Photoinduced Trifluoromethylation with CF3Br as a Trifluoromethyl Source: Synthesis of α-CF3-Substituted Ketones.

An efficient and novel photoinduced trifluoromethylation employing CF3Br as a trifluoromethyl source is described. With commercially accessible fac-Ir(III)(ppy)3 as the catalyst, radical trifluoromethylation between O-silyl enol ether and CF3Br occurs successfully. This method provides various α-CF3-substituted ketones with a broad substrate scope in good yields under mild reaction conditions.

Introduction

CF3-containing molecules are of great academic and industrial importance with widespread applications in pharmaceuticals, agrochemicals, and functional materials.[1] As such, the strong demand for CF3-containing molecules has nourished various synthetic methods.[2]

Specifically, radical trifluoromethylation has recently received considerable attention along with the rapid development of mild methods to generate trifluoromethyl radical species.[3] Diverse trifluoromethylation reagents including Togni reagents,[4] Umemoto reagents,[5] Ruppert–Prakash reagents (Me3SiCF3),[6] Langlois reagent (CF3SO2Na),[7] CF3SO2Cl,[8] CF3I,[9] and CF3Br[10] have been developed as radical trifluoromethyl sources successively. Among these reagents, CF3Br is the only one that combines the advantages of low cost and high atom economy.

CF3Br is a stable, nontoxic, and easily available industrial product. Until now, CF3Br has been mainly used as an extinguishant[11] as well as a primary starting material of several prominent trifluoromethylation reagents in the fluoride chemistry.[12] In a sharp contrast, only a handful of studies on trifluoromethylation directly using CF3Br as the CF3 source have been reported. This can be mainly attributed to the following reasons: first, low solubility and gas–liquid mass transfer efficiency seriously hinder the development of CF3Br as the CF3 source. Second, CF3Br has a high reduction potential (−1.7 V vs SCE) so that it can only be reduced under strong reduction conditions.[13] More recently, two ingenious catalytic strategies that allow CF3Br to participate directly in radical trifluoromethylation have been developed (Scheme a): (i) Outstanding studies on transition metal catalytic trifluoromethylation by Beller’s,[10a] Wang’s[10c] and Li’s[10d] groups have shown that judicious selection of metals and ligands together with the assistance of other reaction conditions enables reduction of CF3Br to CF3 radicals, thus triggering radical trifluoromethylation. (ii) Visible-light-induced hydrotrifluoromethylation of alkenes and alkynes by using CF3Br as the trifluoromethyl reagent was reported by Professor Zhang’s group in 2018.[10b] This method not only provides an easy access to convert CF3Br to CF3 radicals but also offers a new opportunity for the application of CF3Br in a green, safe, and efficient way.

Scheme 1

Radical Trifluoromethylation by Using CF3Br

Inspired by the high efficiency of photocatalysis, we envisioned that more trifluoromethylation reactions between CF3Br and radical receptors would occur through the photocatalytic process. Among numerous radical receptors, O-silyl enol ethers can be easily accessible from the corresponding carbonyl compounds and show great reactivity for CF3 radicals. Furthermore, the products α-CF3-substituted ketones are important fragments of marketed drugs such as Alpelisib and Elexacaftor. Therefore, we would like to investigate the photoinduced trifluoromethylation of O-silyl enol ether with CF3Br as the trifluoromethyl source to produce α-CF3-carbonyl compounds and get more detail insights into the mechanism for the production of trifluoromethyl radicals from CF3Br under visible light irradiation (Scheme b).

Results and Discussion

In our initial studies, we performed the trifluoromethylation of 1-phenyl-1-trimethylsiloxyethylene with CF3Br in the presence of 1 mol % fac-Ir(III)(ppy)3 and 3 equiv of diisopropyl ethyl amine (DIPEA) in THF under 10 W blue-LED irradiation (460 nm) (Table , entry 1). The reaction gave two dimer products 4a and 4b as well as a large amount of acetophenone 5 without desired product 3a (Scheme a). Then, DIPEA was removed from the reaction system, and the desired product 3a was obtained only in trace amounts (Table , entry 2). Subsequently, substrates with different trialkylsilyl groups were screened (Table , entries 3–4). To our delight, the bulky triisopropylsilyl (TIPS) group could effectively increase the yield to 25%. The reason is the bulky TIPS group can prevent the photoinduced formation of acetophenone from O-silyl enol ether (Scheme b).[14] To prevent the formation of acetophenone, several aprotic solvents were screened in which MeCN afforded the best result (50% yield, Table , entries 5–9). Meanwhile, diverse light sources were also evaluated, but 10 W blue LED remained the best option (Table , entries 10 and 12). Afterward, the concentration of CF3Br, the loading of the photocatalyst, and the volume of the flask were investigated (Table , entries 13–17), and we finally increased the yield of 3a to 85% (1.5 atm CF3Br, 1 mol % fac-Ir(III)(ppy)3, with 50 mL Schlenk flasks). In addition, no product was observed in the absence of the photocatalyst or light (Table , entries 18 and 19). For more detailed data, refer to the Supporting Information.

Table 1

Optimization of Reaction Conditionsa

entry	SiR3	additive	cat. (mol %)	light source	solvent	yieldb (%)
1	TMS	DIPEA	1%	10 W blue LED	THF	0
2	TMS		1%	10 W blue LED	THF	trace
3	TES		1%	10 W blue LED	THF	13%
4	TIPS		1%	10 W blue LED	THF	25%
5	TIPS		1%	10 W blue LED	dioxane	trace
6	TIPS		1%	10 W blue LED	DMF	5%
7	TIPS		1%	10 W blue LED	DMSO	5%
8	TIPS		1%	10 W blue LED	hexane	N.D.
9	TIPS		1%	10 W blue LED	CH3CN	50
10	TIPS		1%	5 W blue LED	CH3CN	37
11	TIPS		1%	15 W blue LED	CH3CN	50
12	TIPS		1%	10 W white LED	CH3CN	25
13c	TIPS		1%	10 W blue LED	CH3CN	69
14c	TIPS		2%	10 W blue LED	CH3CN	64
15c	TIPS		3%	10 W blue LED	CH3CN	60
16c,d	TIPS		1%	10 W blue LED	CH3CN	80
17c,e	TIPS		1%	10 W blue LED	CH3CN	85
18c,e	TIPS			10 W blue LED	CH3CN	N.D.
19c,e	TIPS		1%		CH3CN	N.D.

Reaction conditions: 1a (0.5 mmol, 1.0 equiv), CF3Br (1.0 atm), photocatalyst (0.005 mmol, 1 mol %) in a 20 mL quartz tube, room temperature, 10 W blue LED.

Isolated yield.

Replace the air in the container with CF3Br three times and then fill it with CF3Br (1.5 atm).

25 mL flask.

50 mL Schlenk flask. N.D.: not detected.

Scheme 2

Formation of Byproducts

The optimized condition was then applied to the reactions of various O-silyl enol ethers derived from aryl methyl ketones to examine the generality of this transformation. As summarized in Scheme , the aryl silyl enol ethers with electron-donating substituents on their phenyl moiety were all suitable to produce the corresponding products 3a–3i in 53–85% yields. The substrates with weak electron-withdrawing groups such as F, Cl, Br, and −CO2Me on the phenyl rings provided the products 3j–3o in slightly lower yields. When strong electron-withdrawing groups such as nitro and cyano groups were attached to the phenyl rings, there were no products observed. O-Silyl enol ethers bearing α-naphthyl, benzothienyl, and thienyl were also successfully transformed into corresponding products 3p–3r in 45–72% yields. Yet, α-pyridyl silyl enol ether failed to give the product. The substrates containing druglike scaffolds, such as clofibrate and ibuprofen derivatives, were also amenable for this process, providing 3s and 3t with 33 and 74% yields, respectively.

Scheme 3

Substrate Scope

Reaction conditions: 1a (0.5 mmol, 1.0 equiv), CF3Br (1.5 atm), photocatalyst (0.0025 mmol, 0.5 mol %), in a 50 mL Schlenk flask, room temperature, and 10 W blue LED (460 nm). N.D.: not detected.

Afterward, O-silyl enol ethers derived from other aryl alkyl ketones, indanones, and aliphatic ketones were explored under the optimized reaction condition (Scheme ). A large number of O-silyl enol ethers from aryl alkyl ketones were trifluoromethylated efficiently and afforded the corresponding products in 27–77% yields (7a–7h). O-Silyl enol ethers derived from indanones with electron-donating and electron-withdrawing substitutions on the aryl ring could also work smoothly, achieving 7i–7p in 41–63% yields. As for aliphatic O-silyl enol ethers, only trace products can be monitored from 19F NMR, but they could not be separated (7q–7s).

Scheme 4

Substrate Scope

Reaction conditions: 1a (0.5 mmol, 1.0 equiv), CF3Br (1.5 atm), photocatalyst (0.0025 mmol, 0.5 mol %), in a 50 mL Schlenk flask, room temperature, and 10 W blue LED (460 nm). N.D.: not detected.

To gain an insight into the reaction mechanism, several control experiments were performed. The addition of a radical-trapping reagent TEMPO (1 mmol) to the reaction system led to a suppression of the formation of 3a, while TEMPO-CF3 adduct 6 was detected by 19F NMR in 12.5% yield (Figure a). When TEMPO reacted with CF3Br without substrate 1a, TEMPO-CF3 adduct 6 could also be detected by 19F NMR in 35% yield (Figure b). These experiments indicated that the CF3 radical was produced from CF3Br under the reaction conditions, and then, it participated in the subsequent reaction. Further fluorescence quenching experiments of the excited fac-Ir(III)(ppy)3* with different concentrations of 1a and CF3Br also indicated that the excited fac-Ir(III)(ppy)3* could be quenched by CF3Br (Figure c). Moreover, the detected dimer products 4a and 4b showed the existence of radical intermediate A (Scheme ).

Figure 1

Mechanistic studies. (a,b) Radical-trapping experiments. (c) Fluorescence quenching experiments.

Scheme 5

Proposed Reaction Mechanism

On the basis of the above results and the literature,[10b] the possible reaction mechanism is proposed in Scheme . Under the blue LED irradiation, the photocatalyst fac-Ir(III)(ppy)3 was excited to fac-Ir(III)(ppy)3*, which was oxidatively quenched by CF3Br and resulted in CF3 radicals and bromine ions. Meanwhile, fac-Ir(III)(ppy)3* is oxidized to fac-Ir(IV)(ppy)3. The CF3 radical attacked the terminal carbon on the C=C bond of O-silyl enol ether to form intermediate A, which was then oxidized by fac-Ir(IV)(ppy)3 to generate intermediate B. After desilylation of the intermediate B, the α-trifluoromethylketone product 3 was produced. If DIPEA is present in the reaction system, fac-Ir(IV)(ppy)3 would be reduced first by DIPEA to give fac-Ir(III)(ppy)3, and the intermediate A is left to couple with itself to form dimer products 4a and 4b.

Conclusions

In summary, the successful use of CF3Br as a trifluoromethyl source in photoinduced trifluoromethylation of O-silyl enol ethers has been achieved under mild and practical conditions. Enabled by the photocatalysis, the C–Br bond of CF3Br can be cleaved to deliver CF3 radicals, which provides facile access to utilize CF3Br as the trifluoromethyl source. Moreover, this method provides a series of α-CF3-substituted ketones, which have great potential applications in the pharmaceutical and agrochemical fields.

Experimental Section

Materials and Methods

Reagents were obtained from commercial suppliers and used without further purification unless otherwise noted. O-Silyl enol ethers 1a–t and 6a–s were synthsized according to the lierature.[15] NMR experiments were carried out in deuterated chloroform (CDCl3). 1H NMR, 13C{1H} NMR, and 19F NMR spectra were recorded at 400 or 600 MHz, 100 or 150 MHz, and 376 MHz spectrometers, respectively. High-resolution mass spectra were recorded on a Micro TOF-QII mass instrument (ESI).

General Procedure for Synthesis of α-CF3-Substituted Ketones

O-Silyl enol ethers 1 (0.5 mmol), fac-Ir(III)(ppy)3 (0.5 mol %), and CH3CN (3 mL) were added into a 50 mL Schlenk flask. The mixture was degassed with CF3Br gas (1.5 atm observed from a barometer). The reaction mixture was stirred at room temperature under a blue LED lamp (hν = 460 nm) for 7–36 h. The reaction mixture was then concentrated in vacuo, and the residue was purified by silica gel chromatography with PE/DCM (10:1 to 1:1) as the eluent to afford pure product 3.

Note:It is worth mentioning that some of the products are a bit volatile.