Zeolite membranes offer a promising alternative for energy-efficient separation due to their nanoporous frameworks, tunable pore sizes, and high thermal/chemical stability. These naturally occurring aluminosilicate materials are nanoporous, exhibiting a near continuum of pore sizes that allow for separation of target gas molecules. [2–5] However, optimizing these materials requires atomic-level insights into adsorption, diffusion, and framework dynamics—challenges that remain experimentally intractable.
Here, we will present a new machine-learned interatomic potential for bulk MFI zeolite based on the ChIMES physics-informed machine learned interatomic model and development framework.[6–8] Our training strategy incorporates strained and compressed structural configurations to capture framework flexibility, enabling robust predictions of bulk properties (e.g., radial distribution functions, vibrational power spectra) that align closely with DFT benchmarks. This work will focus on thermal expansion behavior, a critical factor in membrane stability under industrial conditions, by leveraging large-scale molecular dynamics simulations to probe temperature-dependent structural changes in MFI zeolites. These simulations aim to predict thermal and dynamic properties critical to the intelligent design of membranes for industrial applications. Large-scale molecular dynamics simulations can enable understanding of temperature-dependent structural changes in MFI zeolites, providing atomic-level insights into adsorption dynamics and pore channel responses to thermal stress. These simulations predict industrially relevant properties, such as framework rigidity and diffusion pathways, which are pivotal for designing efficient separation membranes.
The model also demonstrates promising transferability to topologically similar zeolite frameworks (e.g. MEL), suggesting broad applicability for studying zeolite-based separation systems. By connecting atomicscale simulations with macroscopic material properties, this work provides a foundation for designing improved zeolite membranes for energy-efficient ammonia separation in industrial processes.
References
[1] Collin Smith, Alfred K. Hill, and Laura Torrente-Murciano. “Current and future role of Haber–Bosch ammonia in a carbon-free energy landscape”. In: Energy Environ. Sci. 13 (2 2020), pp. 331–344. DOI:10.1039/C9EE02873K. URL: http://dx.doi.org/10.1039/C9EE02873K.
[2] Hisao Inami, Chie Abe, and Yasuhisa Hasegawa. “Development of Ammonia Selectively Permeable Zeolite Membrane for Sensor in Sewer System”. In: Membranes 11.5 (2021). ISSN: 2077-0375. DOI:10.3390/membranes11050348. URL: https://www.mdpi.com/2077-0375/11/5/348.
[3] Miguel Palomino et al. “New Insights on CO2Methane Separation Using LTA Zeolites with Different Si/Al Ratios and a First Comparison with MOFs”. In: Langmuir 26.3 (2010). PMID: 19757816, 1910–1917. DOI:10.1021/la9026656. eprint: https://doi.org/10.1021/la9026656. URL: https://doi.org/10.1021/la9026656.
[4] Jin Shang et al. “Temperature controlled invertible selectivity for adsorption of N2 and CH4 by molecular trapdoor chabazites”. In: Chem. Commun. 50 (35 2014), pp. 4544–4546. DOI: 10.1039/C4CC00269E. URL: http://dx.doi.org/10.1039/C4CC00269E.
[5] Jin Shang et al. “Discriminative Separation of Gases by a “Molecular Trapdoor” Mechanism in Chabazite Zeolites”. In: Journal of the American Chemical Society 134.46 (2012). PMID: 23110556, pp. 19246 DOI: 10.1021/ja309274y. eprint: https://doi.org/10.1021/ja309274y. URL: https://doi.org/10.1021/ja309274y.
[6] Rebecca Lindsey. ChIMES Calculator. URL: https://github.com/rk-lindsey/chimes_calculator.
[7] Rebecca Lindsey. ChIMES Generator. URL: https://github.com/rk-lindsey/chimes_lsq.
[8] Rebecca Lindsey. ChIMES Active Learning Driver. URL: https://github.com/rk-lindsey/al_driver.