Predictive simulation of gas adsorption in fixed-beds and limitations due to the ill-posed Danckwerts boundary condition
Date of Award
Doctor of Philosophy (PhD)
Mechanical and Aerospace Engineering
Gases--Absorption and adsorption., Gas-solid interfaces., Thermodynamics., Zeolites.
The 1-D axially dispersed plug flow model is a mathematical model widely used for the simulation of adsorption processes. Lumped mass transfer coefficients such as the Glueckauf linear driving force (LDF) term and the axial dispersion coefficient are generally obtained by fitting simulation results to the experimental breakthrough test data. An approach is introduced where these parameters, along with the only free parameter in the energy balance equations, are individually fit to specific test data that isolates the appropriate physics. It is shown that with this approach this model provides excellent simulation results for the CO2 on zeolite 5A sorbent/sorbate system; however, for the H2O on zeolite 5A system, non-physical deviations from constant pattern behavior occur when fitting dispersive experimental results with a large axial dispersion coefficient. A method has also been developed that determines a priori what values of the LDF and axial dispersion terms will result in non-physical simulation results for a specific sorbent/sorbate system when using the one-dimensional axially dispersed plug flow model. A relationship between the steepness of the adsorption equilibrium isotherm as indicated by the distribution factor, the magnitude of the axial dispersion and mass transfer coefficient, and the resulting non-physical behavior is derived. This relationship is intended to provide a guide for avoiding non-physical behavior by limiting the ￼￼￼magnitude of the axial dispersion term on the basis of the mass transfer coefficient and distribution factor.
Knox, James Clinton, "Predictive simulation of gas adsorption in fixed-beds and limitations due to the ill-posed Danckwerts boundary condition" (2016). Dissertations. 108.