Pd 2 L 4 Metal–Organic Cages with Perfluoroalkyl Edges: Ligand Design and Synthesis

Open AccessCCS ChemistryRESEARCH ARTICLES11 Sep 2025Pd2L4 Metal–Organic Cages with Perfluoroalkyl Edges: Ligand Design and Synthesis Guillermo Fiorini, Noël Pairault, Cristian Vicent, Ramón Martínez-Máñez, Nathan McClenaghan and Vicente Martí-Centelles Guillermo Fiorini Instituto Interuniversitario de Investigación de Reconocimiento Molecular y Desarrollo Tecnológico (IDM), Universitat Politècnica de València, Universitat de València, 46022 Valencia , Noël Pairault Institut des Sciences Moléculaires, University of Bordeaux/CNRS, Talence 33405 , Cristian Vicent Serveis Centrals d'Instrumentació Científica, Universitat Jaume I, 12071 Castelló de la Plana , Ramón Martínez-Máñez *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Instituto Interuniversitario de Investigación de Reconocimiento Molecular y Desarrollo Tecnológico (IDM), Universitat Politècnica de València, Universitat de València, 46022 Valencia Departamento de Química, Universitat Politècnica de València, 46022 Valencia CIBER de Bioingeniería Biomateriales y Nanomedicina, Instituto de Salud Carlos III, 46022 Valencia Unidad Mixta de Investigación en Nanomedicina y Sensores, Instituto de Investigación Sanitaria La Fe (IISLAFE), Universitat Politècnica de València, 46026 Valencia Unidad Mixta UPV-CIPF de Investigación en Mecanismos de Enfermedades y Nanomedicina, Centro de Investigación Príncipe Felipe, Universitat Politècnica de València, 46012 Valencia , Nathan McClenaghan *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Institut des Sciences Moléculaires, University of Bordeaux/CNRS, Talence 33405 and Vicente Martí-Centelles *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Instituto Interuniversitario de Investigación de Reconocimiento Molecular y Desarrollo Tecnológico (IDM), Universitat Politècnica de València, Universitat de València, 46022 Valencia Departamento de Química, Universitat Politècnica de València, 46022 Valencia CIBER de Bioingeniería Biomateriales y Nanomedicina, Instituto de Salud Carlos III, 46022 Valencia Online publication date: 11 Sep 2025https://doi.org/10.31635/ccschem.025.202505924 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookXBlueskyLinked InEmail Molecular cages are versatile supramolecular architectures with applications in guest encapsulation, catalysis, and sensing. This study reports the first Pd2L4 cage where vertices are bridged by -(CF2)n- chains. Cage synthesis is achieved by the self-assembly of a ligand comprising a perfluorinated alkyl chain grafted with terminal pyridines and Pd(II). While the synthesis with a ligand comprising α,ω-monosubstituted pyridines partially produces the desired cage, the further design of two ligands with methoxy groups at the meta or para positions relative to the N-atom greatly enhances the coordination strength between the metal centre and the pyridine group. The para-substituted ligand quantitatively yields the most stable Pd2L4 cage. These findings highlight the influence of electronic effects on coordination-driven self-assembly and provide a pathway for designing robust and functional metal–organic cages containing perfluorinated edges. Additionally, we performed host–guest experiments that show a good affinity with environmentally concerning contaminant perfluorooctanoic acid. In summary, we developed Pd2L4 metal–organic cages with perfluoroalkyl edges that show good stability and promising host–guest properties. Download figure Download PowerPoint Introduction Supramolecular architectures such as molecular cages provide synthetic cavities to bind guest molecules effectively through multiple noncovalent interactions.1,2 Their use has been extended in recent years thanks to the development of a number of synthetic methods to prepare metal–organic as well as purely organic cages, resulting in a wide range of structures with size- and shape-dependent host–guest properties.3–8 The straightforward synthesis of molecular cages and the efficient encapsulation of guests have captured the attention of supramolecular chemists, who have developed numerous systems for encapsulating molecules for different applications such as the stabilization of species,9–12 separation processes,13,14 catalysis,15–21 biological applications,22–28 and environmental applications,29–32 and sensors,33–35 among many others.36–39 Typically, metal–organic cages are prepared by the reaction of ligands with metals, forming metal–ligand bonds.6,40,41 Geometric information for the cage is, in most cases, encoded by the building blocks so the cage is assembled via coordination-driven self-assembly. Highly symmetrical, rigid ligands and metal ions with predictable coordination geometries are typically employed since they provide well-defined geometries. These self-assembly processes are usually performed under thermodynamic control where the metal–ligand bonds can be reversibly broken and formed, thus ensuring error proofing to obtain the most thermodynamically stable structure. In such cases, selecting the right metal ion is crucial, as it must have a coordination geometry that matches the desired cage shape and strike a balance between kinetic lability—allowing bond rearrangements for error correction during self-assembly—and sufficient stability to maintain the integrity of the final cage.40,42 The formation of Pd2L4 cages involves the reaction of two Pd(II) ions and four ditopic ligands, typically containing pyridine groups at the ends, through metal–pyridine coordination bonds. An extended number of Pd2L4 cages have been designed and extensively studied,8,43–51 including their chiral properties.52,53 The enduring popularity of this cage type is due, at least in part, to its predictable assembly behavior from just a few components. For the efficient synthesis of Pd2L4 cages, the ligands must be ditopic to allow coordination at two separate sites with metal ions, forming square planar geometries to form a lantern-shaped structure with a small cavity.54 However, in this research area, no fluorinated metal–organic cages based on Pd and pyridine-containing ligands have been reported to our knowledge, due to the reduced coordination strength of pyridine ligands when fluorine atoms are present in the molecular structure, acting as electron-withdrawing groups. The closest examples were reported by Lusby and coworkers who prepared a hemi-cage system featuring -CF3 groups in a meta position from the pyridine45 and by Casini and coworkers, who obtained a 18F-labeled Pd2L4 cage with ammonium trifluoroborate groups conjugated to the centre of an azide-modified ligand.28 There are also examples of purely organic cages that incorporate fluorine atoms or short -(CF2)n- chains in their structure as substituents in aromatic rings.55–60 However, none of these examples uses -(CF2)n- chains as a fundamental structural component of the cage. Some potential advantages of coordination cages featuring perfluoroalkyl spacers include the structural rigidity of the chains imparted by the bulky fluorine atoms and the highly hydrophobic nature of these moieties. Aiming to overcome this lack of fluorinated metal–organic cages, we report herein the design of the first Pd2L4 cage incorporating a ligand featuring a -(CF2)n- chain. Ligands consisting of a central -(CF2)n- chain and two pyridine moieties at its ends were developed. Cage assembly requires the ligand to align both pyridine rings in the same direction, which involves a 180° rotation of one of the pyridine units, resulting in a racemic helical chirality to the ligands. Figure 1 shows a simplified scheme of the synthesis of the fluorinated metal–organic cage C1. The prepared cages were utilized in host–guest experiments, and they demonstrated strong affinity for perfluorooctanoic acid (PFOA), an environmentally concerning contaminant. Figure 1 | First approximation for the formation of a metal–organic cage comprising perfluoroalkyl chains (C1). Download figure Download PowerPoint Experimental Methods Synthesis of perfluorohexane bis-pyridine ligand (L1) 3-Iodopyridine (686 mg, 3.35 mmol) was dissolved in dry and deoxygenated dimethyl sulfoxide (DMSO) (2 mL), under an argon atmosphere. Cu(0) (1.72 g, 27.1 mmol) was added, and the reaction mixture was stirred at room temperature (RT) for 10 min in the dark. 1,6-Diiodoperfluorohexane (759 mg, 1.37 mmol) was added at RT, and the reaction mixture was heated to 120 °C. After 19 h stirring at 120 °C in the dark, the reaction mixture was cooled to RT and quenched with 2 M ethylenediaminetetraacetic acid (EDTA)-NH3(aq.) (100 mL). The aqueous layer was extracted with Et2O (3 × 50 mL), and the combined organic layer was washed with 0.1 M EDTA-NH3(aq.) (70 mL) and brine (70 mL). The organic layer was dried over Na2SO4, filtered and concentrated in vacuo. Chromatography (SiO2, petrol-Et2O 100:0 to 3:7) afforded L1 as a white powder (462 mg, 74%). Ligand L1 characterization is described in Supporting Information Section 2.1 and Figures S1–S6. Synthesis of m-OMe-perfluorohexane bis-pyridine ligand (L2) 3-Bromo-5-methoxypyridine (411 mg, 2.19 mmol) was dissolved in dry and deoxygenated DMSO (8 mL), under an argon atmosphere. Cu(0) (1.17 g, 18.4 mmol) was added, and the reaction mixture was stirred at RT for 10 min in the dark. 1,6-Diiodoperfluorohexane (0.52 g, 0.93 mmol) was added at RT, and the reaction mixture was heated to 120 °C. After 19 h stirring at 120 °C in the dark, the reaction mixture was cooled to RT and quenched with 1 M EDTA-NH3(aq.) (100 mL). The aqueous layer was extracted with Et2O (3 × 50 mL), and the combined organic layer was washed with 0.1 M EDTA-NH3(aq.) (70 mL) and brine (70 mL). The organic layer was dried over Na2SO4, filtered and concentrated in vacuo. Column chromatography (SiO2, petrol-Et2O 100:0 to 3:7) afforded L2 as an off-white powder (283 mg, 59%). Ligand L2 characterization is described in Supporting Information Section 2.2 and Figures S7–S12. Synthesis of perfluorohexane bis-p-methoxypyridine ligand (L3) 3-Iodo-4-methoxypyridine (514 mg, 2.19 mmol) was dissolved in dry and deoxygenated DMSO (8 mL), under an argon atmosphere. Cu(0) (1.17 g, 18.4 mmol) was added, and the reaction mixture was stirred at RT for 10 min in the dark. 1,6-Diiodoperfluorohexane (0.51 g, 0.92 mmol) was added at RT, and the reaction mixture was heated to 120 °C. After 18 h stirring at 120 °C in the dark, the reaction mixture was cooled to RT and quenched with 1 M EDTA-NH3(aq.) (100 mL). The aqueous layer was extracted with Et2O (3 × 50 mL), and the combined organic layer was washed with 0.1 M EDTA-NH3(aq.) (50 mL) and brine (50 mL). The organic layer was dried over Na2SO4, filtered and concentrated in vacuo. Column chromatography (SiO2, petrol-Et2O 1:1 to 0:100) afforded bis-pyridine L3 as a white powder (263 mg, 56%). Ligand L3 characterization is described in Supporting Information Section 2.3 and Figures S13–S18. Synthesis of para-OMe-perfluorooctane bis-pyridine (L4) 3-Iodo-4-methoxypyridine (514 mg, 2.19 mmol) was dissolved in dry and deoxygenated DMSO (8 mL), under argon atmosphere. Cu(0) (1.17 g, 18.4 mmol) was added, and the reaction mixture was stirred at RT for 10 min in the dark. 1,8-Diiodoperfluorooctane (0.60 g, 0.94 mmol) was added at RT, and the reaction mixture was heated to 120 °C. After 18 h stirring at 120 °C in the dark, the reaction mixture was cooled to RT and quenched with 1 M EDTA-NH3(aq.) (100 mL). The aqueous layer was extracted with Et2O (3 × 50 mL), and the combined organic layer was washed with 0.1 M EDTA-NH3(aq.) (50 mL) and brine (50 mL). The organic layer was dried over Na2SO4, filtered and concentrated in vacuo. Column chromatography (SiO2, petrol-Et2O 1:1 to 0:100) afforded bis-pyridine L4 as a white powder (309 mg, 53%). Ligand L4 characterization is described in Supporting Information Section 2.4 and Figures S19–S24. Synthesis of metal–organic cage C1 without isolation L1 (10.0 mg, 21.9 μmol) was dissolved in DMSO-d6 (0.4 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (5.2 mg, 12 μmol) was dissolved in DMSO-d6 (0.3 mL) and added dropwise to the L1 solution. The reaction flask was rinsed three times with 0.1 mL of DMSO-d6, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by nuclear magnetic resonance (NMR) spectroscopy. Cage C1 without isolation characterization is described in Supporting Information Section 2.5.1 and Figures S25–S31. Synthesis of metal–organic cage C1 with isolation L1 (25.0 mg, 55 μmol) was dissolved in CH3CN (0.3 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (12.2 mg, 27 μmol) was dissolved in CH3CN (0.2 mL) and added dropwise to the L1 solution. The reaction flask was rinsed two times with 0.1 mL of CH3CN, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. Et2O (5 mL) was added to the resulting mixture, inducing the precipitation of a white solid. The precipitate was collected by vacuum filtration, washed twice with 2 mL of Et2O, and dried under vacuum to afford C1 as a white solid (24.5 mg, 75%). Cage C1 with isolation characterization is described in Supporting Information Section 2.5.2 and Figures S32–S35. Synthesis of metal–organic cage C2 without isolation L2 (10.0 mg, 19 μmol) was dissolved in DMSO-d6 (0.4 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (4.8 mg, 11 μmol) was dissolved in DMSO-d6 (0.3 mL) and added dropwise to the L2 solution. The reaction flask was rinsed three times with 0.1 mL of DMSO-d6, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy. Cage C2 without isolation characterization is described in Supporting Information Section 2.6.1 and Figures S36–S42. Synthesis of metal–organic cage C2 with isolation L2 (24.9 mg, 48 μmol) was dissolved in CH3CN (0.3 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (10.8 mg, 24 μmol) was dissolved in CH3CN (0.2 mL) and added dropwise to the L2 solution. The reaction flask was rinsed two times with 0.1 mL of CH3CN, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy. Et2O (5 mL) was added to the resulting mixture, inducing the precipitation of a white solid. The precipitate was collected by vacuum filtration, washed twice with 2 mL of Et2O and dried under vacuum to afford C2 as a white solid (19.1 mg, 60%). Cage C2 with isolation characterization is described in Supporting Information Section 2.6.2 and Figures S43–S46. Synthesis of metal–organic cage C3 without isolation Ligand L3 (10.0 mg, 19 μmol) was dissolved in DMSO-d6 (0.4 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (4.7 mg, 11 μmol) was dissolved in DMSO-d6 (0.3 mL) and added dropwise to the L3 solution. The reaction flask was rinsed three times with 0.1 mL of DMSO-d6, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy. Cage C3 without isolation characterization is described in Supporting Information Section 2.7.1 and Figures S47–S53. Synthesis of metal–organic cage C3 with isolation Ligand L3 (25.1 mg, 49 μmol) was dissolved in CD3CN (0.3 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (10.9 mg, 25 μmol) was dissolved in CD3CN (0.3 mL) and added dropwise to the L3 solution. The reaction flask was rinsed twice with 0.1 mL of CD3CN, and the washings were combined with the reaction mixture. The solution was stirred at room temperature overnight. Et2O (5 mL) was added to the resulting mixture, inducing the precipitation of a white solid. The precipitate was collected by vacuum filtration, washed twice with 2 mL of Et2O and dried under vacuum to afford C3 as a white solid (26.8 mg, 84%). Cage C3 with isolation characterization is described in Supporting Information Section 2.7.2 and Figures S54–S57. Synthesis of metal–organic cage C4 without isolation L4 (10.0 mg, 16 μmol) was dissolved in DMSO-d6 (0.3 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (3.7 mg, 8 μmol) was dissolved in DMSO-d6 (0.2 mL) and added dropwise to the L4 solution. The reaction flask was rinsed three times with 0.1 mL of DMSO-d6, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy. Cage C4 without isolation characterization is described in Supporting Information Section 2.8.1 and Figures S58–S64. Synthesis of metal–organic cage C4 with isolation Ligand L4 (50.0 mg, 81 μmol) was dissolved in CD3CN (1.2 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (mg, μmol) was dissolved in CD3CN (0.3 mL) and added dropwise to the L4 solution. The reaction flask was rinsed twice with 0.1 mL of CD3CN, and the washings were combined with the reaction mixture. The solution was stirred at room temperature overnight. Et2O (5 mL) was added to the resulting mixture, inducing the precipitation of a white solid. The precipitate was collected by vacuum filtration, washed twice with 5 mL of Et2O and dried under vacuum to afford C4 as a white solid (41.6 mg, 68%). Cage C4 with isolation characterization is described in Supporting Information Section 2.8.2 and Figures S65–S68. Competitive C1 and C3 cage formation L1 (10.0 mg, 22 μmol) was dissolved in DMSO-d6 (0.1 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (4.9 mg, 11 μmol) was dissolved in DMSO-d6 (0.2 mL) and added dropwise to the L1 solution. The reaction flask was rinsed two times with 0.1 mL of DMSO-d6, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy, returned to the reaction flask, and rinsed two times with 0.1 mL of DMSO-d6. L3 (11.3 mg, 22 μmol) was dissolved in DMSO-d6 (0.1 mL) and added dropwise to the solution in the reaction flask. It was also rinsed two times with 0.1 mL of DMSO-d6. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy (see Supporting Information Section 2.9.1 and Figure S69). Competitive C1 and C3 cage formation—kinetic experiments Cage C1 (5.92 mg, 2.5 μmol) was dissolved in DMSO-d6 (500 μL) in the presence of 1,4-dimethoxybenzene (10 mM) as an internal standard. Separately, a stock solution of ligand L3 (86 mM) was prepared in DMSO-d6. 115.5 μL of this solution (10 μmol, 4.0 equiv) was added to the cage solution. The resulting mixture was analyzed directly by 1H NMR spectroscopy at several time intervals: 5, 15, and 30 min, 1, 2, 5, 12, and 24 h (see Supporting Information Section 2.10.1 and Figures S73 and S74). Competitive C1 and C2 cage formation L1 (10.1 mg, 22 μmol) was dissolved in DMSO-d6 (0.1 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (4.9 mg, 11 μmol) was dissolved in DMSO-d6 (0.2 mL) and added dropwise to the L1 solution. The reaction flask was rinsed two times with 0.1 mL of DMSO-d6, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy, returned to the reaction flask, and rinsed two times with 0.1 mL of DMSO-d6. L2 (11.4 mg, 22 μmol) was dissolved in DMSO-d6 (0.1 mL) and added dropwise to the solution in the reaction flask. It was also rinsed two times with 0.1 mL of DMSO-d6. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy (see Supporting Information Section 2.9.2 and Figure S70). Competitive C1 and C2 cage formation—kinetic experiments Cage C1 (5.71 mg, 2.4 μmol) was dissolved in DMSO-d6 (500 μL) in the presence of 1,4-dimethoxybenzene (10 mM) as an internal standard. Separately, a stock solution of ligand L2 (82 mM) was prepared in DMSO-d6. 116.3 μL of this solution (9.5 μmol, 4.0 equiv) was added to the cage solution. The resulting mixture was analyzed directly by 1H NMR spectroscopy at several time intervals: 5, 15, and 30 min, 1, 2, 5, and 12 h (see Supporting Information Section 2.10.2 and Figures S75 and S76). Competitive C2 and C3 cage formation L2 (10.0 mg, 19 μmol) was dissolved in DMSO-d6 (0.1 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (5.1 mg, 12 μmol) was dissolved in DMSO-d6 (0.2 mL) and added dropwise to the L2 solution. The reaction flask was rinsed two times with 0.1 mL of DMSO-d6, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy, returned to the reaction flask and rinsed two times with 0.1 mL of DMSO-d6. L3 (10.0 mg, 19 μmol) was dissolved in DMSO-d6 (0.1 mL) and added dropwise to the solution in the reaction flask. It was also rinsed two times with 0.1 mL of DMSO-d6. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy (see Supporting Information Section 2.9.3 and Figure S71). Competitive C2 and C3 cage formation—kinetic experiments Cage C2 (6.48 mg, 2.5 μmol) was dissolved in DMSO-d6 (500 μL) in the presence of 1,4-dimethoxybenzene (10 mM) as an internal standard. Separately, a stock solution of ligand L3 (86 mM) was prepared in DMSO-d6. 114.8 μL of this solution (10 μmol, 4.0 equiv) was added to the cage solution. The resulting mixture was analyzed directly by 1H NMR spectroscopy at several time intervals: 5, 15, and 30 min, 1, 2, 5, and 12 h (see Supporting Information Section 2.10.3 and Figures S77 and S78). Competitive C3 and C4 cage formation L3 (10.0 mg, 19 μmol) was dissolved in DMSO-d6 (0.1 mL). In a separate flask, Pd(CH3CN)4(BF4)2 (4.2 mg, 9.5 μmol) was dissolved in DMSO-d6 (0.2 mL) and added dropwise to the L3 solution. The reaction flask was rinsed two times with 0.1 mL of DMSO-d6, and the washings were added to the reaction mixture. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy, returned to the reaction flask, and rinsed two times with 0.1 mL of DMSO-d6. L4 (11.7 mg, 19 μmol) was dissolved in DMSO-d6 (0.1 mL) and added dropwise to the solution in the reaction flask. It was also rinsed two times with 0.1 mL of DMSO-d6. The solution was stirred at room temperature overnight. The resulting mixture was analyzed directly by NMR spectroscopy (see Supporting Information Section 2.9.4 and Figure S72). Competitive C3 and C4 cage formation—kinetic experiments Cage C4 (7.35 mg, 2.4 μmol) was dissolved in DMSO-d6 (500 μL) in the presence of 1,4-dimethoxybenzene (10 mM) as an internal standard. Separately, a stock solution of ligand L3 (86 mM) was prepared in DMSO-d6. 113.0 μL of this solution (9.7 μmol, 4.0 equiv) was added to the cage solution. The resulting mixture was analyzed directly by 1H NMR spectroscopy at several time intervals: 5, 15, and 30 min, 1, 2, 5, 12, and 24 h (see Supporting Information Section 2.10.4 and Figure S79). Host–guest experiments of cages C3 and C4 with small molecules A stock solution of cage C3 or C4 (3.75 mM) was prepared in DMSO-d6. Aliquots of 500 μL were transferred into five individual NMR tubes. To each tube, 100 equiv (with respect to the cage) of various small organic guest molecules were added. The selected guests included dichloromethane (CH2Cl2), benzene, bromochloromethane (BrCH2Cl), and chloroform (CHCl3). No guest encapsulation was observed as no chemical shift perturbations of the cage signals occurred (see Supporting Information Sections 2.11.1 and 2.12.1 and Figures S80 and S83 for complete results). Host–guest experiments of cages C3 and C4 with perfluorooctanoic acid A stock solution of cage C3 or C4 (3.75 mM) and another of concentrated PFOA (93.75 mM) were prepared in DMSO-d6. The 1H titration was carried out on a Bruker AV400 spectrometer (Bruker Corporation, Billerica, Massachusetts, United States) at 300 K. Increasing amounts of PFOA solution were added to an initial sample of 500 μL of cage C3 or C4 solution. The 1H spectra were recorded between the additions of 0.2 to 5 or 4 equiv of guest molecule with respect to the cages C4 or C3, respectively (see Supporting Information Sections 2.11.2 and 2.12.2 and Figures S81, S82, and S84 for complete results and calculations). Results and Discussion Synthesis of cages To obtain the fluorinated cage C1, we first designed ligand L1, containing two terminal meta-substituted pyridine moieties connected with a 6-carbon perfluorinated chain (Scheme 1). The synthesis was performed via a Cu-catalyzed Ullmann-type coupling of 3-iodopyridine with 1,6-diiodoperfluorohexane in DMSO at 120 °C under inert atmosphere for 19 h. L1 was isolated by extraction with diethyl ether and purified by silica gel column chromatography, affording the product as a white solid in 74% yield (see Supporting Information Sections 2.1–2.4 for complete ligand characterization). Scheme 1 | Synthetic scheme of ligands L1–L3. Download figure Download PowerPoint After obtaining ligand L1, the Pd2L4 metal–organic cage C1 was synthesized via self-assembly of L1 and Pd2+ (Scheme 2). The reaction was carried out in DMSO-d6, which proved to be the best solvent to monitor the reaction by 1H NMR spectroscopy, maintaining all reagents, intermediates, and products in solution. For this, ligand L1 and Pd(CH3CN)4(BF4)2 were dissolved separately in DMSO-d6 and then combined in a 2:1 ligand:metal molar ratio. The mixture was stirred at room temperature overnight to allow the coordination-driven self-assembly of cage C1. Scheme 2 | Synthetic scheme of metal–organic cages C1–C3. Download figure Download PowerPoint The overnight 1H NMR spectrum of the reaction mixture reveals that the cage is formed as the main species. However, a portion of the ligand remains in equilibrium with C1. As judged from the integral values from the 1H NMR spectrum, 77% of cage and 23% of free ligand was observed, indicating an incomplete reaction (Figure 2a, see Supporting Information Section 2.5 for additional information). Based on these results, we hypothesise that the incomplete formation of cage C1 is due to a low coordination strength of pyridine moiety due to the electron-withdrawing effect of the -(CF2)n- chain, which decreases the electron density available on the nitrogen atom, resulting in the weakening of the Pd-pyridine coordination bond. To address this limitation and improve the stability of the Pd–pyridine bond, we designed ligands L2 and L3 by introducing a methoxy group in the meta- and para-positions of the pyridine (Scheme 1). These modifications aimed to increase the nitrogen atom electron density and enhance its coordination strength with palladium. Figure 2 | 1H NMR spectra (500 MHz, aromatic region, DMSO-d6, 298 K) of the self-assembly reaction mixture performed in DMSO-d6. Comparison between (a) free L1 (orange) and the overnight reaction mixture of C1 (red); (b) free L2 (cyan) and the overnight reaction mixture of C2 (blue); (c) free L3 (light green) and the overnight reaction mixture of C3 (green). Download figure Download PowerPoint We synthesized ligands L2 and L3 in a 59% and 56% yields, respectively, following a similar procedure to that developed for L1, using 3-bromo-5-methoxypyridine or 3-iodo-4-methoxypyridine (Scheme 1, see Supporting Information Sections 2.2 and 2.3 for complete characterization). Then cages C2 and C3, were obtained following the same protocol used for cage C1, that is, a 2:1 ligand: Pd(CH3CN)4(BF4)2 molar ratio in DMSO-d6, overnight at room temperature (Scheme 2). The overnight 1H NMR spectra of the reaction mixtures of C2 and C3 reveal increased cage formation compared to C1 (Figure 2a–c). Quantitative analysis of the integral values of the 1H NMR signals indicates that C2 achieves 87% of cage formation, with 13% of free ligand in equilibrium. In contrast, cage C3 exhibits 100% cage formation with no detectable free ligand signals. These results strongly suggest that C3 is the most stable of the three cages. After proving the correct formation of the cages in DMSO-d6, attempts to isolate and purify the cages were unsuccessful. Additions of nonpolar solvents such as CH2Cl2/Et2O to the DMSO-d6 reaction mixtures did not result in substantial precipitation of the cages. To improve product C1– C3 isolation, we changed the reaction solvent to CH3CN, allowing product precipitation by the addition