To optimize the reaction conditions, 2-formylbenzoic acid (1a) and benzoylacetic acid (2a) were chosen as model substrates. As summarized in Table 1, the reaction did not occur without the use of a base, indicating that a promoter is essential for an effective transformation (Table 1, entry 1). We then turned our attention to the examination of a series of base catalysts for the present annulation protocol. To our delight, the reaction proceeded smoothly to give the desired 3-phenacylphthalide (3a) in 32% yield with Et3N (20 mol %) as a catalyst at 65 °C (Table 1, entry 2). To improve the efficiency of the reaction, a detailed optimization study was performed with various bases, such as tertiary, secondary and primary amines. When tertiary amines were used as the catalyst, very low yields of 3a were obtained (Table 1, entries 3–5). Unfortunately, a sluggish reaction was observed in the presence of pyrrolidine (Table 1, entry 6). Gratifyingly, the desired product was obtained in the presence of primary amines in good yields (Table 1, entries 7–9). In addition, the use of several inorganic bases delivered no catalytic activation (Table 1, entries 10 and 11). Notably, a higher reaction temperature did not increase the yield of 3a (Table 1, entry 12). Overall, the best result was achieved when 20 mol % of p-anisidine was used as the catalyst in glycerol at 65 °C for 0.5 h, providing 3a in 80% yield (Table 1, entry 9).
aGeneral reaction conditions: 1a (0.5 mmol), 2a (1.0 mmol) and base (0.1 mmol) in glycerol (3 mL) at 65 °C for 0.5 h. DMAP = 4-dimethylaminopyridine. bThe yields indicated are the isolated yields after column chromatography. cNo reaction. dThe reaction was carried out at 90 °C.
With the optimal conditions in hand, we started to explore the scope and limitations of this one-pot cascade aldol/cyclization system. Typical results are shown in Scheme 2. The annulation proceeded smoothly with β-keto acids 2 bearing diverse arene substituents to provide the corresponding isobenzofuran-1(3H)-ones in moderate to excellent yields. For para-substituted β-keto acids 2b–f, both electron-donating and electron-withdrawing groups such as methyl (2b), methoxy (2c), and halogens (2d–f) afforded the products in relatively high reaction yields. The use of meta-aryl substituted β-keto acids provided the products in good yields, whereas β-keto acid 2i, bearing a methyl group at the ortho-position, was evaluated to produce the desired compound 3i in moderate yield due to steric hindrance. Moreover, 3-(naphthalen-2-yl)-3-oxopropanoic acid (2j) turned out to be a good substrate, and the corresponding product was obtained in 76% yield. It is worth mentioning that even a β-keto acid bearing a heteroaromatic ring (2k) afforded the desired product in 85% yield. Unfortunately, acetoacetic acid did not undergo the reaction due to its low reactivity. Finally, the substituted 2-formylbenzoic acid 1b (5,6-(OCH3)2) was also tested for this transformation, and the target product 3l was obtained with 67% yield.
After investigating the scope of the synthesis of 3-substituted phthalides, the recyclability of glycerol was investigated for the reaction shown in Table 2. After completion of the condensation, the mixture was extracted with ethyl acetate. The glycerol phase was directly reused for further reactions. We were pleased to observe that the yield of 3a was almost consistent after four runs.
aThe yields indicated are the isolated yields after column chromatography. bGeneral reaction conditions: 1 (0.5 mmol), 2a (1.0 mmol) and p-anisidine (0.1 mmol) in glycerol (3 mL) at 65 °C for 0.5 h. cRecovered glycerol was used.
On the basis of the results described above and other previous work [29,43], a plausible mechanism for this reaction has been tentatively proposed (Scheme 3). Firstly, 2-formylbenzoic acid (1a) is attacked preferably by benzoylacetic acid (2a) in the presence of a base to afford the aldol intermediate A. Next, the subsequent facile decarboxylation and lactonization of intermediate A leads to 3-phenacylphthalide (3a).
This manuscript is dedicated to Prof. Dieter Enders on the occasion of his 70th birthday and for his seminal contributions to organic chemistry.
Transformations of 2-formylbenzoic acid provide direct access to a series of heterocyclic organic compounds such as phthalides and isoindolinones. Here, we use (+)-cinchonine as a catalyst in conjunction with nonafluoro-tert-butanol as a hydrogen-bond donor to afford enantiomerically enriched acylated 3-hydroxyphthalides with up to 99% yield and 90% ee through dynamic kinetic resolution. Moreover, various 3-alkoxyphthalides as well as 2-alkyl-3-hydroxy-1-isoindolinones were synthesized from 2-formylbenzoic acid.
Keywordsacylation - chiral auxiliary - cinchona alkaloids - dynamic kinetic resolution - organocatalysis
- Supporting information for this article is available online at http://dx.doi.org/10.1055/s-0036-1589404.
- Supporting Information