Simulating Microwave-Heated Open Systems: Tuning Competitive Sorption in Zeolites

Simulating adsorptive expansion of zeolites: application to biomass-derived solutions in contact with silicalite

We have constructed and applied an algorithm to simulate the behavior of zeolite frameworks during liquid adsorption. We applied this approach to compute the adsorption isotherms of furfural-water and hydroxymethyl furfural (HMF)-water mixtures adsorbing in silicalite zeolite at 300 K for comparison with experimental data. We modeled these adsorption processes under two different statistical mechanical ensembles: the grand canonical (V-Nz-μg-T or GC) ensemble keeping volume fixed, and the P-Nz-μg-T (osmotic) ensemble allowing volume to fluctuate. To optimize accuracy and efficiency, we compared pure Monte Carlo (MC) sampling to hybrid MC-molecular dynamics (MD) simulations. For the external furfural-water and HMF-water phases, we assumed the ideal solution approximation and employed a combination of tabulated data and extended ensemble simulations for computing solvation free energies. We found that MC sampling in the V-Nz-μg-T ensemble (i.e., standard GCMC) does a poor job of reproducing both the Henry's law regime and the saturation loadings of these systems. Hybrid MC-MD sampling of the V-Nz-μg-T ensemble, which includes framework vibrations at fixed total volume, provides better results in the Henry's law region, but this approach still does not reproduce experimental saturation loadings. Pure MC sampling of the osmotic ensemble was found to approach experimental saturation loadings more closely, whereas hybrid MC-MD sampling of the osmotic ensemble quantitatively reproduces such loadings because the MC-MD approach naturally allows for locally anisotropic volume changes wherein some pores expand whereas others contract.

Experimental evidence of selective heating of molecules adsorbed in nanopores under microwave radiation

We have performed in situ quasielastic neutron scattering (QENS) measurements on zeolite-guest systems under microwave irradiation, for comparison with corresponding simulations. Both experiment and simulation reveal selective heating of methanol in silicalite, but little to no heating of benzene in silicalite. Effective translational and rotational temperatures extracted from QENS data under microwave heating were found to depend on microwave power. In agreement with simulation, QENS rotational temperatures significantly exceed translational ones at high microwave power, thus providing the first microscopic proof for athermal effects in microwave-driven nanopores.