Recent Advances for the Direct Introduction of the CF2Me Moiety

Fluorine-containing molecules are compounds of interest in materials as well as in pharmaceutical and agrochemical industries. Therefore, developments in that research field are tremendous and a special focus was dedicated to the design and the study of emergent fluorinated groups. In particular, the CF2Me residue is attractive as it could be used for example as a bioisostere of the methoxy group. Despite the clear asset that represents this fluorinated moiety and in complement to the traditional approaches used to construct the CF2Me residue from existing functional groups, the quest for direct methodologies for the 1,1-difluoroethylation reaction of molecules has triggered a strong interest from the scientific community. This Mini-review will focus on the recent advances toward the design of reagents and their applications for the direct 1,1-difluoroethylation of various classes of compounds.


INTRODUCTION
The design of tools and unprecedented synthetic pathways in organofluorine chemistry is of prime importance to extend the current tool box (Landelle et al., 2013;Liang et al., 2013;Besset et al., 2014Besset et al., , 2016Egami and Sodeoka, 2014;Merino and Nevado, 2014;Belhomme et al., 2015;Champagne et al., 2015;Ni and Hu, 2016;Lemos et al., 2018;Song et al., 2018). With more than 40% of agrochemicals and 25% of pharmaceuticals having at least one fluorine atom, this research field is very active and has witnessed a strong interest from academic and industry laboratories (Purser et al., 2008;Fujiwara and O'Hagan, 2014;Ilardi et al., 2014;Wang et al., 2014;Gillis et al., 2015;Ni et al., 2015;Meanwell, 2018). Taking benefit from the unique features of the fluorine atom and fluorinated groups, the biological and physical properties of a molecule might be modulated at will (O'Hagan, 2008). Therefore, it appeared as crucial to further develop efficient transformations for the introduction of fluorinated moieties onto complex molecules and to design and study new fluorine-containing moieties. Among these emergent fluorinated groups, a special attention was paid to the 1,1-difluoroethyl group. Present in various compounds of interest, the CF 2 Me residue appears as a fluorinated bioisostere of the alkoxy ethers. Indeed, the replacement of the oxygen atom by a CF 2 residue makes the molecules more metabolically stable and the presence of a CF 2 Me moiety impacts the spatial geometry (difference of conformational preference of an OMe vs. CF 2 Me group) although keeping similar electronic and steric properties (Zhou et al., 2013). In addition, the metabolic stability of the molecules might be modulated by replacing a benzylic methylene residue by a CF 2 one as it was the case for a urea transporter B (UTB) inhibitor (II), used for edema (Anderson et al., 2012). In addition, these compounds bearing this fluorinated moiety already demonstrated interesting properties, like in the case of edema and malaria treatment, for instance (Coteron et al., 2011;Anderson et al., 2012). Consequently, over the last years, the landscape of this research field has been impacted by key contributions from several research groups, who have pioneered the synthesis and application of original reagents to construct C-CF 2 Me bonds. The synthesis of CF 2 Me-containing molecules mainly relied on the fluorination of (1) carbonyl derivatives (or analogs i.e., thiocarbonyl compounds), (2) benzylic positions, (3) alkynes, or (4) alkenes and was well-documented (for selected examples: Markovskij et al., 1973;Middleton, 1975;York et al., 1996;Lal et al., 1999;Reddy et al., 2005;Yamauchi et al., 2008;Umemoto and Singh, 2012;Xia et al., 2013;Ilchenko et al., 2014;Okoromoba et al., 2014;Xu et al., 2014;Ma et al., 2015;Koperniku et al., 2016;Hua et al., 2017;Li et al., 2017;Zhao et al., 2017;Iacono et al., 2018;Tomita et al., 2018). In sharp contrast, the design of new reagents or methodologies for the direct incorporation of this fluorinated moiety onto molecules is still underdeveloped. Recently Li, Dong et al. reviewed the synthesis of (1,1-difluoroethyl)arenes based on the construction of the CF 2 Me motif and few examples describing its direct incorporation onto arenes were reported (Li et al., 2018).
The aim of this mini-review is to showcase and discuss the recent advances made on novel synthetic strategies for the direct introduction of the emergent CF 2 Me group onto various classes of molecules (arenes, aliphatic, and carbonyl compounds). Therefore, in this review, we will highlight the new technological solutions based on the design of original reagents and methodologies to build up C-CF 2 Me bonds. Note that the approaches employed to construct the CF 2 Me group will not be discussed. Taking these considerations in mind, a first strategy to SCHEME 1 | 1,1-Difluoroethylation of carbonyl derivatives 2 with the TESCF 2 Me reagent. [a] 1 H NMR yield of the crude reaction mixture. access to CF 2 Me-containing molecules relies on the introduction of the CF 2 Me group by using nucleophilic reagents, whereas a second approach deals with the construction of a C-CF 2 Me bond according to a radical pathway.
1,1-DIFLUOROETHYLATION OF MOLECULES USING A NUCLEOPHILIC CF 2 ME-CONTAINING REAGENT Transition Metal-Free Reactions to Access CF 2 Me-Containing Molecules In this section, key advances for the direct 1,1-difluoroethylation of carbonyl derivatives via a transition metal-free process will be depicted.
TESCF 2 Me and TMSCF 2 Me as CF 2 Me Sources In their quest for new fluorine-containing reagents, Prakash et al. (Mogi et al., 2007) reported the synthesis of a 1,1difluoroethylated reagent, the TESCF 2 Me (Scheme 1). The latter resulted from the reaction between TESCl and (1,1difluoroethyl)phenylsulfone 1 in the presence of magnesium metal in 35% yield on a gram scale. The precursor 1 was itself prepared in three steps from thiophenol (Langlois, 1988;Mogi et al., 2007). Note that the nature of the solvents (THF/HMPA, 1:1) was crucial to ensure the full conversion of the sulfone 1 into the desired TESCF 2 Me reagent. When this reagent was reacted with aromatic aldehydes 2, the corresponding 1,1-difluoroethylated secondary alcohols 3 were obtained in moderate to good 19 F NMR yields (50-77%). The reaction was tolerant to electron-donating groups (3b,c) and halogen (3d). The reaction was also sensitive to steric hindrance since when the reaction was conducted with a sterically hindered aldehyde or ketone, only traces of products were obtained. Finally, an enolizable aldehyde and ketone and an α,β-unsaturated SCHEME 2 | (1,1-Difluoroethyl)trimethylsilane as CF 2 Me source for the functionalization of carbonyl derivatives.
aldehyde were not suitable substrates in that transformation and constituted the main limitations.

Use of a CF 2 Me-Substituted Phosphonium Salt
About 10 years later, Xiao and co-workers reported a new methodology to access CF 2 Me-containing molecules by means of a nucleophilic source . As an alternative to the powerful TESCF 2 Me reagent, which has a low boiling point and which required a tedious multi-step synthesis [four steps from thiophenol (Mogi et al., 2007)], Lin and Xiao designed a nucleophilic reagent based on a phosphonium salt. The shelfstable 1,1-difluoroethyl phosphonium salt was prepared in threesteps from the commercially available and inexpensive PPh 3 and EtBr, in a 29% overall yield (Scheme 3). After the formation of the ethyl phosphonium salt, a subsequent two-step fluorination sequence with NFSI, allowed the formation of the desired reagent on a gram-scale.
This reagent was then successfully applied for the 1,1difluoroethylation of (het)aromatic aldehydes and the 4-phenyl acetophenone 10. Electron-rich aryl aldehydes were smoothly converted into the desired secondary alcohols in moderate to high yields, as illustrated with the para-phenyl derivative 11a and the para-methoxy derivative 11b. A slight decrease of yield was observed when the reaction was performed from the sterically hindered electron-rich aldehyde 10c. Aryl aldehydes bearing an electron-withdrawing substituent (i.e., CF 3 , CN, halogens) were also suitable substrates 10d-h for the reaction. In addition, the substitution pattern had no impact on the reaction outcome since the para-, meta-, and ortho-CF 3 substituted aryl alcohols 11d,e,h were obtained in good yields (68, 73, and 81%, respectively). Note that the 3-quinolinecarboxaldehyde 10i was functionalized in a good yield. However, only traces of the desired 1,1-difluoroethylated product were detected when starting from an aliphatic aldehyde and a lower 19 F NMR yield (21%) was observed when the reaction was carried out with 4-phenyl acetophenone. To further demonstrate the potential of the methodology, the authors applied their methodology to the functionalization of the N-tosyl imines 12a-f and the corresponding products 13 were obtained in moderate to good yields (39-66%). It is worth mentioning that due to the lower reactivity of N-tosyl imines 12 compared to aldehydes 10, 1,1-difluoroethylated amines 13 were obtained in lower yields than 1,1-difluoroethylated alcohols 11. A plausible mechanism was suggested by the authors. After reaction of the Cs 2 CO 3 promoter with the 1,1-difluoroethyl phosphonium salt, the intermediate I would be formed. This latter might then undergo a decarboxylation reaction to afford Ph 3 PO and release a nucleophilic CF 2 Me residue from the cleavage of the P-CF 2 bond. Then, this species might react with 10 or 12 to afford the desired 1,1-difluoroethyated compound 11 or 13.
Transition Metal-Promoted Reactions to Access CF 2 Me-Containing Derivatives As a complementary strategy, the development of transition metal promoted 1,1-difluoroethylation of molecules offered an efficient synthetic pathway to build up C-CF 2 Me bonds. SCHEME 3 | Application of the 1,1-difluoroethyl phosphonium salt for the 1,1-difluoroethylation reaction of carbonyl derivatives. SCHEME 4 | Access to the CF 2 Me-containing molecule 14 via a two-step process. 1,10-Phen. = 1,10-Phenanthroline.

Access to 1,1-Difluoroethylated Derivatives From Organozinc Reagents
The group of Dilman (Zemtsov et al., 2014) depicted a two-step process for the synthesis of 1,1-difluoroethylated derivatives, thanks to the in situ formation of the MeCF 2 ZnX species. The reaction of the difluorocarbene with MeZnI formed the desired organozinc reagent. The latter was engaged in a copper-catalyzed allylation reaction and allowed the synthesis of the corresponding CF 2 -containing molecules. With this methodology, a single example of a CF 2 Me-containing molecule 14 was prepared in 66% yield (Scheme 4).

Copper-Mediated 1,1-Difluoroethylation Reaction Using TMSCF 2 Me as a Fluorinated Source
In 2016, Hu and co-workers (Li et al., 2016) described the coppermediated 1,1-difluoroethylation reaction of diaryliodoniums triflate 15 using CuCF 2 Me (Scheme 5). The CuCF 2 Me reagent was in situ generated from the reaction of CuCl with TMSCF 2 Me (Mogi et al., 2007) in the presence of tBuOK. Then, the synthesis of (1,1-difluoroethyl)arenes 16 was carried out. In presence of 1.5 equivalents of the in-situ generated CuCF 2 Me and 0.5 equivalents of Et 3 N·3HF, as a crucial additive, a panel of electron-rich diaryliodoniums salts were smoothly converted into the desired CF 2 Me-containing arenes 16a,b,g in good yields. The reaction conditions were also tolerant toward diaryliodoniums salts bearing an electron-withdrawing group such as halogen, ketone, ester, aldehyde, and nitro groups to furnish the 1,1-difluoroethylated arenes 16c-f,h in moderate to high yields. Sterically hindered diaryliodonium salts 15i and 15j were also suitable substrates under these reaction conditions and furnished the desired products 16ij in good yields. The potential of this strategy was further demonstrated by the 1,1-difluoroethylation of analogs of relevant compounds such as estrone 15k and the anti-inflammatory drug naproxen 15l. Concerning the mechanism, the authors ruled out a radical pathway and proposed the following mechanism pathway: formation of the intermediate Cu(III) species I, which would result from the oxidative addition of the CuCF 2 Me species with the diaryliodonium salt 15, followed by a final reductive elimination step to furnish the expected product 16. SCHEME 6 | Cobalt-catalyzed 1,1-difluoroethylation of ArMgBr 17 with BrCF 2 Me.
[a] Yields shown in parenthesis were determined by 19 F NMR.

Cobalt-Catalyzed 1,1-Difluoroethylation Reaction of Aryl Grignard Reagent Using BrCF 2 Me
A complementary synthetic route toward the synthesis of 1,1-difluoroethylated arenes was reported by Yamakawa and Ohtsuka (Ohtsuka and Yamakawa, 2016). The cobalt-catalyzed 1,1-difluoroethylation of aryl Grignard derivatives 17a-g using BrCF 2 Me was developed (Scheme 6). Various aryl Grignard derivatives were functionalized using two sets of reaction conditions. Note that the nature of the ligand and the solvent played an important role in each catalytic system. Aryl Grignard reagents bearing electron-donating and electron-withdrawing groups at the para position were functionalized leading to the corresponding products 18 in low to good yields. It turned out that the substitution pattern had an impact on the reaction outcome since the derivative 18g bearing a methoxy group at the ortho-position was obtained in a lower yield compared to the compound 18b and 18f with both reaction conditions (A and B). Note that, except for product 18c, the isolated yields were rather lower than the ones determined by 19 F NMR, presumably due to volatility issue of the fluorinated products.

DIRECT INTRODUCTION OF THE CF 2 ME MOIETY VIA A RADICAL PROCESS
In this section, the recent breakthroughs for the direct and selective incorporation of the CF 2 Me residue onto molecules via a radical process will be discussed. Note that this strategy was mainly used to access 1,1-difluoroethyl-containing heteroarenes.

DFES-Na as the 1,1-Difluoroethyl Source
In 2013, a pioneer work was reported by the group of Baran (Zhou et al., 2013). They designed the synthesis of the sodium difluoroethylsulfinate (DFES-Na) and investigated its application to the functionalization of various heterocycles. The DFES-Na reagent was obtained in two steps from the Hu's reagent and was prepared on a large scale (>100 g). Under oxidative and robust conditions (TBHP, water as cosolvent under air), various classes of heteroarenes 19 were functionalized (21 examples) in the presence of ZnCl 2 and TsOH . H 2 O. The transformation turned out to be functional group tolerant and moderate to good selectivity was observed.
In addition, the authors demonstrated the possible direct radical functionalization of Michael acceptors 21 and thiol derivatives 22 (Scheme 7).
This reagent was then used by the group of Vincent (Ryzhakov et al., 2017) for the direct introduction of the CF 2 Me residue on protected indoles 25. The reaction led to the corresponding fluorinated spirocyclic indolines 26 under oxidative conditions (2 examples, Scheme 8). SCHEME 7 | 1,1-Difluoroethylation of heteroarenes and extension to other classes of compounds with the DFES-Na reagent.
Frontiers in Chemistry | www.frontiersin.org CF 2 Me-Containing Sulfones as the 1,1-Difluoroethyl Source In 2015, the group of Dolbier (Zhang et al., 2015) reported the synthesis of fluorinated phenanthridines 28 under visible light photoredox catalysis. Starting from biphenyl isocyanides 27, the 1,1-difluoroalkylation occurred using an Ir photocatalyst via a tandem addition/cyclization/oxidation sequence. Although the study mainly focused on the difluoromethylation reaction, the introduction of the 1,1-difluoroethyl radical was readily performed using the MeCF 2 SO 2 Cl reagent as precursor of the CF 2 Me radical (Scheme 9). Using this reaction manifold, four CF 2 Me-containing phenanthridines were synthesized in good yields (up to 83%). The following mechanism was proposed by the authors: first, the generation of the radical CF 2 Me from the reduction of the MeCF 2 SO 2 Cl reagent with the excited Ir catalyst followed by its addition on the isocyanides 27. Then a cyclization would lead to the corresponding radical A, which would be oxidized into the species B, regenerating the Ir-catalyst. A final deprotonation of B would yield to the expected products 28.
In 2017, the group of Dolbier (Zhang et al., 2017) developed a radical fluoroalkylation of unactivated alkenes under photoredox catalysis to build up fluorinated tetralin derivatives. In that context, a single example of the direct introduction of the CF 2 Me group to build up the corresponding carbocyclic compound under Ir catalysis was depicted. The expected product 33 was obtained in 64% yield using the MeCF 2 SO 2 Cl reagent (Scheme 11).

SUMMARY AND OUTLOOK
In this Mini-review, we discussed the recent advances made for the direct and selective introduction of the valuable CF 2 Me group. As alternative to the traditional synthetic routes, these direct approaches represented efficient and new retrosynthetic pathways. In this context, technological solutions based on the design of new tools (original strategies and reagents) for the 1,1-difluoroethylation of several classes of compounds were designed. Two strategies were employed based either on the use of nucleophilic reagents or precursors of CF 2 Me radical sources. In the former case, transition metal free approaches as well as transition metal-mediated or -catalyzed transformations were depicted. The complementary synthetic pathway relied on the use of a radical process to access CF 2 Me-containing molecules using carefully designed reagents. Beyond these impressive achievements, further developments for the direct introduction of the CF 2 Me residue, especially for the late-stage functionalization of complex (bioactive) molecules will be of prime importance to enlarge the existing toolbox. Thanks to the pivotal importance of the organofluorine chemistry in various research fields, we strongly believe that this Mini-review will offer new perspectives to further study and apply this original fluorinated moiety.

AUTHOR CONTRIBUTIONS
EC and TB collected the literature data related to this review article. EC, TP, PJ, and XP wrote sections of the manuscript. TB wrote the first draft of the manuscript. All authors contributed to the final version of the manuscript and approved the submitted version.