catalysts Palladium-Catalyzed Dehydrogenative C-2 Alkenylation of 5-Arylimidazoles and Related Azoles with Styrenes

: The construction of carbon–carbon bonds by direct involvement of two unactivated carbon–hydrogen bonds, without any directing group, ensures a high atom economy of the en-tire process. Here, we describe a simple protocol for the Pd(II)/Cu(II)-promoted intermolecular cross-dehydrogenative coupling (CDC) of 5-arylimidazoles, benzimidazoles, benzoxazole and 4,5-diphenylimidazole at their C-2 position with functionalized styrenes. This speciﬁc CDC, known as the Fujiwara–Moritani reaction or oxidative Heck coupling, also allowed the C-4 alkenylation of the imidazole nucleus when both 2 and 5 positions were occupied. ( 2a ) the experimental conditions of entry 7, Table 1. Hence, 18 and 2.0 equiv of 2a were reacted in the presence of 5 mol% Pd(OAc) 2 and 3.0 equiv of Cu(OAc) 2 in EtCOOH (Method A , Scheme 7). To our delight, after stirring at 120 ◦ C for 24 h, the GLC conversion of 18 was 83%, and we were able to isolate ( E )-1-methyl-2-styryl-1 H -imidazole ( 19 ) in 45% yield conﬁrming the expected C-2 selectivity.


Introduction
Transition metal-catalyzed carbon-carbon bond-forming reactions that occur by the breaking of carbon-hydrogen bonds are attracting increasing interest in modern synthetic organic chemistry since this approach does not require any pre-activation of the starting materials [1][2][3][4][5][6][7][8][9][10][11]. When compared with conventional cross-coupling methodologies that require the use of organic halides and/or preformed organometallic reagents, this strategy, known as cross dehydrogenative coupling (CDC), allows the obtainment of a high degree of atom economy and structural complexity in the target molecule, while ensuring high chemoselectivity. In addition, unlike traditional cross-couplings, the possibility of avoiding the use of metals and halogens in stoichiometric quantities reduces the production of inorganic waste.
Although this reaction was first reported in 1967 [12], and thus historically precedes the development of the Mizoroki-Heck alkenylation [18,19], problems related to poor regioselectivity and the need to use oxidants have in the past limited its application in favor of both the aforementioned Mizoroki-Heck alkenylation and the traditional cross-coupling procedures, and also the most recent direct alkenylation of aromatic C-H bonds, also catalyzed by transition metals, involving alkenyl halides [20][21][22][23][24][25][26]. If problems associated with the use of oxidants in stoichiometric quantities can be overcome by the latest electrochemical approaches [27][28][29][30], the achievement of high regioselectivity is still often an issue to be solved. In this regard, however, it is important to note that when the reaction is conducted using a heteroarene as a partner, the presence of one or more heteroatoms leads to an innate distinction among the different C-H bonds, thus allowing, with appropriate optimization of the reaction conditions, the selective involvement of a specific Csp2-H bond.
Due to our continuous interest in the development of methods for the palladiumcatalyzed regioselective C-H functionalization of azoles and in their application to the preparation of new organic materials [31][32][33][34][35][36], we recently decided to evaluate the Fujiwara-Moritani reaction as an atom economy way to achieve the preparation of styryl-substituted imidazoles. Our interest in this investigation was also given by the fact that while several procedures are reported for the dehydrogenative alkenylation of indoles, pyrroles, and oxazoles [4,11], to the best of our knowledge only two papers reported the synthesis of styryl-substituted imidazoles by dehydrogenative alkenylation, both using only unfunctionalized styrene as the coupling partner.
Due to our continuous interest in the development of methods catalyzed regioselective C-H functionalization of azoles and in thei preparation of new organic materials [31][32][33][34][35][36], we recently decided t wara-Moritani reaction as an atom economy way to achieve the pr substituted imidazoles. Our interest in this investigation was also gi while several procedures are reported for the dehydrogenative alke pyrroles, and oxazoles [4,11], to the best of our knowledge only two synthesis of styryl-substituted imidazoles by dehydrogenative alken only unfunctionalized styrene as the coupling partner.
In 2018, Joo and co-workers described a protocol for the regiosel tion of 1-substituted imidazoles (Scheme 2) [38]. The optimized cond coupling of 1-methylimidazole with 2.0 equiv of styrene in the presen ladium(II) acetate (Pd(OAc)2, 2.0 equiv potassium pivalate (KOPiv) cetamide (DMA) at 120 °C for 24 h under an oxygen atmosphere. Wh imidazoles were used as the reaction partners, the authors found it b coupling using copper (II) acetate (Cu(OAc)2) as the stoichiometric ox ygen, in dioxane at 100 °C for 15 h.
In 2018, Joo and co-workers described a protocol for the regioselective C-5 alkenylation of 1-substituted imidazoles (Scheme 2) [38]. The optimized conditions involved the coupling of 1-methylimidazole with 2.0 equiv of styrene in the presence of 10 mol% palladium(II) acetate (Pd(OAc) 2 , 2.0 equiv potassium pivalate (KOPiv) in N,N-dimethylacetamide (DMA) at 120 • C for 24 h under an oxygen atmosphere. When 1,2-disubstituted imidazoles were used as the reaction partners, the authors found it better to perform the coupling using copper (II) acetate (Cu(OAc) 2 ) as the stoichiometric oxidant instead of oxygen, in dioxane at 100 • C for 15 h.
In this paper, we are pleased to summarize our efforts in finding an effective and simple protocol for the dehydrogenative alkenylation of imidazole derivatives, which allowed us to develop a simple procedure for the dehydrogenative alkenylation of 5-aryl-1methylimidazoles and some related azoles with functionalized styrenes (Scheme 3). The optimized reaction conditions involve the use propanoic acid as the solvent at 120 • C, in the presence Pd(OAc) 2 as the pre-catalyst and Cu(OAc) 2 as the oxidant.
In this paper, we are pleased to summarize our efforts in f simple protocol for the dehydrogenative alkenylation of imidazo lowed us to develop a simple procedure for the dehydrogenativ 1-methylimidazoles and some related azoles with functionalized optimized reaction conditions involve the use propanoic acid as the presence Pd(OAc)2 as the pre-catalyst and Cu(OAc)2 as the ox Scheme 3. Our protocol for the cross-dehydrogenative Pd(II)/Cu(II)-med azoles.

Screening of the Reaction Conditions
At the onset of our study, we decided to test the efficiency trying a dehydrogenative alkenylation of 5-(4-methoxyphenyl) (1a) with styrene (2a), chosen as model reaction partners. Hence were reacted in the presence of 10 mol% Pd(TFA)2, 15 mol% 1,10 TFA (Scheme 4). After stirring the reaction mixture for 16 h at 130 °C in toluen GLC conversion of 1a was observed. Moreover, the required alke
Catalysts 2021, 11, x In this paper, we are pleased to summarize our efforts in finding an effect simple protocol for the dehydrogenative alkenylation of imidazole derivatives, w lowed us to develop a simple procedure for the dehydrogenative alkenylation o 1-methylimidazoles and some related azoles with functionalized styrenes (Scheme optimized reaction conditions involve the use propanoic acid as the solvent at 12 the presence Pd(OAc)2 as the pre-catalyst and Cu(OAc)2 as the oxidant.
Interestingly, under acidic conditions the formation of the 2-alkyimidazole 4 was not observed in the crude reaction mixture, proving that propionic acid as solvent cleanly increases the regioselectivity of the carbopalladation step of the mechanistic pathway (see later). Interestingly, under acidic conditions the formation of the 2-alkyimidazole 4 was not observed in the crude reaction mixture, proving that propionic acid as solvent cleanly increases the regioselectivity of the carbopalladation step of the mechanistic pathway (see later).
With the aim of evaluating the influence of the carboxylic acid, we then carried out the Ag(I)-promoted coupling using acetic acid and pivalic acid as the reaction solvent (entries 2 and 3, Table 1). However, both the acidic solvents revealed less effectiveness in promoting the alkenylation when compared with propionic acid, scoring 24 and 23% GLC yields, respectively. As recently reported [40], the efficiency of C-H activation reactions carried out using palladium catalysts with carboxylate ligands strictly depends on the pKa of the carboxylic acid used as the solvent. It is in fact necessary to find a balance between the generation of an active catalyst and the N-3 protonation of the imidazole nucleus with its consequent deactivation. In our case, the pKa of propionic acid (4.87) is intermediate between that of pivalic acid (5.05) and acetic acid (4.76), which means that acetic acid gave a higher percentage of unreactive imidazolium salt, while pivalic acid is not enough acid to generate an active catalyst.
Notably, while the use of silver(I) salts different from AgOAc gave GLC yields ranging from 32 to 38% (entries 4-6, Table 1), when the alkenylation was performed in the presence of 3.0 equiv of copper(II) acetate (Cu(OAc) 2 ), a relevant increase in the GLC yield of 3a was observed, and the C-2 alkenylated product was isolated in a satisfactory 56% yield (entry 7, Table 1). From the crude reaction mixture, we were also able to isolate the side-product 5a in a 13% yield. Interestingly, under acidic conditions the formation of the 2-alkyimidazole 4 was not observed in the crude reaction mixture, proving that propionic acid as solvent cleanly increases the regioselectivity of the carbopalladation step of the mechanistic pathway (see later).  3 AP% is the area percent of the products in the GLC chromatogram. AP% values are uncorrected for the differences in GLC response factors. 4 Compound 5a was also isolated in 13% yield. 5 This reaction was carried out using 0.5 mmol of 2a. 6 This reaction was performed at 80 °C (oil bath temperature). 7 This reaction was carried out under a dioxygen atmosphere. 8 This reaction was carried out using PdCl2 (0.025 mmol) as pre-catalyst. 9 This reaction was carried out using Pd(acac)2 (0.025 mmol) as pre-catalyst. In an attempt to reduce the amount of undesired double alkenylated imidazole 5a, we lowered the amount of styrene to 1.0 equiv, but a parallel lowering of the 3a yield without a significant increase in the selectivity was observed (entry 8, Table 1).
Lowering the reaction temperature from 120 • C to 80 • C led to a complete recovery of the reactants (entry 9, Table 1), and a similar negative result was observed when the coupling was performed under a dioxygen atmosphere (entry 10, Table 1).
None of the other typical copper(II) salts tested gave results comparable with that obtained when Cu(OAc) 2 was used. CuO gave 3a in 31% GLC yield (entry 11, Table 1), while no reaction was observed when CuCl 2 was employed as the oxidant (entry 12, Table 1). The use of two typical organic oxidants, i.e., NMO and PhI(OAc) 2 gave unsatisfactory results (entries 13 and 14, Table 1).
In order to reduce the amount of propionic acid, we tried also the alkenylation involving 1a and 2a using 1:1 (v:v) mixtures of propionic acid with, respectively, DMF or NMP (entries 17-18, Table 1), but the presence of an organic solvent depletes the formation of the required alkenylimidazole 3a.

Scope of the Pd-Catalyzed Dehydrogenative Alkenylation of Imidazoles and Related Azoles
Considering the results of the preliminary screening, the scope and limitations of this regioselective C-2 dehydrogenative alkenylation under the experimental conditions of entry 7, Table 1, were then evaluated by us. Hence, 5-aryl-1-methylimidazoles 1a-f and styrenes 2a-g ( Figure 1) were reacted in the presence of 5 mol% Pd(OAc) 2 and 3.0 equiv of Cu(OAc) 2 in EtCOOH at 120 • C ( Table 2).
That electron-poor substituents negatively influence the coupling is evidenced also when styrenes 2f and 2g were employed as reaction partners, and it seems synergic with the effect exerted by electron-withdrawing groups at C-5 on the imidazole counterpart. Actually, while an acceptable 44% isolate yield was observed when imidazole 1a was reacted with 4-nitrostyrene 2g (entry 6, Table 2), a more significant reduction in the chemical yield was recorded when the 4-chlorophenyl substituted imidazole 1c reacted with the electron-poor styrenes 2f and 2g (entries 14 and 15, Table 2).  1 Reaction conditions: 1 (0.5 mmol), 2 (1.0 mmol), Pd(OAc) 2 (0.025 mmol), Cu(OAc) 2 (1.5 mmol), EtCOOH (5.0 mL) for 24 h at 120 • C (oil bath temperature) under an argon atmosphere, unless otherwise reported. 2 Isolated yield. 3 After 24 h the GLC conversion of 1 was >95% unless otherwise noted. 4 AP% is the area percent of the products in the GLC chromatogram. AP% values are uncorrected for the differences in GLC response factors. 5 The GLC conversion of 1c was 49%. 6 The coupling was carried out for 72 h. 7 The GLC conversion of 1c was 45%. 8 The GLC conversion of 1c was 56%. 9 The GLC conversion of 1e was 53%.
That electron-poor substituents negatively influence the coupling is evidenced also when styrenes 2f and 2g were employed as reaction partners, and it seems synergic with the effect exerted by electron-withdrawing groups at C-5 on the imidazole counterpart. Actually, while an acceptable 44% isolate yield was observed when imidazole 1a was reacted with 4-nitrostyrene 2g (entry 6, Table 2), a more significant reduction in the chemical yield was recorded when the 4-chlorophenyl substituted imidazole 1c reacted with the electron-poor styrenes 2f and 2g (entries 14 and 15, Table 2).
Regarding the results summarized in Table 2, it is also important to note that the efficiency of the coupling strongly depends also on the relative stability of the substituted styrenes 2 in the acid medium. It is in fact well known that electron-rich styrenes, such as 4-methoxystyrene 2b, are highly susceptible to polymerization in an acidic environment, while electron-poor analogues such as 4-nitrostyrene 2g are almost inert under the same conditions [41]. For this reason, it is not possible to make a clear correlation between the nature of the coupling partners 1 and 2 and the observed isolated yields of compounds 3.
We were pleased to find that the reaction conditions summarized in Table 1, entry 7, are also well suited for the C-2 dehydrogenative alkenylation of 1-methyl-1H-benzimidazole (6) and 1H-benzimidazole (7). As summarized in Scheme 5, 1-methyl-2-styrylbenzimidazole 8 and 2-styrylbenzimidazole 9 were isolated in a satisfactory 87 and 64% yield, respectively. In contrast, the reaction involving benzoxazole 10 with styrene gave the required 2-styrylbenzoxazole 11 in a lower isolated yield (30%) (Scheme 5), while no product was observed when benzothiazole was submitted to the dehydrogenative alkenylation (result not shown).
while electron-poor analogues such as 4-nitrostyrene 2g are alm conditions [41]. For this reason, it is not possible to make a clea nature of the coupling partners 1 and 2 and the observed isolate We were pleased to find that the reaction conditions summ are also well suited for the C-2 dehydrogenative alkenylation o azole (6) and 1H-benzimidazole (7). As summarized in Scheme zimidazole 8 and 2-styrylbenzimidazole 9 were isolated in a satis respectively. In contrast, the reaction involving benzoxazole 10 quired 2-styrylbenzoxazole 11 in a lower isolated yield (30%) (S uct was observed when benzothiazole was submitted to the dehy (result not shown).
Considering also that the C4-H bond seems to be reactive when the other two positions on the imidazole ring are occupied due to the formation of side-products 5, we also tried to force the C-4 alkenylation by using 5-(4-methoxyphenyl)-1,2-dimethyl-1H-imidazole (16) as a typical 2,5-disubstituted imidazole. Fortunately, when the reaction was carried out using 2-methyl substituted imidazole 16 and 5.0 equiv of 2a, the expected C4-alkenylated imidazole 17 was recovered in a 50% isolated yield (Scheme 6).

of 18
GLC conversion of 18 was 83%, and we were able to isolate (E)-1-methyl-2-styryl-1H-imidazole (19) in 45% yield confirming the expected C-2 selectivity. In contrast, when the same coupling was carried out using the Ong protocol [37], i.e., reacting 16 and 2.5 equiv of 2a in a closed vessel for 16 h at 130 °C in toluene in the presence of 10 mol% Pd(TFA)2, 15 mol% 1,10-Phen and 2.0 equiv AgTFA, the GLC conversion of 16 was lower (70%), and the required imidazole 19 was observed in only 33% isolated yield (vs. a reported 67% yield [37]) (Method B, Scheme 7). It is worth mentioning that also in this case GLC-MS analysis of the crude reaction mixture evidences the presence of the side-product 22, a structural analogue to imidazole 4 already observed when the same reaction was performed with 1-methylbenzimidazole 1a (Scheme 4), in a 77:23 GLC ratio with 19.
Based on previous reports [2,11,13,42,43] and according to the results described here, a plausible reaction mechanism is summarized in Figure 2. In contrast, when the same coupling was carried out using the Ong protocol [37], i.e., reacting 16 and 2.5 equiv of 2a in a closed vessel for 16 h at 130 • C in toluene in the presence of 10 mol% Pd(TFA) 2 , 15 mol% 1,10-Phen and 2.0 equiv AgTFA, the GLC conversion of 16 was lower (70%), and the required imidazole 19 was observed in only 33% isolated yield (vs. a reported 67% yield [37]) (Method B, Scheme 7). It is worth mentioning that also in this case GLC-MS analysis of the crude reaction mixture evidences the presence of the side-product 22, a structural analogue to imidazole 4 already observed when the same reaction was performed with 1-methylbenzimidazole 1a (Scheme 4), in a 77:23 GLC ratio with 19.
As already noted for Pd/Cu-mediated direct arylation reactions of 1,3-azoles with aryl halides [44][45][46], it is thought that an initial N-3 protonation or complexation with copper enhances the acidity of the C2-H bond, allowing a fast and regioselective palladation to give the imidazole intermediate A. The subsequent regioselective carbopalladation yields the intermediate B, which decomposes through β-elimination to generate the desired product 3 and Pd(0). Finally, the reoxidation of Pd(0) to Pd(II) by Cu(II) closed the catalytic cycle.
Based on previous reports [2,11,13,42,43] and according to the results described here, a plausible reaction mechanism is summarized in Figure 2. yield (vs. a reported 67% yield [37]) (Method B, Scheme 7). It is worth mentioning that also in this case GLC-MS analysis of the crude reaction mixture evidences the presence of the side-product 22, a structural analogue to imidazole 4 already observed when the same reaction was performed with 1-methylbenzimidazole 1a (Scheme 4), in a 77:23 GLC ratio with 19.
Based on previous reports [2,11,13,42,43] and according to the results described here, a plausible reaction mechanism is summarized in Figure 2. As already noted for Pd/Cu-mediated direct arylation reactions of 1,3-azoles with aryl halides [44][45][46], it is thought that an initial N-3 protonation or complexation with copper enhances the acidity of the C2-H bond, allowing a fast and regioselective palladation to give the imidazole intermediate A. The subsequent regioselective carbopalladation

Procedure for the Screening of the Reaction Conditions for the Pd-Catalyzed Dehydrogenative C2-Alkenylation of 5-(4-Methoxyphenyl)-1-Methyl-1H-Imidazole (1a) with Styrene (2a) Using Carboxylic Acids as Reaction Solvents
A mixture of 5-(4-methoxyphenyl)-1-methyl-1H-imidazole (1a) (94 mg, 0.5 mmol), styrene (2a) (0.12 mL, 104 mg, 1.0 mmol), palladium pre-catalyst (0.025 mmol), oxidant (1.5 mmol), in the selected solvent (5 mL) was stirred for 24 h at 120 • C. After cooling to room temperature, when an Ag(I) oxidant was used the crude reaction mixture was diluted with AcOEt, and PPh 3 was added as internal standard. When a Cu(II) salt was used as oxidant, the crude reaction mixture was diluted with AcOEt and poured into a saturated aqueous NH 4 Cl solution. The resulting mixture was basified with a few drops of aqueous NH 4 OH, stirred in the open air for 0.5 h, and then extracted with AcOEt and with CH 2 Cl 2 . The organic extract was washed with water, dried, filtered, and PPh 3 was added as internal standard.
All the resulting mixtures were analyzed by GLC, GC-MS, and UPLC-MS. Table 1 summarizes the results of this screening.

(E)-5-(4-Methoxyphenyl)-2-(4-methoxystyryl)-1-methyl-1H-imidazole (3b)
The crude reaction product, which was obtained by Pd-catalyzed reaction of 1a with 2b (entry 2, Table 2 The crude reaction product, which was obtained by Pd-catalyzed reaction of 1a with 2c (entry 3, Table 2 The crude reaction product, which was obtained by Pd-catalyzed reaction of 1a with 2d (entry 4, Table 2 The crude reaction product, which was obtained by Pd-catalyzed reaction of 1a with 2e (entry 5, Table 2 The crude reaction product, which was obtained by Pd-catalyzed reaction of 1a with 2g (entry 6, Table 2 The crude reaction product, which was obtained by Pd-catalyzed reaction of 1b with 2a (entry 7, Table 2 The crude reaction product, which was obtained by Pd-catalyzed reaction of 1b with 2b (entry 8, Table 2

Conclusions
In this work, we developed a simple and efficient Pd(II)/Cu(II)-promoted dehydrogenative alkenylation of 5-arylimidazoles, 4,5-diphenylimidazole, benzimidazoles and benzoxazole with functionalized styrenes. Starting from a preliminary screening of the role of oxidant, catalyst precursors, solvents, and reaction temperature on the efficiency and selectivity of the alkenylation of 5-(4-methoxyphenyl)-1,2-dimethyl-1H-imidazole (1a) with styrene (2a) we were able to identify reaction conditions suitable for the simple preparation of several 2-alkenyl-substituted azoles. We believe that our findings may represent an important clue for late-stage functionalization protocols [56][57][58] involving imidazoles, because no pre-activation of the reactive bonds is required. Further studies on the application of this interesting methodology to the synthesis of new heteroaromatic organic fluorophores are underway.