English

• Due to their unique physiochemical properties and therapeutic efficacies in drug discovery and development [1-5], fluorine-containing motifs have drawn considerable attention from synthetic chemists over the past decade [6-10]. As a consequence, diversified electrophilic, nucleophilic or free-radical fluorination and fluoroalkylation reagents have been developed to address their synthesis through converting fluoride-free starting materials to fluorine-containing chemicals in a highly efficient manner [11-15]. However, these reactions usually proceed with specially pre-functionalized substrates and under harsh conditions, thus giving limited types of fluorinated products with relatively low step-/atom-economy and functional group (FG) tolerance. Besides, recently environmental enforcement has given a great challenge for both the supply chain and the future use of these developed fluorinated reagents since environmentally harmful raw materials, solvents and/or additives were often involved for their preparation and application [16-18]. These serious situations greatly hindered their application. Therefore, further exploration of new, general and environmentally benign fluorinated reagents/methodologies for building fluorine-containing motifs remains one of the hottest topics.

  Taking advantages of the exceptional advantage of C—H functionalization strategy in terms of activity, selectivity, atom-/step-economy, substrate scope and functional group tolerance [19-28], more recently, several easily available fluorinated π-compounds including perfluoroalkenes [29-32], gem–difluoroalkenes [33-40], gem–difluoromethylene alkynes [41-45], monofluoroalkynes [46] and gem–difluorocyclopropenes [47,48] have been developed as versatile coupling partners (CPs) to address this issue via transition metal (TM)-catalyzed C—H functionalization (Scheme 1a). Indeed, such fluorinated π-coupling partners-involved C—H activation mode often lead to the incorporation of the fluorine fragment into the structural backbone of the product, thereby delivering the corresponding fluorine-containing motifs with improved structural diversity and FG tolerance. Despite these notable progresses made in the organic integration of TM-catalyzed C—H activation with fluorinated π-compound CPs towards their synthesis, all the protocols developed to date remain limited to fluorinated alkyne/alkene-mediated reactions. Given the significance of fluorine-containing core in nature, thus, there is an urgent need to further expand the new fluorinated CPs with innovative C—H activation reaction mode to construct fluorine-containing motifs for meeting their increasing demands.

  Scheme 1

  

  Scheme 1.  TM-catalyzed C—H functionalization with fluorinated π-compounds.

  DownLoad: Full-Size Img PowerPoint

  Inspired by the above information, we turn our attention to the allene species since it features impressive reactivity and has been successfully documented in recent C—H activation reactions as a class of versatile synthons [49-52]. However, owing to the presence of two orthogonal double bonds in the allene moiety, the chemo- and regioselectivity control remains a challenging topic for allene-mediated C—H activation reactions (Scheme 1b) [53,54]. In general, the substituent bearing a big steric and/or strong electronic effect is needed to be introduced for realizing the desired control of the selectivity, which would overshadow their further derivatization [55-58]. In view that the use of difluorinated π-compounds often leads to specific migratory insertion driven by the unique fluorine effect in the stage of C—H activation/C—C coupling [40,59,60], we envisioned that a reasonable strategy for achieving the chemo-/regioselectivity control of the allene moiety might be recognized by introducing the proper difluorine-containing fragment into the allene scaffold. Taking all these into account, we decided to design a new gem–difluorine-containing allene species to coordinate the reaction characteristics of the allene moiety and the gem–difluorine-enabled fluorine effect for the chemo-/regioselectivity control. To the best of our knowledge, such an assumption involving gem–difluorine-functionalized allenes has not been revealed to date in the field of C—H activation. Herein, we would like to report the first use of gem–difluoromethylene allenes for the regiocontrolled assembly of isoquinolone and pyridinone derivatives bearing a biologically important monofluoroalkene side chain [5,61] at the 3-position with exclusive Z-selectivity via Rh(Ⅲ)-catalyzed C—C coupling/C—N cyclization (Scheme 1c).

  To test our hypothesis, we commenced our investigation by employing N-pivaloyloxybenzamide 1a with gem–difluoromethylene allene 2a as model substrates. Preliminary screening demonstrated that the desired C—H coupling occurred under the classic Cp^*Rh(Ⅲ) catalytic system, resulting in the formation of the interesting 3-fluoroalkenyl isoquinolone framework 3aa (see Table S1 in Supporting information for details). To our surprise, this transformation featured an unconventional allene insertion mode (via 1,2 double bond insertion) in comparison with previously reported TM-catalyzed C—H functionalization reactions involving monosubstituted allenes (via 2,1 or 2,3 double bond insertion) [53,54]. The introduction of the gem–difluoromethylene fragment might account for this phenomenon, thereby giving a fruitful expansion for the reaction manifold of allene substrates. Encouraged by this finding, we were next intrigued to probe the optimal reaction conditions. Preliminary examination of different TM catalysts gave inferior results, and the brief screening of the solvent and additive revealed that [Cp^*RhCl₂]₂/KOPiv in TFE was the optimal composition, affording 3aa in 75% yield. The switch of other amide substrates bearing different oxidizing directing groups (ODGs) gave relatively low efficiency. After systematically screening of other experimental parameters such as catalyst loading, the reaction temperature and the ratio of starting materials, to the end, we were pleased to identify the optimized reaction conditions and furnished the desired product 3aa in 81% isolated yield.

  Having established the optimal reaction conditions, we next explored the compatibility and generality of the developed protocol. As shown in Scheme 2a, a diverse array of N-pivaloyloxybenzamides were first examined to couple with gem–difluoromethylene allene 2a and showed good FG tolerance. Various commonly encountered substituents including alkyl (3ba and 3ca), methoxyl (3da), methylthio (3ea), halogens (3fa-3ha), trifluoromethyl (3ia), ester (3ja), cyano (3ka), nitro (3la) and phenyl (3ma) group were all tolerable to afford the desired cyclization products in moderate to good yields. The structure of compound 3aa was further confirmed by X-ray crystallography and revealed a specific (Z)-selective configuration of the fluoroalkenyl moiety (CCDC: 2126006). Notably, ortho-substituted substrates that have been proven to be inefficient in most C—H activation cases were also feasible to participate in this transformation, delivering the corresponding fluoroalkenyl tethered isoquinolones in synthetically useful yields (3na-3pa). With meta-substituted N-pivaloyloxybenzamides being the substrates, the desired C—C coupling/C—N cyclization occurred effectively with a preference at the less-hindered site (3qa-3ta). Gratefully, the coupling of gem–difluoromethylene allene 2a with complex N-pivaloyloxybenzamides derived from pharmaceutical ingredients such as probenecid (anti-gout agent), roflumilast (PDE-4 inhibitor) and pleconaril analog (antiviral agent) took place smoothly to yield the desired isoquinolone derivatives 3ua-3wa in decent isolated yields, demonstrating the synthetic potential of the developed protocol for late-stage C—H modification of bioactive molecules.

  Scheme 2

  

  Scheme 2.  Scope for (hetero)aromatic and vinylic C—H functionalization with gem–difluoromethylene allenes. Reaction conditions: 1 (0.2 mmol), 2 (0.2 mmol), [Cp^*RhCl₂]₂ (2.5 mol%) and KOPiv (1 equiv.) in TFE (0.1 mol/L) at room temperature for 24 h under air; isolated yields were reported. ^a The yield was obtained with 5 mol% of [Cp^*RhCl₂]₂ catalyst loading. ^b Dihydropyridin-2(1H)-one derivative 4ea' was detected as the side-product, see Supporting information for details.

  DownLoad: Full-Size Img PowerPoint

  Given the distinctive regioselectivity and FG compatibility demonstrated in benzamide substrates, we were next intrigued to extend the developed protocol to heteroaromatic and vinylic N-pivaloyloxy amides (Scheme 2b). As predicted, N-(pivaloyloxy)furan-2-carboxamide and N-(pivaloyloxy)thiophene-2-carboxamide were proved to be feasible for this transformation, affording the desired pyridinone derivatives 4aa and 4ba in moderate yields. Vinylic N-pivaloyloxy amides were also available substrates, leading to the formation of desired C—C coupling/C—N cyclization products 4ca-4ea effectively. Subsequent examination of different gem–difluoromethylene allenes was conducted to further probe the robustness of this transformation (Scheme 2c). Interestingly, the results revealed that several 1,3-disubtituted allenes were also available CPs, furnishing the corresponding products 3ab-3ad in decent yield with identically specific regioselectivity demonstrated by monosubstituted allenes, revealing that the presence of gem–difluorine moiety played a crucial role for determining the chemo-/regioselectivity control. Moreover, gem–difluoromethylene allenes derived from bioactive natural products, e.g., cyclamen aldehyde and citronellal, were also good reactants for the developed transformation, delivering the desired fluorinated isoquinolones 3ae and 3af in 66% and 73% yields, respectively.

  To further probe the reaction mechanism, in particular, to clarify the unconventional regioselectivity of the developed transformation, the detailed DFT calculations were next carried out using Gaussian 09 by selecting the five-membered rhodacycle INT-1 as the starting point (zero value of energy) [37,62,63]. In consideration of the electron-withdrawing effect of the gem–difluoromethylene fragment, we envisioned that the allenic C—H bond cleavage might occur owing to its relatively high acidity. As shown in Fig. S1 (Supporting information), the coordination of gem–difluoromethylene allene to INT-1 resulted in the formation of intermediate INT-2a with a free energy of −1.1 kcal/mol. Of note, the distinctive short distance of 2.16 Å between the allenic terminal hydrogen and the carbonyl oxygen of OPiv was observed in INT-2a, thus providing a possible allenic C—H bond activation step via concerted metalation deprotonation process. In this view, a Rh-assisted hydrogen transfer process probably occurred via TS-1a (ΔG^≠ = 29.4 kcal/mol), affording the intermediate INT-3a with a free energy of 17.6 kcal/mol. The direct C—C bond reductive elimination from INT-3a was ruled out due to the relatively high energy barrier of 36.6 kcal/mol (from INT-2a to TS-2a′), while an alternative oxidative addition proceeded via TS-2a (ΔG^≠ = 26.8 kcal/mol) to give the Rh(V) intermediate INT-4a along with the cleavage of N—O bond. Further C—C bond formation from INT-4a occurred facilely via TS-3a (ΔG^≠ = −2.7 kcal/mol), delivering the key ortho-allenylation intermediate INT-7b with a free energy of −49.9 kcal/mol. To sum up, the unconventional allenic C—H bond activation process was involved with the overall energy barrier of 30.5 kcal/mol (from INT-2a to TS-1a).

  Having clarified the unexpected allenic C—H bond activation process with a relatively high energy barrier, we were next intrigued to calculate the Gibbs energy profiles of other alternative reaction paths. Due to the equipment of two double bonds, classic double bond insertion was usually proposed for allene-involved C—H functionalization reactions. Herein we compared different insertion modes of the allene moiety into the C-Rh bond and the results were shown in Fig. 1. The coordination of 1,2-double bond in the gem–difluoromethylene allene to INT-1 followed by the migratory insertion via TS-1b was more reasonable owing to the relatively low energy barrier of 15.1 kcal/mol (from INT-2b to TS-1b), delivering the seven-membered rhodacycle INT-3b with a free energy of −8.8 kcal/mol (Fig. 1A). As a comparison, other insertion modes involving contrary regioselectivity and different double bonds were ruled out since the high energy barriers of 17.1 (from INT-2a to TS-1a′), 19.2 (from INT-1 to TS-1b) and 19.8 (from INT-1 to TS-1c) kcal/mol were involved. These results were in line with the experimental observation that specific 1,2-regioselective allene insertion products were formed.

  Figure 1

  

  Figure 1.  Computational mechanistic studies for the allene insertion process. (A) Computed Gibbs energy profiles; (B) Calculated frontier-molecular-orbital (FMO) diagram for 2a and INT-1, isovalue = 0.05; (C) Non-covalent interactions of TS-1b/1a′: Geometries of TS-1b (a), IGMH maps of TS-1b (b), Geometries of TS-1a′ (c), IGMH maps of TS-1a′ (d), isosurfaces of δ_g^inter = 0.004 a.u.; green surfaces represent the weak interaction.

  DownLoad: Full-Size Img PowerPoint

  Moreover, the different regioselectivity for allene insertion was next illustrated by the calculated frontier molecular orbitals (Fig. 1B). The gem–difluoromethylene allene 2a was selected as a model molecule, of which the HOMO-2 (−7.34 eV) and HOMO-3 (−7.58 eV) orbitals were closely related to the 1,2- and 2,3-double bond, respectively. Analyzing the HOMO and LUMO energies of the allene 2a and the rhodacycle INT-1, it clearly revealed that the 1,2-double bond insertion was thermodynamically more favorable due to the relatively small HOMO-2/LUMO gap in comparison with 2,3-double bond insertion mode. To further clarify the difference between 1,2-/2,1-double bond insertion, and probe the potential role of the gem–difluoromethylene in controlling the observed regioselectivity, we then carried out detailed DFT calculations to investigate the non-covalent interactions of the transition states TS-1b/1a′ [64]. As shown in Fig. 1C, both the carbonyl oxygen atom and the fluorine atom acted as hydrogen bond acceptors in TS-1b to form potent non-covalent weak interactions, thus rendering TS-1b in a relatively stable conformation. As a comparison, no obvious non-covalent interaction between the gem–difluoromethylene fragment and the rhodacycle species in TS-1a′ was found, which probably accounted for the relatively high energy barrier of TS-1a′ (17.1 vs. 15.1 kcal/mol).

  With the mechanistic insight into a unique 1,2-double bond insertion established, the detailed reaction path for the formation of the final 3-fluoroalkenyl isoquinolone framework from intermediate INT-3b was then investigated (Fig. 2). Selective β-H elimination via TS-2b (ΔG^≠ = 2.8 kcal/mol) afforded the allene intermediate INT-5b with a free energy of 1.2 kcal/mol, which further underwent a HOPiv-assisted hydrogen transfer via TS-3b (ΔG^≠ = 12.9 kcal/mol) with an energy barrier of 21.7 kcal/mol (from INT-3b to TS-3b), delivering the Rh(I) species INT-6b. Subsequent oxidative addition via TS-4b (ΔG^≠ = 5.2 kcal/mol) led to the cleavage of N—O bond, thus re-oxidizing the Rh(I) species to the Rh(Ⅲ) intermediate INT-7b with an obvious exothermic process. By comparison, the highest energy barrier from INT-1 to INT-7b involving the allene insertion process was 21.7 kcal/mol, which was obviously lower than that of the pathway shown in Fig. 1 (21.7 vs. 30.5 kcal/mol), suggesting that the allene insertion mode should be more reasonable than allenic C—H bond activation. Alternatively, other reaction pathways from INT-3b including sequential oxidative addition/β-H elimination via TS-2b′ (ΔG^≠ = 1.0 kcal/mol) and TS-3b′ (ΔG^≠ = 17.4 kcal/mol) and direct C—N reductive elimination via TS-2b′′ (ΔG^≠ = 15.4 kcal/mol) were also excluded due to the high energy barriers of 32.6 (from INT-4b′ to TS-3b′) and 24.2 (from INT-3b to TS-2b′′) kcal/mol. Furthermore, the cyclization from INT-7b proceeded via the insertion of allene double bond into the N-Rh bond, affording the desired isoquinolone skeleton INT-9b via TS-5b (ΔG^≠ = −35.5 kcal/mol). Finally, the selective β-F elimination was calculated and revealed that the product bearing a Z-type configuration was more favorable both thermodynamically (−91.2 vs. −89.2 kcal/mol) and dynamically (14.8 vs. 15.4 kcal/mol), which was in good agreement with our experimental results.

  Figure 2

  

  Figure 2.  Computed Gibbs free energy changes of the reaction pathway for the formation of 3-fluoroalkenyl isoquinolone from INT-3b.

  DownLoad: Full-Size Img PowerPoint

  To gain more insight into the reaction mechanism, several experimental studies were further conducted. As shown in Scheme 3a, deuterium-labeling experiments in TFE-d₃ revealed that obvious deuterium incorporation was detected at the ortho-position of the ODG, suggesting a reversible C—H metalation process. In addition, the developed Rh(Ⅲ)-catalyzed C—H coupling of N-pivaloyloxybenzamide 1a with gem–difluoromethylene allene 2a occurred smoothly in TFE-d₃ to give the desired product 3aa without obvious deuteration at both the phenyl ring and the 3-fluoroalkenyl moiety, implying that the allene migratory insertion was fast and the fluoroalkenyl substituent was formed via β-F elimination rather than the protonolysis of C-Rh bond. Control experiments using allenic ketone 5 or monofluoroallene 7 as the CP were next conducted to define the role of the gem–difluoromethylene fragment. The results revealed that a mixture of regioisomers was obtained for both cases, illustrating that the equipment of the gem–difluoromethylene functionality in the allene substrate was crucial for realizing the specific regioselectivity (Scheme 3b). Further derivatization of product 3aa was next carried out to probe the synthetic potential of this protocol (Scheme 3c). Exposure of 3aa to m-CPBA led to the formation of dione derivative 9 via selective oxidation of the monofluorided alkene moiety, and the treatment of 3aa with different halohydrocarbons delivered the N-substituted isoquinolones smoothly, albeit with a Z/E mixture. Taken together, these results further strengthen the synthetic potential of the developed protocol for the construction of intriguing isoquinolone frameworks.

  Scheme 3

  

  Scheme 3.  Experimental mechanistic studies and product derivatization.

  DownLoad: Full-Size Img PowerPoint

  On the basis of the above experimental and computational results, we proposed a HOPiv-assisted redox-neutral Rh(Ⅲ)-Rh(I)-Rh(Ⅲ) catalytic cycle involving a tandem regioselective allene 1,2-insertion/β-H elimination/hydrogen transfer/oxidative addition/cyclization/cis-β-F elimination sequence for the developed transformation [65-67] (Scheme 4). Initially, the active catalyst Cp^*Rh(OPiv)₂ was formed readily via ligand exchange, followed by a reversible C—H metalation to afford the five-membered rhodacycle A. Subsequent regioselective migratory insertion of the allene 1,2-double bond into C-Rh bond delivered the intermediate B with an energy barrier of 15.1 kcal/mol, which underwent β-H elimination with an energy barrier of 11.6 kcal/mol to give the allene intermediate C. Sequential hydrogen transfer/oxidative addition process proceeded to furnish the intermediate D, which involved an energy barrier of 21.7 kcal/mol. Next, the intramolecular cyclization occurred via selective insertion of the allene double bond into the N-Rh bond with an energy barrier of 14.4 kcal/mol, and resulted in the construction of the isoquinolone framework. Finally, selective cis-β-F elimination to quench the C-Rh bond with an energy barrier of 14.8 kcal/mol led to the final formation of the product 3aa and the active Rh(Ⅲ) catalyst in a redox-neutral manner. The detailed DFT calculations demonstrated that the non-covalent weak interaction network between the gem–difluoromethylene part and the OPiv moiety attributed to the unconventional and specific regioselectivity.

  Scheme 4

  

  Scheme 4.  Proposed catalytic cycle.

  DownLoad: Full-Size Img PowerPoint

  In conclusion, by virtue of newly-developed gem–difluoromethylene allenes as versatile CPs, an efficient and regiocontrolled Rh(Ⅲ)-catalyzed C—H activation/C—C coupling/C—N cyclization cascade has been developed successfully for the construction of 3-fluoroalkenyl isoquinolones and pyridinones, in which a unique 1,2-migratory insertion of the allene unit was involved. Either (hetero)aromatic or vinylic amides with the embedment of various functional groups were proven to be viable substrates for the established transformation. Through combined DFT calculations and experimental mechanistic studies, the role of the gem–difluoromethylene moiety, the key function of the KOPiv additive as well as the origin of the chemo-/regioselectivity have been clarified accordingly. Further investigations on the scope and application of the innovative gem–difluoromethylene allene species in assembling othexr intriguing fluorine-containing motifs via TM-catalyzed C—H activation are in progress.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  We thank the National Natural Science Foundation of China (NSFC, Nos. 21877020, 22007020), Natural Science Foundation of Guangdong Province (No. 2019A1515010935) and Science and Technology Program of Guangzhou (No. 202102020615) for financial support on this study.

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2022.107849.

  
  
• 1. [1]

     J. Han, A.M. Remete, L.S. Dobson, et al., J. Fluorine Chem. 239 (2020) 109639. doi: 10.1016/j.jfluchem.2020.109639

  2. [2]

     T. Liang, C.N. Neumann, T. Ritter, Angew. Chem. Int. Ed. 52 (2013) 8214–8264. doi: 10.1002/anie.201206566

  3. [3]

     Y. Zhou, J. Wang, Z. Gu, et al., Chem. Rev. 116 (2016) 422–518. doi: 10.1021/acs.chemrev.5b00392

  4. [4]

     E.P. Gillis, K.J. Eastman, M.D. Hill, D.J. Donnelly, N.A. Meanwell, J. Med. Chem. 58 (2015) 8315–8359. doi: 10.1021/acs.jmedchem.5b00258

  5. [5]

     N.A. Meanwell, J. Med. Chem. 61 (2018) 5822–5880. doi: 10.1021/acs.jmedchem.7b01788

  6. [6]

     P. Xiao, X. Pannecoucke, J.P. Bouillon, S. Couve-Bonnaire, Chem. Soc. Rev. 50 (2021) 6094–6151. doi: 10.1039/d1cs00216c

  7. [7]

     M. Reichel, K. Karaghiosoff, Angew. Chem. Int. Ed. 59 (2020) 12268–12281. doi: 10.1002/anie.201913175

  8. [8]

     Y. Zhu, J.L. Han, J.D. Wang, et al., Chem. Rev. 118 (2018) 3887–3964. doi: 10.1021/acs.chemrev.7b00778

  9. [9]

     Q. Cheng, T. Ritter, Trends Chem. 1 (2019) 461–470. doi: 10.1016/j.trechm.2019.04.001

  10. [10]

      S. Hara, Stereoselective synthesis of mono-fluoroalkenes, in: J. Wang (Ed.), Stereoselective Synthesis of Mono-fluoroalkenes, Stereoselective Alkene Synthesis, Topics in Current Chemistry, vol. 327, Springer, Berlin, Heidelberg, 2012, pp. 59–86.

  11. [11]

      Y.Y. See, M.T. Morales-Colon, D.C. Bland, M.S. Sanford, Acc. Chem. Res. 53 (2020) 2372–2383. doi: 10.1021/acs.accounts.0c00471

  12. [12]

      M. Li, X.S. Xue, J.P. Cheng, Acc. Chem. Res. 53 (2020) 182–197. doi: 10.1021/acs.accounts.9b00393

  13. [13]

      R. Szpera, D.F.J. Moseley, L.B. Smith, A.J. Sterling, V. Gouverneur, Angew. Chem. Int. Ed. 58 (2019) 14824–14848. doi: 10.1002/anie.201814457

  14. [14]

      Z. Feng, Y.L. Xiao, X. Zhang, Acc. Chem. Res. 51 (2018) 2264–2278. doi: 10.1021/acs.accounts.8b00230

  15. [15]

      S. Barata-Vallejo, M.V. Cooke, A. Postigo, ACS Catal. 8 (2018) 7287–7307. doi: 10.1021/acscatal.8b02066

  16. [16]

      J. Han, L. Kiss, H. Mei, et al., Chem. Rev. 121 (2021) 4678–4742. doi: 10.1021/acs.chemrev.0c01263

  17. [17]

      S. Caron, Org. Process Res. Dev. 24 (2020) 470–480. doi: 10.1021/acs.oprd.0c00030

  18. [18]

      A. Harsanyi, G. Sandford, Green Chem. 17 (2015) 2081–2086. doi: 10.1039/C4GC02166E

  19. [19]

      B. Zhao, B. Prabagar, Z. Shi, Chem 7 (2021) 2585–2634. doi: 10.1016/j.chempr.2021.08.001

  20. [20]

      R. Jana, H.M. Begama, E. Dinda, Chem. Commun. 57 (2021) 10842–10866. doi: 10.1039/d1cc04083a

  21. [21]

      S. Shabani, Y.Z. Wu, H.G. Ryan, C.A. Hutton, Chem. Soc. Rev. 50 (2021) 9278–9343. doi: 10.1039/d0cs01441a

  22. [22]

      L. Guillemard, N. Kaplaneris, L. Ackermann, M.J. Johansson, Nat. Rev. Chem. 5 (2021) 522–545. doi: 10.1038/s41570-021-00300-6

  23. [23]

      J. Zhang, X. Lu, C. Shen, et al., Chem. Soc. Rev. 50 (2021) 3263–3314. doi: 10.1039/d0cs00447b

  24. [24]

      J. Wen, Z. Shi, Acc. Chem. Res. 54 (2021) 1723–1736. doi: 10.1021/acs.accounts.0c00888

  25. [25]

      D. Sun, D.N. Confair, J.A. Ellman, Acc. Chem. Res. 54 (2021) 1766–1778. doi: 10.1021/acs.accounts.1c00027

  26. [26]

      J. Mas-Roselló, A.G. Herraiz, B. Audic, A. Laverny, N. Cramer, Angew. Chem. Int. Ed. 60 (2021) 13198–13224. doi: 10.1002/anie.202008166

  27. [27]

      S.Y. Hong, Y. Hwang, M. Lee, S. Chang, Acc. Chem. Res. 54 (2021) 2683–2700. doi: 10.1021/acs.accounts.1c00198

  28. [28]

      J. Huang, F. Liu, F. Du, L. Zeng, Z. Chen, Green Synth. Catal. (2022) https://doi.org/10.1016/j.gresc.2022.06.002. doi: 10.1016/j.gresc.2022.06.002

  29. [29]

      N. Li, Y. Wang, L. Kong, J. Chang, X. Li, Adv. Synth. Catal. 361 (2019) 3880–3885. doi: 10.1002/adsc.201900220

  30. [30]

      D. Zell, U. Dhawa, V. Mueller, et al., ACS Catal. 7 (2017) 4209–4213. doi: 10.1021/acscatal.7b01208

  31. [31]

      Y. Li, F. Xie, Y. Liu, X. Yang, X. Li, Org. Lett. 20 (2018) 437–440. doi: 10.1021/acs.orglett.7b03775

  32. [32]

      B. Shu, S.Y. Chen, N.X. Deng, et al., Org. Chem. Front. 8 (2021) 4445–4451. doi: 10.1039/d1qo00462j

  33. [33]

      P. Tian, C. Feng, T.P. Loh, Nat. Commun. 6 (2015) 7472. doi: 10.1038/ncomms8472

  34. [34]

      H.J. Tang, L.Z. Lin, C. Feng, T.P. Loh, Angew. Chem. Int. Ed. 56 (2017) 9872–9876. doi: 10.1002/anie.201705321

  35. [35]

      L. Kong, B. Liu, X. Zhou, F. Wang, X. Li, Chem. Commun. 53 (2017) 10326–10329. doi: 10.1039/C7CC06048C

  36. [36]

      W.W. Ji, E. Lin, Q. Li, H. Wang, Chem. Commun. 53 (2017) 5665–5668. doi: 10.1039/C7CC02105D

  37. [37]

      J.Q. Wu, S.S. Zhang, H. Gao, et al., J. Am. Chem. Soc. 139 (2017) 3537–3545. doi: 10.1021/jacs.7b00118

  38. [38]

      J. Hu, X. Han, Y. Yuan, Z. Shi, Angew. Chem. Int. Ed. 56 (2017) 13342–13346. doi: 10.1002/anie.201708224

  39. [39]

      J. Hu, Y. Zhao, Z. Shi, Nat. Catal. 1 (2018) 860–869. doi: 10.1038/s41929-018-0147-9

  40. [40]

      S. Koley, R.A. Altman, Isr. J. Chem. 60 (2020) 313–339. doi: 10.1002/ijch.201900173

  41. [41]

      H. Gao, S. Lin, S. Zhang, et al., Angew. Chem. Int. Ed. 60 (2021) 1959–1966. doi: 10.1002/anie.202013052

  42. [42]

      Y. Yang, N. Li, J. Zhao, et al., Adv. Synth. Catal. 363 (2021) 3600–3606. doi: 10.1002/adsc.202100441

  43. [43]

      T. Li, C. Zhou, X. Yan, J. Wang, Angew. Chem. Int. Ed. 57 (2018) 4048–4052. doi: 10.1002/anie.201712691

  44. [44]

      F. Romanov-Michailidis, B.D. Ravetz, D.W. Paley, T. Rovis, J. Am. Chem. Soc. 140 (2018) 5370–5374. doi: 10.1021/jacs.8b02716

  45. [45]

      C.Q. Wang, L. Ye, C. Feng, T.P. Loh, J. Am. Chem. Soc. 139 (2017) 1762–1765. doi: 10.1021/jacs.6b12142

  46. [46]

      X. Zhong, S. Lin, H. Gao, et al., Org. Lett. 23 (2021) 2285–2291. doi: 10.1021/acs.orglett.1c00418

  47. [47]

      H. Xu, W. Chen, M. Bian, et al., ACS Catal. 11 (2021) 14694–14701. doi: 10.1021/acscatal.1c04508

  48. [48]

      Y. He, L. Tian, X. Chang, et al., Chin. Chem. Lett. 33 (2022) 2987–2992. doi: 10.1016/j.cclet.2022.01.068

  49. [49]

      J.M. Alonso, P. Almendros, Chem. Rev. 121 (2021) 4193–4252. doi: 10.1021/acs.chemrev.0c00986

  50. [50]

      G. Li, X. Huo, X. Jiang, W. Zhang, Chem. Soc. Rev. 49 (2020) 2060–2118. doi: 10.1039/c9cs00400a

  51. [51]

      X.L. Han, P.-. P. Lin, Q. Li, Chin. Chem. Lett. 30 (2019) 1495–1502. doi: 10.1016/j.cclet.2019.04.027

  52. [52]

      J.L. Mascareñas, I. Varela, F. López, Acc. Chem. Res. 52 (2019) 465–479. doi: 10.1021/acs.accounts.8b00567

  53. [53]

      B. Yang, Y. Qiu, J.E. Backvall, Acc. Chem. Res. 51 (2018) 1520–1531. doi: 10.1021/acs.accounts.8b00138

  54. [54]

      J. Ye, S. Ma, Acc. Chem. Res. 47 (2014) 989–1000. doi: 10.1021/ar4002069

  55. [55]

      R.K. Shukla, A.M. Nair, S. Khan, C.M.R. Volla, Angew. Chem. Int. Ed. 59 (2020) 17042–17048. doi: 10.1002/anie.202003216

  56. [56]

      S.-. G. Wang, Y. Liu, N. Cramer, Angew. Chem. Int. Ed. 58 (2019) 18136–18140. doi: 10.1002/anie.201909971

  57. [57]

      X. Vidal, J.L. Mascareñas, M. Gulías, J. Am. Chem. Soc. 141 (2019) 1862–1866. doi: 10.1021/jacs.8b12636

  58. [58]

      R. Zeng, S. Wu, C. Fu, S. Ma, J. Am. Chem. Soc. 135 (2013) 18284–18287. doi: 10.1021/ja409861s

  59. [59]

      S. Liu, L. Zhu, T. Zhang, et al., Org. Lett. 23 (2021) 1489–1494. doi: 10.1021/acs.orglett.1c00223

  60. [60]

      H. Wu, X. Li, X. Tang, C. Feng, G. Huang, J. Org. Chem. 83 (2018) 9220–9230. doi: 10.1021/acs.joc.8b01229

  61. [61]

      G. Landelle, M. Bergeron, M.O. Turcotte-Savard, J.F. Paquin, Chem. Soc. Rev. 40 (2011) 2867–2908. doi: 10.1039/c0cs00201a

  62. [62]

      M. Bian, H. Mawjuda, H. Gao, et al., Org. Lett. 22 (2020) 9677–9682. doi: 10.1021/acs.orglett.0c03734

  63. [63]

      S. Vásquez-Céspedes, X. Wang, F. Glorius, ACS Catal. 8 (2018) 242–257. doi: 10.1021/acscatal.7b03048

  64. [64]

      T. Lu, Q. Chen, J. Comput. Chem. 43 (2022) 539. doi: 10.1002/jcc.26812

  65. [65]

      S. Rakshit, C. Grohmann, T. Besset, F. Glorius, J. Am. Chem. Soc. 133 (2011) 2350–2353. doi: 10.1021/ja109676d

  66. [66]

      G. Liu, Y. Shen, Z. Zhou, X. Lu, Angew. Chem. Int. Ed. 52 (2013) 6033–6037. doi: 10.1002/anie.201300881

  67. [67]

      X.X. Ma, J.B. Liu, F. Huang, C.Z. Sun, D.Z. Chen, Catal. Sci. Technol. 8 (2018) 3590–3598. doi: 10.1039/C8CY00481A

• 

  Scheme 1  TM-catalyzed C—H functionalization with fluorinated π-compounds.

  下载: 全尺寸图片 幻灯片
  

  Scheme 2  Scope for (hetero)aromatic and vinylic C—H functionalization with gem–difluoromethylene allenes. Reaction conditions: 1 (0.2 mmol), 2 (0.2 mmol), [Cp^*RhCl₂]₂ (2.5 mol%) and KOPiv (1 equiv.) in TFE (0.1 mol/L) at room temperature for 24 h under air; isolated yields were reported. ^a The yield was obtained with 5 mol% of [Cp^*RhCl₂]₂ catalyst loading. ^b Dihydropyridin-2(1H)-one derivative 4ea' was detected as the side-product, see Supporting information for details.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Computational mechanistic studies for the allene insertion process. (A) Computed Gibbs energy profiles; (B) Calculated frontier-molecular-orbital (FMO) diagram for 2a and INT-1, isovalue = 0.05; (C) Non-covalent interactions of TS-1b/1a′: Geometries of TS-1b (a), IGMH maps of TS-1b (b), Geometries of TS-1a′ (c), IGMH maps of TS-1a′ (d), isosurfaces of δ_g^inter = 0.004 a.u.; green surfaces represent the weak interaction.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Computed Gibbs free energy changes of the reaction pathway for the formation of 3-fluoroalkenyl isoquinolone from INT-3b.

  下载: 全尺寸图片 幻灯片
  

  Scheme 3  Experimental mechanistic studies and product derivatization.

  下载: 全尺寸图片 幻灯片
  

  Scheme 4  Proposed catalytic cycle.

  下载: 全尺寸图片 幻灯片
•