Programmable synthesis of difluorinated hydrocarbons from alkenes through a photocatalytic linchpin strategy

The introduction of difluoromethylene moieties into organic molecules has garnered significant attention due to their profound influence on the physicochemical and biological properties of compounds. Nonetheless, the existing approaches for accessing difluoroalkanes from readily available feedstock chemicals remain limited. In this study, we present an efficient and modular protocol for the synthesis of difluorinated compounds from alkenes, employing the readily accessible reagent, ClCF2SO2Na, as a versatile “difluoromethylene” linchpin. By means of an organophotoredox-catalysed hydrochlorodifluoromethylation of alkenes, followed by a ligated boryl radical-facilitated halogen atom transfer (XAT) process, we have successfully obtained various difluorinated compounds, including gem-difluoroalkanes, gem-difluoroalkenes, difluoromethyl alkanes, and difluoromethyl alkenes, with satisfactory yields. The practical utility of this linchpin strategy has been demonstrated through the successful preparation of CF2-linked derivatives of complex drugs and natural products. This method opens up new avenues for the synthesis of structurally diverse difluorinated hydrocarbons and highlights the utility of ligated boryl radicals in organofluorine chemistry.

7][28][29][30] For instance, the Ye group successfully activated electron-decient C-H bonds by employing a quinuclidineborane (LB 1) as the HAT reagent to realize a radical hydroalkylation reaction of unactivated alkenes in 2021 (Fig. 1a). 31owever, the XAT process triggered by ligated-boryl radicals was only known to activate alkyl iodides and bromides for decades, [32][33][34][35][36] until recently, when our group successfully activated the C-Cl bond (BDE cal.= 87 kcal mol −1 ) from chlorodi-uoromethane (ClCF 2 H, Freon-22) under blue light irradiation, using commercially available and inexpensive trimethylamineborane (LB 2) as the XAT reagent (Fig. 1b). 37At the same time, the Wang group reported a three-step process for sequential C-Cl bond functionalization of activated trichloromethyl groups with the choice of an appropriate ligated borane reagent in each step (LB 3-5) (Fig. 1c). 38ncouraged by these recent scientic advancements, we aimed to establish a protocol for ligated boryl radical-promoted C-Cl bond functionalization of C(sp 3 )-CF 2 -Cl substrates to provide a novel and streamlined method for accessing gem-diuorinated aliphatic backbones, which were previously challenging to obtain.0][41] The proposed sequence involves two radical carbon-carbon bond formation transformations and has the potential to connect two alkene fragments through the valuable CF 2 -unit (Fig. 1d).0][51] Herein, we collaborated with colleagues from Pzer to prepare ClCF 2 SO 2 Na in kilogram-scale.By merging radical chlorodiuoromethylation of alkenes with a subsequent ligated boryl radical-facilitated hydro-diuoroalkylation of different alkenes, we were able to easily access a wide variety of internal gem-diuoro alkanes under mild visible-light irradiation conditions.Furthermore, photoinduced hydrodechlorination and base-promoted elimination reactions enabled the synthesis of a broad range of terminal gem-diuorinated alkanes and alkenes (Fig. 1d).

Development of hydrochlorodiuoromethylation of alkenes
Our investigation began with the development of a radical chlorodiuoromethylation protocol for alkenes with ClCF 2 -SO 2 Na.A screening of photocatalysts, thiols, and solvents established that the desired product 1j was obtained in up to 86% isolated yield by conducting the reaction with Mes-Acr-Me + ClO 4 − as the photocatalyst and methyl thiosalicylate as the HAT mediator in a mixed solvent of CHCl 3 /CF 3 CH 2 OH (9/1, 0.2 M, see the ESI Table 1 † for optimization details).As shown in Fig. 2A, this protocol accommodated a wide range of unactivated mono-, di-, and trisubstituted alkenes (1a-1s) with good to high yields.Notably, the reaction conditions tolerated various functional groups, including those bearing silicon (1b), phosphine (1c), amine (1d), silyl ether (1k and 1m), free hydroxyl group (1l), aldehyde (1i), tosylate (1n), hetero/aliphaticcyclic rings (1f, 1g, and 1r), alkyne (1k), and bromide (1p).Additionally, a substrate bearing two olen motifs was fully chlorodiuoromethylated at both sites in a 71% yield (1t). Notably, selected complex alkenes derived from natural products or drug molecules were found to be well-tolerated under standard conditions, affording the corresponding diuorinated derivatives (1u-1ae) in good yields.
When the scope of this protocol was extended to styrenes, however, low yields were observed for the target chlorodi-uoromethylation products due to undesired polymerization side reactions in the presence of the acridinium photocatalyst.Notably, cyanoarenes have been previously shown to serve as effective organophotocatalysts for the hydrouoroalkylation of styrenes, with the solvent choice playing a critical role. 52Taking this into account, we investigated various organic photocatalysts and solvents, ultimately nding that the desired reaction could be achieved using 1,2,3,5-tetrakis(carbazol-9-yl)-4,6-dicyanobenzene (4CzIPN) in DMSO (Fig. 2B, see the ESI Table 2 † for screening details).The protocol was shown to be effective for a range of styrenes substituted at ortho-, meta-, or para-positions of the phenyl ring, yielding the desired products in moderate to good yields (1af-1ao).Furthermore, the protocol successfully delivered chlorodiuoromethylation products from substrates including 2-naphthyl (1ap), heteroaromatic (1aq), cyclic internal alkene (1ar), 1,2-disubstituted (1ar-1at) and 1,1disubstituted (1au) styrene derivatives.Importantly, the reaction could be easily scaled up to gram-scale with minimal loss (1k, 1af, 1ag, and 1aj), highlighting the practical synthetic utility of this protocol.In addition, relatively complex styrenes derived from estrone or phenylalanine are also accommodated in this reaction to give 1av and 1aw in moderate yields.

Synthesis of gem-diuorinated alkanes from alkenes
With a series of chlorodiuoromethyl alkanes in hand, our study then focused on the selective activation of the C-Cl bond by choosing compound 1k as the model substrate with unactivated alkenes.A systematic optimization of the reaction parameters, including photocatalysts, ligated boranes (XAT reagents), disuldes (HAT reagents), solvents, and light source, revealed that the desired product 2a could be obtained in 90% yield by utilizing quinuclidine-ligated borane LB 6 in combination with a simple phenyl thiol under blue light irradiation conditions (see the ESI Table 3 † for optimization details).Fig. 3 illustrates the assembly of CF 2 -linked aliphatic hydrocarbons through an organophotoredox-catalysed radical chlorodi-uoromethylation coupled with ligated boryl radical-enabled hydrodiuoroalkylation, starting from ClCF 2 SO 2 Na.As a proof of concept study, we briey explored the scope of ClCF 2 -alkanes (1k, 1af, 1ag, and 1aj) to connect with unactivated alkenes.The study found that a wide range of mono-substituted alkenes with various functional groups, such as tosylate (2b), free hydroxyl group (2c), ester (2e and 2k), amide (2f), carbonyl group (2j and 2x), chloride (2t), and free carboxylic acid (2p), were all compatible to deliver the corresponding diuorinated products in moderate to high yields.Phosphorus (2d), nitrogen (2i), oxygen (2u), and silicon (2o and 2v) atom-tethered alkenes were also competent reaction partners.Furthermore, cyclic structures, including carbazole (2g and 2l), furan (2m), oxetane (2n), cyclododecane (2r) and adamantane (2y), underwent the reaction smoothly.It should be noted that the reaction between ethylene, the largest-volume and cost-effective industrial chemical that has an annual production of over 150 million tons, and 1y led to the formation of gem-diuoroalkane (2q) with a yield of 46%.Additionally, the linchpin strategy demonstrated effective utilization in the late-stage functionalization of complex alkenes derived from pharmaceutical molecules and natural products.Unactivated alkenes containing the core structures of norborene (2z), camphene (2aa), carprofen (2ab), or sclareol (2ac) smoothly reacted with 1y with good yields.Importantly, it was feasible to use the linchpin Lower conversions accounted for the moderate yields of some substrates (i.e., 2b, 2i and 2y), and dechlorination of intermediate 1 would be the major process aer prolonging the reaction time.Alternatively, low solubility of some alkene starting materials or alkyl-CF 2 Cl intermediates in t BuCN might result in moderate yields of some products (i.e., 2ab, 2ad-2ag).
To further demonstrate the synthetic application of the ClCF 2 -alkane intermediates, various transformations have been explored (Fig. 4).Initially, in the absence of an alkene, hydrodechlorination triggered by quinuclidine-ligated borane (LB 6) was achieved by employing a more sterically hindered hydrogen atom donor, bis(2,4,6-triisopropylphenyl) disulde (TRIPS) 2 (Fig. 4A).Substrates derived from both unactivated alkenes and styrenes were found to be suitable for this transformation (3a-3d).Moreover, ClCF 2 -alkanes obtained from natural products, such as oxaprozin, ethylparaben and sclareol, were also efficiently converted to the corresponding products through hydrodechlorination with moderate yields (3e-3g).Furthermore, gem-diuoroalkenes (3h and 3i) were obtained in variable yields by simply treating ClCF 2 -alkane 1 with t BuOK (Fig. 4B). 53,54Notably, ClCF 2 -alkanes derived from styrenes reacted with t BuOK to give conjugated (E)-b-diuoromethyl styrenes (3j-3l) in high yields attributed to the C]C bond migration to more conjugated systems (Fig. 4C).In addition, inspired by recent advances in E to Z photoisomerization of alkenes through an energy transfer process, 55 adding Ir(ppy) 3 as a triplet photosensitizer in a one-pot two-step manner led to the formation of (Z)-b-diuoromethyl styrenes as the major products (3m-3o) with a Z/E ratio of up to 4 : 1 (Fig. 4D).

Mechanistic considerations
In order to gain a better understanding of the C-Cl bond activation mechanism, a series of experiments were conducted (Fig. 5).First, the isolation of byproduct quinuclidine-BH 2 Cl 4 (conrmed by the X-ray crystallographic analysis) provides support for the proposed key XAT process (Fig. 5a).A radical clock experiment using b-pinene resulted in the formation of the ring-opened product 5 in 57% yield (Fig. 5b).In addition, a radical trapping experiment was also performed using 2,2,6,6tetramethylpiperidine-1-oxyl (TEMPO), which completely inhibited the formation of the diuoroalkylation product 2a.Instead, the TEMPO-intercepted species 6 was detected by high resolution mass spectrometry (HRMS) (Fig. 5c).These results strongly suggest the involvement of alkyl-CF 2 radicals in the reaction process.Furthermore, two parallel reactions were conducted using quinuclidine-BH 2 Cl and quinuclidine-BD 2 Cl (about 60% deuterated) with excess deuterium oxide as an additive (Fig. 5d).The results showed that product 3b, with much higher deuterium incorporation, was observed in the latter case (36% vs. 81% deuterium incorporation, respectively).This nding suggests that the hydrogen atom of the CF 2 Hproduct was most likely derived from quinuclidine-BH 3 through a HAT process.In addition, the very low value of quantum yield (F = 0.0013) supports the possibility of an incage mechanism (See the ESI † for details).Stern-Volmer quenching experiments indicated that the excited photocatalyst could be quenched more effectively by PhSSPh than by ligated borane or the alkyl-CF 2 Cl intermediate (See the ESI † for details).
A tentative mechanism is proposed based on all the experimental data and previous literature reports 31,37,38   electron transfer (PCET) process.Subsequently, the nucleophilic boryl radical activates the C-Cl bond of substrate 1 through a XAT process, resulting in the formation of the corresponding CF 2 -alkyl radical (R 1 -CF 2 c).The R 1 -CF 2 c undergoes intermolecular radical addition with the alkene substrate, followed by a HAT process with thiophenol, to generate the desired gem-diuoroalkane 2. Alternatively, the CF 2 -alkyl radical may undergo HAT with amine-borane, leading to the formation of diuoromethyl alkane 3 in the absence of alkenes.

Conclusions
In summary, we have successfully developed a general and widely applicable method for synthesizing diverse diuorinated alkanes and alkenes by leveraging easily accessible ClCF 2 SO 2 Na as a practical diuoromethylene linchpin.This strategy involves an organophotoredox-catalysed hydrochlorodiuoromethylation reaction, followed by a tertiary amine-borane-triggered XAT process under blue light irradiation conditions.Our approach offers several advantages, including broad substrate scope, excellent functional group tolerance, metal-free character, mild reaction conditions, and CF 2 -link-derivatization of complex bioactive alkenes, which demonstrate the potential utilities of this linchpin protocol.Our ongoing research involves merging ligated boryl radicals with transition metal catalysis and exploring its potential applications.