60th Annual Report on Research 2015

Overview:

Our group is currently developing reactions involving functional groups that are underutilized in modern synthetic organic chemistry. Specifically, we have found that N-acylphthalimides (cf. 1, Scheme 1) react with aldehydes by a net insertion process to give N-phthalimido-O-acyl-N,O-acetals (cf. 2). This unprecedented process occurs in the presence of lithium cation and a non-nucleophilic tertiary amine. The acyl substituent in the resulting product can be displaced by nucleophilic substitution under acidic conditions to form an alpha-branched phthalimide. This report will describe our initial efforts related to reaction discovery, substrate scope determination, and mechanism elucidation. Additionally, application of this methodology to the acid-mediated synthesis of several alpha-branched phthalimides will be discussed.

Summary of Experimental Efforts:

N-Acylphthalimides are an underutilized functional group in organic chemistry. We were initially interested in studying the potential for N-acylphthalimides to undergo an aldol reaction. We hypothesized that lithium cation would coordinate to the imide carbonyls of an N-acylphthalimide (cf. 3, Figure 1) and withdraw electron density from the acyl group such that a tertiary amine base would be capable of deprotonating the alpha-hydrogen atom (cf. 3 to 4). The resulting lithium enolate (cf. 4) could form a new carbon-carbon bond via a net aldol reaction. To test this hypothesis, a mixture of isobutyraldehyde and N-acetylphthalimide (1) was treated with LiBF₄ and triethylamine. To our surprise, the expected aldol adduct 5 was not observed. Instead, N,O-acetal 2a was isolated in 66% yield. Inspecting the structure of the product suggested that isobutyraldehyde had undergone a net C–N bond insertion. The exotic nature of the transformation and the moderate efficiency of the unoptimized conditions led us to pursue this unprecedented transformation more thoroughly.

In addition to isobutyraldehyde, other aldehydes readily participated in the “insertion” reaction to form N,O-acetal products 2b–l (Table 1). It is noteworthy that the reaction is tolerant of alkenes (cf. 2g), hindered alcohols (cf. 2h and 2z), reactive carbocycles (cf. 2l), and aromatic substituents (cf. 2j, 2k, 2n–z). However, use of benzaldehyde-derived substrates produced the corresponding phenyl-containing N,O-acetals (e.g., 2n) in low yields. Additionally, no N,O-acetal product was observed with sterically hindered aliphatic aldehydes or electron-rich benzaldehydes. This limitation was overcome when 4-dimethylaminopyridine (DMAP) was used in place of diisopropylethylamine (DIPEA). Under these revised conditions, pivaldehyde (to give product 2m) and many aromatic aldehydes (to give products 2n–2z) were found to participate efficiently under the revised reaction conditions.

Having established that the reaction occurs with a broad substrate scope, we turned our efforts to elucidating the mechanism of the reaction. To determine if the reaction was occurring via a process that involved acylphthalimide cleavage and phthalimide dissociation, a crossover experiment was performed by subjecting a mixture of isobutyraldehyde and N-acylphthalimides 6 and 7 to conditions that would promote the “insertion” reaction (Figure 2). A combination of ¹H NMR and Hi-Res LC/MS analysis definitively showed that four N,O-acetals (i.e., 8–11) were formed and that crossover had occurred. The crossover observed between 6 and 7 suggests that the acylphthalimide is cleaved under the reaction conditions to form a reactive intermediate that serves as a strong acylating agent. This could explain why yields improved when 4Å MS were added to the reaction mixture because adventitious water was likely engaging the intermediate acylating agent. Use of N-acetylphthalimide-d₃ gave rise to the corresponding N,O-acetal with 30% hydrogen incorporation, which suggests the formation of a ketene intermediate. However, preliminary ab initio computations suggest this pathway to be high in energy. Additionally, N-benzoylphthalimide readily participates under the optimal reaction conditions despite having no alpha-hydrogen atoms.

The synthetic utility of the N-phthalimido-O-acyl-N,O-acetals arising from the newly discovered transformation has also been evaluated. We have demonstrated that this class of N,O-acetals is capable of undergoing acid-mediated nucleophilic substitution at the acyl substituent (cf. 12–14, Figure 3). For example, treating N,O-acetal 2a with aluminum trichloride or titanium tetrachloride gives rise to 14a. Similarly, treating a mixture of tetrabutylammonium bromide or iodide and N,O-acetal 2a with BF₃•OEt₂ leads to formation of N,halo-acetals 14b and 14c, respectively. Carbon-carbon bonds can also be formed via similar methodology. Specifically, the nucleophilic alkene of isopropenyl acetate readily engages the putative iminium intermediate (cf. 13) to form, after deacylation, ketone 14d. Similarly, nucleophilic benzenoids are capable of reacting with N,O-acetals under Lewis or Brønsted acid conditions to form alpha-branched aryl phthalimides (cf. 14e) in a net Friedel–Crafts reaction.

In summary, we have discovered an unprecedented method for preparing N,O-acetals from N-acylphthalimides. The process involves insertion of an aldehyde pi-bond into a C–N bond of an acylphthalimide. The reaction occurs under mild conditions and is applicable to a broad range of substrates. We are in the final stages of manuscript preparation and will submit these results for publication shortly. Mechanistic studies are ongoing, and a fellow Ripon College professor, Joseph Scanlon, has begun evaluating possible mechanistic pathways using ab initio DFT computations. A total of nine undergraduate students have made substantial contributions to this project. In March 2015, aspects of this work were presented at the American Chemical Society National Meeting in Denver, Colorado via two student posters and one oral presentation by Patrick Willoughby.