High fidelity numerical simulation of ethylene epoxidation packed bed

Abstract

Packed bed reactors with a low tube-to-particle diameter ratio are mainly used in strong exothermic/endothermic reaction systems. However, the traditional plug flow packed bed reactor model is unsuitable for reactors with low N value (i.e., tube-to-particle diameter ratio) due to channelling near the wall and reflux in the bed. In this work, a high fidelity numerical model of packed bed reactor with different macro morphologies of particles was established by coupling the heterogeneous packed bed model, particle internal diffusion model, component migration equation, and reaction kinetics. Taking the packed bed reactor for ethylene epoxidation as an example, the industrial simulation of catalyst particles with different macro morphology was carried out by using this method. The total porosity and porosity distribution of the bed in this model are in good agreement with empirical formulas, with the errors in bed pressure drop and ethylene conversion rate being less than 15%. By using this model to simulate the flow field, temperature field, and internal diffusion of particles in a packed bed reactor with high fidelity, the concentration distribution of each component can be predicted. By comparing the pressure drop, temperature rise, and ethylene conversion rate of packed bed reactors filled with catalyst particles of different macroscopic shapes, it was found that HC2 particle-packed beds have a lower pressure drop and a higher ethylene conversion rate.