Pyridine-enabled copper-promoted cross dehydrogenative coupling of C(sp2)–H and unactivated C(sp3)–H bonds

Pyridine-enabled cross dehydrogenative coupling of sp2 C–H bonds of polyfluoroarenes and unactivated sp3 C–H bonds of amides was achieved.


Introduction
Transition metal-promoted direct functionalization of unactivated C-H bonds is a highly valuable approach for the selective construction of C-C bonds, and considerable efforts have been devoted into this research area over the past couple of decades. 1 Within this reaction category, ligand-assisted cross dehydrogenative coupling (CDC) is of current interest, and signicant progress has been achieved in recent years. 2 Compared with the conventional cross coupling reactions, this method enables the direct manipulation of aromatic and aliphatic C-H bonds by obviating the pre-installation of the functional groups. Moreover, the ligand acts as a directing group to ensure the high siteselectivity. In the process, a noble metal species such as palladium, rhodium, or ruthenium is oen employed as a catalyst. From an economical point of view, avoidance of the use of the precious catalyst in the process would be highly desirable. Towards this effort, Miura and co-workers reported the rst copper-promoted cross dehydrogenative coupling of 2-phenylpyridines and benzoxazoles in 2011 (Scheme 1a). 3 It was then found that azine-N-oxides, benzamides, indoles, naphthylamines, and 2-pyridones were also effective substrates. 4 Despite being a highly efficient method for the construction of C-C bonds, this process does not allow for the site-selective direct functionalization of unactivated sp 3 bonds coupling with an arene bearing a directing group, thus restricting the product diversity. 5,6 Inspired by the bidentate directing group-assisted unactivated sp 3 C-H bond activation process developed by Daugulis' group, 7 we envisaged that attachment of a bidentate directing group to an aliphatic acid may potentially overcome this drawback. 8 With this design, we have examined and report here the copper-promoted cross dehydrogenative coupling of aliphatic amides 9 and polyuoroarenes 10 (Scheme 1b), which provides an efficient access to alkyl-substituted per-uoroarenes, an important structural motif in pharmaceuticals and agrochemicals. 11 It is worth mentioning that this is the rst example of ligand-directed copper-promoted cross dehydrogenative coupling reaction by employing polyuoroarenes as the coupling partners.
that pyridine could also act as a ligand in the process, and thus promote the reaction, a screening of nitrogen-containing potential bidentate ligands was further carried out. As shown in Table 1, several ligands such as TMEDA, 2,2 0 -bipyridine, and 1,10-phenanthroline could promote the process, but none of these molecules is as effective as pyridine (entries 8-10). Next, different copper sources were examined, and it was found that CuOAc was the only other species, but with less efficiency (entry 13). Then, the effects of an oxidant towards the reaction were examined, and it turned out that di-tert-butyl peroxide was optimal, providing 3a in 44% yield (entry 17). It was also noted that a higher yield could be obtained under atmospheric nitrogen (entry 18). Following the above investigation, we carried out an extensive screening of the solvents, and the reaction was signicantly improved with the solvent mixture of DME and 1,4-dioxane (entry 22). Furthermore, an excellent yield was observed with increased amounts of pyridine (entry 24). It was also noted that the reaction yield was dramatically decreased with reduced amounts of the copper species (entry 25), presumably due to the competitive coordination of pyridine or tert-butanolate released from di-tert-butyl peroxide to copper.
Moreover, it is clear that pyridine is required for this process since no apparent product formation occurred in the absence of pyridine (entry 26). As expected, a high site-selectivity was observed with a preference for the C-H bond of the a-methyl over that of the a-methylene, bor g-methyl group, 9 which is believed to arise from the steric effect and preference for the formation of a ve-membered ring intermediate over the six-or seven-membered ring intermediate in the cyclometalation step.
To gain some mechanistic insights into this reaction, a series of deuterium-labeling experiments were carried out (Scheme 2). Considering that t BuOH is generated from ( t BuO) 2 as a byproduct in the process, stoichiometric amounts of t BuOD was added to the reaction system for this study. It was noted that an apparent H/D exchange occurred with pentauorobenzene (2a) with or without 2-ethyl-2-methyl-N-(quinolin-8-yl)butanamide (1b), indicating that C-H bond cleavage of uorobenzene is a reversible step. Furthermore, either a small or trace amount of [D]-2a was observed in the absence of Cu(OAc) 2 or pyridine, while the obvious H-D scrambling occurred without ( t BuO) 2 . It should be mentioned that H-D scrambling could also be promoted by CuOAc instead of Cu(OAc) 2 in the absence of ( t BuO) 2 . These results suggest that the copper species promotes the sp 2 C-H bond cleavage with the pyridine as a base and ligand to facilitate the process, and a pyridine-coordinated aryl copper II or aryl copper I intermediate may be involved in the reaction. 4e-i In the study, it was found that H/D exchange did not happen with 1b in the absence of 2a, indicating that an aryl copper intermediate may be involved in the sp 3 C-H bond cleavage step of the amide.
We further carried out deuterium-labeling experiments with [D 3 ]-1b. As shown in Scheme 3, a H/D exchange was not observed with either this substrate or the product, suggesting that sp 3 C-H bond cleavage is an irreversible step. In addition, a second order kinetic isotope effect was observed with 1b in the process, indicating that cyclometalation of the amide is not the rate-limiting step.
On the basis of the above observed results and the previous reports, 4,9,12 a plausible mechanism for this reaction is proposed (Scheme 4). It is believed that this process begins with the reversible C-H cupration of a uoro(hetero)arene with Cu(OAc) 2 in the presence of pyridine. Coordination of amide 1 to this Cu II species followed by a ligand exchange step gives rise to the Cu II intermediate B. Subsequent oxidation of the Cu II species B generates the Cu III intermediate C, which undergoes an intramolecular cyclometalation step to provide the Cu III complex D. Reductive elimination of this intermediate followed by a ligand dissociation process affords the product 3 and a Cu I species.

Conclusions
In summary, copper-promoted pyridine-enabled cross dehydrogenative coupling of aromatic sp 2 C-H bonds and unactivated aliphatic sp 3 C-H bonds was developed with high efficiency and good functional group tolerance. In this process, high regioselectivity was observed with sp 2 C-H bond functionalization, favoring an sp 2 C-H bond between two C-F bonds of (hetero)arenes. In addition, a predominant preference for functionalizing the sp 3 C-H bonds of a-methyl groups over those of the a-methylene, bor g-methyl groups was observed with aliphatic amides. Mechanistic studies suggested that sp 2 C-H bond cleavage is a reversible step while sp 3 C-H bond cleavage is an irreversible but not the rate-limiting step. Interestingly, it was also found that sp 3 C-H bond cleavage is dependent on sp 2 C-H bond cleavage. The detailed mechanistic studies and potential synthetic applications of this process are currently under investigation in our laboratory.

Scheme 4 Plausible reaction mechanism.
This journal is © The Royal Society of Chemistry 2015