화학공학소재연구정보센터
Chemical Engineering Science, Vol.57, No.19, 4227-4242, 2002
Limitations of the LDF/equimolar counterdiffusion assumption for mass transport within porous adsorbent pellets
Mass transfer resistance plays an important role in the performance of a periodic adsorption process under rapid cycling conditions. In this study, we examine the limitations of the Fick plus equimolar counterdiffusion (F+EC) and linear driving force (LDF) model for simulating the enrichment of oxygen from air in comparison to the dusty gas model (DGM) under rapid pressure swing adsorption (RPSA) conditions. A conservative, finite volume approach to the solution of the governing differential equations is developed and validated for an adsorbent pellet. Two variations on the RPSA boundary conditions at the pellet surface are investigated. The first considers the square-wave change in concentration from adsorption to desorption investigated with the RPSA-LDF model of Nakao and Suzuki (J. Chem. Eng. Jpn. 16(1983)114). The second considers the partial pressure variation as predicted from an adsorption simulator, representative of the true conditions experienced over a two-step RPSA cycle. From both cases, the impact of bulk gas motion within the pores (DGM) resulted in deviations exceeding 30% on the predicted working capacity of the sieve. This identifies the equimolar counterdiffusion assumption as a significant limitation for predicting performance with macro-mesoporous adsorbents. Along with bulk flow, an additional 10-50% deviation resulted from the RPSA-LDF model incorrectly predicting working capacity (in relation to the F+EC) for the case where boundary conditions do not follow a step change. To propose additional cycle-time-corrected correlations and/or intrapellet concentration profiles when approaching the RPSA limit appears futile given the range of operating conditions expected over a true cycle. The level of radial discretisation within the pellet also appears to be more sensitive for the F+EC as opposed to the DGM approach. These trends were observed for dimensionless cycle times exceeding the traditionally excepted critical value of 0.1, highlighting the importance of a DGM approach in describing mass transfer when approaching the RPSA limit.