High-capacity sieving of C₃F₆ and C₃F₈ by a copper-based MOF with interconnected gourd-shaped channels

Abstract

Targeting the challenging purification of electronic-grade C₃F₈, we report the size-sieving separation of C₃F₆ and C₃F₈ by a robust copper-based metal–organic framework, CuHTPO, that features a distinctive interconnected “gourd”-shaped pore architecture. This compound completely excludes C₃F₈ while strongly adsorbing C₃F₆, achieving an adsorption capacity as high as 71.3 cm³ g⁻¹ at 298 K and 100 kPa. Dynamic breakthrough experiments demonstrate the direct production of ultra-high-purity C₃F₈ (>5N) from a C₃F₆/C₃F₈ (10 : 90, v/v) gas mixture. The underlying size-sieving-based separation mechanism is corroborated by in situ infrared spectroscopy and density functional theory calculations.

---

Introduction

Perfluoropropane (C₃F₈), owing to its unique carbon-to-fluorine ratio (C/F = 0.375) and favorable thermodynamic properties, exhibits an excellent process window in HARC (high aspect ratio contact) etching and CVD (chemical vapor deposition) chamber cleaning.¹ However, as process nodes approach near-zero defect tolerance, the purity requirement for C₃F₈ has been elevated to above 99.999% (5N).^2–5 Beyond the electronics industry, C₃F₈ plays an irreplaceable role in the medical field particularly in complex vitreoretinal surgeries, where similarly stringent purity standards are required.⁶ The presence of any toxic impurities may lead to irreversible damage to the retinal neuroepithelium or unintended elevation of intraocular pressure,^7,8 potentially resulting in severe clinical complications. C₃F₆ is one of the most ubiquitous impurities encountered during both industrial purification and exhaust gas recovery of C₃F₈.^9–13 Owing to the small boiling point difference between C₃F₆ and C₃F₈ (ΔT_b ≈ 7.3 K), achieving >5N purity via cryogenic distillation requires exceptionally high reflux ratios and oversized distillation columns, resulting in prohibitive energy consumption and capital costs.^14–16 Consequently, the development of efficient and low-energy separation strategies for C₃F₆ removal from C₃F₈ is of critical industrial importance.

Adsorptive separation utilizing porous materials has emerged as a compelling alternative to cryogenic distillation, driven by its superior energy efficiency and lower capital expenditure.¹⁷ Conventional adsorbents, exemplified by activated carbon and zeolites, have demonstrated the potential for physical discrimination of C₃F₆ and C₃F₈.^18–23 Metal–organic frameworks (MOFs), exhibiting remarkable tunability in pore size, topology, and surface chemistry, hold particularly promise for precise separation of physicochemically similar molecules.^24–31 Given the ultrahigh purity requirement of electronic-grade C₃F₈, molecular sieving is widely considered the most desirable separation mechanism for removing trace C₃F₆ and directly producing high-purity C₃F₈. However, achieving precise size-dependent discrimination is frequently accompanied by a significant reduction in adsorption capacity. Recent strategies involving framework flexibility,^31,32 pore environment engineering,^33,34 electrostatic modulation,³⁵ and biomimetic design³⁶ have largely mitigated this limitation, nevertheless, sieving-based MOFs with high adsorption capacity remains highly needed.

In this work, we demonstrate high-capacity sieving of C₃F₆ and C₃F₈ using a MOF featuring large cages accessible through narrow windows. This architecture allows for the maximal packing of guest molecules within the internal voids while maintaining strictly defined apertures for size-selective discrimination.^37–42 The robust copper-based MOF, CuHTPO (H₃TPO = tris-(4-carboxylphenyl) phosphine oxide), featuring interconnected “gourd”-shaped pore channels, effectively overcomes the “trade-off” between adsorption capacity and size-sieving precision. CuHTPO completely excludes C₃F₈ while its large internal cavities provide abundant adsorption sites for C₃F₆. As a result, it exhibits negligible uptake of C₃F₈ but a record-high C₃F₆ adsorption capacity of 71.3 cm³ g⁻¹ at 298 K and 100 kPa. Dynamic breakthrough experiments further demonstrate the direct production of high-purity C₃F₈ (>5N) from a C₃F₆/C₃F₈ (10 : 90, v/v) mixture with excellent cycling stability, while in situ infrared spectroscopy and density functional theory calculations provide insight into the underlying separation mechanism.

Results and discussion

CuHTPO was synthesized via a slightly modified procedure based on a previously reported method,⁴³ with detailed synthetic protocols provided in the SI. Briefly, Cu(NO₃)₂·3H₂O and H₃TPO were subjected in a mixed solvent of N,N-dimethylformamide/H₂O/methanol, and the subsequent solvothermal reaction afforded crystals of CuHTPO (Fig. S1 and S2). It crystallizes in the orthorhombic crystal system with the space group Pbcn. The framework is constructed from phosphine oxide ligand HTPO²⁻ coordinated to classical paddlewheel-type Cu₂(COO)₄ secondary building units (SBUs) (Fig. 1a). Each HTPO²⁻ ligand laterally bridges two copper clusters through two carboxylate groups, while a phosphine oxide moiety coordinates monodentately to a copper center along the axial direction; notably, the non-coordinated carboxylic acid groups engage in hydrogen-bonding interactions with carboxylate groups from neighboring ligands (Fig. 1b). CuHTPO assembles into a robust three-dimensional network, featuring pore channels extending along the a axis that exhibit a characteristic gourd-like architecture composed of alternating large cavities of approximately 6.9 Å and narrow pore apertures of about 5.5 Å (Fig. 1c). Distinct from conventional one-dimensional gourd-shaped channels,^44,45 the pores of CuHTPO form a two-dimensional interconnected “gourd”-shaped channel system propagating along the ac plane, in which large cavities are interconnected by four narrow apertures (Fig. 1d). This unique pore topology not only mitigates the intrinsic trade-off between selectivity and adsorption capacity in molecular sieving but also provides multiple and efficient diffusion pathways for guest molecules within the framework.

Fig. 1 Crystal structure of CuHTPO. (a) 3D structure built from H₃TPO and Cu₂(COO)₄. (b and c) Cavities and 2D channels of CuHTPO. Color scheme: Cu, blue; O, red; C, gray; P green.

The phase purity of as-synthesized CuHTPO was confirmed by powder X-ray diffraction (PXRD). As shown in Fig. S3, the PXRD patterns of the as-synthesized and methanol-exchanged samples match well with the simulated pattern derived from single-crystal data. Notably, the activated sample and the sample after adsorption measurements retain identical diffraction features, indicating full preservation of crystallinity throughout activation and adsorption processes. Thermogravimetric analysis (TGA) of the as-synthesized sample reveals continuous mass loss upon heating, whereas the methanol-exchanged sample exhibits an extended plateau up to 320 °C (Fig. S4). Furthermore, the chemical and thermal stability of CuHTPO were systematically investigated. The material was immersed in various organic solvents and aqueous solutions with pH values ranging from 3 to 9 for seven days. PXRD analyses reveal that the framework structure is largely preserved in different organic solvents (Fig. S5), while maintaining good structural integrity in mildly acidic aqueous media (Fig. S6). This stability is particularly relevant considering that industrial production of C₃F₈ typically introduces trace amounts of acidic impurities,⁴⁶ rendering the crude product weakly acidic. Therefore, materials exhibiting resistance to mildly acidic environments are better suited for practical separation processes in such systems. In addition, variable-temperature in situ PXRD measurements under a nitrogen atmosphere demonstrate that the crystalline structure of CuHTPO remains essentially intact up to 320 °C (Fig. S7). Combined with its stability in various solvents, this robustness enables the use of more rigorous yet efficient solvent-exchange and activation procedures. The permanent porosity of CuHTPO was evaluated by N₂ adsorption at 77 K, revealing two types of pores with diameters of 5.5 and 6.8 Å, respectively (Fig. 2a and S8), in good agreement with the crystal structure. The adsorption displays an N₂ uptake of 239 cm³ g⁻¹, and a corresponding Brunauer–Emmett–Teller (BET) surface area of 1000.3 m² g⁻¹ (Fig. S9).

Fig. 2 (a) N₂ adsorption isotherms of CuHTPO measured at 77 K (inset: pore size distribution derived from the DFT method). (b) Single-component adsorption isotherms of C₃F₆ and C₃F₈ on CuHTPO measured at different temperatures. (c) Adsorption kinetics of C₃F₆ and C₃F₈ obtained at 298 K and a partial pressure of 0.5 bar. (d) Comparison of the C₃F₆/C₃F₈ uptake ratio and the C₃F₆ uptake at 298 K and 100 kPa for CuHTPO and representative benchmark material.

The robust framework and precisely defined pore dimensions of CuHTPO prompted us to evaluate its adsorption and separation performance toward C₃F₆ and C₃F₈. Single-component adsorption isotherms for C₃F₆ and C₃F₈ were measured at 273, 298, and 308 K (Fig. 2b). Across this temperature range, CuHTPO exhibits negligible adsorption of C₃F₈, whereas a typical type-I adsorption profile is observed for C₃F₆. At 298 K and 100 kPa, CuHTPO delivers a substantial C₃F₆ uptake of 71.6 cm³ g⁻¹ (3.2 mmol g⁻¹). These results clearly demonstrate that CuHTPO is capable of completely excluding the bulkier C₃F₈ while efficiently adsorbing C₃F₆, highlighting its potential for purifying C₃F₈ by removing trace C₃F₆. Notably, the C₃F₆ adsorption capacity of CuHTPO surpasses previously reported molecular-sieving-based benchmark MOFs for C₃F₆/C₃F₈ separation, including Ca-tcpb³¹ (44.8 cm³ g⁻¹), Ni(INA)₂-NH₂ (ref. 33) (56.7 cm³ g⁻¹), Zr-PMA³⁴ (59.1 cm³ g⁻¹), Zn-bzx-CF₃ (ref. 35) (47.0 cm³ g⁻¹), and CoFA³⁶ (44.8 cm³ g⁻¹) (Fig. 2d and Table S3). Adsorption kinetics measurements reveal that CuHTPO takes up C₃F₆ quickly, whereas C₃F₈ remains essentially being excluded, confirming its selective molecular exclusion behavior (Fig. 2c). The calculated diffusion time constant for C₃F₆ (3.3 × 10⁻³ s⁻¹) surpasses that of C₃F₈ (6.4 × 10⁻⁵ s⁻¹) by a factor of over 50 (Fig. S10). We attribute this precise size-sorting to the optimal pore window of the large-cavity-small-aperture architecture of CuHTPO.

To quantitatively evaluate the binding strength of C₃F₆ and C₃F₈ within the CuHTPO framework, differential scanning calorimetry (DSC) measurements were performed at 298 K. The adsorption enthalpy (ΔH_ads) for C₃F₆ was determined to be 43.02 kJ mol⁻¹, whereas the value for C₃F₈ was negligible (Fig. 3a). Furthermore, the isosteric heat of adsorption (Q_st) calculated from the C₃F₆ isotherms at various temperatures is ≈38.7 kJ mol⁻¹ (Fig. S11–S16), generally matching the adsorption enthalpy measured by DSC. On this basis, the dynamic separation performance of CuHTPO toward the C₃F₆/C₃F₈ mixture was further evaluated at 298 K using a fixed-bed column packed with CuHTPO under a feed composition of C₃F₆/C₃F₈ (10 : 90, v/v). As shown in Fig. 3b, C₃F₈ eluted immediately upon introduction of the gas mixture, whereas C₃F₆ was retained in the column and did not break through until approximately 240 min g⁻¹, thereby enabling the production of ultrapure C₃F₈ (>99.999%) with a high yield of 323.65 cm³ g⁻¹. Furthermore, subsequent adsorption–desorption breakthrough cycling experiments, in which the column was regenerated under a He flow at 423 K, revealed nearly identical breakthrough profiles over three consecutive cycles (Fig. 3c), demonstrating the excellent recyclability and cycling stability of CuHTPO.

Fig. 3 (a) Differential scanning calorimetry (DSC)-derived adsorption enthalpies of CuHTPO toward C₃F₆ and C₃F₈. (b) Binary breakthrough curves of a C₃F₆/C₃F₈ (10 : 90, v/v) mixture recorded at 298 K using a fixed-bed column packed with CuHTPO; The purity of C₃F₈ exceeded 99.999% before the breakthrough of C₃F₆. (c) Comparison of three consecutive dynamic separation cycles. (d) C₃F₆ adsorption–desorption cycling performance of CuHTPO over 10 consecutive cycles at 298 K.

In addition, to systematically assess its long-term recyclability, temperature-swing adsorption–desorption cycling tests were conducted, during which the sample was exposed to pure C₃F₆ at 298 K followed by regeneration under an N₂ atmosphere at 373 K. Notably, no discernible loss in uptake capacity was observed over 10 consecutive cycles (Fig. 3d), highlighting the structural stability and reusability of CuHTPO.

To gain deeper insights into the adsorption mechanism of perfluorinated gases within the CuHTPO framework, in situ infrared (IR) spectroscopy measurements were performed (Fig. 4a and b). Upon activation, approximately 20 Torr of pure C₃F₆ or C₃F₈ was introduced into the IR cell, and difference spectra at various time intervals were obtained by subtracting the activated-state spectrum from the adsorption spectra. As shown in Fig. 4a and b, upon introduction of C₃F₆, a red shift of approximately 5 cm⁻¹ was observed in the 1193–1178 cm⁻¹ region, which can be assigned to the asymmetric C–F stretching vibration (ν_as) of C₃F₆, while a distinct negative band appears around 1000 cm⁻¹, attributed to the in-plane bending vibration of aromatic C–H (δ), indicating that this mode is sensitive to guest occupancy and suggesting the presence of specific host–guest interactions dominated by C–H⋯F hydrogen bonding between C₃F₆ and the framework; in contrast, upon exposure to C₃F₈, no significant changes were observed in the aromatic C–H in-plane deformation region and only gas-phase C–F vibrational features were detected near 1000 cm⁻¹ These results directly demonstrate, at the molecular level, that CuHTPO discriminates C₃F₆ from C₃F₈ through a molecular sieving mechanism.

Fig. 4 IR difference spectra of CuHTPO upon loading C₃F₆ and C₃F₈ (≈20 Torr) at 298 K: (a) referenced to the activated sample under vacuum, with gas-phase signals subtracted and (b) time-resolved spectra collected at 20 min intervals. (c) The optimal adsorption sites of C₃F₆ on CuHTPO. (d) DFT-calculated energy profiles and relative energies for C₃F₆ and C₃F₈.

Density functional theory (DFT) calculations further corroborated the conclusions drawn from the in situ IR experiments. As illustrated in Fig. 4c, C₃F₆ preferentially resides in the corner regions of the pore, where the dominant interactions arise from multiple C–F⋯H hydrogen bonds formed between fluorine atoms of C₃F₆ and hydrogen atoms on the phenyl rings of the framework, with F⋯H interaction distances ranging from 2.538 to 3.153 Å (indicated by black dashed lines). To further validate the molecular sieving mechanism of CuHTPO toward C₃F₆ and C₃F₈, diffusion energy barrier calculations were performed. The calculated energy barrier for C₃F₆ is 22.2 kJ mol⁻¹, whereas that for C₃F₈ is as high as 90.6 kJ mol⁻¹ (Fig. 4d), confirming that the diffusion of C₃F₈ is essentially prohibited. In addition, the calculated binding energy of C₃F₆ is 40.76 kJ mol⁻¹, which is in good agreement with the results obtained from DSC measurements. The charge density difference analysis reveals pronounced electron redistribution around the pore apertures and metal–ligand coordination sites upon C₃F₆ adsorption, indicating strong host–guest interactions accompanied by significant molecular polarization (Fig. S18). Consistently, electrostatic potential calculations show an overall positively charged framework and a negatively charged C₃F₆ molecule, confirming a favorable electrostatic affinity between the guest and the framework (Fig. S19). These theoretical results are in excellent agreement with the experimental adsorption behavior, further confirming that the optimal pore aperture of CuHTPO enables effective size-exclusive molecular sieving for C₃F₆/C₃F₈ separation.

Conclusions

We have demonstrated here the high-capacity size-sieving separation of C₃F₆ and C₃F₈ by a stable and robust copper-based metal–organic framework, CuHTPO. The unique coordination mode between the phosphine oxide-carboxylate ligand and copper nodes not only significantly enhances the structural stability of CuHTPO, but also constructs a distinctive interconnected “gourd”-shaped pore architecture, in which large cavities are connected through narrow pore apertures. This pore configuration effectively breaks the trade-off between selectivity and adsorption capacity commonly observed in conventional molecular-sieving MOFs, while simultaneously maintaining excellent framework robustness. Binary dynamic breakthrough experiments demonstrate that CuHTPO can stably produce ultra-high-purity C₃F₈ (>99.999%) over multiple cycles. Furthermore, in situ infrared spectroscopy combined with density functional theory calculations elucidates the interaction strengths of C₃F₆ and C₃F₈ within CuHTPO at the molecular level and reveals the underlying separation mechanism. This work provides important structure–performance relationship insights for the development of high-performance adsorbents for C₃F₆/C₃F₈ separation under stringent purity requirements.

Author contributions

H. Wang and L. Yu conceived and designed the project. M. Gao provided conceptual guidance. Z. Wang synthesized the compounds and performed PXRD analysis, stability tests, adsorption measurements, and breakthrough experiments. M.-Y. Zhou performed theoretical calculations. S. Li and X. Zhou provided technical support for the breakthrough tests. Y. Wu and F.-A. Guo conducted the in situ infrared spectroscopy experiments. Z. Wang, S. Mao, and L. Yu wrote the first draft of the manuscript. All authors contributed to the discussion of the results and the revision of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

CCDC 2544357 contains the supplementary crystallographic data for this paper.⁴⁷

Data associated to the article are available in the supplementary information (SI). Supplementary information: experimental methods, PXRD analysis, TGA curves, additional adsorption isotherms, calculation adsorption selectivity and heat for CuHTPO. See DOI: https://doi.org/10.1039/d6sc01756h.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (22478251, 22508261), Guangdong Science and Technology Program (2024TQ08A672), Shenzhen Science and Technology Program (JCYJ20250604135818024), Shenzhen Polytechnic University Research Fund (6024310024K) and Postdoctoral Foundation of Shenzhen Polytechnic University (6024331002K).

Notes and references

1. K. Kim, C. Oh, N.-E. Lee, J. Kim, J. Bae, G. Yeom and S. Yoon, J. Vac. Sci. Technol. B, 2004, 22, 483–488.
2. M. B. Chang and J.-S. Chang, Ind. Eng. Chem. Res., 2006, 45, 4101–4109.
3. F. Illuzzi, H. Thewissen and J. Integr, Environ. Sci., 2010, 7, 201–210.
4. W.-T. Tsai, H.-P. Chen and W.-Y. Hsien, J. Loss Prev. Process Ind., 2002, 15, 65–75.
5. R. S. Prabhakar, T. C. Merkel, B. D. Freeman, T. Imizu and A. Higuchi, Macromolecules, 2005, 38, 1899–1910.
6. S. Chang, H. A. Lincoff, D. J. Coleman, W. Fuchs and M. E. Farber, Ophthalmology, 1985, 92, 651–656.
7. N. C. Steinle, D. S. Dhoot, C. Q. Ruiz, A. A. Castellarin, D. J. Pieramici, R. F. See, S. C. Couvillion, M. a. A. Nasir and R. L. Avery, Retina, 2017, 37, 643–650.
8. I. A. Rodrigues, A. N. Stangos, D. A. McHugh and T. L. Jackson, Am. J. Ophthalmol., 2013, 155, 270–276.
9. M. Basu, Understanding Fluoride and Fluorocarbon Toxicity: An Overview, Springer Nature Singapore, Singapore, 2024.
10. R. D. Chambers, Fluorine in organic chemistry, John Wiley & Sons, New York, 2004.
11. Y. Oda, R. Suzuki, T. Mori, H. Takahashi, H. Natsugari, D. Omata, J. Unga, H. Uruga, M. Sugii and S. Kawakami, Int. J. Pharm., 2015, 487, 64–71.
12. G. Casini, P. Loiudice, S. De Cillà, P. Radice and M. Nardi, Int. J. Retina Vitreous, 2016, 2, 10.
13. S. M. McClintic, J. D. Grodsky and G. G. Emerson, Retin. Physician, 2025, 22, 8–10.
14. X. Cao, R. Liu, Y. Lu, S. Jia and X. Yuan, Sep. Purif. Technol., 2021, 279, 119813.
15. D. S. Sholl and R. P. Lively, Nature, 2016, 532, 435–437.
16. F. Fang and J. Joffe, J. Chem. Eng. Data, 1966, 11, 376–379.
17. J.-R. Li, J. Sculley and H.-C. Zhou, Chem. Rev., 2012, 112, 869–932.
18. P. Hohenberg and W. Kohn, Phys. Rev., 1964, 136, B864.
19. W. Kohn and L. J. Sham, Phys. Rev., 1965, 140, A1133.
20. M. Menk, C. Bläker, C. Pasel and D. Bathen, Microporous Mesoporous Mater., 2025, 402, 113988.
21. Z. Du, Z. Wu, X. Li, W. Zhang, J. Huang, Y. Wu, L. Zhu and J. Xiao, Sep. Purif. Technol., 2026, 383, 136249.
22. Y. Fu, L. Sheng, W. Xia, G. Hai, J. Yan, L. Chen, Q. Yang, Z. Zhang, Q. Ren and Z. Bao, Ind. Chem. Mater., 2025, 3, 567–577.
23. X. Wei, S. Du, Y. Zheng, J. Peng, G. miao, X. Meng, X. Li, X. Li and J. Xiao, AlChE J., 2026, 72, e70155.
24. O. M. Yaghi and H. Li, J. Am. Chem. Soc., 1995, 117, 10401–10402.
25. B. M. Ji, D. Y. Zhang, R. Liang, G. H. Kang, Q. Y. Zhu and D. S. Deng, Cryst. Growth Des., 2021, 21, 482–489.
26. P. A. P. Mendes, P. Horcajada, S. Rives, H. Ren, A. E. Rodrigues, T. Devic, E. Magnier, P. Trens, H. Jobic, J. Ollivier, G. Maurin, C. Serre and J. A. C. Silva, Adv. Funct. Mater., 2014, 24, 7666–7673.
27. S. Kitagawa, Chem. Soc. Rev., 2014, 43, 5415–5418.
28. H. Furukawa, K. E. Cordova, M. O'Keeffe and O. M. Yaghi, Science, 2013, 341, 1230444.
29. L. Lan, S. J. Wei, W. Xia, Q. Zhang, Y. Liu, C. Bao, M. Cheng, Z. Y. Qian, L. L. Wang, Y. L. Li, M. Feng, Z. Bao and T. L. Hu, Angew Chem. Int. Ed. Engl., 2026, 72, e6037453.
30. H.-F. Ma, X.-P. Fu, H.-P. Xiao, L. Chen, Q.-Y. Liu and Y.-L. Wang, Sci. China Chem., 2026 DOI:10.1007/s11426-025-3177-3.
31. W. Xia, Y. Yang, L. Sheng, Z. Zhou, L. Chen, Z. Zhang, Z. Zhang, Q. Yang, Q. Ren and Z. Bao, Sci. Adv., 2024, 10, eadj6473.
32. L. Sheng, W. Xia, Y. Fu, J. Yan, Z. Zhou, F. Zheng, F. Shen, L. Chen, Z. Zhang, Q. Yang, Q. Ren and Z. Bao, ACS Mater. Lett., 2025, 7, 2080–2087.
33. X. Lv, M. Zheng, Z. Jiang, H. Huang, Z. Bao and C. Zhong, Chem. Eng. J., 2025, 522, 168005.
34. W. Xia, Z. Zhou, C. Xia, L. Chen, L. Sheng, F. Zheng, Z. Zhang, Q. Yang, Q. Ren and Z. Bao, Angew. Chem., Int. Ed., 2025, 64, e202503505.
35. M. Zheng, W. Xue, T. Yan, Z. Jiang, Z. Fang, H. Huang and C. Zhong, Angew. Chem., Int. Ed., 2024, 63, e202401770.
36. W. Xia, Z. Zhou, L. Sheng, L. Chen, F. Shen, F. Zheng, Z. Zhang, Q. Yang, Q. Ren and Z. Bao, Nat. Commun., 2024, 15, 8716.
37. J.-R. Li, R. J. Kuppler and H.-C. Zhou, Chem. Soc. Rev., 2009, 38, 1477–1504.
38. H. Wang and J. Li, Acc. Chem. Res., 2019, 52, 1968–1978.
39. X.-W. Zhang, D.-D. Zhou and J.-P. Zhang, Chem, 2021, 7, 1006–1019.
40. H. Wang, Y. Liu and J. Li, Adv. Mater., 2020, 32, 2002603.
41. L. Li, J. G. Bell, S. Tang, X. Lv, C. Wang, Y. Xing, X. Zhao and K. M. Thomas, Chem. Mater., 2014, 26, 4679–4695.
42. B. Van de Voorde, B. Bueken, J. Denayer and D. De Vos, Chem. Soc. Rev., 2014, 43, 5766–5788.
43. W. R. Lee, D. W. Ryu, W. J. Phang, J. H. Park and C. S. Hong, Chem. Commun., 2012, 48, 10847–10849.
44. L. Yu, X. Dong, Q. Gong, S. R. Acharya, Y. Lin, H. Wang, Y. Han, T. Thonhauser and J. Li, J. Am. Chem. Soc., 2020, 142, 6925–6929.
45. L. Yu, S. Ullah, H. Wang, Q. Xia, T. Thonhauser and J. Li, Angew Chem. Int. Ed. Engl., 2022, 61, e202211359.
46. Y. W. Alsmeyer, W. V. Childs, R. M. Flynn, G. G. I. Moore and J. C. Smeltzer, Electrochemical Fluorination and Its Applications, Springer US, Boston, 1994.
47. CCDC 2544357: Experimental Crystal Structure Determination, 2026,  DOI:10.5517/ccdc.csd.cc2rdm1v.

---