(660g) Multiscale Modeling of the Breakthrough Behavior of Adsorption Columns

The present work is associated with the development of a comprehensive theoretical approach to model the breakthrough behavior of adsorption columns used to separate gaseous components. The adsorption of gaseous species onto the surface of a porous solid is a stochastic process and is simulated using kinetic Monte Carlo methods. The molecules attach and detach form the sites of the surface through three micro-processes: adsorption or deposition, desorption, and migration of surface diffusion. Adsorption columns usually operate under transient, non-steady state conditions. Analysis of the transient behavior of a gas mixture flowing through a column filled with adsorbent yields the so-called breakthrough time which is the time required to achieve a desired degree of separation. Modeling of the column comprises conservation statements for the species (mass balances) that contain the amounts adsorbed onto the porous material. The conservation equations can be solved numerically using a finite difference/element technique in which the equations are discretized in space. The amount deposited onto the adsorbent is calculated from the kinetic Monte Carlo simulations. The process is repeated recursively until breakthrough behavior is achieved. The results of the kinetic Monte Carlo simulations are compared against experimental adsorption isotherms for methane, carbon dioxide, and nitrogen as well as for binary mixtures of these gases on microporous activated carbon. Furthermore, the calculated breakthrough times are verified experimentally using a custom-made dynamic column experiment to determine the outlet concentration. The multiscale model will be used in the sweetening (separation of sour gases) and inert rejection (separation of nitrogen) of natural gas.