Direct titanium-mediated conversion of ketones into enamides with ammonia and acetic anhydride.

N-Acyl enamides are useful compounds in organic synthesis. In the realm of catalytic asymmetric hydrogenation, they are among the most exhaustively studied class of substrates, and provide access to valuable chiral amine building blocks. These substrates have also demonstrated broad utility in catalytic asymmetric C C bond forming processes such as aza–ene, Michael, Friedel–Crafts, cycloaddition, and arylation reactions. Despite the extensive applications of Nacyl enamides, their preparation remains challenging. The direct condensation of acetamide with ketones, while attractive in its simplicity, proceeds either in low yields or not at all for the majority of ketone substrates. The Pd-catalyzed cross-coupling of vinyl electrophiles with amides and the Heck reaction of N-vinylacetamide with aryl halides often require additional steps for preparation of coupling precursors and employ a costly transition metal catalyst. The addition of alkyl magnesium or alkyl lithium reagents to nitriles followed by trapping with Ac2O or AcCl has limited functional-group tolerance and requires low reaction temperatures. By far the most commonly employed procedure is the two-step conversion of ketones through ketoximes (Scheme 1). This reaction was first described in 1975 by Barton and coworkers. After a first step of oxime formation, the ketoxime was then treated with Ac2O and either pyridine at reflux, Cr(OAc)2, or Ti(OAc)3 for reductive acylation. [13] Subsequently, numerous alternative reducing agents were developed. The most commonly employed reductant is Fe powder, which was first demonstrated by Barton and Zard in 1985 and subsequently developed by Burk and Zhang in 1998. From a large scale perspective, the use of highenergy hydroxylamine, generating a high-energy oxime intermediate, and reducing the oxime at high temperatures present safety concerns. In addition, the workup of the Fe process is often tedious, requiring filtration of large amounts of inorganic salts. While several alternatives to Fe metal have emerged recently, these still rely on the same overall two-step process through a ketoxime. Our own requirements for large-scale synthesis of N-acetyl enamides for asymmetric hydrogenation prompted us to develop a more direct and process-friendly alternative in which hydroxylamine is replaced with ammonia. Herein we describe a direct, redoxfree synthesis of enamides from ketones, ammonia, and Ac2O mediated by Ti(OiPr)4. In addition, we introduce the use of edte (N,N,N’,N’-tetrakis(2-hydroxyethyl)ethylenediamine) to effect water solubilization of the Ti and allow a simple extractive workup. Our strategy for enamide synthesis was based on condensation of a ketone with ammonia to give an N-unsubstituted imine or enamine, followed by N-acetylation on addition of Ac2O. The imine formation presented a challenge due to the volatility of ammonia, which excluded the common method for imine formation by azeotropic distillation for removal of water. Therefore the condensation with NH3 at room temperature in the presence of various dehydrating agents was explored. Acetophenone 1 was treated with a dehydrating agent (2 equiv) and an ammonia source at room temperature for 24 h, followed by quenching with Et3N and Ac2O (Table 1). Little or no enamide 2 was observed with conventional desiccants and ammonia (entries 1–4). The use of sodium tetraborate (Na2B4O7) or boric anhydride (B2O3) in THF or NMP gave modest conversion to product (entries 5– 7). This prompted screening of other boron reagents, and the discovery that certain trialkyl borates, in combination with NH4Br/Et3N as the ammonia source, gave moderate conversions to product. The most effective boron reagent was 2isopropoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (entry 11), which gave a 59 % conversion to 2. The best results were obtained by using titanium alkoxides, however, with Scheme 1. Conventional two-step enamide synthesis and the direct Timediated method.