Synthesis of α-Ketoamides by Copper-Catalyzed Reactions of Phenylacetic Acids with N,N-Dialkylformamides

Abstract A new synthetic approach for α-ketoamides was achieved by copper-catalyzed coupling reactions between N,N-dialkylformamides and phenylacetic acids. A variety of substrates were well tolerated to give yields of 46–87% (17 examples). A 13C-labeling experiment revealed that the carbonyl group of the products came from the phenylacetic acid. GRAPHICAL ABSTRACT

INTRODUCTION a-Ketoamides constitute key frameworks of many biologically active agents in natural products such as the immunosuppressive drug rapamycin, [1] and they could serve as precursors of pharmacologically important structures such as oxazolidinones, [2] b-lactams, [2,3] and chiral a-hydroxyamides. [4] The importance of the a-ketoamide motif necessitates the development of efficient methods for their preparation. Representative methods for the construction of a-ketoamides include amidation of a-ketoacids and a-ketoacylhalides, [5] transition-metal-catalyzed double carbonylative amination of aryl halides, [6] and reaction of isocyanide with aromatic acyl chloride or anhydride followed by hydrolysis of the resulting a-ketoimidoyl chloride. [7] Besides these methods, Cu-catalyzed oxidative approaches have also been reported, [8] such as cross-coupling reactions between formamide and a-oxocarboxylic acids (path 1, Scheme 1), [8a,8b] the reactions between aryl acetaldehydes or a-carbonyl aldehyde with amines (path 2, Scheme 1), [8c,8d] amidation of terminal alkynes with anilines (path 3, Scheme 1), [8e] and reactions of aryl methyl ketones with amines (path 4, Scheme 1). [8f,8g] More recently, Mupparapu and coworkers developed an elegant procedure for the construction of a-ketoamides involving highly reactive iminium ions as an intermediate and using dimethylsulfoxide (DMSO) as the oxidant. [8h] However, this protocol suffered from limited substrate scope in terms of primary amines and was not suitable for the synthesis of N,N-dialkyl-substitued ketoamides. Nevertheless, most of these reports are restricted by precursor availability, harsh conditions, multistep processes, or utilization of expensive transition-metal catalysts. Therefore, the development of mild, convenient, and efficient methods toward a-ketoamides is still in challenging, especially for the synthesis of N,N-disubstituted a-ketoamides.
In continuation of our recent work on copper-catalyzed amidation reactions using dimethylformamide (DMF) as an amide source, [9] herein is reported coppercatalyzed oxidative reactions of the formation of alpha-ketoacid from alphaarylacetic acids, and followed by the reaction with N,N-dialkylformamides to afford a-ketoamides. [10] RESULTS AND DISCUSSION Initially, we chose phenylacetic acid 1a and DMF 2 as model substrates to optimize the reaction conditions under an oxygen atmosphere. As shown in Table 1, the control experiment showed that the reaction of phenylacetic acid with DMF failed to give the expected product 3a in the absence of catalyst (entry 1, Table 1). Meanwhile, 12% yield of product 3a was obtained after addition of CuBr and phenanthroline (entry 2, Table 1). Subsequent studies revealed that Cu 2 O was the best copper source Scheme 1. Methods for the copper-catalyzed preparation of a-ketoamides.

Entry
Catalyst   Table 1). The reaction was further improved by increasing the substrate concentration to afford 3a in 78% yield (entry 23, Table 1). Decreasing the temperature resulted in lower yield (entry 24, Table 1). Therefore, the optimized reaction conditions for the model reaction were Cu 2 O (10 mol%), 1,10-phenanthroline (20 mol%), DTBP (3.0 equiv), PivOH (2 equiv), and DMF (1 mL) at 130°C for 24 h. Under the optimized reaction conditions, the scope of substituted phenylacetic acids was then investigated. As shown in Table 2, the reactions between phenylacetic acids and DMF proceeded smoothly to give the desired products in moderate to good yields ranging from 52% to 87%. Some substituents such as methyl, methoxyl, and halogens could be well tolerated during catalysis. Phenylacetic acids with electrondonating groups seemed to be more beneficial to the reaction than those containing electron-withdrawing groups. For example, p-tolylacetic acid gave the corresponding product in 87% yield, while p-nitrophenylacetic acid resulted in only 52% yield (entries 7 and 8, Table 2). Meanwhile, steric hindrance seemed to have fewer effects on the results. For example, phenylacetic acids with methyl group substituted either on para, meta, or ortho position gave similar yields around 85% (entries 2, 7, and 10, Table 2). Notably, 1-naphthaleneacetic acid could also be reacted well with DMF to give corresponding product in 78% yields (entry 12, Table 2).
Encouraged by these promising results, N, N-diethylformamide could also be applied as substrate instead of DMF for this reaction. As shown in Table 3, moderate yields were obtained in these cases, indicating the potential of this protocol for the synthesis of N-substituted a-ketoamides (Table 3). It was reported that DMF could be utilized as a precursor to provide -CONMe 2 [8a] or -NMe 2 [8b] units in the reaction of a-oxocarboxylic acid with DMF. To elucidate the possible reaction pathway of this work, several control experiments were carried out as shown in Scheme 2. First, benzaldehyde was used instead of phenylacetic acid under the standard conditions, and desired product 3 could not be obtained, which indicated that benzaldehyde is not formed as an intermediate [Scheme 2, Eq. (1)]. Further evidence of a 13 C-labeling experiment unambiguously confirmed that the carbonyl group of the product would come from the phenylacetic acid, rather than DMF [Eq. (2); see supporting information for details], which made it seem that this reaction might involve a different pathway compared with previous work on copper-catalyzed aerobic decarboxylation of phenylacetic acid by Song et al. [10] Furthermore, after addition of 2 equiv. of 2,2,6,6-tetramethylpiperidine N-oxyl (TEMPO) to the reaction under standard conditions, the transformation was remarkably suppressed, suggesting that a free N,N-dimethylamine radical might be involved in the reaction [Scheme 2, Eq. (3)]. [8b,8i,11] On the basis of these and previously reported results, [8b,9] a possible reaction pathway was proposed as shown in Scheme 3. Initially, phenylacetic acid is oxidized to 2-oxo-2-phenylacetic acid, [10b] and then 2-oxo-2-phenylacetic acid forms a product with the N,N-dimethylamine radical, [8b] which is formed through decomposition of DMF under the reaction conditions. [8i,11] This is different from previous studies [12] that showed the amide could be obtained by the decarboxylation of carbamic anhydride formed from the direct coupling of acid with fomamide.

CONCLUSIONS
In summary, we have discovered a simple method for the synthesis of a-ketoamides via copper-catalyzed coupling of formamides with phenylacetic acids. 13 C-labeling experiment confirmed that the carbonyl group of the product came from the phenylacetic acid. This protocol also provides a practical synthetic method for the construction of N-substituted a-ketoamides.

FUNDING
We are grateful to the Natural Science Foundation of China (Nos. 21272161, 21372163, 21472128, and J1103315) and the Ministry of Education of China (No. 20120181110050) for financial support.

SUPPLEMENTAL MATERIAL
Supplemental data for this article can be accessed on the publisher's website.