Jiao
Yu
a,
Jin-Hong
Lin
a,
Yu-Cai
Cao
b and
Ji-Chang
Xiao
*a
aKey Laboratory of Organofluorine Chemistry, Shanghai Institute of Organic Chemistry, University of Chinese Academy of Sciences, Chinese Academy of Science, 345 Lingling Road, Shanghai 200032, China. E-mail: jchxiao@sioc.ac.cn
bState Key Laboratory of Polyolefins and Catalysis, Shanghai Key Laboratory of Catalysis Technology for Polyolefins, Shanghai Research Institute of Chemical Industry Co. Ltd, China
First published on 5th September 2019
Described here is the radical-mediated hydrodifluoromethylation of alkenes that occurs under transition-metal-free conditions by using the phosphonium salt [Ph3P+CF2H]Br− under the irradiation of 26 W household compact fluorescent light. The reaction lends itself to a convenient protocol for the installation of a C–CF2H bond while maintaining good functional group tolerance.
The straightforward strategies for HCF2 incorporation include the insertion of difluorocarbene into X-H bonds3 (X = C, N, O, S, etc.) and direct difluoromethylation with a HCF2 reagent.2,4 Difluorocarbene insertion is highly effective for the installation of an XCF2H moiety (X = N, O, or S), but for the formation of a C–CF2H group by this strategy, a reactive substrate or a strong base has to be used.5 Over the last few years, a large number of direct difluoromethylation approaches have been developed, such as nucleophilic,6 radical,7 electrophilic8 and transition-metal-promoted reactions,9 enabling convenient construction of C–CF2H bonds. However, the reported methods usually suffer from the use of a transition metal that may limit their biomedical applications, or difluoromethylation reagents that are volatile or difficult to prepare. In addition, a further tedious procedure may be required to remove the undesired auxiliary group X from XCF2 moieties to form HCF2-products. It is therefore highly desirable to develop mild protocols for direct difluoromethylation under transition-metal-free conditions by using an easy-to-handle reagent.
Hydrodifluoromethylation of alkenes is an attractive approach for the installation of a Csp3–CF2H bond. In 2014, Hao described a hydrodifluoromethylation of terminal alkenes with TMSCF2CO2Et, a process in which an excess of silver source is essential (Scheme 1, eqn (1)).10 In 2015, Qing disclosed a two-step sequence involving visible-light-induced hydrobromodifluoro-methylation with ozone-depleting CF2Br2 and the subsequent reductive debromination to convert the BrCF2 to HCF2 groups (Scheme 1, eqn (2)).11 Shortly after, they further reported the hydrodifluoromethylation using phosphonium reagents, [Ph3P+CF2Br Br−]12 and [Ph3P+CF2H Br−],13 under photocatalytic conditions (eqn (3)). Dolbier et al. found that the photocatalyzed hydrodifluoromethylation with HCF2SO2Cl could also occur well (eqn (3)).14 Gouverneur reported a visible-light-promoted hydrodifluoromethylation of alkenes with HCF2CO2H (eqn (4)).15 This operationally simple reaction did not require a photocatalyst, affording the hydrodifluoromethylation products in good yields. Although the above approaches are quite efficient, their synthetic utility may be compromised by the disadvantages such as the need for an additional step to remove an auxiliary group,10,11 or the use of an expensive photocatalyst12–14 or a strong oxidant.15 In continuation of our studies on fluoroalkylation,6d,16 we investigated the feasibility of performing direct hydrodifluoromethylation of alkenes under radical mediated conditions. Herein, we report the visible light induced hydrodifluoromethylation of alkenes using the phosphonium salt [Ph3P+CF2H]Br− under the irradiation of with 26 W household compact fluorescent light (CFL) (eqn (5)).
We have previously shown that the phosphonium salt, [Ph3P+CF2H]Br− (2),6d could be easily prepared from the phosphobetaine Ph3P+CF2CO2−, a reagent developed by us,17via a convenient decarboxylative protonation. Since salt 2 could be reduced to generate a difluoromethyl radical,18 various reducing agents were examined in our initial attempts at the hydrodifluoromethylation of alkene 1a with 2 (Table 1, entries 1–4). However, almost no desired product was detected by using metals as reducing agents (entries 1–3). We then probed Hantzsch ester 3, as it was known to act as an efficient reducing agent when irradiated with visible light.19 To our delight, 16% yield was obtained by irradiating the reaction with household 26 W CFL in the presence of a Hantzsch ester (entry 4). A brief survey of the reaction solvents (entries 4–7) revealed that 1,2-dichloroethane (DCE) was a superior choice (entry 7). The yield was increased in the presence of NaHCO3 (entry 8 vs. 7), but other bases seemed ineffective (entries 9–12 vs. 7). Apparently, a hydrogen source was needed in this hydrodifluoromethylation reaction. The 41% yield (entry 8) indicated that one of the reagents used also served as a hydrogen source. Since it was difficult to increase the yield further, a second hydrogen source was then added (entries 13–16). The use of Ph2SiH2 increased the yield to 50% (entry 14). The molar ratios of the reagents were screened (entries 17–20). An increase in the yield was observed by increasing the loading of phosphonium salt 2 (entry 20).
Entry | [H] source | Base | Ratiob | Yieldc (%) |
---|---|---|---|---|
a Reaction conditions: 1a (0.2 mmol), 2, 3, 4 and base in 1,2-dichloroethane (3.5 mL) under the irradiation of household 26 W CFL at room temperature for 22 h. b Molar ratio of 1a:2:3:4:base. c The yields were determined by 19F NMR; ND = not detected. d CH3CN was used as the reaction solvent. e Mg was used as the reductant instead of a Hantzsch ester. f Zn was used as the reductant instead of a Hantzsch ester. g Fe was used as the reductant instead of a Hantzsch ester. h DMF was used as the reaction solvent. i DCM was used as the reaction solvent. | ||||
1d,e | — | — | 1:3:3:0:0 | ND |
2d,f | — | — | 1:3:3:0:0 | Trace |
3d,g | — | — | 1:3:3:0:0 | ND |
4d | — | — | 1:3:3:0:0 | 16 |
5h | — | — | 1:2:2:0:0 | 4 |
6i | — | — | 1:2:2:0:0 | 25 |
7 | — | — | 1:2:2:0:0 | 28 |
8 | — | NaHCO3 | 1:2:2:0:4 | 41 |
9 | — | KH2PO4 | 1:2:2:0:4 | 17 |
10 | — | CH3COOK | 1:2:2:0:4 | 28 |
11 | — | Na2C2O4 | 1:2:2:0:4 | 16 |
12 | — | Cs2CO3 | 1:2:2:0:4 | 28 |
13 | (TMS)3SiH | NaHCO3 | 1:2:2:4:4 | 15 |
14 | Ph2SiH2 | NaHCO3 | 1:2:2:4:4 | 50 |
15 | Et3SiH | NaHCO3 | 1:2:2:4:4 | 45 |
16 | Bu3SnH | NaHCO3 | 1:2:2:4:4 | 6 |
17 | Ph2SiH2 | NaHCO3 | 1:2:2:4:2 | 31 |
18 | Ph2SiH2 | NaHCO3 | 1:3:2:4:4 | 57 |
19 | Ph2SiH2 | NaHCO3 | 1:4:2:4:4 | 70 |
20 | Ph2SiH2 | NaHCO3 | 1:4:2:2:4 | 71 |
With the optimized reaction conditions in hand (Table 1, entry 20), we then investigated the substrate scope of the visible-light-induced hydrodifluoromethylation of alkenes (Scheme 2). The process could be applied to a wide range of alkenes, and various functional groups could be tolerated, including ester, carbonyl, halides, heterocycles, hydroxyl, sulfonyl, and imide groups. Besides monosubstituted alkenes, disubstituted terminal alkenes could also be converted smoothly into the desired products (5k). However, in the case of trisubstituted alkenes, low regioselectivity was observed, and complex mixtures were obtained. For substrates with electron-withdrawing carbonyl and sulfonyl groups, although the carbonyl-substituted alkene showed a low reactivity (5r), the sulfonyl alkene was transformed into the desired product in a good yield (5s). While the reaction of an enamine derivative gave the product in a good yield (5t), the hydrodifluoromethylation of an aryl alkene under the present reaction conditions led to a low conversion (5u).
More experimental evidence was collected to gain insight into the mechanism of this hydrodifluoromethylation reaction. The addition of a radical scavenger, TEMPO (2,2,6,6-tetramethylpiperidin-1-oxyl), completely suppressed the formation of the desired product, and the TEMPO-CF2H byproduct was produced in 69% yield (Scheme 3, eqn (1)). The TEMPO-CF2H adduct was also detected from the reaction run in the absence of a substrate (eqn (2)), indicating that a radical mechanism is operative. The desired product was generated in a very low yield when the reaction was conducted in the dark (eqn (3)), suggesting that light is essential for this process. After the reaction was complete, CF2H2, Ph3P and pyridine 6 were produced as the byproducts (eqn (4)). In particular, Ph3P and 6, formed from [Ph3P+CF2H]Br− and the Hantzsch ester, respectively, were isolated in high yields, indicating that [Ph3P+CF2H]Br− was reduced while the Hantzsch ester was oxidized in the reaction. The Stern–Volmer luminescence quenching experiments revealed that the phosphonium salt [Ph3P+CF2H]Br− could quench the excited Hantzsch ester (see the ESI† for experimental details), suggesting that electron transfer between these two reagents might have occurred.
Scheme 3 Mechanistic investigations. aThe yields were determined by 19F NMR spectroscopy based on 1a (0.2 mmol); bisolated yield based on [Ph3P+CF2H Br−]; cisolated yield based on the Hantzsch ester. |
On the basis of the above results, we proposed the reaction mechanism as shown in Scheme 4. Under the irradiation of light, a single electron transfer from the Hantzsch ester to the phosphonium salt [Ph3P+CF2H]Br− generates radical cation Int1 and [Ph3PCF2H]˙ radical. The release of Ph3P from [Ph3PCF2H]˙ radical provides HCF2˙, which is easily trapped by an alkene to form intermediate Int3. The HCF2˙ radical may also be quenched by Int1 or Ph2SiH2 to give CF2H2. Radical Int3 would abstract a hydrogen atom from Ph2SiH2 or radical cation Int1 to deliver the desired product 5. The reaction of Int3 with Int1 leads to the formation of pyridinium cation Int4, which is neutralized by NaHCO3 to afford compound 6.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9qo00919a |
This journal is © the Partner Organisations 2019 |