A Highly Active Cobalt Catalyst System for Kumada Biaryl Cross-Coupling

Abstract

A highly active cobalt catalyst system has been developed for the cross-coupling reactions of arylmagnesium reagents and aryl bromides. In the presence of 1 mol% CoCl₂, 2 mol% IPr·HCl and 2 mol% NaO ^t Bu, a wide range of (hetero)biaryls are prepared in 51–99% yields at room temperature within a short reaction time.

Transition-metal-catalyzed cross-coupling reactions for carbon–carbon bond formation have become an indispensable tool in contemporary organic synthesis.[1] Catalysts based on palladium and nickel are frequently employed in these processes on both laboratory and commercial scales. However, palladium is a precious metal with a low crustal abundance while nickel has high toxicity, which taints its use in consumer goods and healthcare products. Thus, the search for alternative catalysts based on iron and cobalt for cross-coupling reactions has received considerable attention.[2] [3]

The iron- and cobalt-catalyzed Kumada reactions of aryl halides and aryl Grignard reagents are of particular interests since the biaryl structural motif is ubiquitous in natural products, compounds of medicinal values, and functional materials.[4] It is, however, not trivial to control the product selectivity in these reactions as aryl Grignard reagents can undergo self-coupling rather readily in the presence of an iron or cobalt catalyst.[3b] [5] In a seminal report, Nakamura et al. showed that aryl halides and aryl Grignard reagents could be coupled in high yields at 60–80 °C in the presence of a catalyst system comprising an iron or cobalt fluoride salt, an N-heterocyclic carbene (NHC) precursor, and an ­alkylmagnesium reagent (Scheme [1, a]).[6] The presence of fluoride was critical to attain a good product selectivity as the employment of other metal salts (e.g., metal chlorides) led to a significant homocoupling of arylmagnesium reagents. Presumably, this ‘fluoride effect’ arises from the ability of fluoride to stabilize the metal center against reduction by the Grignard reagents, ultimately suppressing the homocoupling pathway.[6b] We subsequently investigated the counterion effect in the iron-catalyzed reaction, and realized that iron(III) alkoxide/NHC and iron(II) triflate/NHC catalyst systems also facilitate this type of coupling in high selectivity.[7] We have now identified a fluoride-free and highly active cobalt catalyst system for Kumada biaryl cross-coupling. Under our conditions, the reactions of aryl bromides and arylmagnesium reagents generally proceed rapidly at room temperature to afford a wide range of (­hetero)biaryls in moderate to excellent yields (Scheme [1, b]).

During our study of the cross-coupling reaction of bromobenzene (1a) and p-tolylmagnesium bromide (2a) in the presence of CoCl₂ (3 mol%) and the free carbene IPr (9 mol%), we found that an excellent yield of 3a was obtained even at room temperature as determined by GC analysis (Table [1], entry 1).[8] Replacement of IPr with a combination of IPr·HCl/NaO ^t Bu led to a comparable result as 3a was formed in 99% yield together with a small amount of the homocoupled product 3a′ (Table [1], entry 2). Compared to IPr, the saturated counterpart SIPr was slightly less efficient, resulting in 92% of 3a (Table [1], entry 3). Reduction in steric bulk of the NHC-N-substituents led to a significant decrease in the cross-coupling selectivity (i.e., IMes vs. IPr, ICy vs. IAd and I ^t Bu; Table [1], entries 4–7). While CoI₂ (Table [1], entry 8) and Co(acac)₃ (Table [1], entry 9) were comparable to CoCl₂ as the cobalt source, CoF₂·4H₂O was less active, resulting in only a partial conversion of 1a even after 16 h (Table [1], entry 10). These results suggest that the NHC ligand plays a major role in determining the product selectivity under these conditions, whereas the counterion effect is much less pronounced. A 2:1 ligand-to-metal ratio was optimal, and the catalyst loading could be reduced to 1 mol% CoCl₂, 2 mol% IPr·HCl and 2 mol% NaO ^t Bu without any erosion in the reaction yield (Table [1], entries 11 and 12). Furthermore, the reaction of 1a reached completion within 1 h (Table [1], entry 12).

Table 1 Cobalt-Catalyzed Cross-Coupling of Bromobenzene 1a and p-Tolylmagnesium Bromide 2a ^a

Entry	[Co]	NHC·HX	Conv. (%)	3a (%)b	3a′ (%)b,c
1d	CoCl2	IPr	100	95	13
2	CoCl2	IPr·HCl	100	99	8
3	CoCl2	SIPr·HCl	100	92	16
4	CoCl2	IMes·HCl	96	78	17
5	CoCl2	I t Bu·HCl	97	64	24
6	CoCl2	ICy·HCl	73	28	32
7	CoCl2	IAd·HCl	92	66	22
8	CoI2	IPr·HCl	100	99	11
9	Co(acac)3	IPr·HCl	100	99	10
10	CoF2·4H2O	IPr·HCl	89	85	5
11e	CoCl2	IPr·HCl	100	85	9
12f	CoCl2	IPr·HCl	100	99	9

^a Reactions conditions: 1.2 equiv 2a, 3 mol% CoCl₂, 9 mol% NHC·HX, 9 mol% NaO ^t Bu, r.t., 16 h.

^b Determined by GC using dodecane as an internal standard.

^c Calculated based on 2a.

^d 3 mol% CoCl₂ and 9 mol% IPr were employed.

^e 3 mol% CoCl₂, 3 mol% IPr·HCl and 3 mol% NaO ^t Bu were employed.

^f 1 mol% CoCl₂, 2 mol% IPr·HCl and 2 mol% NaO ^t Bu were employed. The reaction was run for 1 h.

The reaction of chlorobenzene also proceeded at room temperature to give 3a in 77% yield (Table [2], entry 1). Running the reaction at elevated temperatures helped improve the yield of 3a considerably (Table [2], entries 2 and 3). Attempts to reduce the ligand-to-metal ratio or the catalyst loading led to a reduction in the reaction yield (Table [2], entries 4 and 5). A significant decomposition was observed in the reaction of iodobenzene (Table [2], entry 6), presumably via protodehalogenation.

Table 2 Cobalt-Catalyzed Cross-Coupling Reaction of Phenyl Chloride and Iodide with 2a ^a

Entry	X	Temp (°C)	Conv. (%)	3a (%)b	3a′ (%)b,c
1	Cl	r.t.	98	77	16
2	Cl	40	100	83	21
3	Cl	60	100	90	17
4d	Cl	60	100	80	16
5e	Cl	60	47	39	23
6	I	r.t.	100	26	34

^aReactions conditions: 1.2 equiv 2a, 3 mol% CoCl₂, 9 mol% IPr·HCl, 9 mol% NaO ^t Bu, 16 h.

^b Determined by GC using dodecane as an internal standard.

^c Calculated based on 2a.

^dReactions conditions: 1.2 equiv 2a, 3 mol% CoCl₂, 6 mol% IPr·HCl, 6 mol% NaO ^t Bu, 16 h.

^eReactions conditions: 1.2 equiv 2a, 1 mol% CoCl₂, 2 mol% IPr·HCl, 2 mol% NaO ^t Bu, 16 h.

The scope of aryl bromides and aryl Grignard reagents was evaluated employing 1 mol% CoCl₂, 2 mol% IPr·HCl and 2 mol% NaO ^t Bu at room temperature (Scheme [2], conditions A). The reactions reached completion within 1 h in most cases except for highly electron-rich substrates (3e,h,q). In the case of 3e and 3q, a prolonged reaction time was needed to ensure a good conversion. For 3h, the reaction failed to reach a good conversion even after 16 h, and a higher catalyst loading of 3 mol% CoCl₂, 6 mol% IPr·HCl, and 6 mol% NaO ^t Bu (conditions C) was needed. Under these optimized conditions, moderate to excellent yields of the cross-coupled products could be obtained. Grignard reagents incorporating electron-donating (3d,e,g,h) and electron-withdrawing (3j,k) substituents are well-tolerated under cobalt catalysis to give the biaryl products in high yielding. The ­cobalt-catalyzed reaction was also applicable to the synthesis of ortho,ortho'-disubstituted biaryls (3l,m). On the other hand, the reaction of mesityl bromide (3n) failed to convert, even at 60 °C. The reaction of 3-bromofuran with 4-anisylmagnesium bromide proceeded to give 3o in 91% yield. A significant decomposition was observed in the reaction of 2- and 3-bromothiophene (3p,q) to give the products in moderate yields.

In general, the reactions of aryl chlorides resulted in lower yields of the biaryl products compared to those of aryl bromides (Scheme [2], conditions B). Chlorobenzene was converted in good yields in the reactions with aryl Grignard reagents incorporating electron-donating or electron-withdrawing substituents (3a,i,k). The reaction of electron-rich 1-chloro-4-methoxybenzene proceeded to give 3c in 56% yield whereas the electron-deficient 1-chloro-4-fluorobenzene was coupled with 4-anisylmagensium bromide to give 3f in 83% yield.

We further explored the coupling of substrates incorporating sensitive functional groups (i.e., nitrile and ester). 4-Bromobenzonitrile (1r) readily decomposed under the reaction conditions due to nucleophilic attack of the organomagnesium 2a on the nitrile as determined by GC-MS analysis (Table [3], entry 1). Running the reaction at 0 °C (Table [3], entry 2), or in the presence of LiCl/ZnCl₂ additives (Table [3], entries 3 and 4), which were previously found to beneficial in the nickel-catalyzed Kumada reactions,[9] did not help improve the yield of 3r. Similarly, the reaction of methyl 3-bromobenzoate with 2a under cobalt catalysis resulted in only ca. 10% of the desired product as determined by GC-MS analysis.

Table 3 Cobalt-Catalyzed Cross-Coupling Reaction of 4-Bromobenzonitrile with 2a ^a

Entry	Additive	Temp (°C)	Conv. (%)	3r (%)b
1	none	r.t.	98	7
2	none	0	96	8
3	LiCl (3 equiv)	r.t.	75	7
4	LiCl (1.2 equiv) + ZnCl2 (10 mol%)	r.t.	70	7

^aReactions conditions: 1.2 equiv 2a, 3 mol% CoCl₂, 9 mol% IPr·HCl, 9 mol% NaO ^t Bu, 16 h.

^b Determined by GC using dodecane as an internal standard.

In conclusion, a catalyst system comprising CoCl₂/­IPr·HCl/NaO ^t Bu is highly efficient for the Kumada biaryl cross-coupling reaction. The coupling selectivity was significantly influenced by the NHC ligand employed, whereas counterion effect is much less pronounced. Under these conditions, the reactions of aryl bromides and arylmagnesium reagents proceed rapidly at room temperature to afford a wide range of biaryl products.

• References and Notes

• 1a de Meijere A. Diederich F. In Metal-Catalyzed Cross-Coupling Reactions . Wiley-VCH; New York: 2004. 2nd ed
  Search in Google Scholar
• 1b Miyaura N. In Cross-Coupling Reactions: A Practical Guide . Springer; Berlin: 2002
  Search in Google Scholar

For recent reviews on iron-catalyzed cross-coupling reactions, see:
• 2a Adams CJ. Bedford RB. Carter E. Gower NJ. Haddow MF. Harvey JN. Huwe M. Cartes MA. Mansell SM. Mendoza C. Murphy DM. Neeve EC. Nunn J. J. Am. Chem. Soc. 2012; 134: 10333
  CrossrefPubMedSearch in Google Scholar
• 2b Nakamura E. Hatakeyama T. Ito S. Ishizuka K. Ilies L. Nakamura M. Iron-Catalyzed Cross-Coupling Reactions, in Organic Reactions 2014; 83: 1
  Search in Google Scholar
• 2c Kuzmina OM. Steib AK. Moyeux A. Cahiez G. Knochel P. Synthesis 2015; 47: 1696
  CrossrefSearch in Google Scholar
• 2d Bauer I. Knölker H.-M. Chem. Rev. 2015; 115: 3170
  CrossrefPubMedSearch in Google Scholar
• 2e Guérinot A. Cossy J. Top. Curr. Chem. 2016; 374: 49
  CrossrefPubMedSearch in Google Scholar

For reviews on cobalt-catalyzed cross-coupling reactions, see:
• 3a Gosmini C. Bégouin J.-M. Moncomble A. Chem. Commun. 2008; 44: 3221
  Search in Google Scholar
• 3b Cahiez G. Moyeux A. Chem. Rev. 2010; 110: 1435
  CrossrefPubMedSearch in Google Scholar
• 4a Magano J. Dunetz JR. Chem. Rev. 2011; 111: 2177
  CrossrefPubMedSearch in Google Scholar
• 4b Slagt VF. de Vries AH. M. de Vries JG. Kellogg RM. Org. Process Res. Dev. 2010; 14: 30
  CrossrefSearch in Google Scholar
• 4c Ivica C. Synthesis of Biaryls . Elsevier; Oxford: 2004
  Search in Google Scholar

For examples on iron-catalyzed homocoupling of arylmagnesium reagents, see:
• 5a Nagano T. Hayashi T. Org. Lett. 2005; 7: 491
  CrossrefPubMedSearch in Google Scholar
• 5b Cahiez G. Chaboche C. Mahuteau-Betzer F. Ahr M. Org. Lett. 2005; 7: 1943
  CrossrefPubMedSearch in Google Scholar
• 5c Cahiez G. Moyeux A. Buendia J. Duplais C. J. Am. Chem. Soc. 2007; 129: 13788
  CrossrefPubMedSearch in Google Scholar

For an example on cobalt-catalyzed homocoupling of arylmagnesium reagents, see:
• 5d Kharasch MS. Fuchs CF. J. Am. Chem. Soc. 1941; 63: 2316
  CrossrefSearch in Google Scholar
• 6a Hatakeyama T. Nakamura M. J. Am. Chem. Soc. 2007; 129: 9844
  CrossrefPubMedSearch in Google Scholar
• 6b Hatakeyama T. Hashimoto S. Ishizuka K. Nakamura M. J. Am. Chem. Soc. 2009; 131: 11949
  CrossrefPubMedSearch in Google Scholar

Also, see:
• 6c Agrawal T. Cook SP. Org. Lett. 2014; 16: 5080
  CrossrefPubMedSearch in Google Scholar
• 7a Chua Y.-Y. Duong HA. Chem. Commun. 2014; 50: 8424
  CrossrefPubMedSearch in Google Scholar
• 7b Chua Y.-Y. Duong HA. Chem. Commun. 2016; 52: 1466
  CrossrefPubMedSearch in Google Scholar
• 7c Wu W. Teng Q. Chua Y.-Y. Huynh HV. Duong HA. Organometallics 2017; 36: 2293
  CrossrefSearch in Google Scholar
• 8 General Experimental Procedure In a 25 mL Schlenk flask, a mixture of CoCl₂ (2 mg, 0.015 mmol, 1 mol%), NaO ^t Bu (3 mg, 0.03 mmol, 2 mol%) and IPr·HCl (13 mg, 0.03 mmol, 2 mol%) in THF (1.0 mL) was stirred under argon at rt for 1 h. A solution of bromobenzene (1a, 236 mg, 1.5 mmol, 1.0 equiv) in 0.5 mL THF was added followed by a solution of p-tolylmagnesium bromide (2a, 2.3 mL, 0.78 M in THF, 1.8 mmol, 1.2 equiv). The reaction progress was monitored by GC using dodecane as an internal standard. Once completed, the reaction mixture was quenched with saturated aqueous NH₄Cl solution (10 mL) and extracted with ethyl acetate (20 mL) three times. The combined organic layers were dried over anhydrous magnesium sulfate and concentrated in vacuo. The resulting crude mixture was purified by silica gel column chromatography to give 3a (237 mg, 94%) as a white solid. Analytical Data for Compound 3a ¹H NMR (400 MHz, CDCl₃): δ = 7.68–7.59 (m, 2 H), 7.58–7.51 (m, 2 H), 7.47 (dd, J = 8.5, 6.9 Hz, 2 H), 7.38 (d, J = 7.3 Hz, 1 H), 7.30 (d, J = 7.9 Hz, 2 H), 2.45 (s, 3 H). ¹³C{1 H} NMR (101 MHz, CDCl₃): δ = 141.3, 138.5, 137.1, 129.6, 128.8, 127.2, 127.1, 127.0, 126.9, 21.2.
• 9a Xie L.-G. Wang Z.-X. Chem. Eur. J. 2010; 16: 10332
  CrossrefPubMedSearch in Google Scholar
• 9b Liu N. Wang Z.-X. J. Org. Chem. 2011; 76: 10031
  CrossrefPubMedSearch in Google Scholar

• References and Notes

• 1a de Meijere A. Diederich F. In Metal-Catalyzed Cross-Coupling Reactions . Wiley-VCH; New York: 2004. 2nd ed
  Search in Google Scholar
• 1b Miyaura N. In Cross-Coupling Reactions: A Practical Guide . Springer; Berlin: 2002
  Search in Google Scholar

For recent reviews on iron-catalyzed cross-coupling reactions, see:
• 2a Adams CJ. Bedford RB. Carter E. Gower NJ. Haddow MF. Harvey JN. Huwe M. Cartes MA. Mansell SM. Mendoza C. Murphy DM. Neeve EC. Nunn J. J. Am. Chem. Soc. 2012; 134: 10333
  CrossrefPubMedSearch in Google Scholar
• 2b Nakamura E. Hatakeyama T. Ito S. Ishizuka K. Ilies L. Nakamura M. Iron-Catalyzed Cross-Coupling Reactions, in Organic Reactions 2014; 83: 1
  Search in Google Scholar
• 2c Kuzmina OM. Steib AK. Moyeux A. Cahiez G. Knochel P. Synthesis 2015; 47: 1696
  CrossrefSearch in Google Scholar
• 2d Bauer I. Knölker H.-M. Chem. Rev. 2015; 115: 3170
  CrossrefPubMedSearch in Google Scholar
• 2e Guérinot A. Cossy J. Top. Curr. Chem. 2016; 374: 49
  CrossrefPubMedSearch in Google Scholar

For reviews on cobalt-catalyzed cross-coupling reactions, see:
• 3a Gosmini C. Bégouin J.-M. Moncomble A. Chem. Commun. 2008; 44: 3221
  Search in Google Scholar
• 3b Cahiez G. Moyeux A. Chem. Rev. 2010; 110: 1435
  CrossrefPubMedSearch in Google Scholar
• 4a Magano J. Dunetz JR. Chem. Rev. 2011; 111: 2177
  CrossrefPubMedSearch in Google Scholar
• 4b Slagt VF. de Vries AH. M. de Vries JG. Kellogg RM. Org. Process Res. Dev. 2010; 14: 30
  CrossrefSearch in Google Scholar
• 4c Ivica C. Synthesis of Biaryls . Elsevier; Oxford: 2004
  Search in Google Scholar

For examples on iron-catalyzed homocoupling of arylmagnesium reagents, see:
• 5a Nagano T. Hayashi T. Org. Lett. 2005; 7: 491
  CrossrefPubMedSearch in Google Scholar
• 5b Cahiez G. Chaboche C. Mahuteau-Betzer F. Ahr M. Org. Lett. 2005; 7: 1943
  CrossrefPubMedSearch in Google Scholar
• 5c Cahiez G. Moyeux A. Buendia J. Duplais C. J. Am. Chem. Soc. 2007; 129: 13788
  CrossrefPubMedSearch in Google Scholar

For an example on cobalt-catalyzed homocoupling of arylmagnesium reagents, see:
• 5d Kharasch MS. Fuchs CF. J. Am. Chem. Soc. 1941; 63: 2316
  CrossrefSearch in Google Scholar
• 6a Hatakeyama T. Nakamura M. J. Am. Chem. Soc. 2007; 129: 9844
  CrossrefPubMedSearch in Google Scholar
• 6b Hatakeyama T. Hashimoto S. Ishizuka K. Nakamura M. J. Am. Chem. Soc. 2009; 131: 11949
  CrossrefPubMedSearch in Google Scholar

Also, see:
• 6c Agrawal T. Cook SP. Org. Lett. 2014; 16: 5080
  CrossrefPubMedSearch in Google Scholar
• 7a Chua Y.-Y. Duong HA. Chem. Commun. 2014; 50: 8424
  CrossrefPubMedSearch in Google Scholar
• 7b Chua Y.-Y. Duong HA. Chem. Commun. 2016; 52: 1466
  CrossrefPubMedSearch in Google Scholar
• 7c Wu W. Teng Q. Chua Y.-Y. Huynh HV. Duong HA. Organometallics 2017; 36: 2293
  CrossrefSearch in Google Scholar
• 8 General Experimental Procedure In a 25 mL Schlenk flask, a mixture of CoCl₂ (2 mg, 0.015 mmol, 1 mol%), NaO ^t Bu (3 mg, 0.03 mmol, 2 mol%) and IPr·HCl (13 mg, 0.03 mmol, 2 mol%) in THF (1.0 mL) was stirred under argon at rt for 1 h. A solution of bromobenzene (1a, 236 mg, 1.5 mmol, 1.0 equiv) in 0.5 mL THF was added followed by a solution of p-tolylmagnesium bromide (2a, 2.3 mL, 0.78 M in THF, 1.8 mmol, 1.2 equiv). The reaction progress was monitored by GC using dodecane as an internal standard. Once completed, the reaction mixture was quenched with saturated aqueous NH₄Cl solution (10 mL) and extracted with ethyl acetate (20 mL) three times. The combined organic layers were dried over anhydrous magnesium sulfate and concentrated in vacuo. The resulting crude mixture was purified by silica gel column chromatography to give 3a (237 mg, 94%) as a white solid. Analytical Data for Compound 3a ¹H NMR (400 MHz, CDCl₃): δ = 7.68–7.59 (m, 2 H), 7.58–7.51 (m, 2 H), 7.47 (dd, J = 8.5, 6.9 Hz, 2 H), 7.38 (d, J = 7.3 Hz, 1 H), 7.30 (d, J = 7.9 Hz, 2 H), 2.45 (s, 3 H). ¹³C{1 H} NMR (101 MHz, CDCl₃): δ = 141.3, 138.5, 137.1, 129.6, 128.8, 127.2, 127.1, 127.0, 126.9, 21.2.
• 9a Xie L.-G. Wang Z.-X. Chem. Eur. J. 2010; 16: 10332
  CrossrefPubMedSearch in Google Scholar
• 9b Liu N. Wang Z.-X. J. Org. Chem. 2011; 76: 10031
  CrossrefPubMedSearch in Google Scholar

• Supplementary Material