Excited-state palladium-catalyzed defluorinative arylation of polyfluoroarenes

Abstract

Visible-light-induced excited-state palladium catalysis enables defluorinative C(sp²)–C(sp²) cross-coupling of polyfluoroarenes with aryl boronic acids under blue LED irradiation at room temperature. The Pd(OAc)₂/BrettPhos catalyst selectively activates C(sp²)–F bonds without external photocatalysts, affording multifluorinated biaryls in good to excellent yields through photoinduced formation of Ar–Pd(II) intermediates.

---

Fluorinated aromatic compounds are widely encountered in pharmaceuticals, agrochemicals, and functional materials with enhanced metabolic stability, lipophilicity, and electronic modulation.^1–3 Fluorinated biaryl motifs, in particular, are key structural elements in bioactive molecules and advanced materials, driving significant interest in efficient and selective functionalization of polyfluoroarenes.⁴ Among the various strategies available, defluorinative functionalization of polyfluoroarenes has emerged as a powerful approach for constructing C–C bonds while simultaneously introducing valuable fluorinated motifs.⁵ However, the activation of C(sp²)–F bonds remains challenging due to their high bond dissociation energy and the strong electronegativity of fluorine.⁶ Traditional approaches often require harsh reaction conditions, specialized reagents, or prefunctionalized substrates, which limit their synthetic utility (Scheme 1a).^7–11

Scheme 1 (a) Conventional and photocatalytic strategies for defluorinative cross-coupling of polyfluoroarenes. (b) This work: excited-state Pd-catalyzed site-selective defluorinative arylation of polyfluoroarenes.

In recent years, visible-light-induced strategies have enabled the generation of aryl radicals from polyfluoroarenes through selective C–F bond cleavage, thereby allowing diverse defluorinative transformations.^12,13 The Weaver group reported the selective arylation and alkenylation of C(sp²)–F bonds in polyfluoroarenes, using electron-rich arenes and alkynes as coupling partners in the presence of an iridium photocatalyst under visible light irradiation (Scheme 1a).^14,15 Building on this approach, our group developed a dual catalysis strategy that combines iridium photocatalyst with a nickel catalyst to enable the cross-coupling of polyfluoroarenes with aryl bromides under light (Scheme 1a).¹⁶ Subsequently, the Chu group reported a synergistic approach combining palladium and iridium photocatalysts for the reductive cross-electrophile coupling of polyfluoroarenes with aryl triflates and bromides; however, the reaction required elevated temperatures of 70–75 °C (Scheme 1a).¹⁷ Despite these advances, these systems typically rely on external photocatalysts and multi-catalyst reaction platforms.

More recently, visible-light-induced Pd-catalysis has emerged, enabling palladium to participate in photochemical processes beyond conventional ground-state catalytic cycles.^18–21 However, visible-light-induced defluorinative cross-coupling reactions employing Pd catalysis are unexplored.^22–29 Inspired by these studies, we envisioned that photoexcited Pd species could enable site-selective defluorinative arylation of polyfluoro- and perfluoroarenes via ionic or radical pathways, allowing direct C(sp²)–C(sp²) bond formation with readily available coupling partners (Scheme 1b).^30–32

We began our investigation by examining the coupling of pentafluorobenzene (1a) with phenylboronic acid (2a) under visible-light irradiation. Initial studies revealed that Pd(OAc)₂ (2.5 mol%) and BrettPhos (5 mol%) in the presence of Cs₂CO₃ and 1,4-dioxane under blue LED irradiation efficiently promoted the defluorinative arylation to afford product 3a in 76% yield. Evaluation of various phosphine ligands established BrettPhos as uniquely effective, whereas commonly employed ligands such as PPh₃, BINAP, XPhos, RuPhos, and DavePhos resulted in trace or no product formation. Screening of solvents such as MeCN, toluene, and THF resulted in significantly diminished yields, while DMA or DMSO suppressed the reactivity. Among several palladium sources tested, Pd(TFA)₂ and Pd(hfacac)₂ delivered the product in moderate yield, while PdCl₂, Pd(dppf)Cl₂, Pd(PPh₃)₂Cl₂, and Pd(PPh₃)₄ were found ineffective. Evaluating different bases revealed that Cs₂CO₃ was superior, whereas other inorganic and organic bases provided moderate yields. Increasing the catalyst loading to Pd(OAc)₂ (5 mol%) and BrettPhos (10 mol%) improved the yield to 86%, while reducing the reaction time from 24 h to 6 h furnished the product in 88% yield. Control experiments confirmed that both visible light and the Pd/BrettPhos catalytic system are essential for the transformation, supporting a visible-light-induced palladium-catalyzed C–F bond activation process (see SI for further details).

With the optimized reaction conditions in hand, we next explored the scope of excited-state Pd-catalyzed defluorinative arylation with respect to both aryl boronic acids and polyfluoroarenes (Scheme 2). A range of substituted aryl boronic acids participated smoothly to afford the corresponding fluorinated biaryl products in moderate to excellent yields. Aryl boronic acids bearing ortho-substituents on the aromatic ring were compatible, affording the corresponding defluorinative cross-coupled products 3b and 3c in 44% and 89% yields, respectively. Substrates containing para-substituents, including electron-donating groups (–Me, –^tBu, –OMe, and –OPh) as well as electron-withdrawing substituents (–Ph, –F, –CO₂Et, –CF₃, –CN) were well tolerated to yield the products 3d–3l in 60–95% yields. Boronic acids substituted at the meta-position were also suitable coupling partners, providing the corresponding products 3m–3n in moderate to good yields (67–80%). Furthermore, di- and trisubstituted aryl boronic acids reacted smoothly under the standard conditions to furnish the C(sp²)–C(sp²) cross-coupled products 3o–3s without substantial loss of reactivity. Interestingly, across the range of boronic acids examined, the reaction exhibited high regioselectivity (2,3,5,6-tetrafluoro-1,1'-biaryl), with the corresponding regioisomeric products (3′, 2,3,4,6-tetrafluoro-1,1'-biaryl) formed in less than 5% yield.

Scheme 2 Scope of the reaction with different aryl boronic acids and polyfluoroarenes. 3′ is the corresponding regioisomeric product.

Having established the generality with respect to aryl boronic acids, we next evaluated the compatibility of different polyfluoroarenes and perfluoroarenes (Scheme 3). 1,2-Difluorobenzene (1b) and 1,2,3,4-tetrafluorobenzene (1c) underwent smooth defluorinative arylation to furnish the corresponding biaryl products 3t and 3u in 67% and 45% yields, respectively. Importantly, substrates bearing nucleophile-sensitive functional groups, such as –CN substituent (1d) were well tolerated under the reaction conditions, delivering the corresponding products 3v–3w in 79–97% yields. In addition, several perfluoroarenes (1e–1f) participated smoothly in the reaction, through Pd-catalyzed selective C(sp²)–F bond activation and delivering the corresponding biaryl products 3x–3z in 37–93% yields. Furthermore, a multifluorinated arene (1g) efficiently underwent double arylation to furnish the diarylated polyfluoroarene 3aa in an excellent yield of 85%. When an ester-substituted polyfluoroarene (1h) was employed, complete site selectivity was not achieved, leading to the formation of both ortho- and para-substituted product (3ab) in 76% yield.

Scheme 3 Scope of the reaction with different aryl boronic acids and polyfluoroarenes. 3′ is the corresponding regioisomeric product.

Encouraged by these results, we next investigated whether chelating substituents could influence the selectivity of the defluorination process. In particular, we hypothesized that nitro groups could act as directing groups to promote ortho-selective C–F bond activation. To test this hypothesis, fluorinated nitroarenes were subjected to the optimized reaction conditions (Scheme 4).

Scheme 4 Scope of the reaction with different aryl boronic acids and polyfluoroarenes.

Gratifyingly, these substrates underwent efficient coupling to give the corresponding ortho-defluorinative biaryl products 4a–4h in good to excellent yields. For example, para-substituted aryl boronic acids (2e and 2h) as well as a trisubstituted boronic acid (2r) reacted smoothly with nitroarene 1i, providing the corresponding ortho-defluorinated products 4a–4c in excellent yields (84–97%). Similarly, tetrafluoronitrobenzene (1j) reacted efficiently with various aryl boronic acids to afford the desired biaryl products 4d–4h in 51–62% yields. Furthermore, the methodology could also be extended to 2,4-dinitrofluorobenzene (2,4-DNFB) (1k), which gave the corresponding defluorinative coupling product 4i, albeit in a modest 25% yield. Overall, these results demonstrate the visible-light-driven functional-group tolerance, while allowing a highly selective defluorinative C(sp²)–C(sp²) cross-coupling of polyfluoroarenes.

To gain insight into the reaction mechanism, radical trapping experiments were performed using TEMPO and BHT (see SI for further details). The addition of 1.0 equivalents of TEMPO significantly inhibited the product formation, however, no TEMPO–aryl adducts were detected by GC–MS or NMR analysis. This suggests that the reaction does not proceed through freely diffusing aryl radical intermediates and that TEMPO disrupts the photoexcited Pd-catalytic cycle either via Pd-coordination or excited state quenching.^33,34 In line with this, BHT had no significant effect on the reaction which further excludes the involvement of free radical intermediates. To verify that the ligated Pd(0) complex acts as the photo absorbing species, UV-Vis absorption studies were conducted. The BrettPhos-ligated Pd(0) complex exhibited a broad absorption band in the 360–500 nm region, which overlaps well with the emission profile of the blue LEDs used in the reaction, supporting its role as a photoactive catalytic species. Notably, a red shift in the absorption was observed by increasing pentafluorobenzene concentration which suggests interaction between the Pd complex and the substrate that facilitates photoinduced C–F oxidative addition. Light on/off experiments indicate that continuous light irradiation is required for catalytic turnover, thereby excluding the possibility of a radical chain propagation process and confirming that light plays a direct role in driving the catalytic cycle.

Based on these observations and literature precedents, a plausible catalytic cycle is proposed (Scheme 5). Initially, coordination of BrettPhos to Pd(OAc)₂ generates the active ligated Pd(0)-species, which upon visible-light absorption is promoted to a triplet excited state. The resulting photoexcited Pd-complex is proposed to undergo oxidative addition into the C(sp²)–F bond of the polyfluoroarene, generating a Pd(II)–aryl intermediate A. Subsequent ligand exchange with Cs₂CO₃ affords intermediate B, which then undergoes transmetalation with the aryl boronic acid to form a diaryl Pd(II) intermediate C. Finally, reductive elimination under light irradiation furnishes the desired defluorinative C(sp²)–C(sp²) cross-coupled product, thereby regenerating the active Pd(0) catalyst for the next catalytic cycle.

Scheme 5 Proposed mechanism.

The observed regioselectivity can be rationalized by the preferential oxidative addition of the photoexcited Pd(0)/BrettPhos complex into the most electrophilic C(sp²)–F bond of the polyfluoroarene. In the case of pentafluorobenzene, the C(sp²)–F bond para to the C(sp²)–H bond is expected to be the most activated due to the cumulative electronic influence of the neighboring ortho- and meta-fluorine substituents. Accordingly, the regioselectivity observed in this transformation is consistent with the well-established selectivity patterns reported for nucleophilic aromatic substitution (S_NAr) reactions of polyfluoroarenes. In addition, the steric environment imparted by the BrettPhos ligand may further disfavor competing oxidative addition pathways, thereby enhancing regioselectivity. For nitro-substituted polyfluoroarenes, the strongly electron-withdrawing nitro group is proposed to activate the adjacent ortho-C(sp²)–F bonds, resulting in a switch from para-selective to ortho-selective defluorination, as observed experimentally.

In conclusion, we have developed a visible-light-induced Pd-catalyzed defluorinative C(sp²)–C(sp²) cross-coupling of polyfluoroarenes with aryl boronic acids. Using Pd(OAc)₂/BrettPhos under blue LED irradiation, the transformation proceeds efficiently at room temperature without the need for an external photocatalyst, providing a mild and operationally simple approach for the synthesis of multifluorinated biaryl products. The protocol exhibits broad substrate scope and good functional-group tolerance, while maintaining high regioselectivity in C–F bond activation. Importantly, nitro substituents enables ortho-selective defluorination, demonstrating the potential of directing groups controlled site selectivity.

Arnab Dey: methodology, investigation, validation, formal analysis, writing – original draft. Cătălin C. Anghel: investigation, validation, formal analysis. Rajesh Kancherla: methodology, writing – review and editing, supervision, project administration. Magnus Rueping: conceptualization, methodology, supervision, funding acquisition, project administration, resources, writing – review and editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d6cc03082.

Acknowledgements

This publication is based upon work supported by King Abdullah University of Science and Technology (KAUST) under Award No. RFS-OFP2023-5577 and RFS-OFP2023-5517.

References

1. R. Berger, G. Resnati, P. Metrangolo, E. Weber and J. Hulliger, Chem. Soc. Rev., 2011, 40, 3496–3508.
2. N. A. Meanwell, J. Med. Chem., 2018, 61, 5822–5880.
3. S. Purser, P. R. Moore, S. Swallow and V. Gouverneur, Chem. Soc. Rev., 2008, 37, 320–330.
4. A. Das and N. Chatani, ACS Catal., 2021, 11, 12915–12930.
5. V. Palani, M. A. Perea and R. Sarpong, Chem. Rev., 2022, 122, 10126–10169.
6. K. Wang and W. Kong, ACS Catal., 2023, 13, 12238–12268.
7. L. V. Hooker and J. S. Bandar, Angew. Chem., Int. Ed., 2023, 62, e202308880.
8. Z.-J. Luo, H.-Y. Zhao and X. Zhang, Org. Lett., 2018, 20, 2543–2546.
9. T. Ahrens, J. Kohlmann, M. Ahrens and T. Braun, Chem. Rev., 2015, 115, 931–972.
10. N. R. Judge, A. Logallo and E. Hevia, Chem. Sci., 2023, 14, 11617–11628.
11. X.-Y. Huang, X.-X. Mao, C. Dai, M.-Y. Gu, D. Ge, Z.-L. Shen, K. Guo, C.-J. Li and X.-Q. Chu, JACS Au, 2026, 6, 1602–1616.
12. M. O. Zubkov and A. D. Dilman, Chem. Soc. Rev., 2024, 53, 4741–4785.
13. C. Cheng, Y.-F. Huang, L.-J. Wang, M.-J. Luo and Q. Xiao, Adv. Synth. Catal., 2025, 367, e70102.
14. S. Senaweera and J. D. Weaver, J. Am. Chem. Soc., 2016, 138, 2520–2523.
15. A. Singh, C. J. Fennell and J. D. Weaver, Chem. Sci., 2016, 7, 6796–6802.
16. A. Dewanji, R. F. Bülow and M. Rueping, Org. Lett., 2020, 22, 1611–1617.
17. J. Qin, S. Zhu and L. Chu, Organometallics, 2021, 40, 2246–2252.
18. S. Sarkar, K. P. S. Cheung and V. Gevorgyan, Angew. Chem., Int. Ed., 2024, 63, e202311972.
19. K. P. S. Cheung and V. Gevorgyan, Acc. Chem. Res., 2025, 58, 861–876.
20. R. Kancherla, K. Muralirajan, A. Sagadevan and M. Rueping, Trends Chem., 2019, 1, 510–523.
21. G. M. Torres, Y. Liu and B. A. Arndtsen, Science, 2020, 368, 318–323.
22. N. Kvasovs, J. Fang, F. Kliuev and V. Gevorgyan, J. Am. Chem. Soc., 2023, 145, 18497–18505.
23. K. Muralirajan, R. Kancherla, B. Maity, S. Karuthedath, F. Laquai, L. Cavallo and M. Rueping, Nat. Commun., 2023, 14, 6622.
24. R. Kancherla, K. Muralirajan and M. Rueping, Chem. Sci., 2022, 13, 8583–8589.
25. X. Jia, Z. Zhang and V. Gevorgyan, ACS Catal., 2021, 11, 13217–13222.
26. I. Abdiaj, L. Huck, J. M. Mateo, A. de la Hoz, M. V. Gomez, A. Díaz-Ortiz and J. Alcázar, Angew. Chem., Int. Ed., 2018, 57, 13231–13236.
27. K. Pal, R. Kancherla, S. Dutta, S. Zhumagazy, B. Maity, L. Cavallo and M. Rueping, Angew. Chem., Int. Ed., 2026, 65, e22979.
28. A. Singh, K. Pal, S. Dutta, A. Dey, R. Kancherla, B. Maity, L. Cavallo and M. Rueping, Angew. Chem., Int. Ed., 2025, 64, e202503446.
29. W. Ali, A. Saha, H. Ge and D. Maiti, JACS Au, 2023, 3, 1790–1799.
30. F.-P. Wu, X.-W. Gu, H.-Q. Geng and X.-F. Wu, Chem. Sci., 2023, 14, 2342–2347.
31. D. S. Lee, P. Y. Choy, C. M. So, J. Wang, C. P. Lau and F. Y. Kwong, RSC Adv., 2012, 2, 9179–9182.
32. Z. Peng, C. Gu, O. Y. Yuen, S. S. Ng and C. M. So, Chem. – Asian J., 2025, 20, e00949.
33. S. Chakrabarti, S. Banerjee and L. M. Mirica, J. Am. Chem. Soc., 2025, 147, 41882–41896.
34. A. Casnati, R. Maggi, G. Maestri, N. Della Ca and E. Motti, J. Org. Chem., 2017, 82, 8296–8303.

---

Footnote

† These authors contributed equally to this work.

---