Modeling and parameter estimation for a fixed-bed adsorption process for CO 2 capture using zeolite 13X

Abstract

Modeling of a fixed-bed adsorption process for CO 2 capture using zeolite 13X was performed through parameter estimation. An empirical mass transfer rate model is proposed as a relevant description of the adsorption dynamics of CO 2/N 2 on zeolite 13X. An efficient objective function for the estimation of the mass transfer rate parameters was devised to avoid the local minimum and to improve the convergence to the desired solution. Langmuir isotherm equation coefficients, bed porosity, average particle size, and heat transfer coefficients were determined in separately designed experiments. The porosity estimate was updated during the estimation of mass transfer rate parameters to account for the effect of pore diffusion on bulk flow. To ensure numerically stable computation, the gradient-directed adaptive predictive collocation method was adopted with a cubic spline interpolation function and far-side boundary conditions. The model was experimentally evaluated in an adsorption bed with a height of 70 cm and an inner diameter of 2.54 cm and shown to predict the dynamic behaviors of the process variables with high accuracy.