Concluding Remarks

Particle-resolved computational fluid dynamics is becoming widely used in the literature of fixed-bed reactors. Some authors have suggested that it may replace conventional reaction engineering models in the future. This talk will put forward some thoughts on the nature and differences between simulation and modeling, with reference to fixed bed transport and reaction processes. The arguments made here will be illustrated by three case studies.

1) Particle-resolved CFD simulations are presented of a thermowell for monitoring temperatures in a fixed bed catalytic reactor for the highly exothermic partial oxidation of o-xylene to phthalic anhydride, for tube-to-particle diameter ratios (N) of 3.96, 5.96 and 7.99 and inner thermowell radius to outer tube radius ratios of 0.0 (no thermowell), 0.05, 0.1 and 0.2. This simulation study is used to show that changes in measured axial temperature occurred because the local configuration of particles in the bed was rearranged, even by the thinnest thermowell, causing large-scale changes in voidage, flow pattern and distribution of temperature and reaction inside the bed. Effects of flow bypassing and thermal conduction along the thermowell also contributed to the observed differences in temperature profile. This case illustrates the direct use of simulation results.

2) 3D CFD fixed bed simulations of fixed bed heat transfer are presented to illustrate the use of simulation results to provide inputs to mathematical models. A 2D effective medium model of fixed bed heat transfer is shown, which uses averaged axial and radial velocity components from the 3D CFD simulations to represent radial thermal convection, in place of the usual effective thermal conductivity and wall heat transfer coefficient. The 2D model temperatures and averaged 3D CFD temperatures agreed for 80 ≤ Re ≤ 1900 and N = 3.96, 5.96 and 7.99. This case illustrates the use of detailed simulations to obtain improved models based on physical instead of lumped parameters.

3) 3D CFD fixed bed simulations of steam methane reforming are used in the last case study to show an example of the use of detailed simulation results to aid in model selection and discrimination. Results are presented for a packed bed of 807 spheres at N = 5.96, under conditions typical of the middle of a steam reformer tube, with specified wall heat flux. The 3D CFD simulations show a detailed picture of temperature, species and reaction rates at the level of a catalyst particle or lower. From the 3D CFD simulations, it could be determined what level of detail was necessary for the effective medium 2D + 1D models to represent chosen features of the more detailed 3D model. The results showed that for this endothermic reaction the cross-sectional average profiles require detailed modeling of the reaction rate distribution (via the radial void fraction) as well as a radially-distributed axial velocity profile, but not a radially-varying effective radial thermal conductivity.

These case studies are intended to throw into relief the differences in philosophy between CFD simulation and reaction engineering modeling.