Palladium-catalyzed heck and suzuki coupling in ...

c 2007 Institute of Chemistry, Slovak Academy of Sciences DOI: 10.2478/s11696-007-0026-3

SHORT COMMUNICATION

Palladium-Catalyzed Heck and Suzuki Coupling in Glycerol A. WOLFSON* and C. DLUGY

Green Processes Center, Chemical Engineering Department, Sami Shamoon College of Engineering, Bialik/Basel Sts. Beer-Sheva, 84100 Israel e-mail: [email protected] Received 20 October 2006; Revised 5 December 2006; Accepted 14 December 2006

The Heck coupling of halobenzenes with various alkenes and the Suzuki cross coupling of halobenzenes with phenylboronic acid were successfully performed in glycerol as the reaction solvent using homogeneous and heterogeneous palladium catalysts. Glycerol is a renewable and recyclable green solvent that is able to dissolve organic substrates, inorganic bases, and palladium complexes, and that allows easy isolation of the reaction product by simple extraction with glycerol-immiscible solvents such as diethyl ether, hexane, and dichloromethane. Keywords: Pd cross coupling, Heck reaction, Suzuki reaction, green solvent, liquid—liquid extraction

The formation of a carbon—carbon bond is a fundamental reaction in organic synthesis. In the past three decades, palladium-catalyzed carbon—carbon coupling reactions of aryl halides with activated alkenes (Heck coupling, Scheme 1) and arylboronic acids (Suzuki coupling, Scheme 2) represent two of the most useful methods for carbon—carbon bond formation using arenes [1—8]. These syntheses also show much promise in many industrial processes, as they are simple, inexpensive, and often high-yielding procedures. Though many highly active palladium complexes were reported in literature, their tedious synthesis, expense, and difficult recovery and recycling are significant disadvantages. Most organic transformations are performed in solution. The solvent brings the reactants and catalysts together and delivers the thermal energy needed for the reaction [9]. In addition, the choice of solvent may affect the reaction activity and selectivity. The chemical, physical, and biological properties of the solvent are also important from the viewpoints of the environment, cost, safety, handling, and product isolation. The choice of the solvent in palladium-catalyzed C—C coupling reactions is crucial since the solvent must dissolve both non-polar and polar organic reactants, inorganic bases, and organometallic complexes

[4, 6]. To meet these requirements, polar aprotic solvents such as DMSO, DMF, N-methylpyrrolidine, dioxin, and acetonitrile are frequently used. Water, although a solvent with virtually no environmental impact, has limited value due to its negligible miscibility with many aryl halides and alkenes, as well as some organometallics [10—14]. Recently, more environmentally friendly and recyclable solvents such as lowmolecular-mass PEG [15—18] and ionic liquids [19— 23] were evaluated as reaction media in both Heck and Suzuki couplings. These solvents allow easy separation of products as well as catalyst recycling. Ionic liquids easily dissolve many non-modified palladium complexes and also stabilize Pd complexes against agglomeration (palladium black formation). Nevertheless, ionic liquid production requires large amounts of hazardous and volatile organic solvents, and their price limits their uses on large scale. Recently we have shown that glycerol, a polar, nontoxic, biodegradable and recyclable liquid manufactured from renewable sources, can be used as a green solvent for various catalytic and non-catalytic organic transformations [24, 25]. It has advantageous physical and chemical properties. Glycerol has a very high boiling point and negligible vapor pressure, is compatible with most organic and inorganic compounds,

*The author to whom the correspondence should be addressed.

228

Unauthenticated Download Date | 9/24/15 11:27 PM

Chem. Pap. 61 (3) 228—232 (2007)

PALLADIUM-CATALYZED HECK AND SUZUKI COUPLING IN GLYCEROL

Table 1. Palladium-Catalyzed Heck and Suzuki Coupling of Iodobenzene with Butyl Acrylate and Phenylboronic Acid in Glycerol a Heck coupling Entry

Catalyst

1 2 3 4 5 6 7 8 9 10 11 12

Suzuki coupling

Base

PdCl2 PdCl2 Pd(OAc)2 Pd(OAc)2 PdCl2 (TPPTS)2 PdCl2 (TPPTS)2 Pd(OAc)2 (TPPTS)2 Pd(OAc)2 (TPPTS)2 PdCl2 (DPPF)b2 Pd/C Pd/C Pd(OAc)2 (TPPTS)2

Et3 N Na2 CO3 Et3 N Na2 CO3 Et3 N Na2 CO3 Et3 N Na2 CO3 Na2 CO3 Et3 N Na2 CO3 Na2 CO3

Time/h

Yield/%

Time/h

Yield/%

4 4 4 4 4 4 4 4 4 4 4 0.083 0.167

74 100 65 72 87 100 56 100 86 40 78 56c 93d

1 1 1 1 1 1 1 1 1 1 1 0.083 0.167

76 90 81 95 83 94 66 75 82 61 88 45c 82d

Coupling of 0.5 mmol Ia with 0.75 mmol IIa or IV, using substrate to catalyst mole ratio 50 and 0.6 mmol base in 5 g glycerol at 80 ◦C (a) or in 20 g glycerol (b). Domestic microwave oven-assisted reactions with the temperature rise from 26 ◦C to 89 ◦C (c) or 117 ◦C (d).

Scheme 1

B(OH)2 0.2

I +

2 mole % Pd glycerol

IV

V

Scheme 2

and does not require special handling or storage. Glycerol dissolves palladium salts and complexes and inorganic bases together with halobenzenes and olefins. The poor miscibility of various hydrophobic solvents such as ethers and hydrocarbons with glycerol allows isolation of dissolved species, e.g. reaction products, by simple extraction. In this study we report on the successful palladium-catalyzed Heck and Suzuki C—C coupling reactions carried out in glycerol. Several simple and commercially available palladium catalysts as well as different substrates were tested. The Heck coupling of iodobenzene and butyl acrylate (Scheme 1) and the Suzuki cross coupling of iodobenzene and phenylboronic acid (Scheme 2) were used as representative reactions. Since biphenyl can

Chem. Pap. 61 (3) 228—232 (2007)

be also synthesized via the Ullmann homo-coupling of phenylboronic acid or halobenzene, two control reactions were done first. Both reactions did not yield biphenyl. Several palladium catalysts, with both triethylamine (Et3 N) and Na2 CO3 as bases, were tested. In a typical procedure, 10 µmol of palladium salt or complex or the corresponding amount of 5 % Pd/C, 0.5 mmol of halobenzene, 0.6 mmol of alkene or phenylboronic acid, and 0.6 mmol of sodium carbonate (all purchased from Aldrich) were added to a vial with 5 g of glycerol (Frutarom Ltd, 99.5 % purity). The mixture was placed in a preheated oil bath at 80 ◦C and magnetically stirred. After 1 h or 4 h, the reaction mixture was cooled and extracted with 3 × 10 mL of diethyl ether. The organic phase was concentrated under reduced pressure and the resulting crude product was analyzed by GC (using an HP-1 column) for conversion. The results are summarized in Table 1. In general, it can be clearly seen that the use of the stronger inorganic base, which is soluble in glycerol, gave higher yields with all catalysts. Both PdCl2 and Pd(OAc)2 gave high IIIa (Heck) and V (Suzuki) yields (entries 1—4). However, extensive formation of palladium black, and hence deactivation of the catalysts, was seen in both reactions. It is known that addition of phosphine ligands, like triphenylphosphine (TPP), stabilizes the palladium catalyst. Though PdCl2 (TPP)2 and Pd(OAc)2 (TPP)2 are slightly soluble in the reaction mixture, addition of a water-soluble phosphine, tris(3-sulfophenyl)phosphine trisodium salt (TPPTS) that is also soluble in glycerol, ensured homogeneous reactions. As illustrated in Table 1 (entries 5—8), both PdCl2 (TPPTS)2 and Pd(OAc)2 (TPPTS)2 were active and less palladium black formation was observed. Using another commercially available palladium complex, [1,1 -


229

A. WOLFSON, C. DLUGY

Table 2. The Effects of Base and Substrate to Catalyst Mole Ratio on the Synthesis of Butyl Cinnamate (IIIa) and Biaryl (V) in Glycerola Yield/% Entry

1 2 3 4 5 6 7 8 9

Base

Et3 N NaCOCH3 NaHCO3 NaHCO3 Na2 CO3 Na2 CO3 Na2 CO3 Na2 CO3 Na2 CO3

Amount/mmol

n(S)/n(C)

0.6 0.6 0.6 0.75 0.6 0.75 0.75 0.75 1

Heck coupling

Suzuki coupling

21 27 50 52 50 52 25 14 53

83 87 92 94 94 95 89 77 95

50 50 50 50 50 50 100 200 50

Coupling of 0.5 mmol Ia with 0.6 mmol IIa using 10 µmol PdCl2 (TPPTS)2 in 5 g glycerol at 80 ◦C for 1 h (a).

bis(diphenylphosphino)ferrocene]dichloro-palladium (II) (PdCl2 (DPPF)2 ) (entry 9) also resulted in high products yields. The solubility of PdCl2 (DPPF)2 in glycerol is lower than the solubilities of TPPTS-based complexes and hence higher amounts of glycerol were used. Finally, replacing the homogeneous catalysts by heterogeneous supported palladium catalyst yielded slightly lower yields (entries 10 and 11). The use of glycerol as a solvent allows easy product separation and catalyst recovery. Glycerol has a high boiling point and very low vapor pressure; hence it is possible to distill many products from the reaction mixture. The high boiling temperature of both butyl acrylate and biphenyl (148 ◦C and 255 ◦C, respectively) makes this separation method less relevant. On the other hand, these products are easily separated in high yields by extraction with glycerolimmiscible solvents such as diethyl ether, petroleum ether, dichloromethane, ethyl acetate, and paraffins such as hexane. Mixing each of these solvents with glycerol resulted in immediate phase separation, even faster than with water. The mutual solubilities of glycerol and each of these solvents, with the exception of ethyl acetate, were negligible and hence no further purification of the solvent was required. As glycerol allows the use of any of several solvents for extractions, the choice of the extracting solvent is based on the yield of product extraction as well as the solvent environmental impact, price, handling, and the easiness of recovery of the product from the extraction solvent. As previously mentioned, DMF and DMSO are common solvents for Heck and Suzuki couplings. The comparative Heck coupling of Ia with IIa using Pd/C as the catalyst and Na2 CO3 as the base, in glycerol and DMF gave comparable yields of IIIa after 1 h at 80 ◦C (88 % in glycerol and 90 % in DMF). However, glycerol is a greener solvent than DMF, and separation of the product by extraction from glycerol was much easier than from DMF. Microwave-promoted heating was recently reported to enhance C—C coupling reactions compared to conventional heating [26—29]. Microwave heating is based 230

on the ability of the solvent to absorb microwave energy and convert it into heat. Polar solvents with hydroxyl substitution such as glycerol are very attractive solvents for this purpose due to their high dielectric constant and high boiling point. Microwave-promoted Heck and Suzuki reactions in glycerol were carried out using Pd(OAc)2 (TPPTS)2 as the catalyst and Na2 CO3 as the base for 5 min (Table 1, entry 12). Experiment was conducted in a domestic microwave oven (Crystal WP900, 900 W) in a glass vessel, which was covered with a watch glass at atmospheric pressure. The catalyst was dissolved in 10 mL of glycerol followed by addition of Ia, IIa/IV, and sodium carbonate to a 50 mL vessel. The vessel was covered with the watch glass, and then the temperature of the reaction mixture was increased during 5 min from 26 ◦C to 89 ◦C or during 10 min from 26 ◦C to 117 ◦C using the microwave heating. The vessel was cooled down to room temperature and the reaction mixture was extracted with diethyl ether for GC analysis. As previously mentioned, palladium-catalyzed C— C coupling reactions require an addition of a base to maintain the catalytic cycle. Hence, the effects of the base type, strength, and amount on the activity of the coupling of Ia and IIa or IV in glycerol were examined (Table 2). As expected, the stronger the added base, the higher the reaction activity. Yet, increasing the base to substrate ratio beyond a stoichiometric amount did not change the product yield. Increasing the substrate to catalyst mole ratio (n(S)/n(C)) by changing the substrate amount and retaining the complex concentration decreased the conversion, without significant influence on the yield (Table 2, entries 5— 7). The scope and limitations of palladium-catalyzed C—C coupling in glycerol were studied with different aryl halides (Table 3, Scheme 1). Replacing butyl acrylate by isobutyl acrylate or styrene also gave high conversions in the Heck synthesis. In addition, employing the less active bromobenzene Ib instead of


Chem. Pap. 61 (3) 228—232 (2007)

PALLADIUM-CATALYZED HECK AND SUZUKI COUPLING IN GLYCEROL

Table 3. Variation of the Product Yield with Time during Palladium-Catalyzed C—C Coupling of Halobenzenes in Glycerola Product yield/% Substrate

Ia Ia Ia Ib Ib Ib Ic Ia Ib Ic

Product

IIIa IIIb IIIc IIIa IIIb IIIc IIIa V V V

1h

4h

22 h

52 55 32 28 25 27 1 94 84 2 (4b )

100 100 83 36 32 79 3 100 95 3 (18b )

(62b )

Coupling of 0.5 mmol I with 0.6 mmol II or IV using 10 µmol PdCl2 (TPPTS)2 and 0.6 mmol Na2 CO3 in 5 g glycerol at 80 ◦C (a) or 10 µmol PdCl2 (DPPF)2 in 20 g glycerol (b).

Table 4. Catalyst Recycling in the Suzuki Cross Coupling of Iodobenzene with Phenylboronic Acid in Glycerola Yield/% No. of cycles

1 2 3

PdCl2 (TPPTS)2

Pd/C

PdCl2 (DPPF)b2

88 55 47

98 61 50

82 80 80

Coupling of 0.5 mmol I with 0.6 mmol IV using 2 mole % palladium and 0.6 mmol Na2 CO3 in 5 g glycerol at 80 ◦C for 1 h (a) or in 20 g glycerol (b).

Ia also gave high yields of IIIa—IIIc. Unfortunately, chlorobenzene Ic was inactive toward Heck synthesis. Similarly, both Ia and Ib were active toward Suzuki cross-coupling reaction with PdCl2 (TPPTS)2 , while employing of Ic did not result in any biphenyl formation. On the other hand, the use of PdCl2 (DPPF)2 in the Suzuki cross-coupling reaction enabled the activation of chlorobenzene, although (as expected) the reaction with chlorobenzene was much slower than with the activated halobenzenes. Finally, the recycling of PdCl2 (TPPTS)2 , PdCl2 (DPPF)2 , and Pd/C in glycerol was examined in the cross coupling of Ia and IV (Table 4). Catalyst recycling was performed after extraction of both iodobenzene and biaryl from the glycerol catalytic phase using diethyl ether. Then, identical amounts of fresh substrates and sodium carbonate were added to the glycerol, and the reactions were re-run under identical conditions. As illustrated by the results in Table 4, both PdCl2 (TPPTS)2 and Pd/C lost some activity after each cycle while PdCl2 (DPPF)2 did not. In the case of PdCl2 (TPPTS)2 decomposition of the complex with formation of palladium aggregates was observed.

Chem. Pap. 61 (3) 228—232 (2007)

To conclude, glycerol can be usefully employed as an alternative green solvent for the Heck coupling and the Suzuki cross coupling. Both organic and inorganic bases are soluble in glycerol and afford high product yields. The product can be easily isolated in high yields by extraction with glycerol-immiscible solvents and the catalyst can be recycled. The performances of both Heck and Suzuki reactions in glycerol were comparable to the performance in DMF under similar conditions. However, recovery of the product from glycerol was much simpler than from DMF. Glycerol is greener and safer than DMF and toluene, and with increasing biodiesel production its price continuously drops. REFERENCES 1. Negishi, E. I., Coperet, C., Ma, S. M., Liou, S. Y., and Liu, F., Chem. Rev. 96, 365 (1996). 2. Crisp, G. T., Chem. Soc. Rev. 27, 427 (1998). 3. Dounay, A. B. and Overman, L. E., Chem. Rev. 103, 2945 (2003). 4. Alonso, F., Beletskaya, I. P., and Yus, M., Tetrahedron 61, 11771 (2005). 5. Suzuki, A., J. Organomet. Chem. 653, 83 (2002). 6. Miyaura, N. and Suzuki, A., Chem. Rev. 95, 2457 (1995). 7. Stambuli, J. P., Kuwano, R., and Hartwig, J. F., Angew. Chem., Int. Ed. Engl. 41, 4746 (2002). 8. Frisch, A. C. and Beller, M., Angew. Chem., Int. Ed. Engl. 44, 674 (2005). 9. Reichardt, C., Solvent Effects in Organic Chemistry. Verlag Chemie, Weinheim, 1979. 10. Herrmann, W. A. and Reisinger, C.-P., in AqueousPhase Organometallic Catalysis (Cornils, B. and Herrmann, W. A., Editors). Wiley-VCH, Weinheim, 1998. 11. Genêt, J. P. and Savignac, M., J. Organomet. Chem. 576, 305 (1999). 12. Manabe, K. and Kobayashi, S., Chem. Eur. J. 8, 4095 (2002). 13. Yan, J., Jin, H. W., and Shan, S., Tetrahedron 62, 5603 (2006). 14. Korolev, D. N. and Bumagin, N. A., Tetrahedron Lett. 47, 4225 (2006). 15. Luo, C. C., Zhang, Y. H., and Wang, Y. G., J. Mol. Catal. A 229, 7 (2005). 16. Chandrasekhar, S., Narsihmulu, C., Sultana, S. S., and Reddy, N. R., Org. Lett. 4, 4399 (2002). 17. Li, J. H., Liu, W. J., and Xie, Y. X., J. Org. Chem. 70, 5409 (2005). 18. Liu, W. J., Xie, Y. X., Yun, L. A., and Li, J. H., Synthesis-Stuttgart 2006, 860. 19. Carmichael, A. J., Earle, M. J., Holbrey, J. D., McCormac, P. B., and Seddon, K. R., Org. Lett. 1, 997 (1999). 20. Howarth, J. and Dallas, A., Molecules 5, 851 (2000). 21. B¨ ohm, V. P. W. and Herrmann, W. A., Chem. Eur. J. 6, 1017 (2000). 22. Szmant, H. H. and Roman, M. N., J. Am. Chem. Soc. 88, 4034 (1966).


231

A. WOLFSON, C. DLUGY

23. Wong, H. T., Han, S. J., and Livingston, A. G., Chem. Eng. Sci. 61, 1338 (2006). 24. Wolfson, A., Dlugy, C., and Shotland, Y., Environ. Chem. Lett., accepted for publication (2006). 25. Wolfson, A., Dlugy, C., Tavor, D., Blumenfeld, J., and Shotland, Y., Tetrahedron: Asymmetry 17, 2043 (2006). 26. Kappe, C. O., Angew. Chem., Int. Ed. Engl. 43, 6250 (2004).

232

27. Tierney, J. and Lidstr¨ om, P., Microwave-Assisted Organic Synthesis. Blackwell, Oxford, 2005. 28. Lidstr¨ om, P., Tierney, J., Wathey, B., and Westman, J., Tetrahedron 57, 9225 (2001). 29. Liu, Y. B., Khemtong, C., and Hu, J., Chem. Commun. 2004, 398.


Chem. Pap. 61 (3) 228—232 (2007)