Metal-free oxidative trifluoromethylation of indoles with CF₃SO₂Na on the C2 position

Abstract

An efficient method of synthesizing 2-trifluoromethylindoles from indoles with easy-to-handle, cheap and low-toxic CF₃SO₂Na under metal-free conditions is described, which selectively introduces trifluoromethyl to indoles on the C2 position. The desired product can be obtained in 0.7 g yield. A radical intermediate may be involved in this transformation.

An efficient method of synthesizing 2-trifluoromethylindoles from indoles with easy-to-handle, cheap and low-toxic CF₃SO₂Na under metal-free conditions is described, which selectively introduces trifluoromethyl to indoles on the C2 position.

Introduction

Indole compounds are widely found in nature and most of them are bioactive and diffusely used in medicine, as food additives and in other fields.¹ For example, indometacin is one of the strongest prostaglandins and synthetic inhibitors (Fig. 1a).² As a special functional group, trifluoromethyl is often applied to materials, pesticides and pharmaceuticals,³ which can enhance the polarity, stability and lipophilicity.⁴ For instance, Prozac is mainly used for the treatment of mental illness and fludelone can effectively inhibit cancer cells (Fig. 1b and c).⁵ Given the importance of indoles and trifluoromethyl groups, combining the two into a single entity would be very interesting. In this context, it is of profound significance to study new and potentially physiologically active indoles containing trifluoromethyl from the perspective of synthetic methodology and application prospects.

Fig. 1. Indoles and trifluorides with biological activities.

The classical methods for the synthesis of trifluoromethyl compounds usually use Umemoto,⁶ Ruppert–Prakash,⁷ Langlois⁸ and Togni⁹ reagents or other reagents.^10,12 For the past few decades, exploring effective ways to obtain 2-trifluoromethylindoles from indoles has gained increasing attention. Rey-Rodriguez et al.¹¹ demonstrated the iron(ii) catalyzed trifluoromethylation of indole under mild reaction conditions (Fig. 2a). Unfortunately, it showed low regioselectivity. Furthermore, Choi et al.¹² described the platinum(ii) complexes catalyzed trifluoromethylation of indole on the C2 position in the presence of CF₃I and visible light; however, CF₃I is difficult to preserve and toxic (Fig. 2b). In the recent years, CF₃SO₂Na has gradually become an environmentally friendly and cheap source of trifluoromethyl in the field of organic synthesis.¹³ Shi¹⁴ offered an efficient copper-catalyzed oxidation method for the trifluoromethylation of C3 position-blocked indoles (Fig. 2c). CF₃SO₂Na has been widely used as a CF₃ source in organic synthesis; however, the metal-free and highly regioselective synthesis of 2-trifluoromethylindoles has rarely been reported. The disadvantages of expensive reagents, poor regioselectivity or difficult-to-handle reagents are often encountered. As part of our ongoing interest in the synthesis of indoles derivatives,¹⁵ we reported herein a tert-butyl hydroperoxide (TBHP)-promoted metal-free and highly regio-selective trifluoromethylation of indole on the C2 position using easy-to-handle and low-toxic CF₃SO₂Na as the trifluoromethyl source.

Fig. 2. Preparation of trifluoromethylindoles from indoles.

Initially, indole (1a) and CF₃SO₂Na were selected as model substrates for the optimization of the reaction conditions (Table 1). When 1a, 1 equiv. of CF₃SO₂Na, 2 equiv. of TBHP and 2 mL CH₃CN were added to a sealed Pyrex test tube under room temperature, 2a was obtained in the yield of 15% (entry 1). Subsequently, higher temperature (80 °C and 140 °C) gave the desired product 2a in higher yields (36% and 45%, respectively) (entries 2–3). Next, we switched the ratio of raw materials. It was found that the amount of CF₃SO₂Na would greatly affect the yield of 2a, which showed that a small loading of CF₃SO₂Na resulted in less desired product and a higher loading of CF₃SO₂Na would result in byproduct 3a (entries 5–6). A higher yield of 2a was obtained in the presence of 3 equiv. of TBHP (entries 3–5). A reaction time of 18 hours was found to be enough for this transformation, which gave the desired product 2a in 66% yield (entry 5). Shortening the reaction time to 16 hours or prolonging the reaction time to 20 hours (even 24 hours) afforded 2a in the yield of 53% and 65% (66%), respectively (entries 5–9). When toluene, DMF and H₂O were used as solvents, 2a was obtained in the yield of 49%, 41% and 30%, respectively (entries 10–11), while no targeting product 2a was obtained with DMSO or 1,4-dioxane as solvents. Besides, we also investigated other trifluoromethylation reagents such as TMSCF₃ (trifluoromethyltrimethylsilane), using which only trace 2a was observed. In brief, CF₃SO₂Na (2.0 equiv.), TBHP (3.0 equiv.), CH₃CN (2 mL), 140 °C and 18 h are the optimized reaction conditions, as shown in entry 5. The two-dimensional NMR (H–H COSEY) plot also supported that trifluoromethylation was on the C2 position of indoles (ESI 5†).

Screening of the reaction conditions^a.

Entry	CF3 reagent(X eq.)	Y eq.	Sol (2 mL)	Temp/°C	Time/h	Yieldb2a/(3a) %
1	CF3SO2Na(2 eq.)	1	CH3CN	25	18	15
2	CF3SO2Na(2 eq.)	1	CH3CN	80	18	36
3	CF3SO2Na(2 eq.)	1	CH3CN	140	18	45
4	CF3SO2Na(2 eq.)	2	CH3CN	140	18	50
5	CF3SO2Na(2 eq.)	3	CH3CN	140	18	66
6	CF3SO2Na(3 eq.)	3	CH3CN	140	18	56(16)
7	CF3SO2Na(2 eq.)	3	CH3CN	140	16	53
8	CF3SO2Na(2 eq.)	3	CH3CN	140	20	65
9	CF3SO2Na(2 eq.)	3	CH3CN	140	24	66
10	CF3SO2Na(2 eq.)	3	Toluene	140	18	49
11	CF3SO2Na(2 eq.)	3	DMF	140	18	41
12	CF3SO2Na(2 eq.)	3	H2O	140	18	30
13	CF3SO2Na(2 eq.)	3	DMSO	140	18	nr
14	CF3SO2Na(2 eq.)	3	1,4-Dioxane	140	18	nr
15	TMSCF3(2 eq.)	3	CH3CN	140	18	Trace

^aUnless otherwise noted, the reaction was carried out with 1a (0.3 mmol) and solvent CH₃CN (2 mL), 140 °C, stirred for 18 h in air.

^bIsolated yields.

With the optimized reaction conditions in hand, we investigated the substrate scopes with respect to different substituents on indole scaffolds shown in Table 2. To our delight, most of the transformations went smoothly and provided the corresponding products in moderate to good yields. When the hydrogen of NH was replaced by methyl, trifluoromethylation took place and gave 2b in the yield of 56%. It should be noted that when the position of the methyl group on the scaffold of indole was changed from C3 to C7, the corresponding products were obtained with the yield of 73% to 57% (2c, 2d, 2g, 2m, 2t). Indoles with the methyl group substituted at C4 or C6 position successfully converted into corresponding products with the yields of 76% (2d) and 75% (2m), respectively, showing a higher activity. Importantly, this approach was compatible with halogen groups such as bromine (2k) and iodine (2l), which presented corresponding products with the yield of 54% and 59%, respectively. Except for the fluorine-containing substituents, it was found that the yield of electron-donating groups was slightly higher than that of the electron-withdrawing groups (2h, 2n, 2r). For example, 7-azaindole was successfully converted into its corresponding trifluoromethylation product 2s with the yield of 64%.

Trifluoromethylation of indoles^a,b.

^aThe reaction was carried with 1 (0.3 mmol), CF₃SO₂Na (2.0 equiv.) and TBHP (3.0 equiv.) were used in CH₃CN (2 mL) at 140 °C in air for 18 h.

^bIsolated yields.

Subsequently, we studied other indole derivatives under optimized conditions (Fig. 3), such as 2-methylindole (1u), indoline (1v) and indazole (1w). The desired product was not obtained, which may indicate that trifluoromethylation occurred on the C2 position and indoline could not be compatible in the present system.

Fig. 3. Trifluoromethylation of other indole derivatives.

Investigations established that a large-scale amplification experiment of 1b (5 mmol) under optimized conditions at 80 °C smoothly gave 2b. We were pleased to observe that 0.507 g of the corresponding product was isolated with the yield being 51% (Fig. 4).

Fig. 4. Gram-scale experiment.

In order to get insights into the reaction mechanism, a series of control experiments were conducted. When 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) (4 equiv.), butylatedhydroxytoluene (BHT) (4 equiv.) or 1,4-benzoquinone (4 equiv.) was respectively added under the standard reaction conditions (Table 1, entry 6), almost no desired products were detected (Fig. 5). In addition, a radical intermediate (molecular weight: 288) was detected by GC-MS when a radical scavenger (BHT) was used. Based on these results, we proposed herein a plausible reaction pathway (Fig. 6).^8a,14 At first, a free radical, ^tBuO˙ (II) generated from TBHP (I) by heating reacts with a trifluorosulfinate anion to provide the free radical (III). Subsequently, a free radical was formed by releasing SO₂ from the radical (III). The intermediate (V) was generated from the radical with substrate 1. TBHP accepted one electron from the intermediate (V) to give the intermediate cation (VI) and release OH⁻. Finally, the product 2 was obtained by releasing a proton from the intermediate cation (VI).

Fig. 5. Control experiments.

Fig. 6. Proposed mechanism for trifluoromethylation of indoles.

Conclusions

In summary, we have developed an original method of synthesizing 2-trifluoromethylindoles from indoles with easy-to-handle, cheap and low-toxic CF₃SO₂Na under metal-free conditions, which selectively proceeded on the C2 position of indoles. High functional group tolerance and moderate to good yield were presented in this transformation. The control experiments provided evidences that the reaction may undergo a free radical pathway. Further applications of 2-trifluoromethylindoles are currently being carried out in our laboratory.

Conflicts of interest

There are no conflicts to declare.