Tuning SF₆ affinity via tailoring pore environments in metal–organic frameworks

Abstract

Herein, an investigation into the impact of subtle binding site changes on SF₆ affinity is presented by systematically altering the pore environment of MOF-808 via post-synthetic modification with aryl and benzylic fluoride groups. All functionalized derivatives outperform the parent material, with aryl fluorides increasing SF₆ uptake fourfold.

---

Sulfur hexafluoride (SF₆), an inert gas with exceptional dielectric and arc-extinguishing properties, has historically been a critical insulating gas in the power industry.^1–3 Despite it being a substantial asphyxiation hazard in enclosed spaces, it remains in use because its properties cannot be replicated by other gas mixtures.^2,4 To reduce demand, it is commonly diluted to between a 10–30% SF₆/N₂ vol% mixture for industrial applications.^4,5 However, despite being more dilute, SF₆ contamination via leakage and improper end-of-life disposal of equipment remains a concern, both from a safety and an economic perspective. These concerns, coupled with the relatively high cost of production of new SF₆, have necessitated the efficient recovery of high purity SF₆ from end-of-life equipment.

SF₆ capture is typically performed using energy-intensive, low temperature liquefaction methods.⁶ An alternative is the use of heterogeneous physisorbent materials. Metal–organic frameworks (MOFs) porous, crystalline network solids assembled from organic linkers and metal nodes—have shown promise for the capture of SF₆.^7–15 However, most reported materials rely primarily on sieving effects, rather than chemical affinity, to capture SF₆.^16–20 Despite the unreactive nature of SF₆, there is opportunity to enhance affinity with MOF pores by tailoring the chemical environment of the pores to increase MOF–SF₆ interactions.

To address this opportunity, herein we present a systematic investigation into the effect of modifying MOF pore environments on SF₆ capture capacity using a series of MOF-808 materials post-synthetically modified with different fluorinated benzoic acid derivatives, Fig. 1. The incorporation of fluorinated groups into MOF pores has been shown to alter the chemical environment, leading to enhanced interactions with adsorbate molecules.²¹ Fluorinated MOFs have shown promise in a variety of gas-capture applications,^21,22 including the separation of non-polar hydrocarbons,²³ trace CO₂ capture,^24,25 and removal of SO₂ from flue gas.²⁷ Further, fluorinated porous polymers have shown promise for SF₆ capture.²⁸ Thus, the fluorobenzoic acid groups are expected to interact favourably with SF₆, which in turn will enhance adsorption.

Fig. 1 Overview of the post-synthetic exchange of capping formates in MOF-808 for functionalized groups (top),^4,26 and the functional groups used in this work (bottom).

To investigate the impact of the fluorines’ electronic environment in conjunction with the degree of fluorination, a set of fluorinated benzoic acids with aryl fluorides, 4-fluorobenzoic acid (F₁) and 3,5-difluorobenzoic acid (F₂) along with a set of fluorinated benzoic acids with benzylic fluorides, 4-(trifluoromethyl) benzoic acid (CF₃) and 3,5-bis(trifluoromethyl)benzoic acid ((CF₃)₂) were selected as model fluorinated groups. Increasing the overall number of fluorines within the pore is expected to increase the number of dispersive interactions with SF₆, leading to increased adsorption, so long as sufficient porosity is maintained; differences in electronic environment between aryl and benzylic fluorines should also alter their relative changes to the pore microenvironment, creating different adsorption sites and thus influence SF₆ uptake.

First, MOF-808 was synthesized and post-synthetically functionalized with each of the fluorobenzoic acids. The resulting MOF-808 fluorinated derivatives were evaluated by powder X-ray diffraction (PXRD), gas sorption, and thermal gravimetric analysis (TGA). The successful incorporation of the fluorinated moieties was confirmed using a combination of digestion ¹H nuclear magnetic resonance (NMR) and Fourier-transform infrared (FT-IR) spectroscopies. Scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy (SEM-EDS) demonstrated the homogeneous distribution of fluorines within the bulk material.

Next, the materials were tested for SF₆ capture. Importantly, uptake could be tuned based on the identity of the fluorinated group installed, with aryl fluorines exhibiting an outsized affinity for SF₆ relative to their incorporation. MOF-808-F₁ exhibited a fourfold increase in SF₆ uptake versus the parent MOF-808, a demonstrable improvement stemming from our pore-engineering strategy.

To gain further insight into the adsorbent–adsorbate interactions at play, we turned to computational modelling. Models for each material, at different degrees of functionalization, were first optimized using density functional theory (DFT). Grand canonical Monte Carlo (GCMC) simulation was then performed to assess the SF₆ uptake for each material. Gratifyingly, the experimental and simulated trends were found to be broadly in agreement. These computational results underscore the impact of functional group identity on the pore microenvironment. Overall, this work highlights the power of synthetically straightforward post-synthetic modification for tailoring the pore environment of MOFs for specific, challenging gas capture applications.

The parent MOF-808 material was synthesized following a modified literature procedure,²⁹ then treated with the four different fluorine containing benzoic acids. The chosen fluorinated benzoic acids were selected due to their varying type and number of fluoride groups; it is expected that aryl fluorides may interact differently with SF₆ than benzylic fluorides, while increasing the degree of fluorination should increase SF₆ uptake. The fluorobenzoic acid derivatives were installed by heating pristine MOF-808 with a given reagent in DMF at 60 °C overnight (see SI for full synthetic details).

PXRD confirmed the desired bulk crystalline structure and phase purity of pristine MOF-808 and functionalized MOF-808 (Fig. 2a). By TGA, no significant decrease in thermal stability was observed upon addition of the fluorinated groups (Fig. S1). The materials exhibit permanent porosity, as evidenced by their N₂ adsorption at 77 K (Fig. 2b). MOF-808 was found to have a Brauner–Emmett–Teller (BET) surface area of 1820 m² g⁻¹, while the functionalized MOFs possessed surface areas between 760–1680 m² g⁻¹. The functionalized MOF-808 materials have decreased surface area compared to the parent material, which is an expected consequence of the functional groups obstructing some of the pore volume. The materials exhibited comparable CO₂ at 298 K adsorption isotherms, suggesting the fluorinated groups did not greatly impact the materials’ interactions with CO₂. Additionally, none of the materials showed appreciable uptake for N₂ at 298 K. The CO₂ and N₂ adsorption at 298 K are presented in Fig. S2.

Fig. 2 (a) PXRD patterns of parent MOF-808 and fluorinated derivatives. (b) N₂ adsorption isotherms at 77 K for MOF-808 and fluorinated derivatives. Filled circles denote adsorption, while hollow circles denote desorption. (c) SEM image of MOF-808-F₁. (d) FT-IR spectra from 1800–400 cm⁻¹ with the carboxylate C=O stretching frequency for each material marked by line and arrow.

Incorporation of the functional groups was quantified by digesting a portion of each material in a mixture of KOD in D₂O, filtering the organics from the insoluble ZrO₂, and analysing the filtrate by ¹H NMR spectroscopy (Fig. S4–S6). The NMR results are summarized in Table 1, highlighting the stoichiometry of the fluorinated functional group derivatives per Zr₆ cluster. All the groups incorporated well, between 2 to 5 groups per cluster. F₂ incorporated the least, at approximately 2 groups per Zr₆ cluster. Conversely, and somewhat surprisingly given its steric bulk, (CF₃)₂, incorporated the most, at approximately 5 groups per cluster. As a qualitative complement, SEM-EDS was also employed to assess the bulk distribution of fluorine and zirconium within each material (Fig. 2c and Fig. S7).

Table 1 Comparison of degree of functionalization per cluster by ¹H NMR

Material	Functional groups per Zr6 cluster^a
a Determined by digestion ^1H NMR.
MOF-808-F1	3.8
MOF-808-F2	2.1
MOF-808-CF3	5.8
MOF-808-(CF3)2	4.9

---

To confirm that the groups were incorporating at the Zr₆ cluster, FT-IR was used (Fig. 2d). If incorporation occurs via exchange of the native capping formate groups, a shift to the carboxylate peak at 1663 cm⁻¹ is expected. Indeed, for all materials, a new carboxylate peak at ∼1655 cm⁻¹ is noticed, while the peak at 1663 cm⁻¹ is no longer observed. The FT-IR data is consistent with incorporation of the fluorinated groups via post-synthetic formate exchange on the clusters, rather than simple impregnation of the pores. A schematic of the post-synthetic incorporation of fluorinated groups is outlined in Fig. 3a.

Fig. 3 (a) Schematic representation of the post-synthetic incorporation of fluorinated groups, showing the DFT-optimized structure for each material at a loading of 3 functional groups per cluster. (b) Experimental SF₆ uptake at 298 K for the parent MOF-808 and fluorinated derivatives. (c) Simulated SF₆ uptake for the parent MOF-808 and fluorinated derivatives at loadings most closely matching experimental loadings.

With the functionalized materials in hand, we next sought to validate their performance as SF₆ adsorbents. Experimental SF₆ adsorption isotherms collected at 298 K revealed substantial differences in capacity for the differently functionalized materials, Fig. 3b. MOF-808-F₁ and MOF-808-F₂ exhibited the highest SF₆ uptakes at saturation, reaching a maximum uptake of 1.6 mmol g⁻¹ and 1.3 mmol g⁻¹, respectively. For MOF-808-F₁, this is a fourfold increase compared to the parent MOF-808. MOF-808-CF₃ and MOF-808-(CF₃)₂ exhibited lower uptakes, less than 1 mmol g⁻¹, though MOF-808-CF₃ had comparable uptake to that of MOF-808-F₁ at low pressures. MOF-808-CF₃ may have favorable adsorption sites at low pressures, yet the accessible pore volume impedes uptake at higher pressures.

Amongst the other four materials, the best performers, MOF-808-F₁ and MOF-808-F₂, are the least functionalized at 3.8 and 2.1 groups per cluster. While these materials are more porous than MOF-808-CF₃ and MOF-808-(CF₃)₂, given that MOF-808 possesses higher surface area, porosity alone does not explain this trend. Rather, the type of fluorinated group plays an integral role in driving SF₆ adsorption. Bound directly to the aromatic phenylene ring, the aryl fluorides in MOF-808-F₁ and MOF-808-F₂ are more engaged with the pi-system of the ring, changing the electronic structure.³⁰ This, in turn, may increase the dispersive interactions with SF₆ at adsorption sites on the ring, leading to enhanced adsorption.^31–33 Similar dispersive SF₆-MOF interactions have been implicated in other MOF systems.^{17,19,20,34,35} Conversely, the −CF₃ groups of MOF-808-CF₃ and MOF-808-(CF₃)₂, being less electronically connected to the ring, do not disrupt the pi-system to generate new adsorption sites. Rather, these groups may rely primarily on adsorption sites at the −CF₃ groups. Overall, a combination of these interactions in conjunction with accessible pore space are the primary factors driving enhanced adsorption.

To further investigate the effects of the different fluorinated groups on SF₆ uptake, atomistic simulations were used. First, DFT calculations were used to optimize the structure of each MOF-808 derivative at varying degrees of functionalization: 1 group per cluster, 2 groups per cluster, and 3 groups per cluster. Then, the optimized structures were used to simulate SF₆ uptake at pressures between 0–760 mmHg (see SI for Computational methods). The simulated isotherms at different loadings for each material are shown in Fig. S8–S11.

Simulated results most closely matching the experimentally observed degree of functionalization for each material (3 groups per cluster for F₁, CF₃, and (CF₃)₂ and 2 groups per cluster for F₂) are overlaid in Fig. 3c. The absolute uptake values for the simulated results generally overperform experiment, given the higher degrees of functionalization determined experimentally. This agrees with previous work comparing force field models for SF₆, which showed that most models overpredict SF₆ uptake at low pressure due to overestimating interaction strength.^36,37 Notably, general trends are preserved, with MOF-808-F₁, MOF-808-F₂, and MOF-808-CF₃ performing best in simulation. However, the ordering is switched at higher pressures. For MOF-808-CF₃ and MOF-808-(CF₃)₂ the experimental loading was higher than the simulated loading of functional groups per nodes; therefore, the experiments reflected a greater reduction in pore volume. This highlights the complex trade-off between increased affinity and reduction in pore volume when adding functional groups. Clearly, myriad factors must be considered when designing functionalized physisorbents for SF₆ capture, including the identity, type, and number of functional groups installed, as well as their impact on the pore structure. The interplay of these factors must be considered in concert to achieve the greatest enhancement in performance.

Herein, a series of fluorinated MOF-808 derivatives were investigated for their ability to adsorb SF₆. All functionalized materials outperformed the parent MOF, and the best performing material, MOF-808-F₁, exhibited a four-fold increase in SF₆ uptake, highlighting the power of simple post-synthetic chemistry to enhance performance. The complex interplay between the identity and number of fluorinated groups critically influences the adsorption capacity, with aryl fluorides exhibiting superior performance compared to benzylic counterparts. While an increase in the number of fluorinated groups generally correlates with improved SF₆ uptake, this relationship is nuanced by the impact of functionalization on the material's porosity. The findings underscore the importance of carefully selecting both the type and quantity of fluorinated groups to optimize the electrostatic environment within the MOF pores, ultimately enhancing gas capture efficiency. This work paves the way for future research aimed at developing tailored MOF materials for effective gas separation applications, contributing to advancements in sustainable practices within industries reliant on SF₆.

This work was supported by the Laboratory Directed Research and Development program at Sandia National Laboratories. This article has been authored by an employee of National Technology & Engineering Solutions of Sandia, LLC under Contract No. DE-NA0003525 with the U.S. Department of Energy (DOE). The employee owns all right, title and interest in and to the article and is solely responsible for its contents. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this article or allow others to do so, for United States Government purposes. The DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan https://www.energy.gov/downloads/doe-public-access-plan. This paper describes objective technical results and analysis. Any subjective views or opinions that might be expressed in the paper do not necessarily represent the views of the U.S. Department of Energy or the United States Government.

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: Table S1 and Fig. S1–S11. See DOI: https://doi.org/10.1039/d5cc05791d.

References

1. C. Crownhart, MIT Technol. Rev., 2024.
2. X. Fang, X. Hu, G. Janssens-Maenhout, J. Wu, J. Han, S. Su, J. Zhang and J. Hu, Environ. Sci. Technol., 2013, 47, 3848–3855.
3. M. Maiss and C. A. M. Brenninkmeijer, Environ. Sci. Technol., 1998, 32, 3077–3086.
4. G. Lee, I. Ahmed, M. A. Hossain, H. J. Lee and S. H. Jhung, Coord. Chem. Rev., 2025, 524, 216325.
5. C. Y. Chuah, Y. Lee and T.-H. Bae, Chem. Eng. J., 2021, 404, 126577.
6. K. Kim, K. S. Kim, J. E. Lee, S. Park, C.-K. Ahn and G.-H. Kim, Sep. Purif. Technol., 2018, 200, 29–35.
7. M. Åhlén, Y. Zhou, D. Hedbom, H. S. Cho, M. Strømme, O. Terasaki and O. Cheung, J. Mater. Chem. A, 2023, 11, 26435–26441.
8. M. Chang, T. Yan, Y. Wei, J.-X. Wang, D. Liu and J.-F. Chen, Chem. Mater., 2022, 34, 9134–9143.
9. L.-G. Ding, B.-J. Yao, W.-L. Jiang, J.-T. Li, Q.-J. Fu, Y.-A. Li, Z.-H. Liu, J.-P. Ma and Y.-B. Dong, Inorg. Chem., 2017, 56, 2337–2344.
10. A. Dong, D. Chen, Q. Li and J. Qian, Small, 2023, 19, 2201550.
11. Y. Hu, L. Wang, R. Nan, N. Xu, Y. Jiang, D. Wang, T. Yan, D. Liu, Y. Zhang and B. Chen, Chem. Eng. J., 2023, 471, 144851.
12. M.-B. Kim, K.-M. Kim, T.-H. Kim, T.-U. Yoon, E.-J. Kim, J.-H. Kim and Y.-S. Bae, Chem. Eng. J., 2018, 339, 223–229.
13. M.-B. Kim, T.-U. Yoon, D.-Y. Hong, S.-Y. Kim, S.-J. Lee, S.-I. Kim, S.-K. Lee, J.-S. Chang and Y.-S. Bae, Chem. Eng. J., 2015, 276, 315–321.
14. J. J. Li, Y. T. Chen, T. Ke, Y. Y. Jin, R. R. Fan, G. H. Xu, L. Yang, Z. G. Zhang, Z. B. Bao, Q. L. Ren and Q. W. Yang, Nat. Commun., 2025, 16, 632.
15. J.-W. Yan, S.-Q. Gang, Z.-Y. Liu, H.-Y. Xu, R. Wang and J.-L. Du, Sep. Purif. Technol., 2023, 327, 124929.
16. H. Wang, L. Shi, Z. Xiong, S. Ma, H. Cao, S. Cai, Z. Qiao, J. Pan and Z. Chen, Chem. Commun., 2024, 60, 2397–2400.
17. F. M. Amombo Noa, O. Cheung, M. Åhlén, E. Ahlberg, P. Nehla, G. Salazar-Alvarez, S. Ershadrad, B. Sanyal and L. Öhrström, Chem. Commun., 2023, 59, 2106–2109.
18. X. Dong and Y. S. Lin, Chem. Commun., 2013, 49, 1196–1198.
19. M. Åhlén, F. M. Amombo Noa, L. Öhrström, D. Hedbom, M. Strømme and O. Cheung, Microporous Mesoporous Mater., 2022, 343, 112161.
20. M. Åhlén, A. Jaworski, M. Strømme and O. Cheung, Chem. Eng. J., 2021, 422, 130117.
21. A. E. Amooghin, H. Sanaeepur, R. Luque, H. Garcia and B. Chen, Chem. Soc. Rev., 2022, 51, 7427–7508.
22. D. M. Venturi and F. Costantino, RSC Adv., 2023, 13, 29215–29230.
23. L. Yan, H.-T. Zheng, L. Song, Z.-W. Wei, J.-J. Jiang and C.-Y. Su, ACS Appl. Mater. Interfaces, 2024, 16, 6579–6588.
24. P. M. Bhatt, Y. Belmabkhout, A. Cadiau, K. Adil, O. Shekhah, A. Shkurenko, L. J. Barbour and M. Eddaoudi, J. Am. Chem. Soc., 2016, 138, 9301–9307.
25. P. Deria, J. E. Mondloch, E. Tylianakis, P. Ghosh, W. Bury, R. Q. Snurr, J. T. Hupp and O. K. Farha, J. Am. Chem. Soc., 2013, 135, 16801–16804.
26. Y.-X. Wang, R. Wang, Z.-X. Wang, Q. Shen, S. Liu and G. Li, Inorg. Chem. Front., 2026 10.1039/D5QI01816A.
27. M. R. Tchalala, P. M. Bhatt, K. N. Chappanda, S. R. Tavares, K. Adil, Y. Belmabkhout, A. Shkurenko, A. Cadiau, N. Heymans, G. De Weireld, G. Maurin, K. N. Salama and M. Eddaoudi, Nat. Commun., 2019, 10, 1328.
28. J. Zhu, D. Luo, Q. Wang, S. Tong, Z. Wang and J. Yan, Chem. Commun., 2024, 60, 12209–12212.
29. H. Furukawa, F. Gándara, Y.-B. Zhang, J. Jiang, W. L. Queen, M. R. Hudson and O. M. Yaghi, J. Am. Chem. Soc., 2014, 136, 4369–4381.
30. T. J. Fuhrer, M. Houck and S. T. Iacono, ACS Omega, 2021, 6, 32607–32617.
31. X. Sun, L. Zhou, J. Chen, Z. Jia, Z. Zhao and Z. Zhao, J. Mater. Chem. A, 2025, 13, 2360–2377.
32. X. Yang, Q. Wang, J. Zhu, J. Yan and S. Guo, ACS Mater. Lett., 2025, 7, 1940–1946.
33. J. Zhu, Q. Wang, Z. Wang and J. Yan, ACS Appl. Polym. Mater., 2025 DOI:10.1021/acsapm.5c01646.
34. D. Hedbom, P. Gaiser, T. Günther, O. Cheung, M. Strømme, M. Åhlén and M. Sjödin, J. Mater. Chem. A, 2025, 13, 1731–1737.
35. S.-M. Wang, X.-T. Mu, H.-R. Liu, S.-T. Zheng and Q.-Y. Yang, Angew. Chem., 2022, 134, e202207066.
36. I. Matito-Martos, J. Álvarez-Ossorio, J. J. Gutiérrez-Sevillano, M. Doblaré, A. Martin-Calvo and S. Calero, Phys. Chem. Chem. Phys., 2015, 17, 18121–18130.
37. N. Gantzler, J. Geary, J. Harvey, R. E. Sikma, R. A. Reyes, C. E. McKeown, N. S. Bobbitt, J. P. Sammon and D. F. Sava Gallis, J. Mater Chem. A, 2026 10.1039/D5TA06940H.

---

Footnote

† Co-first author.

---