(748c) Investigation of CO2 Sorption Mechanisms in Isothermal Columns Via Transient Material and Energy Flow PDE Models

Armaou, A. - Presenter, The Pennsylvania State University
Yang, M., The Pennsylvania State University
Wang, L., Pennsylvania State University
Shahri, S. M. K., Pennsylvania State University
Rioux, R. M., Pennsylvania State University
In recent years, extensive research effort has been dedicated to improving CO2 capture technology due to contribution of CO2 to global climate change. Compared with ammonium solution, CO2 capture sorbents are less caustic and consume less energy during CO2 desorption. To understand and predict the dynamics of CO2 sorption in a fixed bed column, many models have been proposed in the literature. In previous work [1], we proposed a PDE model to circumvent the unphysical assumptions of linear driving force and spatially uniform adsorption rates; unknown parameters were determined using breakthrough curves that were experimentally measured by mass spectrometry. However, the model could not accurately capture the temporal profiles of the heat released due to adsorption (experimentally measured by micro-calorimetry).

The present work attempts to improve the accuracy of our previous PDE model. It relaxes the assumptions of (a) constant concentration of accessible sites and (b) uniform superficial velocity down the column. Using an improved transport description, we consider the accuracy of five reaction mechanisms to describe the experimental observations. Physical quantities are estimated (including CO2 sorption capacity, rate constants and heat of sorption) by minimizing the integral in time of the difference between model predictions and experimental measurements. In previous publication [1] by our group we only investigated the concentration of CO2 while it was being sorbed, operating the column at one temperature. An elementary third order reaction was adequate to explain the column behavior. In this work, we also utilized the concentration profile of CO2 during desorption and the associated calorimetry profiles to provide more data information for estimation. We then evaluated the different reaction mechanisms by comparing (a) the associated PDE models’ predictions with the experimental profiles, (b) the computed accessible sites with the theoretical maximum, (c) heat of sorption as a function of temperature with the range of viable values reported in the literature. We also analyzed (d) the consistency of the rate constants to an Arrhenius model and (e) the model sensitivity to these rate constants.

We investigated five mechanisms. Based on the analysis of the optimization results, we conclude that the dual-site physisorption-chemisorption scheme is the most probable mechanism that describes the experimental data in a physically consistent manner. Since the model is insensitive to two reaction rate constants, design of new experiments and further study are needed to provide arguments for or against this mechanism.


[1] Babaei Pourkargar, D.; Kamali Shahri, S. M.; Rioux, R. M.; Armaou, A. Spatiotemporal Modeling and Parametric Estimation of Isothermal CO2 Adsorption Columns. Industrial and Engineering Chemistry Research 2016, 55, 6443-6453.