Silver-mediated oxidative functionalization of alkylsilanes

A general approach to the functionalization of aliphatic C–Si bonds in the presence of silver salts and oxidants has been reported.

Abstract

A general approach to the functionalization of aliphatic C–Si bonds in the presence of silver salts and oxidants has been reported. This strategy encompasses a range of valuable C–Si transformations, including the direct conversions of a C–Si bond to C–OCF₃, C–OBz, C–OCOCF₃, C–SCF₃, C–SCN, and C–N₃ bonds. Among them, trifluoromethoxylation of alkylsilanes is reported for the first time. In addition, mechanistic studies indicate that this reaction may proceed through a radical mechanism.

Organosilicon compounds have a multitude of applications in basic science, medicine, and industry due to their stability and non-toxicity, and natural abundance of silicon.1 For example, organosilicon compounds have received much attention in the cross-coupling reaction.2 However, the strategic functionalization of aliphatic carbon–silicon bonds is limited,3 for example, the trifluoromethoxy (OCF₃) group and trifluoromethylthio (SCF₃) group are becoming increasingly important in medicinal, agrochemical and materials science due to their strong electron-withdrawing effect and high lipophilicity.4 Thus, the development of efficient methods for the synthesis of OCF₃ and SCF₃ compounds is of great importance.5,6 However, the trifluoromethoxylation and trifluoromethylthiolation of organosilanes are extremely underdeveloped.7 To the best of our knowledge, no trifluoromethoxylation of alkylsilanes has been reported to date. Herein, we sought a strategy that would facilitate the direct conversion of aliphatic C–Si bonds into a variety of functional groups, including the trifluoromethoxylation and trifluoromethylthiolation of alkylsilanes (Scheme 1).

Scheme 1. Silver-mediated oxidative functionalization of alkylsilanes.

Methods for aliphatic C–Si oxidation,8 halogenation,9 and azidation10 have been reported, but new reaction systems are typically required to promote each transformation. In addition, methods for trifluoromethoxylation of organosilanes are rare, and only two examples were reported. In 2008, using trifluoromethyl triflate as the trifluoromethoxylation reagent, the Kolomeitsev group reported the trifluoromethoxylation of arynes from o-trimethylsilylphenyl triflate.7a In 2018, trifluoromethyl benzoate (TFBz) was reported as a new trifluoromethoxylation reagent by the Hu group and was used to prepare trifluoromethyl ether from trifluoromethoxylation–halogenation of arynes, which was in situ generated from o-trimethylsilylphenyl triflate.7g Despite the advances in these methods, trifluoromethoxylation of alkylsilanes has not been reported to date, so the development of a general approach for functionalization including trifluoromethoxylation of alkylsilanes is highly desirable.

Inspired by our previous work of a hypervalent iodine-mediated fluorination of alkylsilanes using fluoride ions as the fluorinating agent,9e we became interested in the possibility of functionalization of alkylsilanes using other nucleophiles such as ^–OCF₃. Recently, trifluoromethyl arylsulfonate (TFMS) as a new trifluoromethoxylation reagent was disclosed by our group, which was used to in situ generate AgOCF₃ in the presence of silver salts and fluoride ions.11 Thus, we envisioned whether the oxidative trifluoromethoxylation of alkylsilanes could be achieved with ^–OCF₃ which was in situ generated in the presence of fluoride ions and TFMS. Initial investigations focused on the reaction of alkylsilane 1 with various fluorine sources in the presence of trifluoromethyl 4-methylbenzenesulfonate (TFMS, 2) (Table 1, see more details in the ESI†). No desired product 3a was observed when Et₃N·3HF, CsF, TBAF or FeF₃ was used as the fluoride ion source (Table 1, entries 1–4). We were delighted to find that 61% yield of the desired product 3a was observed in the presence of AgF (Table 1, entry 5). Different ligands were evaluated, and 3,4,7,8-tetramethyl-1,10-phenanthroline gave the highest yield (Table 1, entries 5–10). Switching to other oxidants such as N-fluorobenzenesulfonimide (NFSI), PhIO, PhI(OAc)₂, and K₂S₂O₈ could not generate 3a (Table 1, entries 11–14). No desired products were detected in control experiments with AgF or the oxidant omitted. After extensive screening of various solvents, different substitutions on TFMS, and temperatures (see more details in the ESI†), the ideal conditions of 4.0 equiv. of AgF, 0.3 equiv. of 3,4,7,8-tetramethyl-1,10-phenanthroline, 3.0 equiv. of Selectfluor, and 5.0 equiv. of TFMS (2) in 7 : 2 (v : v) MeCN/DCM under a N₂ atmosphere at 25 °C were found to produce the highest yields. Encouraged by these results, we successfully extended the present system to a trifluoromethylthiolation of alkylsilanes by changing the ligand and silver salt. After optimization of the reaction conditions, the desired trifluoromethylthiolated product (3b) was obtained in 80% yield using 4.0 equiv. of AgSCF₃, 0.4 equiv. of 4,4′-di-tert-butyl-2,2′-bipyridine, 3.0 equiv. of Selectfluor, and 4.0 equiv. of CsF in 1 : 1 (v : v) MeCN/dioxane under a N₂ atmosphere at 50 °C (Table 1, entry 15, see more details in the ESI†).

Table 1. Optimized reaction conditions ^a .

Entry	–OCF3 or –SCF3	Ligand	Oxidant	Yield (%) c
1	Et3N·3HF, TFMS (2)	3,4,7,8-Me4-Phen	Selectfluor	3a, 0
2	CsF, TFMS (2)	3,4,7,8-Me4-Phen	Selectfluor	3a, 0
3	TBAF, TFMS (2)	3,4,7,8-Me4-Phen	Selectfluor	3a, 0
4	FeF3, TFMS (2)	3,4,7,8-Me4-Phen	Selectfluor	3a, 0
5	AgF, TFMS (2)	3,4,7,8-Me4-Phen	Selectfluor	3a, 61
6	AgF, TFMS (2)	Phen	Selectfluor	3a, 45
7	AgF, TFMS (2)	Neocuproine	Selectfluor	3a, 8
8	AgF, TFMS (2)	4,7-Ph2-Phen	Selectfluor	3a, 52
9	AgF, TFMS (2)	5,6-Dione-Phen	Selectfluor	3a, 2
10	AgF, TFMS (2)	dtbpy	Selectfluor	3a, 56
11	AgF, TFMS (2)	3,4,7,8-Me4-Phen	NFSI	3a, 0
12	AgF, TFMS (2)	3,4,7,8-Me4-Phen	PhIO	3a, 0
13	AgF, TFMS (2)	3,4,7,8-Me4-Phen	PhI(OAc)2	3a, 0
14	AgF, TFMS (2)	3,4,7,8-Me4-Phen	K2S2O8	3a, 0
15 b	AgSCF3	dtbpy	Selectfluor	3b, 80

^aGeneral conditions: 1 (1.0 equiv.), silver salt (4.0 equiv.), TFMS (2) (5.0 equiv.), ligand (0.3 equiv.), oxidant (3.0 equiv.), MeCN/DCM (v/v 7 : 2), 25 °C, N₂.

^bAgSCF₃ (4.0 equiv.), CsF (4.0 equiv.), dtbpy (0.4 equiv.), Selectfluor (3.0 equiv.), MeCN/dioxane (v/v 1 : 1), 50 °C, N₂.

^cYields were determined by ¹⁹F NMR with benzotrifluoride as a standard.

Having established optimized reaction conditions, we then explored the scope of trifluoromethoxylation and trifluoromethylthiolation with structurally diverse alkylsilanes. As displayed in Table 2, a wide range of primary alkylsilanes bearing electron-donating and electron-withdrawing substituents on aryl rings were successfully converted into the desired trifluoromethoxylated and trifluoromethylthiolated products with good isolated yields (4 to 16). Notably, heteroaromatic substrates, such as pyridine, indole, and thiophene, were also successfully employed to provide the corresponding trifluoromethoxylated and trifluoromethylthiolated products (17 to 21). A good range of functional groups including ester, ether, ketone, nitrile, nitro, amide, chloride, bromide, and even iodide were well tolerated under the mild reaction conditions. Moreover, the trifluoromethylthiolation of an alkylsilane with tertiary alcohol also proceeded smoothly (36). These results encouraged the application of this method to more complex small molecules, which gave the corresponding trifluoromethoxylated and trifluoromethylthiolated products in moderate yields (39, 40 and 41), for example, the alkylsilane derived from celecoxib, which is a nonsteroidal anti-inflammatory drug, was converted to the tri-fluoromethoxylated product (40a) in 31% yield or trifluoromethylthiolated product (40b) in 57% yield. In addition, we prepared compounds 5a and 5b at the gram scale under the standard reaction conditions in 44% and 69% isolated yields, respectively, which demonstrates the scalability of this method. However, trifluoromethoxylation and trifluoromethylthiolation of secondary alkylsilanes were observed with low yields (37 and 38), and alkenes were the major byproducts.

Table 2. Substrate scope for silver-mediated oxidative trifluoromethoxylation and trifluoromethylthiolation of alkylsilanes ^a .

^aConditions A: alkylsilanes (1.0 equiv.), AgF (4.0 equiv.), 3,4,7,8-tetramethyl-1,10-phenanthroline (0.3 equiv.), Selectfluor (3.0 equiv.), TFMS (2) (5.0 equiv.), MeCN/DCM (v/v 7 : 2), N₂ atmosphere, 25 °C. Conditions B: alkylsilanes (1.0 equiv.), AgSCF₃ (4.0 equiv.), 4,4′-di-tert-butyl-2,2′-bipyridine (0.4 equiv.), Selectfluor (3.0 equiv.), CsF (4.0 equiv.), MeCN/dioxane (v/v 1 : 1), N₂ atmosphere, 50 °C. Yields of isolated products are given.

^b25 °C was used.

^cMeCN/DCM (v/v 1 : 1) was used.

^dYield was determined by ¹⁹F NMR with benzotrifluoride as a standard.

Encouraged by our success with trifluoromethoxylation and trifluoromethylthiolation of alkylsilanes, we investigated the use of other silver salts to develop the functionalization of alkylsilanes (Table 3). When AgOBz, AgOCOCF₃, or AgSCN was used, the corresponding products (42, 43, and 44) were obtained in moderate yields. Installation of an azide group has proven to be useful in chemical biology, medicinal chemistry, and materials science.12 The use of AgF and TsN₃ enabled azidation of alkylsilane 1 to prepare the desired product 45 in 63% isolated yield. We note that in each system, both the silver salt and oxidant were necessary for productive reactivity. Although these reactions are not fully optimized, they provide a general strategy for the functionalization of alkylsilanes.

Table 3. Substrate scope for silver-mediated oxidative functionalization of alkylsilanes.

Entry	AgFG	Ligand	Yield
1	AgOBz	3,4,7,8-Me4-Phen
2	AgOCOCF3	4,7-MeO2-Phen
3	AgSCN	4,4′-MeO2-bby
4	AgF + TsN3	4,7-MeO2-Phen

^a 1 (1.0 equiv.), AgOBz (4.0 equiv.), CsF (4.0 equiv.), ligand (0.4 equiv.), Selectfluor (3.0 equiv.), MeCN/DCE (v/v 1 : 1), 50 °C, N₂.

^b 1 (1.0 equiv.), AgOCOCF₃ (4.0 equiv.), CsF (4.0 equiv.), ligand (0.4 equiv.), Selectfluor (3.0 equiv.), DMC, 50 °C, N₂.

^c 1 (1.0 equiv.), AgSCN (4.0 equiv.), CsF (4.0 equiv.), ligand (0.4 equiv.), Selectfluor (3.0 equiv.), MeCN/DCE (v/v 1 : 1), 50 °C, N₂.

^d 1 (1.0 equiv.), AgF (4.0 equiv.), TsN₃ (4.0 equiv.), ligand (0.4 equiv.), Selectfluor (II) (3.0 equiv.), MeCN/DCE/EtOH (v/v/v 5 : 5 : 1), 25 °C, N₂.

To gain some insight into the reaction mechanism, we performed some preliminary studies (Scheme 2a). A less than 10% trifluoromethoxylated or trifluoromethylthiolated product was formed when a radical inhibitor 2,6-di-tert-butyl-4-methylphenol (BHT) (8 equiv.) was added. In addition, when 4 equiv. of 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) was added, the TEMPO adduct 46 or 47 was obtained in 25% and 38% isolated yield, respectively. Furthermore, the reaction of alkanesilanes under the standard reaction conditions (conditions B) gave the trifluoromethylthiolated product 48 (37%) along with the 5-exo-cyclization trifluoromethylthiolated product 48′ in 12% yield. Together, these observations strongly suggested that a radical-chain mechanism or single-electron transfer (SET) was involved in the reactions. In addition, a 28% trifluoromethoxylated product was obtained when AgF₂ was used in the absence of AgF and Selectfluor, which indicated that Ag(ii) species could be involved in the reaction (see more details in the ESI†). Finally, a silver mirror was observed in the reaction, which suggested that Ag(0) was generated.

Scheme 2. (a)Mechanistic studies and (b) proposed mechanism.

On the basis of these mechanistic investigations, we proposed the mechanism depicted in Scheme 2b. In the presence of ligand and Selectfluor, AgFG (FG = OCF₃, OBz, OCOCF₃, SCF₃, SCN, N₃) is oxidized to produce Ag(ii) intermediate I;11d then R group transmetalation from silicon to Ag(ii) intermediate I can afford alkylsilver species II. The subsequent single-electron transfer between Ag(ii) and the R group in intermediate II leads to the generation of the R radical and Ag(i) species III. Finally, the FG group transfer from the intermediate III to the R radical generates RFG and Ag(0). At present, we cannot rule out the possibility of an alternative mechanism in which the R radical intermediate is further oxidized to generate an R carbocation intermediate, which is trapped by the FG anion to form the desired product.

Conclusions

In conclusion, we have developed a silver-mediated oxidative functionalization of alkylsilanes. This strategy enables accessing a range of functionalized products directly, thus obviating the need to develop a new methodology for each specific C–Si transformation. Furthermore, the first example of silver-mediated trifluoromethoxylation of alkylsilanes was developed using trifluoromethyl arylsulfonate (TFMS) as the trifluoromethoxylation reagent. Additionally, preliminary mechanistic studies suggested that this reaction may proceed through a radical mechanism.

Conflicts of interest

There are no conflicts to declare.