Suzuki–Miyaura Cross-Coupling of Amides Using Well-Deﬁned, Air- and Moisture-Stable Nickel / NHC (NHC = N-Heterocyclic Carbene) Complexes

: In this Special Issue on N-Heterocyclic Carbenes and Their Complexes in Catalysis , we report the ﬁrst example of Suzuki–Miyaura cross-coupling of amides catalyzed by well-deﬁned, air- and moisture-stable nickel / NHC (NHC = N-heterocyclic carbene) complexes. The selective amide bond N–C(O) activation is achieved by half-sandwich, cyclopentadienyl [CpNi(NHC)Cl] complexes. The following order of reactivity of NHC ligands has been found: IPr > IMes > IPaul ≈ IPr*. Both the neutral and the cationic complexes are e ﬃ cient catalysts for the Suzuki–Miyaura cross-coupling of amides. Kinetic studies demonstrate that the reactions are complete in < 1 h at 80 ◦ C. Complete selectivity for the cleavage of exocyclic N-acyl bond has been observed under the experimental conditions. Given the utility of nickel catalysis in activating unreactive bonds, we believe that well-deﬁned and bench-stable [CpNi(NHC)Cl] complexes will ﬁnd broad application in amide bond and related cross-couplings of bench-stable acyl-electrophiles. evacuation / backﬁlling cycles under high vacuum. Toluene (to reach 0.25 M concentration) was added at room temperature, the reaction mixture was placed in a preheated oil bath at 80 ◦ C, and stirred at 80 ◦ C. After the indicated time, the reaction was cooled down, diluted with CH 2 Cl 2 (10 mL), ﬁltered, and concentrated. The sample was analyzed by 1 H NMR (CDCl 3 , 500 MHz) and GC-MS to obtain conversion, selectivity, and yield using internal standard and comparison with authentic samples. Unless stated otherwise, all compounds have been previously reported. All compounds have been quantiﬁed by 1 H NMR spectroscopy using nitromethane as internal standard (500 MHz, CD 3 Cl). All reactions have been carried out in microwave vials with heavy-wall, Type I, Class A borosilicate. These vials are designed to withstand pressures up to 300 PSI (20 bars) and are equivalent to Fisher-Porter tube.


Introduction
Nickel catalysis has recently garnered significant attention, enabling cleavage of unreactive bonds by this abundant 3D transition metal [1][2][3]. Simultaneously, major advances have been made in amide cross-coupling, wherein highly selective oxidative addition of the N-C(O) bond enables to exploit the traditionally unreactive amides as a novel class of acyl and aryl electrophiles [4][5][6][7][8][9][10]. This unconventional amide bond disconnection is particularly relevant in the view of common presence of amides in natural products, pharmaceuticals, and biopolymers, where the emergence of new catalytic methods has a potentially major impact on the way chemists perceive synthetic routes.
In this context, palladium/NHC (NHC = N-heterocyclic carbene) catalysis using well-defined Pd(II)-NHC precatalysts has been established as the dominant catalytic direction in activating amide N-C(O) bonds for acyl cross-coupling [4,[11][12][13][14]. However, to the best of our knowledge, there are no methods for the use of well-defined, air-and moisture-stable nickel/NHC complexes as efficient precatalysts in amide bond activation. In spite of the advances made by in situ formed Ni(0) catalysts, the lack of air-stability of Ni(cod) 2 severely limits the potential broad applications of the powerful Ni catalysis platform in amide bond activation [15][16][17].
In this Special Issue on N-Heterocyclic Carbenes and Their Complexes in Catalysis, we report the first example of Suzuki-Miyaura cross-coupling of amides catalyzed by well-defined, air-and moisture-stable nickel/NHC (NHC = N-heterocyclic carbene) complexes ( Figure 1). We were attracted to the recent elegant advances made in the design of half-sandwich, cyclopentadienyl [CpNi(NHC)X]

Results
We first examined the cross-coupling of N-acyl-glutarimides as model substrates for the cross-coupling with 4-tolylboronic acid using the readily prepared [CpNi(IPr)Cl] under various conditions (Table 1, Figure 2) (IPr = 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene). Optimization revealed that the desired cross-coupling proceeds in 85% yield in the presence of [CpNi(NHC)Cl] (10 mol%) as catalyst and K2CO3 (3.0 equivalent) as base in toluene as solvent at 80 °C using 4-Tol-B(OH)2 (3.0 equivalent) ( Table 1, entry 1). Interestingly, increasing the reaction temperature to 120 °C had only a minor effect on the cross-coupling (Table 1, entries 2-4). Furthermore, although previous studies suggested the beneficial effect of phosphine ligands on the Suzuki-Miyaura C(sp 2 )-C(sp 2 ) cross-coupling catalyzed by Ni-NHC complexes [32], in our case the addition of phosphine had an inhibitory effect on the cross-coupling (Table 1, entries 5-7). Examination of reaction parameters revealed K2CO3 as the optimal base and toluene as the preferred solvent (

Results
We first examined the cross-coupling of N-acyl-glutarimides as model substrates for the cross-coupling with 4-tolylboronic acid using the readily prepared [CpNi(IPr)Cl] under various conditions (Table 1, Figure 2) (IPr = 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene). Optimization revealed that the desired cross-coupling proceeds in 85% yield in the presence of [CpNi(NHC)Cl] (10 mol%) as catalyst and K 2 CO 3 (3.0 equivalent) as base in toluene as solvent at 80 • C using 4-Tol-B(OH) 2 (3.0 equivalent) ( Table 1, entry 1). Interestingly, increasing the reaction temperature to 120 • C had only a minor effect on the cross-coupling (Table 1, entries 2-4). Furthermore, although previous studies suggested the beneficial effect of phosphine ligands on the Suzuki-Miyaura C(sp 2 )-C(sp 2 ) cross-coupling catalyzed by Ni-NHC complexes [32], in our case the addition of phosphine had an inhibitory effect on the cross-coupling (Table 1, entries 5-7). Examination of reaction parameters revealed K 2 CO 3 as the optimal base and toluene as the preferred solvent ( abstraction with KPF6 according to the procedure Chetcuti [18] showed promising reactivity ( Table  1, entries [23][24], indicating potential application of this class of cationic Ni-NHC catalysts in amide bond cross-coupling in the future. Further, we were particularly interested in evaluating steric demand of NHC ligands on the performance of [CpNi(NHC)Cl] complexes in amide cross-coupling [36,37]. We found that [CpNi(IMes)Cl] is slightly less reactive than [CpNi(IPr)Cl] ( Table 1, entries [25][26]. Furthermore, examination of the highly attractive class of bulky but flexible NHC ligands, IPaul [38] and IPr* [39] revealed [CpNi(IPaul)Cl] and [CpNi(IPr*)Cl] as promising catalysts for N-C bond activation. Of note, [CpNi(IPaul)Cl] is commercially-available, which should facilitate the discovery of future cross-couplings of amide bonds mediated by this precatalyst.  With the optimized catalyst system in hand, we examined the scope of this Suzuki-Miyaura cross-coupling catalyzed by well-defined Ni(II)-NHC precatalysts (Tables 2 and 3, and see supporting information). As shown, the reaction was compatible with electron-donating groups on the boronic acid (3a-c). Steric-hindrance at the ortho-position of the boronic acid was well-tolerated (3d-e). Furthermore, fluorine functionalized boronic acids, such as 3-fluoro and 3-trifluoromethyl (3f-g) could be introduced by this Ni-catalyzed approach. We were further pleased that conjugated arenes, such as naphthalene and biphenyl delivered the desired biaryl ketone products in good yields (3h-i). Only one aliphatic boronic acid was tested, and it was incompatible with the reaction conditions (entry 10). In terms of the amide scope, pleasingly, electron-rich and electron-withdrawing groups were well-tolerated on the amide component (3a,c,j), while the electron-deficient amides appeared to be more reactive (vide infra). Steric hindrance on the ortho-position of the amide was tolerated, albeit it exerted a more pronounced effect than on the boronic acid, consistent with a decreased amide bond twist by ortho-substitution (3d). Furthermore, fluorine-containing amides and heterocyclic amides provided the desired products in good yields (3k-l). It is noteworthy that decarbonylation to give Ar-Ni after loss of CO was not observed [40], consistent with the stability of acyl-Ni(NHC) intermediate.
Next, intermolecular competition experiments were conducted to gain preliminary insight into the reaction (Schemes 1-2). As shown, competitions revealed electron-deficient amides to be significantly more reactive than electron-rich amides (Scheme 1, CF3:MeO = 93:7). In contrast, a comparable reactivity of electron-rich and electron-deficient boronic acids was observed (Scheme 2, MeO:CF3 = 58:42). These preliminary studies are consistent with oxidative addition of the N-C(O) bond as the rate limiting step of the reaction [41]. Further studies on the mechanism are ongoing. With the optimized catalyst system in hand, we examined the scope of this Suzuki-Miyaura cross-coupling catalyzed by well-defined Ni(II)-NHC precatalysts (Tables 2 and 3, and see Supporting Information). As shown, the reaction was compatible with electron-donating groups on the boronic acid (3a-c). Steric-hindrance at the ortho-position of the boronic acid was well-tolerated (3d-e). Furthermore, fluorine functionalized boronic acids, such as 3-fluoro and 3-trifluoromethyl (3f-g) could be introduced by this Ni-catalyzed approach. We were further pleased that conjugated arenes, such as naphthalene and biphenyl delivered the desired biaryl ketone products in good yields (3h-i). Only one aliphatic boronic acid was tested, and it was incompatible with the reaction conditions (entry 10). In terms of the amide scope, pleasingly, electron-rich and electron-withdrawing groups were well-tolerated on the amide component (3a,c,j), while the electron-deficient amides appeared to be more reactive (vide infra). Steric hindrance on the ortho-position of the amide was tolerated, albeit it exerted a more pronounced effect than on the boronic acid, consistent with a decreased amide bond twist by ortho-substitution (3d). Furthermore, fluorine-containing amides and heterocyclic amides provided the desired products in good yields (3k-l). It is noteworthy that decarbonylation to give Ar-Ni after loss of CO was not observed [40], consistent with the stability of acyl-Ni(NHC) intermediate.
Next, intermolecular competition experiments were conducted to gain preliminary insight into the reaction (Schemes 1 and 2). As shown, competitions revealed electron-deficient amides to be significantly more reactive than electron-rich amides (Scheme 1, CF 3 :MeO = 93:7). In contrast, a comparable reactivity of electron-rich and electron-deficient boronic acids was observed (Scheme 2, MeO:CF 3 = 58:42). These preliminary studies are consistent with oxidative addition of the N-C(O) bond as the rate limiting step of the reaction [41]. Further studies on the mechanism are ongoing.        Kinetic studies were performed to gain insight into the reaction profile ( Figure 3). As shown, the reaction reached 75% conversion after 5 min, while 86% and >95% conversion was observed after 30 and 60 min, respectively, consistent with efficient generation of the reactive Ni(0)-NHC catalyst [40,41] under the reaction conditions (TON = 8.5, 10 mol%; TOF = 1.5 min -1 ). Studies on the mechanism are underway and will be reported in due course.
Finally, we were interested to probe the effect of different acyl leaving groups on the cross-coupling (Scheme 3). N-Acyl-glutarimides have emerged as the go-to amides to develop new cross-coupling methods by N-C activation. Furthermore, the present coupling is compatible with N-sulfonyl activation in acyclic amides, such as N,N-Ph/Ts, and N-acyl-succinimides, albeit the cross-coupling product was obtained in lower yield under the present conditions. In contrast, N-Boc-carbamates, were recovered unchanged from the reaction conditions, indicating a potential for chemoselective coupling. Typically, N-Ts amides and N-acyl-succinimides are consumed under the reaction conditions, while other electrophiles were recovered unchanged. Moreover, the C-O cross-coupling is also feasible under the present conditions as demonstrated by the cross-coupling of Opfp ester (pfp = pentafluorophenyl) [42,43]. In contrast, the unactivated phenolic ester was recovered unchanged, consistent with a considerable potential of [CpNi(NHC)Cl] catalysts in chemoselective activation of C(acyl)-O electrophiles. Kinetic studies were performed to gain insight into the reaction profile ( Figure 3). As shown, the reaction reached 75% conversion after 5 min, while 86% and >95% conversion was observed after 30 and 60 min, respectively, consistent with efficient generation of the reactive Ni(0)-NHC catalyst [40,41] under the reaction conditions (TON = 8.5, 10 mol%; TOF = 1.5 min -1 ). Studies on the mechanism are underway and will be reported in due course.

Scheme 2. Competition experiments-boronic acids.
Kinetic studies were performed to gain insight into the reaction profile ( Figure 3). As shown, the reaction reached 75% conversion after 5 min, while 86% and >95% conversion was observed after 30 and 60 min, respectively, consistent with efficient generation of the reactive Ni(0)-NHC catalyst [40,41] under the reaction conditions (TON = 8.5, 10 mol%; TOF = 1.5 min -1 ). Studies on the mechanism are underway and will be reported in due course.
Finally, we were interested to probe the effect of different acyl leaving groups on the cross-coupling (Scheme 3). N-Acyl-glutarimides have emerged as the go-to amides to develop new cross-coupling methods by N-C activation. Furthermore, the present coupling is compatible with N-sulfonyl activation in acyclic amides, such as N,N-Ph/Ts, and N-acyl-succinimides, albeit the cross-coupling product was obtained in lower yield under the present conditions. In contrast, N-Boc-carbamates, were recovered unchanged from the reaction conditions, indicating a potential for chemoselective coupling. Typically, N-Ts amides and N-acyl-succinimides are consumed under the reaction conditions, while other electrophiles were recovered unchanged. Moreover, the C-O cross-coupling is also feasible under the present conditions as demonstrated by the cross-coupling of Opfp ester (pfp = pentafluorophenyl) [42,43]. In contrast, the unactivated phenolic ester was recovered unchanged, consistent with a considerable potential of [CpNi(NHC)Cl] catalysts in chemoselective activation of C(acyl)-O electrophiles.  Finally, we were interested to probe the effect of different acyl leaving groups on the cross-coupling (Scheme 3). N-Acyl-glutarimides have emerged as the go-to amides to develop new cross-coupling methods by N-C activation. Furthermore, the present coupling is compatible with N-sulfonyl activation in acyclic amides, such as N,N-Ph/Ts, and N-acyl-succinimides, albeit the cross-coupling product was obtained in lower yield under the present conditions. In contrast, N-Boc-carbamates, were recovered unchanged from the reaction conditions, indicating a potential for chemoselective coupling. Typically, N-Ts amides and N-acyl-succinimides are consumed under the reaction conditions, while other electrophiles were recovered unchanged. Moreover, the C-O cross-coupling is also feasible under the present conditions as demonstrated by the cross-coupling of Opfp ester (pfp = pentafluorophenyl) [42,43]. In contrast, the unactivated phenolic ester was recovered unchanged, consistent with a considerable potential of [CpNi(NHC)Cl] catalysts in chemoselective activation of C(acyl)-O electrophiles.

Discussion
In summary, we have reported the first example of Suzuki-Miyaura cross-coupling of amides catalyzed by well-defined, air-and moisture-stable nickel/NHC complexes. The reaction delivers biaryl ketones in good yields using inexpensive nickel catalyst with excellent N-C(O) cleavage selectivity cf. endocylic amide bond and acyl vs. decarbonylative coupling. In a broad sense, this report establishes the capacity of highly attractive half-sandwich [CpNi(NHC)Cl] complexes as catalysts for activation of amide N-C(O) bonds. Furthermore, we have established the order of reactivity of NHC ligands in [CpNi(NHC)Cl] complexes as IPr > IMes > IPaul ≈ IPr*, and showed that both neutral and cationic complexes serve as efficient catalysts for amide bond cross-coupling. Reaction profile studies demonstrated that these reactions are complete in < 1 h at 80 °C. In a broader context, the present method should be evaluated in comparison with other known approaches to biaryl ketones from amides [3][4][5][6][7][8][9][10] and acyl electrophiles [15]. The use of Ni catalysis [1][2][3] and the beneficial performance of Ni-NHC complexes [25][26][27][28][29] may accelerate the development of new approaches to activating amide bonds. Considering the utility of nickel catalysis in activation of unreactive bonds, we anticipate that [CpNi(NHC)Cl] complexes will be of interest in activation of bench-stable acyl electrophiles. Further mechanistic studies, as well as efforts to expand the scope of electrophiles in cross-coupling catalyzed by well-defined Ni-NHC complexes are ongoing.

General Information
General methods have been published (See Supporting Information) [11].

General Procedure for [CpNi(IPr)Cl] Catalyzed Cross-Coupling of Amides
In a typical cross-coupling procedure, an oven-dried vial was charged with an amide substrate (neat, 1.0 equivalent), boronic acid (typically, 3.0 equivalent), potassium carbonate (typically, 3.0 equivalent), [CpNi(NHC)Cl] (typically, 10 mol%), placed under a positive pressure of argon or nitrogen, and subjected to three evacuation/backfilling cycles under high vacuum. Toluene (to reach 0.25 M concentration) was added at room temperature, the reaction mixture was placed in a preheated oil bath at 80 °C, and stirred at 80 °C. After the indicated time, the reaction was cooled down, diluted with CH2Cl2 (10 mL), filtered, and concentrated. The sample was analyzed by 1 H NMR (CDCl3, 500 MHz) and GC-MS to obtain conversion, selectivity, and yield using internal standard and comparison with authentic samples. Unless stated otherwise, all compounds have been previously reported. All compounds have been quantified by 1 H NMR spectroscopy using nitromethane as internal standard (500 MHz, CD3Cl). All reactions have been carried out in microwave vials with heavy-wall, Type I, Class A borosilicate. These vials are designed to withstand pressures up to 300 PSI (20 bars) and are equivalent to Fisher-Porter tube.

Discussion
In summary, we have reported the first example of Suzuki-Miyaura cross-coupling of amides catalyzed by well-defined, air-and moisture-stable nickel/NHC complexes. The reaction delivers biaryl ketones in good yields using inexpensive nickel catalyst with excellent N-C(O) cleavage selectivity cf. endocylic amide bond and acyl vs. decarbonylative coupling. In a broad sense, this report establishes the capacity of highly attractive half-sandwich [CpNi(NHC)Cl] complexes as catalysts for activation of amide N-C(O) bonds. Furthermore, we have established the order of reactivity of NHC ligands in [CpNi(NHC)Cl] complexes as IPr > IMes > IPaul ≈ IPr*, and showed that both neutral and cationic complexes serve as efficient catalysts for amide bond cross-coupling. Reaction profile studies demonstrated that these reactions are complete in < 1 h at 80 • C. In a broader context, the present method should be evaluated in comparison with other known approaches to biaryl ketones from amides [3][4][5][6][7][8][9][10] and acyl electrophiles [15]. The use of Ni catalysis [1][2][3] and the beneficial performance of Ni-NHC complexes [25][26][27][28][29] may accelerate the development of new approaches to activating amide bonds. Considering the utility of nickel catalysis in activation of unreactive bonds, we anticipate that [CpNi(NHC)Cl] complexes will be of interest in activation of bench-stable acyl electrophiles. Further mechanistic studies, as well as efforts to expand the scope of electrophiles in cross-coupling catalyzed by well-defined Ni-NHC complexes are ongoing.

General Information
General methods have been published (See Supporting Information) [11].

General Procedure for [CpNi(IPr)Cl] Catalyzed Cross-Coupling of Amides
In a typical cross-coupling procedure, an oven-dried vial was charged with an amide substrate (neat, 1.0 equivalent), boronic acid (typically, 3.0 equivalent), potassium carbonate (typically, 3.0 equivalent), [CpNi(NHC)Cl] (typically, 10 mol%), placed under a positive pressure of argon or nitrogen, and subjected to three evacuation/backfilling cycles under high vacuum. Toluene (to reach 0.25 M concentration) was added at room temperature, the reaction mixture was placed in a preheated oil bath at 80 • C, and stirred at 80 • C. After the indicated time, the reaction was cooled down, diluted with CH 2 Cl 2 (10 mL), filtered, and concentrated. The sample was analyzed by 1 H NMR (CDCl 3 , 500 MHz) and GC-MS to obtain conversion, selectivity, and yield using internal standard and comparison with authentic samples. Unless stated otherwise, all compounds have been previously reported. All compounds have been quantified by 1 H NMR spectroscopy using nitromethane as internal standard (500 MHz, CD 3 Cl). All reactions have been carried out in microwave vials with heavy-wall, Type I, Class A borosilicate. These vials are designed to withstand pressures up to 300 PSI (20 bars) and are equivalent to Fisher-Porter tube.