| Separations using zeolite membranes often require much less use of energy than conventional thermal separations like distillation. Therefore, ultrathin (ca. 1 micrometer) zeolite membranes having a large selectivity and high mass flux are currently generating a great deal of attention. For understanding and controlling the separation selectivity and mass fluxes, detailed information on the molecular scale is needed. Transport through the membrane is often considered as a 3-step process: adsorption from the gas phase into the membrane, intramembrane transport, and desorption from the membrane. Traditional methods to predict separation factors consider transport selectivity (different mobility of guest molecules inside the zeolite) and adsorption selectivity (different occupancy of guest molecules), assuming isothermal conditions. Although a large number of experimental and theoretical studies are focusing on adsorption and diffusion in zeolites, surface- and heat effects have been overlooked to a large extent. However, these effects are crucial for understanding membrane selectivity for the following reasons: (1) Large heat effects are associated with adsorption/desorption of molecules at the entrance/exit of the membrane. It is well known that adsorption and diffusion of hydrocarbon mixtures are strongly temperature dependent. Therefore, one cannot assume isothermal conditions at the surfaces. (2) For ultrathin membranes, the surface resistivity to mass transfer is not necessarily small compared to the intrazeolite resistivity, especially for polar compounds. (3) Assuming isothermal conditions at the boundaries leads to an incorrectly calculated heat flux and therefore an incorrectly calculated entropy production rate. Optimizing the entropy production rate is absolutely crucial for optimizing separation units. However, this is not trivial as the precise interplay between mass transfer, heat transfer and surface resistivity at the microscopic scale is currently unclear. The present proposal aims at equilibrium and non-equilibrium molecular simulations to understand surface- and heat effects of thin zeolite membranes at the microscopic scale, in order to correctly model the mass- and heat transfer required for optimizing the entropy production rate of a membrane separation unit. We will first focus on a simple model systems (separation of propane/butane using a Silicalite membrane) for which equilibrium properties are already available. We will setup non-equilibrium Molecular Dynamics Simulations (NEMD) to find a method to extract the transport coefficients for heat and mass given by Irreversible Thermodynamics. Once the methodology for this is established, we will investigate a system of industrial relevance, i.e. the separation of water/alcohol mixtures using polar and nonpolar zeolites. Furthermore, we would like to obtain a fundamental understanding on transfer coefficients for heat and mass across the interface, and how these coefficients vary with temperature and surface excess concentration. As two-component systems have not been studied in this manner before, significant method development and verification is required. Ultimately, this should lead to a better understanding and design of transport in zeolite membranes, as well as to factual information on transport coefficients; their magnitude and dependence on the surface intensive variables. |
