Predictive ab-initio study on the selectivity of crystalline nano-porous materials with respect to light gases

In recent years major efforts in the membrane technology are focused on design and optimization of low cost materials for gas separation membranes. Special attention is paid to light gases (O2, N2, H2, CO, CO2, H2O) involved in the water gas shift reaction (WGS), for which the sieving mechanism cannot be applied. Due to their adsorption properties inorganic and mixed crystalline nano-porous materials as zeolites and MOFs are of great interest for the underlying separation processes. Although the gas diffusion in zeolites has been widely studied, an analysis concerning the use of ab-initio calculations to evaluate the fundamental parameters characterizing the gas transport equations is difficult to find in the literature. The results of such study is quite general because ab-initio modelling is independent of adjustable quantities. Thus, aim of the present study is to develop a predictive ab-initio procedure to evaluate key quantities defining the gas ideal separation factors of crystalline nano-porous materials. First, we defined the gas separation factor as function of nano-scale features such as adsorption equilibrium constant and activation energy. The developed analytical-numerical formulas are based on the Maxwell-Stefan (MS) and the activated Knudsen (Ka) single component diffusion models. In addition, relationships such as the Van't Hoff and the multi-site Langmuir equations were used in the proposed theoretical procedure. In turn, the research employs ab-initio calculations in the frame of DFT, to obtain the nano-quantities in the above analytical-numerical formulas by means of a quantum embedded cluster approach [1]. This analytical multi-scale approach enables us to study different types of crystalline frameworks which exhibit the same selective mechanism. The study on LTA type zeolite (Ca-A) has been carried out as starting point, because this zeolite can be considered as a model systems. The developed formulas and the calculated ab-initio key quantities give results in good agreement with the available literature data [2, 3, 4]. Moreover, the selectivity with respect to O2 and N2 was predicted as function of temperature, feed and permeate pressures. The outcome shows a smooth dependence of the selectivity in the range of temperature and pressure which have been tested. Thus, the study can be extended to LTA loaded with other cations like Na or to bi-site adsorption isotherm, treating trivial well known system like FAU type zeolite loaded with different cations. Finally, the proposed ab-initio scheme allows to analyse and optimize other crystalline porous materials (e.g. MOFs) used in membranes for purification of WGS gases.