Zhilei Zheng,
Guobo Deng* and
Yun Liang*
Key Laboratory of the Assembly and Application of Organic Functional Molecules, Hunan Normal University, Changsha, Hunan 410081, China. E-mail: gbdeng@hunnu.edu.cn; yliang@hunnu.edu.cn; Fax: +86 731 88872533
First published on 24th October 2016
A simple and convenient copper-catalyzed intermolecular cyclization reaction for the synthesis of quinolines from anilines and terminal acetylene esters has been developed. This methodology constructs the C–N and C–C bonds successively via a cascade process, and provides the desired products in moderate to good yields.
Initially, the reaction of aniline 1a and 1-phenylprop-2-ynyl acetate 2a was chosen as a model reaction to optimize the reaction by diverse catalysts, and solvents. The results were summarized in Table 1. A preliminary study was implemented with the aim of searching an appropriate catalyst to promote the reaction between aniline 1a and 1-phenylprop-2-ynyl acetate 2a. Copper catalyst was chosen as the initial experimental catalyst due to reports revealing copper's friendly applications in such intramolecular and intermolecular cyclic reactions. Firstly, when a solution of 1a and 2a in toluene was stirred in the presence of a catalytic amount of various copper catalysts (such as CuI, CuCl, CuBr, Cu(OAc)2, Cu(OTf)2, CuCl2, entries 1–6, respectively), it was observed that CuBr provided the desired 2-phenylquinoline 3a in 47% isolated yield (entry 3), which was found to be the best among all the considered copper catalysts. Then, the reaction conditions were further investigated, the influence of different solvents on the outcome of the model reaction was screened. It was established that it can be run in a number of solvents (Table 1, entries 7–13). We found that the chlorobenzene was the best solvent (Table 1, entry 11). Next, some ligands and bases were tested (see ESI†). In the end, we increased the amount of catalyst to 100% could not benefit the yield (Table 1, entry 14). Thus, the optimized reaction conditions were as follows: 1a (0.2 mmol), 2a (0.3 mmol), CuBr (20 mol%), in PhCl (2 mL) under air atmosphere at 120 °C.
Entry | Catalyst | Solvent | Yield/3ab |
---|---|---|---|
a Reaction conditions: 1a (0.2 mmol), 2a (0.3 mmol), catalyst (20 mol%), PhCl (2 mL) in sealed Schlenk tube, at 120 °C for 12 h.b Isolated yields.c CuBr (1 equiv.). | |||
1 | CuI | Toluene | Trace |
2 | CuCl | Toluene | 35 |
3 | CuBr | Toluene | 47 |
4 | Cu(OAc)2 | Toluene | Trace |
5 | Cu(OTf)2 | Toluene | 18 |
6 | CuCl2 | Toluene | 20 |
7 | CuBr | DCE | 53 |
8 | CuBr | DMF | 26 |
9 | CuBr | o-Xylene | 57 |
10 | CuBr | 1,4-Dioxane | 48 |
11 | CuBr | PhCl | 60 |
12 | CuBr | DMSO | 17 |
13 | CuBr | NMP | 29 |
14c | CuBr | PhCl | 49 |
With the optimized reaction conditions in hand, we next investigated the scope of this transformation by different aromatic anilines and 1-phenylprop-2-ynyl acetate 2a (Table 2). A variety of different aniline derivatives were smoothly converted to the corresponding 2-arylated quinolines products in moderate yield. First of all, when an electron-donating group at the different position, such as the o-methyl, m-methyl and p-methyl substituted aniline were obtained in 61%, 76%, and 53% yield respectively. However, two regioisomers generated from substrates 1c, in a 1:
1 ratio. Meanwhile, substrates 1e also afforded the desired product in acceptable yields. Then, anilines bearing double substituted groups on the aromatic ring, could moderately react with 1a, and provided the expected products. Although aniline 1a obtained 2-arylated quinoline 3a with a moderate 60% yield, the 4-fluorobenzenamine and 4-iodobenzenamine were less reactive, affording their target products in poor yields. Next, substrates containing carboxyl and amide groups also could lead to the formation of desired products in acceptable yields. Unfortunately, the substrate 4-nitro aniline 1m did not detect the corresponding products 6-nitro-2-phenylquinoline, only attained the intermediate propargyl amine 3m. It revealed that the reaction might undergo an electrophilic cyclic process. These results indicated that an electronic effect on the substituted group played a significant role in the reaction.
To expand the scope of this cyclization methodology, the different substituted naphthylamines and ynyl acetate substrates were screened (Table 3). Initially, both α-naphthylamine and β-naphthylamine were explored, which were also tolerated in this transformation generating corresponding 3n and 3o in 70% and 40% yield. Then, a various of phenylprop-2-ynyl acetates reacted with the α-naphthylamine, both donating groups (such as Me, OMe) and electron-withdrawing (such as F, Cl and Br) were tolerated, and they could smoothly transform into the desired products in moderate to good yields. These results showed that electronic nature of the additional substituent had little influence on regulating the reaction. Substrates 2g with two substituents at the 2 and 5 positions were also viable for constructing 3u in 64% yields. Gratifyingly, the reaction of 1-(thiophen-3-yl)prop-2-yn-1-yl acetate 2h gave 3v in 64% yield, thus making this method useful for the preparation of heterocycle-containing pharmaceuticals and natural products. Finally, 9-phenanthryl substituted ynyl acetate 2i also underwent cyclization reaction, to provide 2-(anthracen-9-yl)benzo[h]quinoline 3w in 49% yield.
To shed light on the possible mechanism of the reaction, several control experiments were carried out, as shown in Scheme 2. Inspired by unpredictable product 3m which was obtained from substrate 1m, N-(1-phenylprop-2-yn-1-yl)aniline (1y) was treated under standard reaction conditions. To our delight, 3a was isolated in 80% yield (Scheme 2a). This result indicated that 1y should be the intermediate among the cyclization process. Subsequently, using substrate 2a′ to react with aniline under standard conditions. To our surprise, the reaction product was 3a, which did not have deuterium at any position of the quinoline ring (Scheme 2b). Next, we conducted the third control experiment in the presence of D2O under standard conditions. Notably, the product 3a′ had the deuterium at the 3 and 4 position of the quinoline ring (Scheme 2c). Meanwhile, N-(1,3-diphenylprop-2-yn-1-yl)aniline (1z) could not be detected the corresponding products under standard conditions (Scheme 2d). According to these control experiments, this result indicated that copper(I) acetylide should be a key intermediate in this intermolecular cyclization reaction.
Based on the above observations and previous mechanistic studies,18 a plausible mechanism is proposed in Scheme 3. First, the terminal acetylene ester 2a undergoes intermolecular nucleophilic attack by the aromatic amine to produce intermediate 1y. CuBr-mediated enhancement of the alkyne acidity by coordination to the carbon–carbon triple bond, enabling the formation of the corresponding copper(I) acetylide A. Then, A generates intermediate B by electrophilic cyclization. Subsequently, B undergoes protonation to produce unstable 2-phenyl-1,2-dihydroquinoline C. Finally, an air oxidation of C affords the desired products 3a′.19
In summary, we have developed a simple method for the synthesis of substituted quinoline derivatives by copper-catalyzed annulation of anilines with terminal acetylene esters. In the reaction, the C–N and C–C bonds are formed via an annulation process. In addition, no ligands, no bases are required and it could be performed under air conditions. Efforts to extend the applications of the transformation in organic synthesis as well as screen for biological activity of these types of compounds are currently underway in our laboratory.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra23858k |
This journal is © The Royal Society of Chemistry 2016 |