 4-bromobenzene with styrene catalyzed by 1a was selected as a model reaction. As shown in Table 1, NaOAc as the base and DMF as the solvent afforded the highest yield (Table 1, entry 8). The relative activities of palladacycles 1a, 1b,2a, and 2b in the coupling reaction of 4-bromobenzene with styrene in the presence of NaOAc as the base in DMF were examined. The results showed that the catalytic activities of these cyclopalladated complexes were similar, but catalyst 1a was slightly more active under the same reaction conditions (entries 8-11).
The couplings of a number of other aryl bromides were conducted in the presence of 0.1 mol% 1a, using the optimized conditions described above; the results are shown in Table 2. The reactions of aryl bromides containing electron- withdrawing groups such as -CN, -CHO, -NO2, and -CF3 gave the coupling products in modest to high yields (entries 1and 8-17). Aryl bromides with electron-rich groups such as -OMe, -NH2, and -CH3 were poor coupling partners or almost inactive under the same reaction conditions (entries 2-7). When heterocyclic compounds such as 1-bromonaphthalene, 3-bromothiophene, and 3-bromopyridine were used as substrates, the cross- coupling products were obtained in modest to excellent yields (entries 18-20). The results showed that complex 1a was an efficient catalyst for Heck cross-coupling reactions of aryl bromides in organic solvents.
The coupling reaction of 4-bromobenzene with styrene, catalyzed by palladacycle 1a, was selected as a model reaction. As shown in Table 3, when NaOAc was used as the base in water, the yield was excellent (entry 8). The relative activities of several palladacycles for the same model reaction, using NaOAc and water, were then studied, and we found that 1a exhibited the highest activity (entries 8-11).
Reaction conditions: PhBr (0.5 mmol), styrene (2 mmol), catalyst 1a (0.1 mol%), base (0.5 mmol), n-Bu4NBr (0.5 mmol), H2O (2.0 mL), under air for 6 h. a GC yields. b Catalyst 1b (0.1 mol%). c Catalyst 2a (0.1 mol%). d Catalyst 2b (0.1 mol%).
We further investigated the Heck coupling reactions between a range of aryl bromides and styrene in water; the results are shown in Table 4. The reactions of electron-deficient aryl bromides with para substituents such as -CN, -CHO, -NO2, and -CF3 gave the coupling products in modest to high yields (entries 1, 10, and 13-15), and those with ortho or meta substituents gave lower yields (entries 8, 9, 11, and 12). Aryl bromides with electron-rich groups such as -OMe, -NH2, and -CH3 exhibited lower reactivities under the same reaction condition (entries 2-4, 6, and 7), but 4-bromoaniline gave the coupling product in 83% yield (entry 5). When heteroaryl bromides such as 3-bromofuran and 3-bromopyridine were used as substrates, the coupling products were obtained in modest to excellent yields, but 1-bromonaphthalene was found to be a poor coupling partner under the same reaction conditions, giving only 16% yield (entries 16-19). Finally, the reactions of aryl chlorides with styrene in water were also investigated. Aryl chlorides gave the corresponding coupling products in low yields (entries 20-24). However, 1-chloro-4-methoxybenzene reacted with styrene to give the coupling product in 80% yield (entry 25).
Suzuki cross-coupling reactions of 4-bromotoluene and phenylboronic acid were carried out using various bases and solvents in the presence of 0.1 mol% of 1a without the protection of an inert gas; the results are shown in Table 5. As can be seen from Table 5, the reaction proceeded smoothly in toluene, and among the tested bases, KOH was found to be the most effective (entry 2). The relative activities of several palladacycles under the same reaction conditions were investigated, using KOH as the base and toluene as the solvent (entries 2-5). The palladacycles all exhibited high catalytic activities for Suzuki coupling reactions under the optimized reaction conditions.
Catalyst 1a was used to catalyze the Suzuki coupling reactions between aryl bromides and phenylboronic acid in toluene; the results are shown in Table 6. Electron-deficient or electron-rich aryl bromides containing groups such as -CN, -CHO, -NO2, -CF3, and -CH3 at the para position gave the coupling products in good yields (entries 2, 5, 6, 9, and 10), and those with ortho or meta substituents gave the coupling products in moderate to good yields (entries 1, 3, 7, and 8). 3-Bromofuran and 3-bromopyridine gave the coupling products in excellent yields, but 1-bromonaphthalene gave a yield of 58% (entries 11-13).
The catalytic properties of the palladacycles in Suzuki coupling reactions in water were investigated. The reaction conditions were screened using the coupling reaction of bromobenzene with phenylboronic acid, using 0.1 mol% of 1a as a model reaction. As shown in Table 7, K3PO4·3H2O, NaHCO3, and K3PO4·7H2O were found to be effective bases, giving high yields in water at 100 °C (entries 4, 7, and 9). When the reaction temperature was reduced from 100 °C to 90 and 80 °C, isolated yields of 97% and 89%, respectively, were obtained (entries 10 and 11). The relative activities of several palladacycles were investigated under the same reaction conditions using K3PO4·3H2O (entries 10-13). The palladacycles all exhibited high catalytic activities for Suzuki coupling reactions under the optimized reaction conditions.
We next examined the scope and limitations of the protocol using aryl bromides and phenylboronic acid in the presence of 0.1 mol% 1a at 90 °C with K3PO4·3H2O as the base. The obtained results are summarized in Table 8. Aryl bromides with electron-donating and electron-withdrawing para substituents all worked well, providing the target products in good to excellent yields after 2 h (entries 2, 4, 6, 8, 12, and 13). The coupling reactions also proceeded smoothly with 3-bromopyridine, 3-bromothiophene, and 1-bromonaphthalene under the same reaction conditions to afford the coupling products in modest to good yields (entries 14-16). However, sterically hindered aryl bromides containing ortho or meta substituents gave the coupled product in slightly lower yields (entries 1, 3, 5, and 7). Interestingly, 2-bromobenzonitrile, 3-bromobenzonitrile, and 4-bromobenzonitrile provided the coupled product in 96%, 76%, and 66% yields, respectively (entries 9-11).
Finally, the reactions of aryl chlorides or iodides with phenylboronic acid in water were also tested. The results are summarized in Table 9. Aryl chlorides gave the corresponding coupling products in low yields (entries 1-4). Iodobenzene reacted with phenylboronic acid to give the coupling product in 79% yield, and 4-iodobenzoic acid was found to be a poor coupling partner (entries 5 and 6). This is probably because 4-iodobenzoic acid was highly reactive and failed to undergo cross-coupling in our catalytic system.
A series of cyclopalladated ferrocenylimine dimers and monomers with ester groups were designed and prepared. We also developed a convenient and efficient methodology for Suzuki and Heck coupling reactions of aryl bromides catalyzed by cyclopalladated ferrocenylimines with ester groups in organic solvents or water under ambient conditions with low catalyst loadings of 0.1 mol%. Suzuki coupling reactions in water at 90 °C in the absence of additives are significant in terms of both economics and the environment.
2014, Vol. 35 Issue (7): 1059-1067
PDF (514KB)
2014, Vol. 35 Issue (7): 1059-1067
PDF (514KB)