Energy Transfer Catalysis Enabled 2,2,2‐Trifluoroethoxy‐Amination of Olefins

Abstract

A thioxanthone‐catalyzed 2,2,2‐trifluoroethoxyamination of olefins is developed via the formation of the corresponding alkoxy and iminyl radicals using unprecedented, easily prepared, and bench‐stable oxime ethers as bifunctional reagents. To bypass possible side reactions (1,2‐Hydrogen Atom Transfer (HAT), H‐abstraction, and β‐scission), the high reactivity of the alkoxy radical is fine‐tuned to promote the selective and challenging formation of a COCH₂CF₃ bond. This reaction, involving a triplet energy transfer process, allows the concomitant formation of a CN and COAlk bond, so far uncharted, using bifunctional oxime ether reagents. Hence, the difunctionalization of a myriad of electron‐rich alkenes selectively afforded the anti‐Markovnikov products with a large functional group tolerance (44 examples, up to 77% yield), offering a straightforward and complementary regioselectivity compared to the existing approaches for the difunctionalization of alkenes with 2,2,2‐trifluoroethanol. Post‐functionalization reactions and mechanistic investigations provided key insights into the reaction mechanism of this transformation.

Introduction

Alkenes are pivotal motifs in organic synthesis, serving not only as essential feedstocks but also as versatile functional groups that enable the development and elaboration of molecular architectures. Among the various reactions devised for the functionalization of alkenes, the 1,2-difunctionalization of olefins,^[1–4] which simultaneously forms two new covalent bonds, is a particularly favored strategy for enhancing molecular complexity.^[5,6] In this context, the aminohydroxylation of alkenes, pioneered by Sharpless,^[7,8] is a compelling approach for the synthesis of vicinal amino alcohols, which are prevalent in both natural products and pharmaceuticals.^[9,10] While numerous methods have been developed to access 1,2-amino alcohols, achieving a switch in selectivity for the selective synthesis of 2-amino 1-alcohols remains a significant challenge. Recently, advances in this direction have been reached under the auspices of photochemistry and particularly energy transfer (EnT)-based transformations.^[11] Leveraging the rich photochemistry of oximes,^[12] particularly the potential for homolytic cleavage of the NO bond in the excited state, this transformation was successfully facilitated. The concomitant generation of a persistent N-centered radical (i.e., an iminyl radical) and a highly electrophilic O-centered radical in the presence of an alkene gave access to products resulting from 1,2-hydroxy-amination (Figure 1). The foundation of this approach relied on the design of the oxime reagent and specifically the nature of the O-centered radical. Consistent with this framework, Glorius reported in 2021 the development of oxime carbonate as a bifunctional reagent.^[13–15] The intrinsic properties of the alkoxycarbonyloxyl group, which displays a significantly lower propensity for decarboxylation (3.8 × 10³ M.s⁻¹) compared to alkyl carboxyl radicals (approximately five to six orders of magnitude higher, around 10⁹ M.s⁻¹), was the key of success to forge the CO bond.^[16,17] Furthermore, the addition rate of alkoxycarbonyloxyl radical to styrene is close to the diffusion-controlled limit.^[18] This reactivity paradigm was subsequently applied to orchestrate complex multicomponent reactions.^[19,20] Later in 2021, Huo and coworkers took advantage of the relatively slow decarboxylation rate of aryl carboxyl radical (10⁶ M s⁻¹) to use carboxylic acid-derived oxime esters in the photoinduced oxyamination of alkenes.^[21] While these previous examples have addressed the selective synthesis of 2-amino-1-alcohols, the incorporation of alkoxy radicals derived from alkyl oxime ethers remains conspicuously absent. The high reactivity of these alkyl radical species introduces several additional challenges,^[22] including rapid rearrangement via 1,2-hydrogen atom transfer (1,2-HAT), β-scission, or hydrogen atom abstraction. In this study, we first presented the synthesis and successful application of alkyl oxime ethers in the oxyamination of alkenes with a complementary regioselectivity compared to the one observed in the case of the use of 2,2,2-trifluoroethanol.^[23–26] By leveraging the unique properties of fluorinated molecules, which are widely utilized in medicinal and agrochemical applications,^[27–30] we have developed a method for the simultaneous introduction of 2,2,2-trifluoroethoxy and imine moieties. Using an original radical precursor for the generation of the OCH₂CF₃ radical by photocatalysis, the construction of the desired COCH₂CF₃ bond was successfully achieved.^[31,32] It is worth to mention that the group of Wang has recently demonstrated the possible construction of COCH₂CF₃ bond using the well-designed N–trifluoroethoxy benzotriazolium triflate. This precursor of the OCH₂CF₃ radical was successfully applied to the trifluoroethoxylation of alkenes and (hetero)arenes.^[33,34] The direct introduction of this fluorinated motif is of significant interest due to its presence in various pharmaceuticals, such as Silodosin and Lansoprazole.

[IMAGE OMITTED. SEE PDF]

Results and Discussion

With the aim of elaborating a new method for the amino 2,2,2-trifluoroethoxylation of alkenes via a radical process, we first designed a general two-step-one-purification synthesis to access a library of original fluorinated oxime ethers (Figure 2). First, the oximes were prepared from the inexpensive and readily available corresponding benzophenone derivatives and hydroxylamine (Figure 2a). Then, without any further purification, the O-alkylation reaction was achieved by reaction of the oximes 1’ with the commercially available 1,1,1-trifluoro-2-iodoethane, as the source of CH₂CF₃ group, affording the bifunctional reagents 1 in good to excellent yields on a 5 mmol scale (61–74% yields). We were delighted to find out that the oxime ether 1a can be easily prepared on a larger scale (40 mmol). To gain further insight into these new bifunctional reagents, relevant physicochemical properties were determined, including their reduction potentials and their UV/visible absorption (Figure 2b).^[35] The non-fluorinated oxime ethers 1i and 1j were also synthesized to compare first their reduction potential and UV/visible absorption, and then their reactivity toward the designed process. Due to these physicochemical properties, all oximes exhibit a low reduction potential (below –2.2 V vs SCE). Therefore, a reductive activation of these reagents would be difficult with the most popular photocatalysts. In addition, these oximes feature a maximum absorption wavelength in the near UV region, which would require a high energy to promote the homolytic cleavage of the NO bond, at the cost of a lower functional group tolerance. Hence, this study clearly supports the development of an EnT-based activation to take advantage of these bifunctional reagents.

[IMAGE OMITTED. SEE PDF]

At the outset of our investigation, we chose oxime ether 1a and the vinyl ether 2a, which is an electron-rich olefin likely to react with the electrophilic alkoxy radical (i.e., •OCH₂CF₃) in situ generated through the homolytic NO bond cleavage. Using the (Ir[dF(CF₃)ppy]₂(dtbpy))PF₆ photosensitizer, which is known to have a high triplet-state energy (E_T = 61.8 kcal mol⁻¹),^[36] the desired difunctionalized anti-Markovnikov product 3a was obtained in 35% yield (Table 1, entry 1) in degassed CH₂Cl₂ under irradiation at 405 nm (LEDs) for 16 h. Increasing the quantity of 2a resulted in a slight enhancement of the yield to 39% (Table 1, entry 2). The concentration of the reaction influenced the outcome of the reaction, and the yield increased to 48% with a higher concentration (0.2 M) (Table 1, entry 3). Interestingly, the addition of a base such as NaOAc was beneficial as 3a was observed in 54% yield (Table 1, entry 4). When the reaction was conducted in the presence of 3 equivalents of 2a and sodium propionate (50 mol%), 3a was isolated in 77% yield (Table 1, entries 5–7).^[37] Finally, the use of other organic photosensitizers with high triplet-state energy, such as the thioxanthone (E_T = 63.4 kcal mol⁻¹)^[36] or the 2-chlorothioxanthone (E_T = 62.1 kcal mol⁻¹),^[38] did not lead to any improvement as the desired product was obtained in lower yields, 51–61% yields, respectively (Table 1, entries 8 and 9). Nevertheless, these results clearly support an EnT-based mechanism for this transformation. The role of both the photocatalyst and light is crucial since no reaction was observed during control experiments (Table 1, entries 10 and 11). Notably, the side product 1a’ has been detected by nuclear magneticresonance (NMR) and high-Resolution mass spectroscopy (HRMS)-- analysis in most of the reactions.^[35] This product resulted from the radical recombination of the iminyl radical and the carbon-centered radical (HOC•HCF₃), issued from the 1,2-HAT process on the trifluoroethoxy radical (Figure 1). The side product 1a’ was obtained in high yield when the reaction was conducted with a nondegassed solvent or without 2a (Table 1, entries 12 and 13). Note that traces of trifluoroethanol were also observed, presumably resulting from a hydrogen atom abstraction by the alkoxy radical. Finally, the reaction was conducted in the absence of a photocatalyst under irradiation at 300 nm (Table 1, entry 14). Surprisingly, no reaction occurred, and the oxime 1a was fully recovered. This result highlights the importance of the photosensitizer for the efficiency of the process.

Table 1 Optimization of the oxyimination of vinyl ether 1a.

Entry	Photocatalyst	Ratio 1a:2a	Concentration	Additive	3a [%]	1a’ [%]
1	[Ir]a)	1:1b)	0.1 M	–	35c)	24c)
2	[Ir]	1:2	0.1 M	–	39	2
3	[Ir]	1:2	0.2 M	–	48	7
4	[Ir]	1:2	0.2 m	NaOAc (1 equiv.)	54	14
5	[Ir]	1:3	0.2 M	NaOAc (1 equiv.)	57	8
6	[Ir]	1:3	0.2 M	Sodium propionate (1 equiv.)	62	10
7	[Ir]	1:3	0.2 M	Sodium propionate (50 mol%)	(77)	10
8	TXT	1:3	0.2 M	Sodium propionate (50 mol%)	51	2
9	Cl-TXT	1:3	0.2 M	Sodium propionate (50 mol%)	61	1
10	–	1:3	0.2 M	Sodium propionate (50 mol%)	0	0
11d)	[Ir]	1:3	0.2m	Sodium propionate (50 mol%)	0	0
12e)	[Ir]	1:3	0.2 M	Sodium propionate (50 mol%)	0	58
13	[Ir]	1:0	0.2 M	Sodium propionate (50 mol%)	0	64
14f)	–	1:3	0.2 M	Sodium propionate (50 mol%)	0	0

With the optimized conditions in hand, we began to examine the scope of the oxime ethers in the difunctionalization reaction. Oxime ethers bearing electron-donating groups or halogens on the aromatic ring were all suitable for this reaction. Indeed, bifunctional reagents 1b, 1c, and 1d gave the desired products 3b, 3c, and 3d in good to very good yields (51, 52, and 71%, respectively). Unsymmetrical oxime ethers 1e–1h were also tested, and the corresponding difunctionalized products were isolated in good yields (52–60%). With unsymmetrical oximes with an aliphatic chain, no reaction or traces of products were detected.^[35] Remarkably, when the ethyl oxime 1i, the hydrogenated analog of the oxime ether 1a, or the methyl oxime 1j was used, no conversion was observed, and the oximes were fully recovered. These observations highlighted the crucial presence of the fluorine atoms. Note that the reproducibility of the results was evaluated by synthesizing compound 3a, achieving an average yield of 76% across four independent reactions (Scheme 1).^[35]

[IMAGE OMITTED. SEE PDF]

The scope of the reaction was then pursued using the oxime ether 1a. Various vinyl ether derivatives were smoothly functionalized, providing the expected 2,2,2-trifluoroethoxylated products 3 in good yields. Indeed, vinyl ethers with a shorter chain than 1a or with a cyclopropane unit were converted into the products 3i and 3j. The presence of a dioxolane residue did not affect the reaction, as the derivative 3k was isolated in a good yield. Alkene with a bulkier group on the chain, like the fluorene-containing vinyl ether 2l gave a lower yield. This methodology was compatible with enol ethers containing a tertiary carbon center at the a-position of the oxygen (2m–2q), giving the corresponding products in moderate to good yields (41–60%). Pleasingly, various functional groups were tolerated in this transformation, such as an ether, a trimethylsilyl, an ester or an alkenyl group (3r–3u and 3aa). Notably, we observed a total selectivity toward the enol ether despite the presence of the alkenyl group at the terminal position of the chain of 2aa. This undoubtedly supports the importance of the electron-rich character of the olefins used in this reaction. R(CH₂)₂O-alkenes 2v, 2w, and 2x substituted with an adamantyl, an aromatic, or a heteroaromatic ring, also reacted smoothly. Then, vinyl ethers bearing longer chains were investigated. To our satisfaction, terminal enol ethers 2y, 2z, 2aa, and 2ab containing a polyfluorinated chain, a phenyl, an alkenyl, or an ester group were easily functionalized to afford the products 3y, 3z, 3aa, and 3ab in good yields (43–66%). Note that phenyl vinyl ether 2ac gave the desired product in 30% yield. Then, the reactivity of benzyl enol ethers was examined. Pleasingly, 2ad was successfully transformed into the functionalized product 3ad in 57% yield. Various other benzyl enol ethers 2ae–2ak bearing electron-donating, electron-withdrawing groups, or halogens (Br, Cl, and F) were functionalized in good yields. Remarkably, changing the substitution pattern of the aromatic ring did not affect the reaction, since not only para, meta, and ortho-substituted but also di- and trisubstituted benzyl vinyl ethers were successfully functionalized. Derivatives from natural products and drugs such as menthol, citronellol, (+)-dehydroisoandrosterone, lithocolic acid, and cholestanol also provided the desired 2,2,2-trifluoroethoxylated products 3al–3ap in moderate to good yields, highlighting the compatibility of the reaction conditions with complex molecular architecture. Subsequently, we extended the scope of olefins to other electron-rich alkenes such as N-vinylamides. Functionalized pyrrolidone and benzamide derivatives 3aq and 3ar were isolated, each in 53% yield. Unfortunately, some substrates were reluctant in our hand; styrene derivatives or other non-activated olefins were not functionalized and were fully recovered (Scheme 2).^[35]

[IMAGE OMITTED. SEE PDF]

To further demonstrate the utility of this methodology, scale-up reactions were performed (Scheme 3a). Using standard conditions on a 3 mmol scale, the desired product 3a was isolated in 61% yield, showing a slightly lower yield compared to the original 0.2 mmol scale. Notably, the expensive iridium-based photocatalyst (Ir[dF(CF₃)ppy]₂(dtbpy))PF₆ can be replaced by the 2-chlorothioxanthone, giving a similar yield on a large scale (59% isolated yield). Next, we explored the reactivity and synthetic diversification of the unedited 2,2,2-trifluoroethoxylated products of this reaction (Scheme 3b). Upon reaction with NaBH₃CN, compound 3a provided the β-trifluoroethoxylated amine 4a. Indeed, while we had planned the selective reduction of the iminyl moiety, the reduction of the ether group seemed favorable during the reduction of the imine. Unfortunately, attempts to control the selectivity of the reduction were unsuccessful. Then, we selectively reduced the ester group of the functionalized product 3ab using LiAlH₄ to afford the compound 4ab containing a terminal alcohol in a very good yield. Finally, we reduced the iminyl part of the product 3aq without the release of the pyrrolidone, giving us the product 4aq in an excellent yield.

[IMAGE OMITTED. SEE PDF]

To get more insights into this transformation, mechanistic investigations were performed (Scheme 4). First, we studied the profile of the reaction by monitoring the quantity of the reagent 1a, the product 3a, the by-product 1a’, and trifluoroethanol (TFE) during the reaction using ¹⁹F NMR analysis (Scheme 4a). The formation of 3a was observed, in line with the consumption of 1a. Then, a reaction using the product 3a as substrate was carried out under standard reaction conditions, and the degradation of the product 3a was observed as only 60% of the product 3a remained at the end of the reaction (Scheme 4b). Traces of TFE generated from the product 3a were also observed, demonstrating a possible formation of the alkoxy radical during the degradation. To preclude the involvement of a single electron transfer in this transformation, the cyclic voltammetry analysis of 1a was conducted, and an irreversible reduction wave at E_p(1a) = −2.32 V vs SCE was observed (Scheme 4c). Hence, considering the redox properties of the iridium-based photocatalyst used in this process (E_1/2([Ir]/[Ir]*) = −0.89 V vs SCE,^[39] a reductive pathway is unlikely. Then, UV/visible absorption of the reagent 1a, the substrate 2a, and the photocatalyst were studied (Scheme 4d). At the operational wavelength (λ_max = 405 nm), only the photocatalyst was capable of absorbing light, thereby proving that the presence of the photosensitizer is crucial for this reaction (Table 1, entry 10). Moreover, both reagents (1a and 2a) did not absorb light at the irradiation wavelength, which supported an EnT pathway. The involvement of a charge transfer complex was also ruled out by UV/visible spectroscopy,^[28] while the lack of absorption of the oxime 1a in the phosphorescence wavelength of the Ir-photosensitizer precludes a Förster EnT pathway. Next, to further support this hypothesis, Stern–Volmer luminescence quenching studies were conducted (Scheme 4e). The data distinctly showed that the bifunctional reagent 1a quenched the excited-state photocatalyst, whereas 2a did not affect the luminescence of the photosensitizer. The use of various radical initiators under thermal conditions in the absence of light also did not lead to the formation of product 3a, showing once again the important role of the photocatalyst and its plausible role as photosensitizer (Scheme 4f). Additionally, in the presence of radical scavengers such as TEMPO or BHT, the reaction was partially or totally inhibited (Scheme 4g). The analysis of the crude reaction mixture led us to identify by HRMS analysis the adduct resulting from the addition of TEMPO to the alkene 2a in place of the iminyl radical. Moreover, the product resulting from the reaction between the iminyl radical and BHT was detected by HRMS analysis. These results strongly support the involvement of both alkoxy radical and iminyl radical, plausibly formed through the homolytic cleavage of the oxime NO bond. The determination of the quantum yield (ϕ = 0.4%), along with on/off experiments, suggested that the reaction does not involve a long radical chain process (Scheme 4h).^[35] To further support an EnT transfer mechanism, the triplet state energy of the oximes and photocatalyst was determined. The energy of the triplet state of the iridium catalyst was measured by phosphorescence (61.6 kcal mol⁻¹) in accordance with the literature data (61.8 kcal mol⁻¹).^[35,36] Similarly, the triplet state energy of 1a was measured at 77 K (E_T = 68.7 kcal mol⁻¹). These measurements showcased a good overlap between the phosphorescence of the iridium catalyst and the phosphorescence of 1a. Hence, a process involving an EnT between triplet ³[Ir]* and the ³[1a]* is likely to occur. Note that a similar triplet state energy was measured for both 1a and 1i, suggesting that their divergent reactivity cannot be explained by their triplet energies.

[IMAGE OMITTED. SEE PDF]

Based on all these experiments, we suggested the following plausible mechanism for this oxyimination process (Scheme 4i). First, visible light irradiation (405 nm) of the photosensitizer [Ir] produced the triplet ³[Ir]* (61.6 kcal mol⁻¹). Then, Dexter triplet-triplet energy transfer (EnT) between ³[Ir]* and the reagent 1 would generate ³[1]*, which then triggers the homolytic cleavage of the NO bond, leading to the persistent iminyl radical I and the electrophilic alkoxy radical II. Subsequently, alkoxy radical II would react rapidly with the electron-rich alkene 2 to form the more stable carbon-centered radical intermediate III. Then, the reaction of this radical III with the iminyl radical I would lead to the desired product 3, according to a radical–radical coupling.^[40] Note that an alternative radical chain mechanism involving the addition of the formed carbon-centered radical III to another oxime molecule can be ruled out in light of the measured quantum yield.^[41,42]

Conclusion

In summary, we reported the synthesis of a new class of bifunctional reagents, which enabled the concomitant introduction of an alkoxy radical and an iminyl radical on olefins under the auspices of EnT catalysis with high atom economy. The preparation of a large panel of oxyaminated products was disclosed in good to excellent yield, showcasing a large functional group tolerance. Mechanistic investigations allowed us to confirm the EnT reaction pathway. A key feature of this work is the controlled intermolecular addition of the alkoxy radical onto an olefin over possible competing side processes, such as 1,2-hydrogen atom transfer (1,2-HAT) or β-scission of the alkoxy radical. This work represents the only known example of a bifunctional reagent that enables the introduction of an alkoxy radical within the course of aminohydroxylation reaction and we believe that this report will spur the discovery of other interesting transformations in this field.

Acknowledgements

This work was partially supported by Normandie Université (NU), the Région Normandie, the Centre National de la Recherche Scientifique (CNRS), Université de Rouen Normandie (URN), INSA Rouen Normandie, Labex SynOrg (ANR-11-LABX-0029), the graduate school for research XL-Chem (ANR-18-EURE-0020 XL CHEM), and Innovation Chimie Carnot (I2C). T.B. thanks the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant no. 758710). F.D. thanks Labex SynOrg (ANR-11-LABX-0029) and the Region Normandy (RIN 50% program) for a doctoral fellowship. T.P. acknowledges the PHOTODERACS program (Normandy Region, grant no. 0015464) for funding. Dr. L. Monsigny is acknowledged for initial studies on the synthesis of the bifunctional reagents, and Dr. E. Nobile is greatly acknowledged for preliminary experiments.

Conflict of Interest

The authors declare no conflict of interest.

Data Availability Statement

The data that support the findings of this study are available in the supplementary material of this article.

References

F. Lucio‐Martínez, W. Chaładaj, Adv. Synth. Catal. 2023, 365, 2092.

H. Egami, M. Sodeoka, Angew. Chem., Int. Ed. 2014, 53, 8294.

E. Merino, C. Nevado, Chem. Soc. Rev. 2014, 43, 6598.

T. Besset, T. Poisson, X. Pannecoucke, Chem. Eur. J. 2014, 20, 16830.

L. M. Surhone, M. T. Tennoe, S. F. Henssonow, Vicinal Difunctionalization, Betascript Publishing 2010.

X.‐W. Lan, N.‐X. Wang, Y. Xing, Eur. J. Org. Chem. 2017, 2017, 5821.

K. B. Sharpless, D. W. Patrick, L. K. Truesdale, S. A. Biller, J. Am. Chem. Soc. 1975, 97, 2305.

M. M. Heravi, T. B. Lashaki, B. Fatthi, V. Zadsirjan, RSC Adv. 2018, 8, 6634.

D. J. Ager, I. Prakash, D. R. Schaad, Chem. Rev. 1996, 96, 835.

J. P. Michael, Nat. Prod. Rep. 1996, 16, 675.

S. Dutta, J. E. Erchinger, F. Strieth‐Kalthoff, R. Kleinmans, F. Glorius, Chem. Soc. Rev. 2024, 53, 1068.

V. Kumar Soni, S. Lee, J. Kang, Y. K. Moon, H. S. Hwang, Y. E. J. YouCho, E. J. Cho, ACS Catal. 2019, 9, 10454.

H.‐M. Huang, P. Bellotti, J. Ma, T. Dalton, F. Glorius, Nat. Rev. Chem. 2021, 5, 301.

T. Patra, M. Das, C. G. Daniliuc, F. Glorius, Nat. Catal. 2021, 4, 54.

S.‐Q Lai, B.‐Y. Wei, J.‐W. Wang, W. Yu, B. Han, Angew. Chem., Int. Ed. 2021, 60, 21997.

J. Chateauneuf, J. Lusztyk, B. Maillard, K. U. Ingold, J. Am. Chem. Soc. 1988, 110, 6727.

R. T. McBurney, A. D. Harper, A. M. Z. Slawin, J. C. Walton, Chem. Sci. 2012, 3, 3436.

D. J. Edge, J. K. Kochi, J. Am. Chem. Soc. 1973, 95, 2635.

G. Tan, F. Paulus, A. Renteria‐Gomez, R. F. Lalisse, C. G. Daniliuc, O. Gutierrez, F. Glorius, J. Am. Chem. Soc. 2022, 144, 21664.

F. Paulus, C. Stein, C. Heusel, T. J. Stoffels, C. G. Daniliuc, F. Glorius, J. Am. Chem. Soc. 2023, 145, 23814.

J. Li, Y. Yuan, X. Bao, T. Sang, J. Yang, C. Huo, Org. Lett. 2021, 23, 3712.

L. Chang, Q. An, L. Duan, K. Feng, Z. Zuo, Chem. Rev. 2022, 122, 2429.

S. Liu, S. Wang, P. Wang, Z. Huang, T. Wang, A. Lei, Nat. Commun. 2022, 13, 4430.

A. D. Kharlamova, A. S. Abel, A. D. Averin, I. P. Beletskaya, Russ. J. Org. Chem. 2022, 58, 1181.

G. Kirby, G. Prestat, F. Berhal, J. Org. Chem. 2023, 88, 4720.

R. Möckel, E. Babaoglu, G. Hilt, Chem. Eur. J. 2018, 24, 15781.

M. Inoue, Y. Sumii, N. Shibata, ACS Omega 2020, 5, 10633.

P. Jeschke, Pest. Manage. Sci. 2024, 80, 3064.

Y. Ogawa, E. Tokunaga, O. Kobayashi, K. Hirai, N. Shibata, iScience 2020, 23, 101467.

T. Fujiwara, D. O’Hagan, J. Fluorine Chem. 2014, 167, 16.

A. Gallego‐Gamo, P. Sarro, Y. Ji, R. Pleixats, E. Molins, C. Gimbert‐Suriñach, A. Vallribera, A. Granados, J. Org. Chem. 2024, 89, 11682

A. Gallego‐Gamo, R. Pleixats, C. Gimbert‐Suriñach, A. Vallribera, A. Granados, Chem. Eur. J. 2024, 30, e202303854.

M. Chen, Y. Lu, Y. Shen, Q. Wang, Org. Lett. 2024, 26, 9586.

M. Chen, J. Ling, K. Cheng, Q. Wang, Org. Lett. 2025, 27, 2241.

See Supporting Information for details.

F. Strieth‐Kalthoff, M. J. James, M. Teders, L. Pitzer, F. Glorius, Chem. Soc. Rev. 2018, 47, 7190.

The significant influence of the base on the reaction outcome has not yet been elucidated. Note that the redox potentials of 1a in the presence of sodium propionate or without base were found unchanged. Both redox potentials were determined using ferrocene as an internal reference and then converted to SCE (See Supporting information for more details).

M. J. Oddy, D. A. Kusza, R. G. Epton, J. M. Lynam, W. P. Unsworth, W. F. Petersen, Angew. Chem., Int. Ed. 2022, 61, e202213086.

G. Tan, M. Das, H. Keum, P. Bellotti, C. G. Daniliuc, F. Glorius, Nat. Chem. 2022, 14, 1174.

D. Leeifert, A. Studer, Angew. Chem., Int. Ed. 2020, 59, 74.

J. Majhi, R. K. Dhungana, J. Am. Chem. Soc. 2022, 144, 15871.

R. Laskar, S. Dutta, J. C. Spies, P. Mukherjee, J. Am. Chem. Soc. 2024, 146, 10899.