Palladium-catalyzed asymmetric carbofluorination of gem-difluorostyrenes with cesium fluoride and allylic acetates

Abstract

Palladium-catalyzed asymmetric carbofluorination of difluorostyrenes with allylic acetates and CsF gives synthetically versatile products with a trifluoromethyl, a vinyl and two different aryl groups directly attached to a scaffold exhibiting vicinal chirality centers in good yields and with high stereoselectivities.

---

Chiral fluorinated organic compounds have found numerous applications in the last 20 years and are nowadays routinely included in drug and agrochemical development projects due to their unique pharmacological and physicochemical properties which often compare favorably with those of nonfluorinated analogs.^1–3 The general importance and the popularity of chiral organofluorines in many academic and industrial laboratories continue to inspire the search for catalytic asymmetric procedures that address the unwavering call for efficient synthetic access to new scaffolds.^4,5 This is typically achieved by stereoselective carbon–carbon bond construction with commercially available fluorinated building blocks or by the introduction of a carbon-fluoride bond into an existing structure with the help of a fluorinating agent.^6–13 Carbofluorination strategies that integrate both transformations into a single reaction and thus allow concomitant C–C and C–F bond formation have also surfaced although asymmetric methods have remained scarce. In 2014, Toste and coworkers showed an early example of a palladium-catalyzed asymmetric fluoroarylation reaction with styrenes carrying a directing group.¹⁴ Since then, intra- and intermolecular variations giving practical access to a variety of fluoroarylation products with excellent regiocontrol and high enantioselectivities have been reported by the same group^15,16 and others.^17,18

Despite these important advances, considerable challenges and opportunities to uncover new chemical reaction space, in particular with regard to catalytic asymmetric fluorinative alkylations, remain. Regioselective carbofluorination of gem-difluorostyrenes has received considerable attention,¹⁹ probably at least in part because it can produce a trifluoromethyl group which is undoubtedly among the most prominent organofluorine moieties in the pharmaceutical development realm.^20,21 Several carbofluorination reactions with difluorostyrenes have been reported to date but they typically afford racemic products or proceed with marginal enantioselectivity, suggesting that this would not be possible by asymmetric palladium catalysis.^22–29

A noteworthy exception is Li's and Zhang's enantioselective palladium(sulfinamide phosphine) catalyzed fluoroarylation which utilizes aryl iodides and stoichiometric amounts of silver fluoride in cyclohexane at 70 °C (Scheme 1).³⁰ Two years ago, our laboratory demonstrated regioconvergent substitution and enantioselective asymmetric alkylation of allylic fluorides,³¹ which was elegantly extended by Companyó's group to organocatalytic reactions with difluorostyrenes.³² This protocol affords high stereoselectivities and gives moderate to good yields but requires a threefold molar excess of the precious difluoroalkene compound. We envisioned that atom-economic asymmetric fluorinative alkylations of gem-difluorostyrenes should be possible if mild homogeneous reaction conditions compatible with nucleophilic fluoride addition and subsequent transition metal catalyzed C–C bond construction could be identified. Herein, we introduce a palladium catalyzed allylic alkylation procedure that achieves this goal and generates two contiguous chirality centers at once with excellent enantioselectivity and good diastereoselectivity by using cesium fluoride in acetonitrile at low temperatures.

Scheme 1 Pd-catalyzed C–F and C–C bond formation with difluorostyrenes.

We began our search for an efficient asymmetric fluoroalkylation protocol by screening the reaction between gemdifluorostyrene, 1a, and 1,3-diphenylallylacetate, 2a, in the presence of 2.5 mol% of [allylPdCl]₂, 5% (R)-BINAP (L1), CsF, and 18-crown-6 at 25 °C (Table 1, entry 1). We were pleased to see that the presence of the crown ether allowed efficient fluoride addition at room temperature, and NMR and chiral HPLC analysis revealed full consumption of 1a and formation of the desired product 3aa in 92% ee and 2.6 : 1 dr. Screening of various solvents showed that the highest conversions are achieved with polar aprotic solvents such as acetonitrile, acetone, EtOAc, benzonitrile, and DMF, which likely aid in the stabilization of the intermediate carbanion formed in situ upon reversible fluoride addition to the difluoroalkene (see SI).²⁴ The best results were obtained using acetonitrile as solvent and we then continued with screening a large group of ligands under these conditions to further improve the diastereoselectivity and yield, which was compromised by the formation of by-product 4. Unfortunately, the amount of 4 increased when we applied L2-L10 in our protocol (Table 1, entries 2–10). However, we noticed that the palladium catalysis is accelerated by Segphos and bis(phosphino)ferrocene ligands which allowed us to decrease the temperature to −40 °C while the reactions were typically still complete within 24 hours (entries 11–18). The formation of 4 was most successfully suppressed when we employed ligands L11, L13, and L17 which we attribute to a faster consumption of the fluorinated carbanion intermediate in the successive C–C bond forming step. Ultimately, we were able to obtain the highest yields and stereoselectivities using ligand L17 (entry 18). Attempts to further improve the diastereoselectivity at −78 °C were unsuccessful and resulted in increased formation of by-product 4. The presence of one equivalent of the crown ether in the final protocol proved important as we observed incomplete conversion and significantly increased formation of 4 in its absence.

Table 1 Optimization of the asymmetric fluoroalkylation of gem-difluorostyrenes^a

Entry	Ligand	Conversion^b (%)	3aa : 4^c	dr^c	ee^d (%)
a Reaction conditions: ligand (5 mol%), [allylPdCl]2 (2.5 mol%), 1a (0.15 mmol), 2a (0.3 mmol), CsF (0.18 mmol), and 18-crown-6 (0.18 mmol) in 1.0 mL of acetonitrile at 25 °C.b Conversion of 1a determined by ^19F NMR spectroscopy of the crude reaction.c Determined by ^19F NMR spectroscopy.d Determined by chiral HPLC using Chiralpak AD-H.e Reaction was run in 1.0 mL dichloromethane.f Reaction was run at −40 °C.g Reaction was run for 24 h. n.d. = not determined.
1	L1	100	5 : 1	2.6 : 1	92
2	L2	100	2 : 1	4.9 : 1	n.d.
3^e	L3	67	2 : 1	1.8 : 1	>99
4^e	L4	66	3 : 1	2.2 : 1	95
5	L5	100	3 : 1	2.0 : 1	n.d.
6	L6	100	1 : 1	2.9 : 1	n.d.
7	L7	77	1 : 1	1.5 : 1	n.d.
8	L8	89	1 : 3	2.7 : 1	n.d.
9	L9	89	1 : 1	2.6 : 1	n.d.
10	L10	100	3 : 1	3.4 : 1	n.d.
11^f	L11	100	10 : 1	3.0 : 1	n.d.
12	L12	100	5 : 1	2.7 : 1	n.d.
13^g	L13	100	12 : 1	4.4 : 1	80
14^g	L14	100	6 : 1	1.5 : 1	n.d.
15^g	L15	79	1 : 1	3.0 : 1	n.d.
16^g	L16	100	1 : 7	3.7 : 1	n.d.
17^g	L17	100	24 : 1	3.5 : 1	n.d.
18^f^g	L17	100	16 : 1	4.3 : 1	91

---

With an optimized procedure in hand, we continued with the evaluation of the substrate scope. We first screened a variety of gem-difluorostyrenes 1a–m with 1,3-diphenylallylacetate 2a (Scheme 2). As expected, 3aa was isolated in 79% yield, 91% ee and 4 : 1 dr. The presence of other ester, ketone and aldehyde functions is perfectly tolerated and 3ba, 3ca and 3da were produced in 92–94% ee and up to 6 : 1 dr. Similar results were obtained with nitro, nitrile, sulfone and methyl derivatives, giving 3ea, 3fa, 3ga and 3ha in 67–79% yield and with good diastereoselectivity and excellent enantioselectivity. We noticed that the rate of fluoride addition to the difluorostyrene was strongly influenced by the electron-withdrawing strength of its aromatic substituents, as evidenced by an increase in the formation of by-product 4 and overalkylation observed with the gem-difluorostyrenes 1e and 1f. However, optimal yields and excellent ee's of 3da, 3ea, 3fa, 3ga and 3ma could still be obtained with slow addition of the difluorostyrene, leading us to believe that the subsequent allylic alkylation is the rate-limiting step in this case. Furthermore, we found that the reaction is compatible with the placement of various halides, a trifluoromethyl group as well as combinations of several of the above-mentioned functionalities in the ortho-, meta- or para-positions in difluorostyrene, and 3ia–3ma were formed with the same protocol in 60–87% yield and comparable ee's and dr's. It is noteworthy that fluoride addition under these conditions did not occur with substrates lacking an electron-withdrawing group, which is consistent with previous findings.²⁴ We were also able to expand the reaction scope by including various substituted allylic acetates 2b–c, producing 3cb, 3cc, and 3cd in 76–88% yield, 91–99% ee, and up to 5 : 1 dr (Scheme 2). A single crystal was obtained by slow evaporation of a solution containing 3ja in dichloromethane and crystallographic analysis revealed (3S,4R)-configuration, thus providing important insights into the sense of asymmetric induction provided by the palladium catalyst derived from L17.

Scheme 2 Reaction scope. See SI for details. ^a48 h. The absolute configuration of 3ja was determined by X-ray analysis. The stereochemistry of all other products is assigned by analogy.

Literature reports and our own findings are in agreement with the mechanism of the asymmetric carbofluorination shown in Scheme 3.^33–36 Coordination of the allylic acetate 2 to the palladium catalyst A gives species B which undergoes oxidative addition to form the η³-allyl palladium complex C. Fluoride addition to the difluorostyrene 1 affords nucleophile D which then attacks C to form the alkyl–alkyl bond in structure E. Dissociation of the π-complex finally regenerates A and yields product 3. The challenges of the asymmetric carbofluorination are not only to control the simultaneous formation of two adjacent chirality centers but also to minimize competitive side reactions. In fact, we found that the by-product 4 can be generated from unconsumed D when the reaction is quenched too early, which suggests that the fluorination with cesium fluoride is faster than the asymmetric C–C bond forming step. Alternatively, 4 can be formed from intermediate D via deprotonation of 3 thus generating F which undergoes overalkylation to G, a problem that we observed in particular with styrenes carrying strong electron-withdrawing groups. We found that this can be effectively mitigated by slow addition of 1 to the reaction mixture which is expected to limit the concentration of D and thus favors the desired reaction between the anionic nucleophile and the cationic allyl-loaded palladium catalyst. The stereochemical outcome of the reaction is in agreement with an outer-sphere mechanism in which the η³-allyl palladium complex C is preferentially attacked by the Re-face of the intermediate carbanion D. This pathway yields the major (R,S,E)-diastereomer of 3, while the attack from the Si-face of D results in the formation of the minor (S,S,E)-diastereomer.

Scheme 3 Proposed mechanism of the asymmetric carbofluorination.

Importantly, the reaction is amenable to scaling, and we were able to prepare more than half a gram of 3ca which was isolated in 65% yield, 7 : 1 dr and 93% ee. As mentioned above, the trifluoromethyl group is of particular importance due to its prominence in many pharmaceuticals, and we anticipated that the incorporation of an allylic group into the gem-difluorostyrene scaffold would facilitate further transformations with the possibility of introducing additional chiral centers. To this end, we carried out a variety of transformations at the double-bond of compound (3S,4R)-3ca (Scheme 4). We were pleased to find that Sharpless asymmetric dihydroxylation with AD-mix-α yielded 5 in 78% yield and 94% ee thus establishing four contiguous chirality centers along the alkyl sidechain. We previously established that the sense of asymmetric induction of the dihydroxylation of similar structures is controlled by the AD catalyst used and not by the substrate chirality and we can therefore assign the structure as (2R,3S,4S,5S)-5.³⁷ Additionally, the epoxidation of 3ca can be carried out efficiently with m-CPBA and we obtained 6 in 77% yield. A single crystal of the major diastereomer was obtained by slow evaporation of a dichloromethane solution. The crystallographic analysis identified 6 as 1-(4-((2R,3S)-1,1,1-trifluoro-3-phenyl-3-((2S,3S)-3-phenyloxiran-2-yl)propan-2-yl)phenyl)ethan-1-one, providing another example with four adjacent chirality centers (see SI). Dihydroxylation of (3S,4R)-3ba with osmium tetroxide and N-methylmorpholine N-oxide (NMMO) followed by diol cleavage with sodium periodate gave aldehyde 7 in 34% overall yield. Finally, we applied (3S,4R)-3la in a Suzuki cross-coupling procedure with 4-methoxyphenylboronic acid and obtained 8 in 89% yield.

Scheme 4 Transformations with 3ca, 3ba and 3la.

In summary, we have developed a catalytic asymmetric fluoroalkylation method that allows efficient C–F and C–C bond formation across an electron-deficient double bond, producing multifunctional products in 53–79% yield, 88–99% ee, and up to 7 : 1 dr. This reaction has several noteworthy features. Two contiguous chirality centers are installed with high stereoselectivity, the initiating C–F bond formation proceeds under mild conditions with stoichiometric amounts of inexpensive cesium fluoride, the palladium-catalyzed reaction is accelerated by a commercially available chiral bis(phosphino)ferrocene ligand and can be scaled up without compromising the stereochemical outcome and yield. The synthetic utility of the products exhibiting a unique arrangement of a trifluoromethyl, a vinyl and two different aryl groups directly attached to a scaffold of vicinal chirality centers is demonstrated by the introduction of additional functionalities either at the aryl ring or at the alkene moiety.

Conflicts of interest

The authors declare no competing financial interest.

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/d6cc02743a.

CCDC 2383107 and 2383108 contain the supplementary crystallographic data for this paper.^38a,b

Acknowledgements

We gratefully acknowledge financial support from the US National Institutes of Health, GM106260. E. N. thanks the Henry Luce Foundation for a Clare Boothe Luce Graduate Fellowship.

References

1. P. Jeschke, ChemBioChem, 2004, 5, 570–589.
2. B. R. Smith, C. M. Eastman and J. T. Njardarson, Med. Chem., 2014, 57, 9764–9773.
3. M. Inoue, Y. Sumii and N. Shibata, ACS Omega, 2020, 5, 10633–10640.
4. K. Müller, C. Faeh and F. Diederich, Science, 2007, 317, 1881–1886.
5. Y. Zhou, J. Wang, Z. Gu, S. Wang, W. Zhu, J. L. Acena, V. A. Soloshonok, K. Izawa and H. Liu, Chem. Rev., 2016, 116, 422–518.
6. T. Liang, C. N. Neumann and T. Ritter, Angew. Chem., Int. Ed., 2013, 52, 8214–8264.
7. D. Cahard, X. Xu, S. Couve-Bonnaire and X. Pannecoucke, Chem. Soc. Rev., 2010, 39, 558–568.
8. X. Yang, T. Wu, J. Phipps and F. D. Toste, Chem. Rev., 2015, 115, 826–870.
9. P. A. Champagne, J. Desroches, J.-D. Hamel, M. Vandamme and J.-F. Paquin, Chem. Rev., 2015, 115, 9073–9174.
10. Y. Zhu, J. Han, J. Wang, N. Shibata, M. Sodeoka, V. A. Soloshonok, J. A. S. Coelho and F. D. Toste, Chem. Rev., 2018, 118, 3887–3964.
11. J. Liu, Q. Yuan, F. D. Toste and M. S. Sigman, Nat. Chem., 2019, 11, 710–715.
12. W. Zhu, X. Zhen, J. Wu, Y. Cheng, J. An, X. Ma, J. Liu, Y. Qin, H. Zhu, J. Xue and X. Jiang, Nat. Commun., 2021, 12, 3957.
13. A. Pons, L. Delion, T. Poisson, A. B. Charette and P. Jubault, Acc. Chem. Res., 2021, 54, 2969–2990.
14. E. P. A. Talbot, T. de, A. Fernandes, J. M. McKenna and F. D. Toste, J. Am. Chem. Soc., 2014, 136, 4101–4104.
15. J. Miró, C. Del  Pozo, F. D. Toste and S. Fustero., Angew. Chem., Int. Ed., 2016, 55, 9045–9049.
16. Z. Liu, L. J. Oxtoby, M. Liu, Z.-Q. Li, V. T. Tran, Y. Gao and K. M. Engle, J. Am. Chem. Soc., 2021, 143, 8962–8969.
17. Y. He, Z. Yang, R. T. Thornbury and F. D. Toste, J. Am. Chem. Soc., 2015, 137, 12207–12210.
18. Y. Xi, C. Wang, Q. Zhang, J. Qu and Y. Chen, Angew. Chem., Int. Ed., 2021, 60, 2699–2703.
19. J. P. Sorrentino and R. A. Altman, Synthesis, 2021, 3935–3950.
20. W. Zhu, J. Wang, S. Wang, Z. Gu, J. L. Aceña, K. Izawa, H. Liu and V. A. Soloshonok, J. Fluorine Chem., 2014, 167, 37–54.
21. A. Abula, Z. Xu, Z. Zhu, C. Peng, Z. Chen, W. Zhu and H. A. Aisa, J. Chem. Inf. Model., 2020, 60, 6242–6250.
22. B. Gao, Y. Zhao, C. Ni and J. Hu, Org. Lett., 2014, 16, 102–105.
23. B. Gao, Y. Zhao and J. Hu, Angew. Chem., Int. Ed., 2015, 54, 638–642.
24. P. Tian, C.-Q. Wang, S.-H. Cai, S. Song, L. Ye, C. Feng and T.-P. Loh, J. Am. Chem. Soc., 2016, 138, 15869–15872.
25. H.-J. Tang, L.-Z. Lin, C. Feng and T.-P. Loh, Angew. Chem., Int. Ed., 2017, 56, 9872–9876.
26. P. E. Daniel, C. I. Onyeagusi, A. A. Ribeiro, K. Li and S. J. Malcolmson, ACS Catal., 2019, 9, 205–210.
27. H. Liu, L. Ge, D.-X. Wang, N. Chen and C. Feng, Angew. Chem., Int. Ed., 2019, 58, 3918–3922.
28. W.-J. Yoo, J. Kondo, J. A. Rodriguez-Santamara, T. V. Q. Nguyen and S. Kobayashi, Angew. Chem., Int. Ed., 2019, 58, 6772–6775.
29. X. Yu, A. Maity and A. Studer, Angew. Chem., Int. Ed., 2023, 62, e202310288.
30. T.-Y. Lin, Z. Pan, Y. Tu, S. Zhu, H.-H. Wu, Y. Liu, Z. Li and J. Zhang, Angew. Chem., Int. Ed., 2020, 59, 22957–22962.
31. J. S. S. K. Formen, C. C. Lynch, E. Nelson, A. Yuan, S. E. Steber and C. Wolf, J. Org. Chem., 2024, 89, 10998–11002.
32. J. Duran, V. G. Moldoveanu, C. Barroso, G. A. Pereira, B. Limburg and X. Companyó, Angew. Chem., Int. Ed., 2026, 65, e202520513.
33. R. Prétôt and A. Pfaltz, Angew. Chem., Int. Ed., 1998, 37, 323–325.
34. B. M. Trost and D. L. Van Vranken, Chem. Rev., 1996, 96, 395–422.
35. G. Helmchen, U. Kazmaier and S. Foerster, Enantioselectiveallylic Substitutions with Carbon Nucleophiles. in Catalytic AsymmetricSynthesis, ed. I. Ojima, John Wiley & Sons, Inc., Hoboken, 3rd edn, 2010, pp. 497–641.
36. O. Pàmies, J. Margalef, S. Cañellas, J. James, E. Judge, P. J. Guiry, C. Moberg, J.-E. Bäckvall, A. Pfaltz, M. A. Pericàs and M. Diéguez, Chem. Rev., 2021, 121, 4373–4505.
37. K. Balaraman and C. Wolf, Angew. Chem., Int. Ed., 2017, 56, 1390–1395.
38. (a) CCDC 2383107: Experimental Crystal Structure Determination, 2026 DOI:10.5517/ccdc.csd.cc2kztft; (b) CCDC 2383108: Experimental Crystal Structure Determination, 2026 DOI:10.5517/ccdc.csd.cc2kztgv.

---