Flow-rate effects in flow-reversal reactors: experiments, simulations and approximations

Abstract

Design and operation of a reactor with flow reversal requires accurate prediction of the domain of operating conditions, and especially the range of flow rates, where the ignited state exist. In this work we compare experimental observations of flow-rate effects during ethylene oxidation on Pt/Al 2O 3, with simulations of this reactor using a kinetic rate expression that was derived elsewhere and with approximate solutions based on instantaneous or very fast reactions. The oxidation of ethylene on supported Pt catalyst, that is employed here as a model reaction, is a complex reaction characterized by self-inhibition (expressed by Langmuir–Hinshelwood kinetics), by strong activation energy and by strong thermal effects that lead to a wide domain of steady-state multiplicity. The analysis of a flow- reversal reactor for such reactions can be approximated using the assumption of an instantaneous or fast reaction as the feed meets the catalyst layer. We suggest several approximations that capitalize on this property and apply them to the structure of our reactor, in which the catalytic bed is imbedded between two inert zones. Adequate agreement between the experimental results and simulations, using homogeneous or heterogeneous reactor model with no adjustable parameters, is demonstrated. The difference between the homogeneous and heterogeneous model predictions is usually small. The approximations show that the most important parameters for predicting the highest temperature are the inert zone properties (conductivity and length).