(25a) A CFD Integrated Multiscale Method for Modeling Ethylene Oxidation Fixed Bed Reactors

 Ethylene oxide is one of the most important petrochemical intermediates with annual production of more than 14 billion ton. Currently, ethylene oxide is exclusively produced through ethylene oxidation over silver catalyst. The reaction is fast and highly exothermic with carbon dioxide as the main by-product. Under industrial conditions selectivity toward ethylene oxide is 65%-80%. Therefore, an increase in selectivity not only reduces greenhouse gases emissions but also results in more efficient use of existing materials and utilities, and eventually a sustainable design.

Extensive research on the ethylene oxidation mechanism over silver catalyst has been done. It is suggested that oxametallacycle is an important intermediate in the mechanism. Oxametallacycle is isomerized to ethylene oxide and acetaldehyde in two different pathways. Acetaldehyde is then oxidized, and produces CO₂ and H₂O, rapidly.

In the present study, to address selectivity loss in an EO fixed bed, computational fluid dynamics is coupled with detailed kinetics and effective pellet diffusivity. This multiscale approach gives a new insight into this partial oxidation reaction. Initially, a reduced microkinetic model is developed based on a 17 step microkinetic model by Stegelmann et al. (2004). The reduced model is implemented in an ANSYS FLUENT user-defined function to calculate reaction rates and surface sites coverages. 3D CFD simulations are carried out on a simple geometry of 9 spherical solid particles inside a fluid zone. Subsequently, the effects of the flow and transport on elementary steps are studied. Our simulations show that under certain operating conditions adsorbed ethylene oxide favors the reverse reaction, and converts to oxametallacycle, which then contributes to carbon dioxide production. This phenomenon results in selectivity loss inside the catalyst particles. In our subsequent work, this method is extended to larger scale models of randomly packed beds of spheres.

In previous studies, it was assumed that the effective diffusivity of species inside the catalyst particle is constant. A simplification of the Dusty Gas Model derived by Hite and Jackson (1977) was used to obtain a Fickian form to model diffusion inside the solid particles.  This simplification required the assumption that species flux ratios are equal to the corresponding stoichiometric ratio. However, this is a questionable assumption in 3D simulations. In this study, to examine the validity of that assumption, an anisotropic diffusivity tensor user-defined function is made and implemented into the CFD method. Hite and Jackson’s model is used without the need for any further assumptions, i.e. the flux for each species is calculated directly in each computational cell. It is observed that the reaction rate and species concentration profiles inside the particles deviate from those calculated by the method with the simplifying assumption. These results show the importance of taking into account the details of diffusion limitations inside the solid particles.

This approach bridges between macro and micro scale modeling, and shows that coupling detailed transport and kinetics is unavoidable to develop an accurate model of catalytic reactions, and achieve a superior understanding of these systems.