Designing materials for efficient carbon capture requires careful design and expertise. Due to their high tunability, selectivity, and porosity, metal-organic frameworks (MOFs) are promising candidates for CO₂ capture. However, challenges still lie in predicting which MOF will perform reliably, especially when moving from computer simulations to real-world synthesis and testing. Screenings on large datasets are a promising avenue as they are generally much faster and more efficient than any group of scientists in a wet lab. However, experimentally verifying computational results is often the pitfall of such projects.

In our recent study, we systematically investigated a family of pyrene-based MOFs for carbon capture, each incorporating a different metal: aluminum (Al), gallium (Ga), indium (In), and scandium (Sc). These materials are expected to crystallize in an orthorhombic space group, where infinite metal rods hold the pyrene ligand in a stacking fashion, creating a very promising active site for CO₂ uptake from wet flue gas¹. While the ligand remains unchanged (i.e., 1,3,6,8-tetrakis(p-benzoic acid)pyrene (TBAPy)), computational predictions suggested that the choice of metal incorporated would significantly influence the stacking distance between pyrene layers¹. Such structural variations would directly impact the CO₂ uptake of the material, highlighting its critical role in tuning adsorption properties. This was the case for Al-TBAPy¹ and Sc-TBAPy², which maintain the orthorhombic crystal structure. However, In-TBAPy³ and Ga-TBAPy^4,5 reveal the presence of additional phases. As reported in its original publication, although In-TBAPy crystallizes in an orthorhombic structure, upon activation and solvent removal from the pores, it transitions to a monoclinic phase³. On the other hand, Ga-TBAPy crystallizes in a monoclinic space group and maintains this structure throughout activation (although a secondary orthorhombic phase also seems to coexist). The Ga-TBAPy structure was solved via micro-electron diffraction (microED/3D-ED), as single crystals were not obtained. The monoclinic structure no longer shows the perfect stacking of the TBAPy ligand as observed in their orthorhombic counterpart (Figure 1).  A rotation of the benzoate groups of the ligand results in a break-up of the linear arrangement of the metal ions, which subsequently leads to a shift of the ligand in the structure. This phenomenon leads to a clear change of the CO₂ binding site, as the pyrene stacks are no longer parallel and much shorter, making it impossible for CO₂ to be adsorbed there.

Integrated lab and computer experiments were also performed. In that regard, we developed four computational models for the orthorhombic structures and two monoclinic models for the In- and Ga-TBAPy structures. Computational CO₂ adsorption isotherms show that the monoclinic structures saturate at much lower pressures than the orthorhombic structures, possibly due to their much lower pore volume. Experimentally, in the low-pressure regime, the uptakes match more closely to the uptakes of the simulated monoclinic structures. At the same time, at higher pressure, they follow more closely the orthorhombic ones, which may suggest structural rearrangements upon CO₂ loading (Figure 2).

A mixed-metal Al_0.50Sc_0.50-TBAPy MOF is also synthesized, as these are the only precursors maintaining the orthorhombic phase before and after activation at the reported conditions. In silico structures with different Al vs Sc arrangements are developed, showing that different metal configurations do not significantly affect this structure's uptake.

Further analysis may be needed to uncover the underlying details fully, but we believe these structural variations affect the MOFs' gas adsorption properties. Accounting for these phases may improve the accuracy of predicted adsorption isotherms, more easily bridging the gap between simulations and experiments.

Feel free to explore the complete study on evaluating and fine-tuning MOF structures for CO₂ adsorption at: https://www.nature.com/articles/s41467-025-56296-w.

References:

1. Boyd, P. G.; Chidambaram, A.; García-Díez, E.; Ireland, C. P.; Daff, T. D.; Bounds, R.; G ladysiak, A.; Schouwink, P.; Moosavi, S. M.; Maroto-Valer, M. M.; others Data-driven design of metal–organic frameworks for wet flue gas CO2 capture. Nature 2019, 576, 253–256.

2. Kinik, F. P.; Ortega-Guerrero, A.; Ebrahim, F. M.; Ireland, C. P.; Kadioglu, O.; Mace, A.; Asgari, M.; Smit, B. Toward Optimal Photocatalytic Hydrogen Generation from Water Using Pyrene-Based Metal-Organic Frameworks. ACS applied materials & interfaces 2021, 13, 57118–57131.

3. Stylianou, K. C.; Heck, R.; Chong, S. Y.; Bacsa, J.; Jones, J. T.; Khimyak, Y. Z.; Bradshaw, D.; Rosseinsky, M. J. A guest-responsive fluorescent 3D microporous metal-organic framework derived from a long-lifetime pyrene core. Journal of the American Chemical Society 2010, 132, 4119–4130.

4. Patricio Domingues, N. Understanding & Application Driven Design of Metal-Organic Frameworks for Carbon Capture. Infoscience EPFL Scientific Publications 2021.

5. Ahlén, M.; Zhou, Y.; Hedbom, D.; Cho, H. S.; Strømme, M.; Terasaki, O.; Cheung, O. Efficient SF₆ capture and separation in robust gallium-and vanadium-based metal–organic frameworks. Journal of Materials Chemistry A 2023, 11, 26435–26441.