Boron Trifluoride Etherate Promoted Regioselective 3-Acylation of Indoles with Anhydrides

An efficient, high-yielding and scalable procedure for the regioselective 3-acylation of indoles with anhydrides promoted by boron trifluoride etherate under mild conditions was reported. This novel protocol provided a simple way to prepare 3-(benzofuran-2-yl) indole in three steps.

Regioselective functionalization of indoles is one of the most important challenges in the field of indole chemistry, especially in the acylation of free (N-H) indoles. Though many strategies, such as the metal-catalyzed intramolecular oxidative coupling reaction for the preparation of 3-acylindoles, have been developed in recent decades [31][32][33][34][35][36][37], the requirement for expensive and complex ligands and mental catalysts in this novel protocol make few of them suitable for lab or industrial preparation today. The Friedel-Crafts reaction was still regarded as the most promising, practical and convenient protocol [38][39][40][41][42]. AlCl3 [38][39][40], SnCl4 [40], TiCl4 [40], ZrCl4 [41] dialkylaluminum chloride [43][44][45][46] were the most commonly used reagents to promote acylation due to their easy availability and high reactivity. However, these common Lewis acids suffered from some limitations. Some of them required additional protection and deprotection steps to eliminate 1-acylation and 1,3-diacylation [41]. In the above developed methods, most of the reagents were poormoisture-tolerant and air-sensitive, and the presence of a metal ion resulted in a laborious and frustrated workup. In addition, most of the metal ions are toxic and must be carefully removed from the products, especially for the drug and pharmaceutical industry. Additionally, environmental awareness has also made them not preferred in this transformation, especially in the large-scale processes. The application of more environmentally benign solid Lewis acids or Brønsted acidic ionic liquids such as modified zeolites or bisulfate in this acylation process have also been reported [47,48], but only very limited substrates have been used, and the preparation of the Brønsted acidic ionic liquid was more complex than the commercial reagents. Accordingly, a metal-free, more environmentally benign regioselective acylation procedure to prepare 3-acylindoles under mild reaction conditions and simple workup is highly desirable. Regioselective functionalization of indoles is one of the most important challenges in the field of indole chemistry, especially in the acylation of free (N-H) indoles. Though many strategies, such as the metal-catalyzed intramolecular oxidative coupling reaction for the preparation of 3-acylindoles, have been developed in recent decades [31][32][33][34][35][36][37], th requirement for expensive and complex ligands and mental catalysts in this nove protocol make few of them suitable for lab or industrial preparation today. The Friedel Crafts reaction was still regarded as the most promising, practical and convenient protoco [38][39][40][41][42]. AlCl3 [38][39][40], SnCl4 [40], TiCl4 [40], ZrCl4 [41] dialkylaluminum chloride [43 46] were the most commonly used reagents to promote acylation due to their easy availability and high reactivity. However, these common Lewis acids suffered from som limitations. Some of them required additional protection and deprotection steps to eliminate 1-acylation and 1,3-diacylation [41]. In the above developed methods, most o the reagents were poor-moisture-tolerant and air-sensitive, and the presence of a meta ion resulted in a laborious and frustrated workup. In addition, most of the metal ions ar toxic and must be carefully removed from the products, especially for the drug and pharmaceutical industry. Additionally, environmental awareness has also made them no preferred in this transformation, especially in the large-scale processes. The application o more environmentally benign solid Lewis acids or Brønsted acidic ionic liquids such a modified zeolites or bisulfate in this acylation process have also been reported [47,48], bu only very limited substrates have been used, and the preparation of the Brønsted acidi ionic liquid was more complex than the commercial reagents. Accordingly, a metal-free more environmentally benign regioselective acylation procedure to prepare 3-acylindole under mild reaction conditions and simple workup is highly desirable.
Herein, we report a high regioselective and scalable protocol (Scheme 1) for the 3 acylatation of indoles with anhydrides in the presence of boron trifluoride etherate, a very common and easy-to-handle Lewis acid that has been widely used in organic reaction [49]. Herein, we report a high regioselective and scalable protocol (Scheme 1) for the 3acylatation of indoles with anhydrides in the presence of boron trifluoride etherate, a very common and easy-to-handle Lewis acid that has been widely used in organic reactions [49]. Regioselective functionalization of indoles is one of the most important challenges in the field of indole chemistry, especially in the acylation of free (N-H) indoles. Though many strategies, such as the metal-catalyzed intramolecular oxidative coupling reaction for the preparation of 3-acylindoles, have been developed in recent decades [31][32][33][34][35][36][37], the requirement for expensive and complex ligands and mental catalysts in this novel protocol make few of them suitable for lab or industrial preparation today. The Friedel-Crafts reaction was still regarded as the most promising, practical and convenient protocol [38][39][40][41][42]. AlCl3 [38][39][40], SnCl4 [40], TiCl4 [40], ZrCl4 [41] dialkylaluminum chloride [43][44][45][46] were the most commonly used reagents to promote acylation due to their easy availability and high reactivity. However, these common Lewis acids suffered from some limitations. Some of them required additional protection and deprotection steps to eliminate 1-acylation and 1,3-diacylation [41]. In the above developed methods, most of the reagents were poor-moisture-tolerant and air-sensitive, and the presence of a metal ion resulted in a laborious and frustrated workup. In addition, most of the metal ions are toxic and must be carefully removed from the products, especially for the drug and pharmaceutical industry. Additionally, environmental awareness has also made them not preferred in this transformation, especially in the large-scale processes. The application of more environmentally benign solid Lewis acids or Brønsted acidic ionic liquids such as modified zeolites or bisulfate in this acylation process have also been reported [47,48], but only very limited substrates have been used, and the preparation of the Brønsted acidic ionic liquid was more complex than the commercial reagents. Accordingly, a metal-free, more environmentally benign regioselective acylation procedure to prepare 3-acylindoles under mild reaction conditions and simple workup is highly desirable.
Herein, we report a high regioselective and scalable protocol (Scheme 1) for the 3acylatation of indoles with anhydrides in the presence of boron trifluoride etherate, a very common and easy-to-handle Lewis acid that has been widely used in organic reactions [49].

Results and Discussion
At the beginning of our studies, acylation of indole 1a with acetic anhydride 2a in different solvents in the presence of BF 3 ·Et 2 O was explored ( that the acylation reaction could occur in DCM, DCE, CHCl3, MeCN or 1,4-dioxane, and that DCM gave the best results (  [11][12][13] were investigated, which revealed that when the ratio of indole, anhydride and BF3·Et2O was 1:1.2:1, the yield of 3-acylindole 3aa achieved 83% (Table 1, entry 12). Another important point is that in the absence of BF3·Et2O (entry 7), no desired product was achieved. Further screening of the reaction temperature showed that room temperature was the best choice (Table 1, entries 14-15). With the optimal reaction conditions, the aliphatic, alicyclic and aryl anhydrides were subjected to investigate the scope of anhydrides in the acylation reaction. The results are summarized in Table 2. The aliphatic, alicyclic and aryl anhydrides could react with indole 1a smoothly to furnish the desired products 3 in good-to-excellent yields ( Table 2, entries 1-6). However, no desired products were observed for 4-chloro and 4-nitro benzoic anhydrides, perhaps due to the solubility of anhydrides in DCM (Table 2, entries 7-8). With the optimal reaction conditions, the aliphatic, alicyclic and aryl anhydrides were subjected to investigate the scope of anhydrides in the acylation reaction. The results are summarized in Table 2. The aliphatic, alicyclic and aryl anhydrides could react with indole 1a smoothly to furnish the desired products 3 in good-to-excellent yields ( Table 2, entries 1-6). However, no desired products were observed for 4-chloro and 4-nitro benzoic anhydrides, perhaps due to the solubility of anhydrides in DCM (Table 2, entries 7-8). Table 2. Scope of anhydrides. entries 7-10) and anhydride (Table 1, entries 11-13) were investigated, which revealed that when the ratio of indole, anhydride and BF3·Et2O was 1:1.2:1, the yield of 3-acylindole 3aa achieved 83% (Table 1, entry 12). Another important point is that in the absence of BF3·Et2O (entry 7), no desired product was achieved. Further screening of the reaction temperature showed that room temperature was the best choice (Table 1, entries 14-15). With the optimal reaction conditions, the aliphatic, alicyclic and aryl anhydrides were subjected to investigate the scope of anhydrides in the acylation reaction. The results are summarized in Table 2. The aliphatic, alicyclic and aryl anhydrides could react with indole 1a smoothly to furnish the desired products 3 in good-to-excellent yields ( Table 2, entries 1-6). However, no desired products were observed for 4-chloro and 4-nitro benzoic anhydrides, perhaps due to the solubility of anhydrides in DCM (Table 2, entries 7-8). To test the scope of the present protocols, various substituents at different positions of indole ring, including the 1-and 2-substitued indoles, were investigated. As shown in Table 3, both electron-donating and electron-withdrawing substituents in indoles gave the corresponding 3-acylindoles in good-to-excellent yields (from 53% to 93%). The position of the substituents and the electronic nature on the indole ring did not play important roles; only the indoles with electron-withdrawing groups afforded a slightly better yield. The 1-methylindole (1b entry1-4) and 2-phenyl-1H-indole (1d entry9-12) needed a longer time to finish the reaction. From Table 3, we can see the aliphatic anhydrides usually gave higher yields than the aryl ones. Furthermore, the structure of the 3-acylation products 3ec was further confirmed by X-ray diffraction analysis (see Supplementary Materials). Table 3. Acylation of substituted indoles with anhydrides.
To test the scope of the present protocols, various substituents at different positions of indole ring, including the 1-and 2-substitued indoles, were investigated. As shown in Table 3, both electron-donating and electron-withdrawing substituents in indoles gave the corresponding 3-acylindoles in good-to-excellent yields (from 53% to 93%). The position of the substituents and the electronic nature on the indole ring did not play important roles; only the indoles with electron-withdrawing groups afforded a slightly better yield. The 1-methylindole (1b entry1-4) and 2-phenyl-1H-indole (1d entry9-12) needed a longer time to finish the reaction. From Table 3, we can see the aliphatic anhydrides usually gave higher yields than the aryl ones. Furthermore, the structure of the 3-acylation products 3ec was further confirmed by X-ray diffraction analysis (see Supplementary Materials). Compared with the aforementioned protocols promoted by the common Lewis acids or dialkylaluminum chloride, the more moisture-tolerant, air-stable and easy-to-handle BF 3 ·Et 2 O provided an efficient entry to 3-acylindoles. It is also worth noting that when the acylation of indole 1a with acetic anhydride was carried out on more than a 10 g (0.1 mol, 11.7 g) scale, the 3-acylation reaction still provided 80% yield (Scheme 2), which would lead these compounds to be applied more easily. Compared with the aforementioned protocols promoted by the common Lewis acids or dialkylaluminum chloride, the more moisture-tolerant, air-stable and easy-to-handle BF3·Et2O provided an efficient entry to 3-acylindoles. It is also worth noting that when the acylation of indole 1a with acetic anhydride was carried out on more than a 10 g (0.1 mol, 11.7 g) scale, the 3-acylation reaction still provided 80% yield (Scheme 2), which would lead these compounds to be applied more easily.

Scheme 2.
Up to ten grams scale preparation for 3aa.

Scheme 2.
Up to ten grams scale preparation for 3aa.

Experimental Section
Unless otherwise noted, all reactions were performed under air atmosphere, and commercial materials and solvents were used directly without further purification. All reagents were weighed and handled in air at room temperature. 1 H-NMR and 13 C-NMR spectra were recorded on Bruker Avance 400 and 600 spectrometers. Chemical shifts are reported in parts per million (δ) referenced to tetramethylsilane (0.0 ppm), chloroform (7.26 ppm or 77.0 ppm) and DMSO (2.5 ppm or 39.5 ppm), respectively. Data for 1 H-NMR and 13 C-NMR spectroscopy are reported as follows: chemical shift (δ ppm), multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, br = broad), coupling constant (Hz), integration. X-ray single crystal diffraction data were recorded on Bruker D8 QUEST and Bruker APEX DUO. High Resolution Mass spectra were taken on an AB QSTAR Pulsar mass spectrometer or Aglient LC/MSD TOF mass spectrometer.

Experimental Section
Unless otherwise noted, all reactions were performed under air atmosphere, and commercial materials and solvents were used directly without further purification. All reagents were weighed and handled in air at room temperature. 1 H-NMR and 13 C-NMR spectra were recorded on Bruker Avance 400 and 600 spectrometers. Chemical shifts are reported in parts per million (δ) referenced to tetramethylsilane (0.0 ppm), chloroform (7.26 ppm or 77.0 ppm) and DMSO (2.5 ppm or 39.5 ppm), respectively. Data for 1 H-NMR and 13 C-NMR spectroscopy are reported as follows: chemical shift (δ ppm), multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, br = broad), coupling constant (Hz), integration. X-ray single crystal diffraction data were recorded on Bruker D8 QUEST and Bruker APEX DUO. High Resolution Mass spectra were taken on an AB QSTAR Pulsar mass spectrometer or Aglient LC/MSD TOF mass spectrometer.

Up to Ten Scale Synthesis of Selected (3aa)
A total of 11.7 g (0.1 mol) of indole 1 was dissolved in 200 mL of DCM and 12.6 g (12.0 mL, 0.12 mol) of acetic anhydride 2a was added. Then, 14.6 g (0.1 mol) of BF 3 . Et 2 O was added dropwise to the stirred mixture at room temperature. After finishing the addition, the reaction mixture was stirred continuously at room temperature until completed. Then, 100 mL of saturated sodium bicarbonate was added and stirred at room temperature for about 0.5 h. The organic layer was separated and the water phase was extracted with DCM (2 × 100 mL). The organic layer was combined, washed with saturated sodium bicarbonate (2 × 100 mL) and dried over Na 2 SO 4 . The solvent was removed and the residue was purified by column chromatography on silica gel or recrystallized from MeOH/H 2 O (5:1) to give 3aa 12.7 g in 80% yield.

Procedure for Synthesis of 4
A solution of synthesized 3aa (10.0 mmol, 1.59 g), NH 2 OH·HCl (20.0 mmol, 1.39 g) and pyridine (30.0 mmol, 2.4 mL) in MeOH (30 mL) was stirred at room temperature for about 18~24 h. The reaction mixture was evaporated to remove MeOH in vacuo, and to the residue was then added to water (50 mL). After extraction with DCM (2 × 50 mL), the combined organic layers were washed with brine, dried over Na 2 SO 4 , and filtered. Volatiles were removed under vacuum to give the oxime 4 as a white solid without any purification to the next step (1.2 g, 69% yield, m.p. 144-146 • C (lit. 147-148 • C)) [68].

Procedure for Synthesis of 5
A Schlenk tube was charged with 4 (0.5 mmol, 87 mg), and DCE (5 mL). t-BuOK (0.75 mmol, 1.5 equiv) was added in one portion at room temperature under a nitrogen atmosphere. The mixture was stirred at room temperature for 5 min. Then, Ph 2 IOTf (0.75 mmol, 220 mg, 1.5 equiv) was added in one portion. The reaction was stirred at room temperature for 4 h. At this time, the DCE was removed under reduced pressure, and the crude product was purified by column chromatography on silica gel using petroleum ether/ethyl acetate 1/6 to 1/3 to provide product 5.

Procedure for Synthesis of 6
A Schlenk tube, open to air, was charged with 5 (0.5 mmol, 125 mg) and 1,4-dioxane (5 mL). A total of 4 M HCl (0.75 mL, 6 equiv) and H 2 O (0.054 mL, 6 equiv) was added in one portion at room temperature. The mixture was stirred at 80 • C. The reaction was monitored by TLC until 5 was consumed completely (8−12 h). At this time, the solvent was removed under reduced pressure, and the residue was washed with saturated sodium bicarbonate (10 mL). Then, after extraction with DCM (3 × 10 mL), the combined organic layers were dried over Na 2 SO 4 and filtered. DCM was removed under reduced pressure, and the crude product was purified by column chromatography on silica gel using petroleum ether/ethyl acetate 1/10 to 1/8 to provide product 6.

Conclusions
In conclusion, we have developed a mild and efficient synthetic method for the BF 3 ·Et 2 O-promoted acylation of free (N-H) indoles with anhydrides. This protocol afforded a variety of 3-acylindoles in good-to-excellent yields with high regioselectivity and was easily up to 10 g scale. 3-Benzofuran-2-yl indole can be synthesized in good yield in three steps. This protocol accomplished the challenging acylation of free (N-H) indoles successfully. Further studies on their synthetic applications are currently underway in our laboratory.