Ligand- and Additive-Free 2-Position-Selective Trifluoromethylation of Heteroarenes Under Ambient Conditions

A highly selective copper-catalyzed trifluoromethylation of indoles is reported with the assistance of a removable directing group. This protocol provides an easy and rapid method to various 2-position-selective trifluoromethylated heteroarenes including indoles, pyrroles, benzofuran, and acetanilide. What is more, the reaction takes place at ambient conditions without any external ligand or additive.


INTRODUCTION
The introduction of trifluoromethyl (CF 3 ) groups into heteroarenes enjoys a privileged role in medicinal chemistry, since it can substantially alter their properties, such as metabolic stability, lipophilicity and ability to penetrate the blood-brain barrier (Shimizu and Hiyama, 2005;Schlosser, 2006;Hagmann, 2008;Boechat and Bastos, 2010;Nie et al., 2011;Wang et al., 2014;Gouverneur and Seppelt, 2015). It has a great potential in the development of new pharmaceutical chemicals (Scheme 1). Thus, the trifluoromethylation of heteroarenes has received tremendous attentions (Sato et al., 2010;Furuya et al., 2011;Tomashenko and Grushin, 2011;Liu et al., 2013;Le et al., 2018). On the other hand, indoles represent ubiquitous structural motifs found in biologically active natural products and pharmaceutical compounds (Lee et al., 2015;Chripkova et al., 2016;Sravanthi and Manju, 2016;Goyal et al., 2018;Kaur et al., 2019). In this regard, direct trifluoromethylation of indoles offers an attractive alternative to the workers in the field of medicinal chemistry and biochemistry.
However, direct trifluoromethylation at the C2-position of indoles under radical trifluoromethylation conditions is quite difficult because of the lack of reaction selectivity at the C2/C3-position and the high reactivity of the CF 3 radical (Scheme 2A; Nagib and MacMillan, 2011). Recently, directing group (DG) has emerged as a powerful tool to achieve regioselective C2-H functionalization of indoles (Nishino et al., 2012;Zhou et al., 2013;Yu et al., 2014). For example, Shi group and Punji group, respectively, achieved trifluoroethylation and difluoroalkylation of indoles at C2 position by introducing a directing group at the N-center of indoles (Scheme 2B; Yan et al., 2017;Soni et al., 2018). Also, Sodeoka group, Cho group, and Shi group accomplished trifluoromethylation of indoles at the C2 position with Togni's reagent, CF 3 I and CF 3 SO 2 Na respectively. However, a substituent at C3 was identified as a crucial factor for the selective trifluoromethylation at C2 (Scheme 2C; Shimizu et al., 2010;Iqbal et al., 2012;Shi et al., 2018). In this context, we envisioned that with the aid of a readily removable N-protecting group, trifluoromethyl group can be introduced to C2 position, which is complementary to the previous work. Herein, we report a copper-catalyzed C2-H trifluoromethylation of N-Boc (t-butyloxy) indoles with CF 3 SO 2 Na under ambient conditions in the absence of any external ligand or additive (Scheme 2D). Notably, the key to our success is the installation of a suitable Boc director on the indole nitrogen atom.

RESULTS AND DISCUSSION
To begin, we chose N-Boc indole (1a) as model substrate. To our delight, when the reaction mixture of N-Boc indole (1a, 0.50 mmol), 2 (1.5 mmol), TBHP (t-butyl hydroperoxide, 70% solution in H 2 O, 2.5 mmol) and CuSO 4 (10 mol%) in DMA (dimethylacetamide, 3 mL) was stirred at 85 • C in air for 1 h, 22% yield of C2-trifluoromethylation product 3a was obtained ( Table 1, entry 1). Trace amount of product could be detected with other solvents, such as DCM (dichloromethane), toluene and THF (tetrahydrofuran) (entries 2-4). The yield could be increased to 46% when acetone was used, and it was elevated to 54% by using MeCN (acetonitrile) (entries 5-6). Subsequently, various metal catalysts were selected. To our delighted, the yield was increased to 65% by employing CuSO 4 •5H 2 O as catalyst (entry 7). Meanwhile, other catalysts such as FeCl 3 , FeCl 2 , Cu(OTf) 2 , and CuCl were screened. Unfortunately, no positive results were obtained (entries 8-11). In addition, the reaction was completely shut down in the absence of metal catalysts (entry 12). Finally, the desired product 3a was obtained in 86% isolated yield when the solvent was reduced to 1 mL (entry 13). The reaction showed low reactivity at room temperature (entry 14). Afterward, the efficiency of different directing groups was investigated. And no desired product was achieved when Ac, Ts and 2-pym were tried (entries 15-17). In addition, the use of the methyl group resulted in a marked decreased in selectivity and yield (entry 18).
With an optimized protocol in hand, the scope and limitation of the title reaction was explored (Scheme 3). With respect to the various indole derivatives, the reaction was found quite general and tolerated by various functional groups. A wide range of 2-trifluoromethylated products with substituent groups such as methyl (3b, 3i), methoxy (3c), acetyl (3d), esters (3e, 3j, 3k), and halogen (3f-3h, 3l-3o) at 4-, 5-and 6-position of indole were produced in moderate to good yields. In particular, halides, such as F, Cl, and Br, were well tolerated, affording the desired 2trifluoromethylated products (3f-3h and 3l-3o) in good yields of 67-89%. However, C7-substituted indoles are not reactive under the optimized reaction conditions, which is presumed due to the steric hindrance. In addition, owing to the strong electronwithdrawing property, the indoles containing cyanide and nitro are not reactive.
To extend the substrate scope of the above reaction, we proceeded to study the trifluoromethylation of other aromatics under the optimized reaction conditions. As shown in Scheme 4, pyrroles reacted smoothly to afford the corresponding 2-trifluoromethylated pyrroles (4a-4d) in good yields. Conventionally, direct deprotonation of benzofuran takes place at the most acidic C2 position (Larbi et al., 2017;Wang et al., 2018). Following C2 deprotonation, we obtained 2-trifluoromethylated benzofuran 4e in 88% through a radical addition mechanism. Notably, benzothiophene was also examined, but only a trace amount of product 4f was detected. Acetanilide, a drug to relieve pain or reduce fever, was also used for the synthesis of 4g. Additionally, we tried other "indole-like" compounds, but the products (4h-4k) were not gained.

MECHANISM
The radical scavenger experiments were conducted to gain some insights into the mechanism of this reaction (Scheme 5).
When radical inhibitors such as TEMPO (2,2,6,6-tetramethyl-1-piperidinyloxy) and BHT (butylated hydroxytoluene) were added, the reaction was suppressed to a great extent. Also, 19 F NMR analysis showed that radical trapping product TEMPO/BHT-CF 3 was formed dominantly. Therefore, we speculated that the high C2 selective is due to the formation of a five membered metallacycle at the C2 position through N-Bocdirected C-H activation (Sandtorv, 2015;Yang et al., 2016).
Based on the analysis of the aforementioned results and previous reports, a plausible mechanism was proposed in Scheme 6 (Langlois et al., 1991;Ji et al., 2011;Zhang et al., 2018;Khan et al., 2019). Initially, the t-butoxy radical, generated from copper metal, reacts with CF 3 SO − 2 to provide •CF 3 SO 2 . This transient intermediate disproportionates, releasing SO 2 and •CF 3 B. Meanwhile, the copper catalyst was introduced to ensure the formation of a five membered metallacycle A at the SCHEME 2 | Different approaches to C2 functionalized indole derivatives (A-D).
C2 position. Subsequently, chelation-assisted C-H metalation of 1a A reacts with B to form C as a key intermediate. After reductive elimination the product 3a was formed and copper(II) catalyst regenerated.

CONCLUSION
In conclusion, we have developed a direct C2-H trifluoromethylation of indoles with the assistance of a removable directing group under ambient conditions. This transformation exhibits high regioselectivity, functional group tolerance and provide a practical method to various trifluoromethylated heteroarenes including indoles, pyrroles, benzofuran, and acetanilide. What is more, control experiments testified that a radical mechanism may be involved in the reaction.

MATERIALS AND METHODS
General 1 H NMR spectra were recorded on Bruker 500 MHz spectrometer and the chemical shifts were reported in parts per million (δ) relative to internal standard TMS (0 ppm) for CDCl 3 . The peak patterns are indicated as follows: s, singlet; d, doublet; dd, doublet of doublet; t, triplet; m, multiplet; q, quartet. The coupling constants, J, are reported in Hertz (Hz). 13 C NMR spectra were obtained at Bruker 126 MHz and referenced to the internal solvent signals (central peak is 77.0 ppm in CDCl 3 ). The NMR yield was determined by 1 H NMR using CH 2 Br 2 as an internal standard. APEX II (Bruker Inc.) was used for ESI-HRMS. Flash column chromatography was performed over silica gel 200-300. All reagents were weighed and handled in air at room temperature. All chemical reagents were purchased from Alfa, Acros, Aldrich, and TCI, J&K and used without further purification.

General Procedure and Characterization Data for Product 3, 4
To a mixture of N-Boc indole 1 (0.5 mmol), CF 3 SO 2 Na 2 (1.5 mmol) and CuSO 4 •5H 2 O (10 mol%), MeCN (1.0 mL) was added in air at room temperature. tert-Butyl hydroperoxide (TBHP, 70% solution in H 2 O, 2.5 mmol) was dropped into the mixture in air at room temperature. The resulting mixture was stirred at 85 • C in air for 1 h. Once the mixture was cooled to room temperature, the solvent was removed under reduced pressure. The crude product was purified by flash column chromatography on silica gel (petroleum ether/CH 2 Cl 2 ) to give product 3 or 4 as colorless oil. The NMR spectra of synthesized compounds are depicted in Supplementary Material.

DATA AVAILABILITY
All datasets generated for this study are included in the manuscript/Supplementary Files.