(714e) Towards Data-Driven Structure-Property Relations for Predicting Adsorption Entropy in Siliceous Zeolites

Zeolites are microporous aluminosilicates with varying pore topology. Evaluating their practicality within adsorption processes depends upon thermodynamic metrics, such as the free energy, which further depends on quantification of the adsorption entropy. Previous works have demonstrated linear scaling relationships between the gas- and adsorbed-phase entropies [1,2], allowing for sensible predictions based on physical parameters. However, such conclusions were based on limited experimental data.

Motivated by these findings, we computationally explore the generality of such relationships across multiple classes of molecules and zeolites. We model our systems by implementing TraPPE forcefields [3] and quantify the adsorbate entropy via Monte-Carlo integrations through the FEASST simulation package [4]. Using a dataset of thirty-seven adsorbates and five siliceous zeolites, we show that linear correlations, to a good first approximation, exist between the gas- and adsorbed-phase entropies. Our correlations were largely dependent on the zeolite’s size, characterized by physical descriptors such as the largest cavity diameter. To further elicit such “structure-topology-thermochemistry” relations, we expand our dataset of adsorbents to include over one thousand combinations between molecules and zeolites, each with unique topological fingerprints. From this expanded dataset, we aim to develop an interpretable data-driven model which may be used to predict the adsorbate entropies of novel systems.

[1] Campbell, C. T.; Sellers, J. R. V. The Entropies of Adsorbed Molecules.J. Am. Chem.Soc.2012,134, 18109–18115.

[2] Dauenhauer, P. J.; Abdelrahman, O. A. A Universal Descriptor for the Entropy of Adsorbed Molecules in Confined Spaces.ACS Cent. Sci.2018,4, 1235–1243.

[3] B.L. Eggimann, A.J. Sunnarborg, H.D. Stern, A.P. Bliss, and J.I. Siepmann. An online parameter and property database for the TraPPE force field, Molec. Simul. 40, 101-105 (2014).

[4] Hatch, H. W., Mahynski, N. A., and Shen, V. K. (2018) FEASST: Free Energy and Advanced Sampling Simulation Toolkit. J. Res. Natl Inst Stan, 123, 123004.