Nickel-Catalyzed Coupling of Azoles with Aromatic Nitriles

Abstract

This manuscript describes the Ni-catalyzed coupling of azoles with aromatic nitriles. The use of BPh₃ promotes these arylations with electronically diverse oxazoles and benzonitriles. While the nickel catalyst is necessary for the arylations of phenyl oxazoles, arylation of benzoxazoles with some nitriles affords the arylated products even in the absence of the Ni-catalyst albeit in lower yield than the catalyzed process. The scope and rates of the Ni-catalyzed process are higher than the uncatalyzed transformation.

Graphical abstract

Diverse aromatic nitriles are relatively inexpensive, bench stable and commercially available.¹ Furthermore, the robust CN group is retained in many transition metal-catalyzed reactions, thereby enabling the late stage derivatization of this functionality in a multi-step synthesis.¹ Despite these advantages, only a few sporadic reports on biaryl synthesis using aromatic nitriles have been published.² These include Ni-catalyzed cross-couplings of aryl nitriles with aryl magnesium,^2b boron^2c or manganese reagents^2d. As part of our program on transition metal catalyzed arylations,³ this manuscript describes the hitherto unknown biaryl formation via Ni-catalyzed direct arylation using aromatic nitriles. These arylations expand the repertoire of electrophiles for Ni-catalyzed C–H arylations, which are thus far limited to the use of aryl halides, phenolic, and ester electrophiles.^4,5,6,7 Importantly, previous reports on Ni-catalyzed direct arylations demonstrate the preferential functionalization of the pivalate, carbamate or chloride leaving groups in the presence of an aryl nitrile.^5c,e Hence, the method described herein may enable selective arylations of the pivalate (or chloride and carbamate) and nitrile leaving groups at different stages in a multi-step synthesis. Furthermore, the arylations detailed in this manuscript are a timely contribution to the recent surge in interest for the development of transformations using earth-abundant metal catalysts.^4,8

The commonly proposed mechanism for nickel-catalyzed direct arylation involves three key steps: oxidative addition, C–H activation, and reductive elimination.^4,9 Although biaryl synthesis via C–H arylation using nitriles has remained elusive, the three elementary steps in the mechanism are individually documented using nickel complexes, thereby supporting the feasibility of the proposed arylations.¹⁰

Our studies commenced with the investigation of reaction parameters for the arylation of phenyl oxazole 1 with 3-trifluoromethylbenzonitrile. These explorations began with the use of catalyst/ligand combinations that have been previously successful for Ni-catalyzed direct arylations.^4,5 As shown in Table 1, product 1a is obtained in moderate yield using Ni(COD)₂/dcype catalyst combination in the presence of Cs₂CO₃ as the base and diglyme as the solvent at 140 or 160 °C (entries 1 and 2). Ligands similar to dcype afford lower or comparable yields of the desired product.¹¹ Use of 1.5 equivalents of either Cs₂CO₃ or K₃PO₄ affords similar yields of 1a (entries 1 and 3). However, increased amount of K₃PO₄ leads to enhanced yields of 1a (entries 3 and 4). Lower catalyst loadings result in diminished yields of 1a over 24 or 48 h (entries 5 and 6). Previous reports on Ni-catalyzed cross-couplings use Lewis acids to accelerate the transformations involving aryl nitriles.¹² Hence, a number of Lewis acids (CuF₂, ZnCl₂, Al-Me₂Cl, BPh₃) were evaluated for the formation of 1a. This study revealed that the use of BPh₃ (40 mol %) affords 1a with lower catalyst loadings (10 mol % Ni(COD)₂) in yields comparable to that obtained in the absence of BPh₃ with 20 mol % Ni(COD)₂ (cf. entries 4 and 10). Use of 40 or 60 mol % BPh₃ affords similar yields of 1a (entries 10 and 11). No product is obtained in the absence of Ni(COD)₂ both in the presence and absence of BPh₃ (entries 8, 9 and 12). To obtain preliminary insight into the role of BPh₃, we monitored the reaction profile of arylations with and without BPh₃ using 10 mol % Ni(COD)₂.11 These studies show that the reaction without BPh₃ levels off at about 30% yield of 1a after 1 h. However, the arylation with BPh₃ keeps progressing even after 3 h. Importantly, there is a very small difference in the initial rates of the two transformations. Taken together these data suggest that BPh₃ might be suppressing catalyst deactivation possibly via sequestering the cyanide anion. This hypothesis is consistent with the lower yields of 1a from the Ni-catalyzed reaction in the presence of KCN (entry 7).¹³

Table 1.

Optimization of Direct Arylation

entry	ligand	base	additive	yield of 1a (%)i
1	dcype	Cs2CO3	none	62
2a	dcype	Cs2CO3	none	59
3	dcype	K3PO4	none	59
4b	dcype	K3PO4	none	71
5b,c	dcype	K3PO4	none	26
6b,c,d	dcype	K3PO4	none	33
7b,e	dcype	K3PO4	KCN	23
8b,f	none	K3PO4	none	0
9b,f	dcype	K3PO4	none	0
10 b,c,g	dcype	K3PO4	BPh3	61
11 b,c,h	dcype	K3PO4	BPh3	63
12 b,f,g	none	K3PO4	BPh3	0

^aConducted at 160 °C.

^b3.0 equiv of K₃PO₄ used.

^cNi(COD)₂ (10 mol %) and dcype (11 mol %) used.

^dReaction time was 48 h.

^eKCN (1.0 equiv) added.

^fGeneral conditions, but with no Ni(COD)₂.

^g40 mol % BPh₃ used.

^h60 mol % BPh₃ used.

ⁱCalibrated GC yields against hexadecane as the internal standard.

The optimal conditions for the formation of 1a with or without BPh₃ (Table 1, entries 4 and 10) can be applied to the arylation of electronically varied oxazoles (Scheme 2). Electron-rich, electron-neutral, and electron-deficient oxazoles afford the corresponding products in comparable yields (cf. 1a–3a). Analogous to the formation of 1a (Table 1), the use of BPh₃ with 10 mol % Ni(COD)₂ affords the arylated products in comparable yields to that obtained using 20 mol % Ni(COD)₂ regardless of azole electronics. Importantly, no product is obtained in the absence of Ni(COD)₂/dcype regardless of the oxazole substrate.

Scheme 2.

Scope of Oxazoles^a

^[a]Isolated yields. ^[b] Ni(COD)₂ (10 mol %) and dcype (11 mol %), and BPh₃ (40 mol %) used.

We next investigated the use of benzoxazoles in these arylations. Not surprisingly, the Ni-catalyzed coupling of 5-methyl benzoxazole with 3-trifluoromethyl benzonitrile affords 11a in good yield using Cs₂CO₃ or K₃PO₄ (Table 2, entries 1–2).

Table 2.

Arylation of 5-methyl benzoxazole^a

entry	ligand	base	time (h)	yield of 11a (%)a
1	dcype	Cs2CO3	24	77
2	dcype	K3PO4	24	71
3b	dcype	Cs2CO3	24	48
4c	none	Cs2CO3	24	27
5c	none	K3PO4	24	23
6	dcype	Cs2CO3	8	79
7c	none	Cs2CO3	8	7
8 b,d	dcype	Cs2CO3	24	74
9 c,d	none	Cs2CO3	24	0

^aCalibrated GC yields against hexadecane as the internal standard.

^b10 mol % Ni(COD)₂ and 11 mol % dcype used.

^cIn the absence of Ni(COD)₂.

^d40 mol % BPh₃ added.

Unlike the coupling of phenyl oxazoles (Scheme 2), product 11a is formed in significant amounts (23–27% yield) ^14,15 in the absence of the nickel catalyst possibly via a S_NAr pathway (entries 4–5).¹⁶ The catalyzed arylation is significantly faster affording 11a in 79% yield over 8 h (versus 7% yield from the non-catalyzed process, cf. entries 6 and 7). These results suggest negligible contribution from the non-catalyzed process to the formation of 11a in the presence of the Ni(COD)₂/dcype. Analogous to the arylations detailed in Scheme 2, the use of BPh₃ (40 mol %) enables the use of lower catalyst loadings (cf. entries 1 and 8).¹¹ Interestingly, however, BPh₃ suppresses the non-catalyzed arylation (cf. entries 4 and 9).¹⁷

These trends for the catalyzed and the non-catalyzed arylations using meta-trifluoromethyl benzonitrile are observed for electronically varied benzoxazoles (Scheme 3). Analogous to arylations in Scheme 2, BPh₃ promotes these arylations regardless of benzoxazole electronics. The non-catalyzed arylation affords the arylated products in yields ranging between 14–68%. In the absence of Ni(COD)₂/dcype, the highest yield (68%) of the arylated product (17a) is observed for the electron-deficient, fluoro-substituted benzoxazole substrate. The extent of the background reaction is diminished for electron-rich and electron-neutral azoles. The nickel catalyst is essential for the formation of product 18a from the arylation of the less-acidic benzothiazole substrate.¹⁸

Scheme 3.

Scope of Benzoxazoles^a

^[a]Isolated yields ^[b] Yields for reactions in the absence of Ni(COD)₂ and dcype ^[c] Yields using BPh₃ (40 mol %) as additive, Ni(COD)₂ (10 mol %) and dcype (11 mol %). ^[d] Calibrated GC yields against hexadecane as the standard. ^[e] dcypt was used in place of dcype.

Electronically diverse aryl and heteroaryl nitriles couple with 5-methyl benzoxazole and oxazoles to afford the products in modest to good yields (Scheme 4). The lowest yield was obtained for the reaction using the ortho-methyl benzonitrile to afford 11j. Interestingly, the identity of the ortho substituent on the nitrile is important since 11h and 11i were obtained in significantly higher yields. Akin to the coupling of benzoxazoles with 3-trifluoromethyl benzonitrile (Scheme 3), the non-catalyzed arylation of benzoxazoles affords significant yields of the biaryls with a few nitriles (Scheme 4). While no clear trends emerge, the nickel catalyst is necessary for the formation of biaryls using nitriles containing an ester (11f) or bearing a methoxy group at the ortho- (11i) or para-positions (11g–h)) relative to the CN functionality. Furthermore, analogous to results in Scheme 2, no product (1b–1d, 1i, 1k) is obtained in the absence of Ni(COD)₂/dcype for arylations with phenyl oxazole regardless of the nitrile substrate. Finally, BPh₃ promotes the arylations of oxazoles and benzoxazoles with electron rich and electron deficient benzonitriles.

Scheme 4.

Scope of Nitriles^a

^[a]Isolated yields ^[b] dcypt used instead of dcype. ^[c] Yields for reactions in the absence of Ni(COD)₂ and dcype. ^[d] Yields using BPh₃ (40 mol %) as additive, Ni(COD)₂ (10 mol %), ligand (11 mol %). ^[e] Calibrated GC yields against hexadecane as the standard. ^[f] K₃PO₄ (3.0 equiv) used as the base.

In summary, this paper describes the first example of Ni-catalyzed arylation using electronically diverse azole and benzonitriles. The use of BPh₃ for these transformations enables the use of lower catalyst loading regardless of the azole or nitrile substrates. While the nickel catalyst is necessary for the arylation of oxazoles, oxadiazole and benzothiazole subtrates, arylations of benzoxazoles can proceed in the absence of Ni(COD)₂/dcype. The extent of the non-catalyzed process is dependent on the acidity of the azole and electronics nature of the benzonitrile substrates. Furthermore, the catalyzed process is significantly faster than the non-catalyzed reaction. Overall, the scope and efficiencies of the Ni-catalyzed arylation are higher than the non-catalyzed process.