Force Field for Molecular Dynamics Computations in Flexible ZIF-8 Framework

Abstract

Correct modeling of framework flexibility plays a major role in obtaining accurate results when performing atomistic simulations of guest molecule diffusion within ZIF crystal structures. Here we present a full set of force field parameters, based on the AMBER database and on previously computed partial charges, well reproducing the ZIF-8 structural properties over a wide range of temperatures and pressures. To test our model, the self-diffusivity for sorbed carbon-dioxide is computed and is found to be in good agreement with experimental measurements. Our results are also compared to the ones obtained with other charge models and are found to be more accurate. Finally, an estimate of the influence on self-diffusion of various simulation details is given.

NMR studies of carbon dioxide and methane self-diffusion in ZIF-8 at elevated gas pressures

“…While the experimental results of IRM and NMR are in the same order Fig. 7 Comparison of self-diffusion coefficients of carbon dioxide determined by PFG NMR (Q, this work) with MD simulations ( , Zheng et al 2012 and•, Pantatosaki et al 2012) and IRM data ( * , Bux et al 2010). The lines serve as guide for the eyes of magnitude, they deviate in the loading dependence for loadings above about 5 CO 2 molecules per cage.…”

“…Using various force fields for ZIF-8 (details see Pantatosaki et al 2012;Zheng et al 2012), molecular dynamic simulations of carbon dioxide and methane in a flexible ZIF-8 framework were performed. For CO 2 in particular, one of (1).…”

“…Experimental methods that are frequently used to measure molecular diffusion in modern MOF materials involve quasi elastic neutron scattering (QENS) (Deroche et al 2011;Yang et al 2011), pulsed field gradient (PFG) NMR (Stallmach et al 2006;Wehring et al 2010), infrared microscopy and interference microscopy (Bux et al 2010). Transport processes in the MOFs are also often studied by molecular dynamic (MD) computer simulations (Liu et al 2011;Keskin et al 2009;Sant et al 2010;Zheng et al 2012;Pantatosaki et al 2010Pantatosaki et al , 2012. Since computer simulations do not require the synthesis of the porous materials for the experiments, they are in particular useful for screening MOF's with respect to potentially interesting gas separation properties.…”

Theoretical model estimation of guest diffusion in metal–organic frameworks (MOFs)

“…Molecular dynamics (MD), based on classical mechanics, is the most widely applied tool for the production of trajectories of guests in the phase space and obtainment of measurable macroscopic properties like diffusion [1]. Although many accurate MD simulations have been reported the results mainly depend on the assumption of inter-and intra-molecular potentials [7][8][9]. A reasonable semi-empirical force field for MOF systems should represent both the stability and the flexibility of the framework, which in principle increases the difficulty of building a force field [7,[9][10][11][12][13].…”

Computational characterization and prediction of metal–organic framework properties

“…The first, and dominant category, is the use of transferable intramolecular force fields for organic molecules (e.g., from the AMBER parameter database [322]) along with hand-tuned metal-organic bending and torsion potentials. The later are then optimized to reproduce known structural properties and vibration frequencies, [323,324,325] or experimental data such as adsorption isotherms and structural transitions. [326,327,328] This approach is not entirely generalizable, and the quality of the resulting force field is only validated on limited experimental data and cannot be systematically improved, as the optimization problem is typically underconstrained (too many parameters fitted on relatively little data).…”