(395i) Classical DFT for Large-Scale Screening of Nanoporous Materials

Metal-organic frameworks (MOFs) are porous crystalline materials formed by coordination bonding of organic ligands and metallic or organometallic centers. Different from typical intermolecular interactions that control the structure of a conventional crystalline solid, the coordination interactions are short-ranged and strongly anisotropic, much like covalent bonds underlying polymerization of monomeric units in a linear polymer. The strength and inherent stability of coordination bonds make the framework structure predictable from the coordination numbers and the geometry of the metal-ligand complexes. Such structure control at the molecular level renders unprecedented fast development of MOF-based materials with diverse and exceptional properties. However, the broad diversity of metallic nodes and organic linkers also imply large variations of MOF structures that overwhelm even most efficient synthesis strategies. Computational methods that allow fast and reliable predictions of physiochemical properties and in silico screening will drastically accelerate the process of MOF materials design and utilization. In this presentation, we introduce classical DFT as an accurate yet computationally efficient method for predicting gas adsorption in MOF materials. The numerical performance of this new method has been calibrated by extensive comparison with simulation results for gas adsorption in slit pores, spherical cages and various nanostructured materials including SWCNT and MCM-41. In addition, we compared the theoretical predictions with results from GCMC simulation at various temperatures and pressures for H₂ storage in a library of over a few thousands of MOF structures (Fig. 1). Because the classical DFT is much faster than GCMC simulation, it promises experimental scientists a convenient yet powerful tool for discovering novel MOF materials.